From f9936e59a4d740849dd4c58fda3c0524226141b7 Mon Sep 17 00:00:00 2001 From: loubigard <111683378+loubigard@users.noreply.github.com> Date: Wed, 14 Aug 2024 16:08:27 +0900 Subject: [PATCH] Tutorial on Continuous Monitoring using ML Pipelines (#3644) # Description A tutorial about creating a pipeline for continuous monitoring. It describes an advanced use case of [running flows in ML pipelines](https://github.com/microsoft/promptflow/tree/main/examples/tutorials/run-flow-with-pipeline). # All Promptflow Contribution checklist: - [x] **The pull request does not introduce [breaking changes].** - [x] **CHANGELOG is updated for new features, bug fixes or other significant changes.** - [x] **I have read the [contribution guidelines](https://github.com/microsoft/promptflow/blob/main/CONTRIBUTING.md).** - [x] **I confirm that all new dependencies are compatible with the MIT license.** - [x] **Create an issue and link to the pull request to get dedicated review from promptflow team. Learn more: [suggested workflow](../CONTRIBUTING.md#suggested-workflow).** ## General Guidelines and Best Practices - [x] Title of the pull request is clear and informative. - [x] There are a small number of commits, each of which have an informative message. This means that previously merged commits do not appear in the history of the PR. For more information on cleaning up the commits in your PR, [see this page](https://github.com/Azure/azure-powershell/blob/master/documentation/development-docs/cleaning-up-commits.md). ### Testing Guidelines - [ ] Pull request includes test coverage for the included changes. Co-authored-by: Philip Gao Co-authored-by: Brynn Yin <24237253+brynn-code@users.noreply.github.com> --- .../README.md | 69 ++ .../simple-summarization/flow.dag.yaml | 45 ++ .../simple-summarization/requirements.txt | 2 + .../summarize_text_content.jinja2 | 7 + .../summarize_text_content__variant_1.jinja2 | 7 + .../components/convert_parquet/conda.yaml | 10 + .../convert_parquet/convert_parquet.py | 116 +++ .../convert_parquet/convert_parquet.yaml | 20 + .../continuous_monitoring_pipeline.ipynb | 742 ++++++++++++++++++ .../monitoring/data/monitoring_dataset.jsonl | 5 + .../monitoring/media/compare_button.png | Bin 0 -> 22556 bytes .../monitoring/media/compare_metrics.png | Bin 0 -> 53695 bytes .../media/continuous_monitoring_pipeline.png | Bin 0 -> 84437 bytes .../monitoring/media/convert_parquet.png | Bin 0 -> 77777 bytes .../monitoring/media/metrics_tab.png | Bin 0 -> 35126 bytes 15 files changed, 1023 insertions(+) create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/README.md create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/flow.dag.yaml create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/requirements.txt create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content.jinja2 create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content__variant_1.jinja2 create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/conda.yaml create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.py create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.yaml create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/continuous_monitoring_pipeline.ipynb create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/data/monitoring_dataset.jsonl create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/compare_button.png create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/compare_metrics.png create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/continuous_monitoring_pipeline.png create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/convert_parquet.png create mode 100644 examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/metrics_tab.png diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/README.md b/examples/flows/integrations/continuous-monitoring-with-pipeline/README.md new file mode 100644 index 00000000000..6d1bb4e0249 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/README.md @@ -0,0 +1,69 @@ +# Continuous Monitoring Pipeline + +This tutorial describes an advanced use case of [running flows in Azure ML Pipelines](https://github.com/microsoft/promptflow/blob/main/examples/tutorials/run-flow-with-pipeline/pipeline.ipynb). +The detailed explanations of the prerequisites and principles can be found in the aforementioned article. +Continuous monitoring is necessary to maintain the quality, performance and efficiency of Generative AI applications. +These factors directly impact the user experience and operational costs. + +We will run evaluations on a basic chatbot flow, then aggregate the results to export and visualize the metrics. +The flows used in this pipeline are described below: +- [Basic Chat](https://github.com/microsoft/promptflow/tree/main/examples/flows/chat/chat-basic) +- [Q&A Evaluation](https://github.com/microsoft/promptflow/tree/main/examples/flows/evaluation/eval-qna-rag-metrics) +- [Perceived Intelligence Evaluation](https://github.com/microsoft/promptflow/tree/main/examples/flows/evaluation/eval-perceived-intelligence) +- [Summarization Evaluation](https://github.com/microsoft/promptflow/tree/main/examples/flows/evaluation/eval-summarization) + +Connections used in this flow: +- `azure_open_ai_connection` connection (Azure OpenAI). + +## Prerequisites + +### Prompt flow SDK: +- Azure cloud setup: + - An Azure account with an active subscription - [Create an account for free](https://azure.microsoft.com/free/?WT.mc_id=A261C142F) + - Create an Azure ML resource from Azure portal - [Create a Azure ML workspace](https://ms.portal.azure.com/#view/Microsoft_Azure_Marketplace/MarketplaceOffersBlade/searchQuery/machine%20learning) + - Connect to your workspace then setup a basic computer cluster - [Configure workspace](https://github.com/microsoft/promptflow/blob/main/examples/configuration.ipynb) +- Local environment setup: + - A python environment + - Installed Azure Machine Learning Python SDK v2 - [install instructions](https://github.com/microsoft/promptflow/blob/main/examples/README.md) - check the getting started section and make sure version of 'azure-ai-ml' is higher than `1.12.0` + +Note: when using the Prompt flow SDK, it may be useful to also install the [`Prompt flow for VS Code`](https://marketplace.visualstudio.com/items?itemName=prompt-flow.prompt-flow) extension (if using VS Code). + +### Azure AI/ML Studio: +Start a compute session. +The follow the installation steps described in the notebook. + +## Setup connections +Ensure that you have a connection to Azure OpenAI with the following deployments: +- `gpt-35-turbo` +- `gpt-4` + +## Run pipeline + +Run the notebook's steps until `3.2.2 Submit the job` to start the pipeline in Azure ML Studio. + +## Pipeline description +The first node reads the evaluation dataset. +The second node is the main flow that will be monitored, it takes the output of the evaluation dataset as a `data` input. +After the main flow's node has completed, its output will go to 3 nodes: +- Q&A Evaluation +- Perceived Intelligence Evaluation +- Simple Summarization + +The Simple Summarization and the main nodes' outputs will become the Summarization Evaluation node's input. + +Finally, all the evaluation metrics will be aggregated and displayed in Azure ML Pipeline's interface. + +![continuous_monitoring_pipeline.png](./monitoring/media/continuous_monitoring_pipeline.png) + +## Metrics visualization +In the node `Convert evaluation results to parquet` Metrics tab, the aggregated metrics will be displayed. + +![metrics_tab.png](./monitoring/media/metrics_tab.png) + +The evolution of the metrics can be monitored by comparing multiple pipeline runs: + +![compare_button.png](./monitoring/media/compare_button.png) + +![compare_metrics.png](./monitoring/media/compare_metrics.png) +## Contact +Please reach out to Lou Bigard () with any issues. \ No newline at end of file diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/flow.dag.yaml b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/flow.dag.yaml new file mode 100644 index 00000000000..2f51243dc11 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/flow.dag.yaml @@ -0,0 +1,45 @@ +$schema: https://azuremlschemas.azureedge.net/promptflow/latest/Flow.schema.json +environment: + python_requirements_txt: requirements.txt +inputs: + answer: + type: string +outputs: + summary: + type: string + reference: ${summarize_text_content.output} +nodes: +- name: summarize_text_content + use_variants: true +node_variants: + summarize_text_content: + default_variant_id: variant_0 + variants: + variant_0: + node: + type: llm + source: + type: code + path: summarize_text_content.jinja2 + inputs: + deployment_name: gpt-35-turbo + model: gpt-3.5-turbo + max_tokens: 128 + temperature: 0.2 + text: ${inputs.answer} + connection: open_ai_connection + api: chat + variant_1: + node: + type: llm + source: + type: code + path: summarize_text_content__variant_1.jinja2 + inputs: + deployment_name: gpt-35-turbo + model: gpt-3.5-turbo + max_tokens: 256 + temperature: 0.3 + text: ${inputs.answer} + connection: open_ai_connection + api: chat diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/requirements.txt b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/requirements.txt new file mode 100644 index 00000000000..7b0dee17c44 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/requirements.txt @@ -0,0 +1,2 @@ +promptflow[azure]>=1.7.0 +promptflow-tools \ No newline at end of file diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content.jinja2 b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content.jinja2 new file mode 100644 index 00000000000..3352c4e8789 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content.jinja2 @@ -0,0 +1,7 @@ +# system: +Please summarize the following text in one paragraph. 100 words. +Do not add any information that is not in the text. + +# user: +Text: {{text}} +Summary: \ No newline at end of file diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content__variant_1.jinja2 b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content__variant_1.jinja2 new file mode 100644 index 00000000000..09d0023df61 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/flows/standard/simple-summarization/summarize_text_content__variant_1.jinja2 @@ -0,0 +1,7 @@ +# system: +Please summarize some keywords of this paragraph and have some details of each keywords. +Do not add any information that is not in the text. + +# user: +Text: {{text}} +Summary: \ No newline at end of file diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/conda.yaml b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/conda.yaml new file mode 100644 index 00000000000..2c0c8ed0040 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/conda.yaml @@ -0,0 +1,10 @@ +name: convert_to_parquet +channels: + - defaults +dependencies: + - python=3.10 + - pip=22.2 + - pip: + - azureml-mlflow==1.56.0 + - pandas==2.2.2 + - pyarrow \ No newline at end of file diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.py b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.py new file mode 100644 index 00000000000..708797947b3 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.py @@ -0,0 +1,116 @@ +from pathlib import Path +import pandas as pd +import mlflow +import argparse +import datetime +from functools import reduce + + +def parse_args(): + # setup argparse + parser = argparse.ArgumentParser() + + # add arguments + parser.add_argument( + "--eval_qna_rag_metrics_output_folder", + type=str, + help="path containing data for qna rag evaluation metrics", + ) + parser.add_argument( + "--eval_perceived_intelligence_output_folder", + type=str, + default="./", + help="input path for perceived intelligence evaluation metrics", + ) + + parser.add_argument( + "--eval_summarization_output_folder", + type=str, + default="./", + help="input path for summarization evaluation metrics", + ) + + parser.add_argument( + "--eval_results_output", + type=str, + default="./", + help="output path for aggregated metrics", + ) + + # parse args + args = parser.parse_args() + + # return args + return args + + +def get_file(f): + f = Path(f) + if f.is_file(): + return f + else: + files = list(f.iterdir()) + if len(files) == 1: + return files[0] + else: + raise Exception("********This path contains more than one file*******") + + +def convert_to_parquet( + eval_qna_rag_metrics_output_folder, + eval_perceived_intelligence_output_folder, + eval_summarization_output_folder, + eval_results_output, +): + now = f"{datetime.datetime.now():%Y%m%d%H%M%S}" + + eval_qna_rag_metrics_file = get_file(eval_qna_rag_metrics_output_folder) + eval_qna_rag_metrics_data = pd.read_json(eval_qna_rag_metrics_file, lines=True) + + eval_perceived_intelligence_file = get_file( + eval_perceived_intelligence_output_folder + ) + eval_perceived_intelligence_data = pd.read_json( + eval_perceived_intelligence_file, lines=True + ) + + eval_summarization_file = get_file(eval_summarization_output_folder) + eval_summarization_data = pd.read_json(eval_summarization_file, lines=True) + + all_dataframes = [ + eval_qna_rag_metrics_data, + eval_perceived_intelligence_data, + eval_summarization_data, + ] + eval_results_data = reduce( + lambda left, right: pd.merge(left, right, on="line_number"), all_dataframes + ) + + eval_results_data["timestamp"] = pd.Timestamp("now") + + eval_results_data.to_parquet(eval_results_output + f"/{now}_eval_results.parquet") + + eval_results_data_mean = eval_results_data.mean(numeric_only=True) + + for metric, avg in eval_results_data_mean.items(): + if metric == "line_number": + continue + mlflow.log_metric(metric, avg) + + +def main(args): + convert_to_parquet( + args.eval_qna_rag_metrics_output_folder, + args.eval_perceived_intelligence_output_folder, + args.eval_summarization_output_folder, + args.eval_results_output, + ) + + +# run script +if __name__ == "__main__": + # parse args + args = parse_args() + + # call main function + main(args) diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.yaml b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.yaml new file mode 100644 index 00000000000..1e7f433342d --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/components/convert_parquet/convert_parquet.yaml @@ -0,0 +1,20 @@ +$schema: https://azuremlschemas.azureedge.net/latest/commandComponent.schema.json +type: command + +name: convert_to_parquet +display_name: Convert evaluation results to parquet +inputs: + eval_qna_rag_metrics_output_folder: + type: uri_folder + eval_perceived_intelligence_output_folder: + type: uri_folder + eval_summarization_output_folder: + type: uri_folder +outputs: + eval_results_output: + type: uri_folder +code: ./ +command: python convert_parquet.py --eval_qna_rag_metrics_output_folder ${{inputs.eval_qna_rag_metrics_output_folder}} --eval_perceived_intelligence_output_folder ${{inputs.eval_perceived_intelligence_output_folder}} --eval_summarization_output_folder ${{inputs.eval_summarization_output_folder}} --eval_results_output ${{outputs.eval_results_output}} +environment: + conda_file: ./conda.yaml + image: mcr.microsoft.com/azureml/inference-base-2004:latest diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/continuous_monitoring_pipeline.ipynb b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/continuous_monitoring_pipeline.ipynb new file mode 100644 index 00000000000..a1e491f81e2 --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/continuous_monitoring_pipeline.ipynb @@ -0,0 +1,742 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Continuous Monitoring Pipeline" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This tutorial describes an advanced use case of [running flows in Azure ML Pipelines](https://github.com/microsoft/promptflow/blob/main/examples/tutorials/run-flow-with-pipeline/pipeline.ipynb). \n", + "The detailed explanations of the prerequisites and principles can be found in the aforementioned article.\n", + "Continuous monitoring is necessary to maintain the quality, performance and efficiency of Generative AI applications. \n", + "These factors directly impact the user experience and operational costs. \n", + "\n", + "We will run evaluations on a basic chatbot flow, then aggregate the results to export and visualize the metrics. \n", + "The flows used in this pipeline are described below:\n", + "- [Basic Chat](https://github.com/microsoft/promptflow/tree/main/examples/flows/chat/chat-basic)\n", + "- [Q&A Evaluation](https://github.com/microsoft/promptflow/tree/main/examples/flows/evaluation/eval-qna-rag-metrics)\n", + "- [Perceived Intelligence Evaluation](https://github.com/microsoft/promptflow/tree/main/examples/flows/evaluation/eval-perceived-intelligence)\n", + "- [Summarization Evaluation](https://github.com/microsoft/promptflow/tree/main/examples/flows/evaluation/eval-summarization)\n", + "\n", + "![continuous_monitoring_pipeline.png](./media/continuous_monitoring_pipeline.png)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.1 Import the required libraries\n", + "## Install the required python packages\n", + "Make sure version of ‘azure-ai-ml’ is higher than 1.12.0" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# import required libraries\n", + "from azure.identity import DefaultAzureCredential, InteractiveBrowserCredential\n", + "from azure.ai.ml import MLClient, load_component, Input, Output\n", + "from azure.ai.ml.constants import AssetTypes\n", + "from azure.ai.ml.dsl import pipeline\n", + "from datetime import datetime" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.2 Configure credential" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "try:\n", + " credential = DefaultAzureCredential()\n", + " # Check if given credential can get token successfully.\n", + " credential.get_token(\"https://management.azure.com/.default\")\n", + "except Exception as ex:\n", + " # Fall back to InteractiveBrowserCredential in case DefaultAzureCredential not work\n", + " credential = InteractiveBrowserCredential()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.3 Get a handle to the workspace" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Get a handle to workspace\n", + "ml_client = MLClient.from_config(credential=credential)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.4.1 Validate Basic Chat Flow\n", + "## Import the flow required packages\n", + "pip install -r ../../../flows/chat/chat-basic/requirements.txt" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!pf flow validate --source ../../../flows/chat/chat-basic" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.4.2 Validate QnA RAG Evaluation Flow\n", + "## Import the flow required packages\n", + "pip install -r ../../../flows/evaluation/eval-qna-rag-metrics/requirements.txt" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!pf flow validate --source ../../../flows/evaluation/eval-qna-rag-metrics" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.4.3 Validate Perceived Intelligence Evaluation Flow\n", + "## Import the flow required packages\n", + "pip install -r ../../../flows/evaluation/eval-perceived-intelligence/requirements.txt" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!pf flow validate --source ../../../flows/evaluation/eval-perceived-intelligence" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.4.4 Validate Summarization flow\n", + "pip install -r ../flows/standard/simple-summarization/requirements.txt" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!pf flow validate --source ./flows/standard/simple-summarization" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 1.4.5 Validate Summarization Evaluation flow\n", + "pip install -r ../../../flows/evaluation/eval-summarization/requirements.txt" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!pf flow validate --source ../../../flows/evaluation/eval-summarization" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 2. Load Chat flow as component" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "chat_flow_component = load_component(\"../../../flows/chat/chat-basic/flow.dag.yaml\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 2.1 Load QnA RAG Evaluation flow as component" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "eval_qna_rag_metrics_component = load_component(\n", + " \"../../../flows/evaluation/eval-qna-rag-metrics/flow.dag.yaml\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 2.2 Load Perceived Intelligence flow as component" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "eval_perceived_intelligence_component = load_component(\n", + " \"../../../flows/evaluation/eval-perceived-intelligence/flow.dag.yaml\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 2.3 Load Summarization flow as component" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "simple_summarization_component = load_component(\n", + " \"./flows/standard/simple-summarization/flow.dag.yaml\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 2.4 Load Summarization Evaluation flow as component" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "eval_summarization_component = load_component(\n", + " \"../../../flows/evaluation/eval-summarization/flow.dag.yaml\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 2.3 Load Parquet Converter\n", + "The Parquet Converter is a [command component](https://learn.microsoft.com/en-us/azure/machine-learning/reference-yaml-component-command?view=azureml-api-2) that aggregates the results of the evaluations node and logs the metrics to ML Pipelines. \n", + "\n", + "![convert_parquet.png](./media/convert_parquet.png)\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "convert_parquet_component = load_component(\n", + " \"./components/convert_parquet/convert_parquet.yaml\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 3.1 Declare input and output" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "data_input = Input(\n", + " path=\"./data/monitoring_dataset.jsonl\",\n", + " type=AssetTypes.URI_FILE,\n", + " mode=\"mount\",\n", + ")\n", + "\n", + "eval_results_output = Output(\n", + " # Provide custom flow output file path if needed\n", + " # path=\"azureml://datastores//paths/\",\n", + " type=AssetTypes.URI_FOLDER,\n", + " # rw_mount is suggested for flow output\n", + " mode=\"rw_mount\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 3.2.1 Run pipeline with single flow component" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Define the pipeline as a function\n", + "@pipeline()\n", + "def continuous_monitoring(\n", + " # Function inputs will be treated as pipeline input data or parameters.\n", + " # Pipeline input could be linked to step inputs to pass data between steps.\n", + " # Users are not required to define pipeline inputs.\n", + " # With pipeline inputs, user can provide the different data or values when they trigger different pipeline runs.\n", + " pipeline_input_data: Input,\n", + " parallel_node_count: int = 1,\n", + "):\n", + " # Declare pipeline step 'flow_node' by using flow component\n", + " chat_flow_node = chat_flow_component(\n", + " # Bind the pipeline input data to the port 'data' of the flow component\n", + " # If you don't have pipeline input, you can directly pass the 'data_input' object to the 'data'\n", + " # But with this approach, you can't provide different data when you trigger different pipeline runs.\n", + " data=pipeline_input_data,\n", + " # Declare which column of input data should be mapped to flow input\n", + " # the value pattern follows ${data.}\n", + " chat_history=\"${data.chat_history}\",\n", + " question=\"${data.question}\",\n", + " # Provide the connection values of the flow component\n", + " # The value of connection and deployment_name should align with your workspace connection settings.\n", + " connections={\n", + " \"chat\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " \"deployment_name\": \"gpt-35-turbo\",\n", + " }\n", + " },\n", + " )\n", + "\n", + " # Provide run settings of your flow component\n", + " # Only 'compute' is required and other setting will keep default value if not provided.\n", + " # If the workspace has been created with Azure AI Studio is inside a hub,\n", + " # a Compute Cluster cannot be used, use a Serverless instance instead.\n", + " chat_flow_node.environment_variables = {\n", + " \"PF_INPUT_FORMAT\": \"jsonl\",\n", + " }\n", + " chat_flow_node.compute = \"serverless\"\n", + " chat_flow_node.resources = {\"instance_count\": parallel_node_count}\n", + " chat_flow_node.mini_batch_size = 5\n", + " chat_flow_node.max_concurrency_per_instance = 2\n", + " chat_flow_node.retry_settings = {\n", + " \"max_retries\": 1,\n", + " \"timeout\": 1200,\n", + " }\n", + " chat_flow_node.error_threshold = -1\n", + " chat_flow_node.mini_batch_error_threshold = -1\n", + " chat_flow_node.logging_level = \"DEBUG\"\n", + "\n", + " # QnA RAG Metrics Evaluation Node\n", + " eval_qna_rag_metrics_node = eval_qna_rag_metrics_component(\n", + " # Bind the pipeline input data to the port 'data' of the flow component\n", + " # If you don't have pipeline input, you can directly pass the 'data_input' object to the 'data'\n", + " # But with this approach, you can't provide different data when you trigger different pipeline runs.\n", + " data=pipeline_input_data,\n", + " # run_outputs connects the output of a previous node\n", + " run_outputs=chat_flow_node.outputs.flow_outputs,\n", + " # Declare which column of input data should be mapped to flow input\n", + " # the value pattern follows ${data.}\n", + " documents=\"${data.documents}\",\n", + " question=\"${data.question}\",\n", + " # Declare which column of previous node output should be mapped to flow input\n", + " # the value pattern follows ${run.outputs.}\n", + " answer=\"${run.outputs.answer}\",\n", + " # Provide the connection values of the flow component\n", + " # The value of connection and deployment_name should align with your workspace connection settings.\n", + " connections={\n", + " \"gpt_groundedness\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " \"deployment_name\": \"gpt-35-turbo\",\n", + " },\n", + " \"gpt_relevance\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " \"deployment_name\": \"gpt-35-turbo\",\n", + " },\n", + " \"gpt_retrieval_score\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " \"deployment_name\": \"gpt-35-turbo\",\n", + " },\n", + " },\n", + " )\n", + "\n", + " # Provide run settings of your flow component\n", + " # Only 'compute' is required and other setting will keep default value if not provided.\n", + " # If the workspace has been created with Azure AI Studio is inside a hub,\n", + " # a Compute Cluster cannot be used, use a Serverless instance instead.\n", + " eval_qna_rag_metrics_node.environment_variables = {\n", + " \"PF_INPUT_FORMAT\": \"jsonl\",\n", + " }\n", + " eval_qna_rag_metrics_node.compute = \"serverless\"\n", + " eval_qna_rag_metrics_node.resources = {\"instance_count\": parallel_node_count}\n", + " eval_qna_rag_metrics_node.mini_batch_size = 5\n", + " eval_qna_rag_metrics_node.max_concurrency_per_instance = 2\n", + " eval_qna_rag_metrics_node.retry_settings = {\n", + " \"max_retries\": 1,\n", + " \"timeout\": 1200,\n", + " }\n", + " eval_qna_rag_metrics_node.error_threshold = -1\n", + " eval_qna_rag_metrics_node.mini_batch_error_threshold = -1\n", + " eval_qna_rag_metrics_node.logging_level = \"DEBUG\"\n", + "\n", + " # Perceived Intelligence Evaluation Node\n", + " eval_perceived_intelligence_node = eval_perceived_intelligence_component(\n", + " # Bind the pipeline input data to the port 'data' of the flow component\n", + " # If you don't have pipeline input, you can directly pass the 'data_input' object to the 'data'\n", + " # But with this approach, you can't provide different data when you trigger different pipeline runs.\n", + " data=pipeline_input_data,\n", + " # run_outputs connects the output of a previous node\n", + " run_outputs=chat_flow_node.outputs.flow_outputs,\n", + " # Declare which column of input data should be mapped to flow input\n", + " # the value pattern follows ${data.}\n", + " question=\"${data.question}\",\n", + " context=\"${data.context}\",\n", + " # Declare which column of previous node output should be mapped to flow input\n", + " # the value pattern follows ${run.outputs.}\n", + " answer=\"${run.outputs.answer}\",\n", + " # Provide the connection values of the flow component\n", + " # The value of connection and deployment_name should align with your workspace connection settings.\n", + " connections={\n", + " \"gpt_perceived_intelligence\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " \"deployment_name\": \"gpt-35-turbo\",\n", + " }\n", + " },\n", + " )\n", + "\n", + " # Provide run settings of your flow component\n", + " # Only 'compute' is required and other setting will keep default value if not provided.\n", + " # If the workspace has been created with Azure AI Studio is inside a hub,\n", + " # a Compute Cluster cannot be used, use a Serverless instance instead.\n", + " eval_perceived_intelligence_node.environment_variables = {\n", + " \"PF_INPUT_FORMAT\": \"jsonl\",\n", + " }\n", + " eval_perceived_intelligence_node.compute = \"serverless\"\n", + " eval_perceived_intelligence_node.resources = {\"instance_count\": parallel_node_count}\n", + " eval_perceived_intelligence_node.mini_batch_size = 5\n", + " eval_perceived_intelligence_node.max_concurrency_per_instance = 2\n", + " eval_perceived_intelligence_node.retry_settings = {\n", + " \"max_retries\": 1,\n", + " \"timeout\": 1200,\n", + " }\n", + " eval_perceived_intelligence_node.error_threshold = -1\n", + " eval_perceived_intelligence_node.mini_batch_error_threshold = -1\n", + " eval_perceived_intelligence_node.logging_level = \"DEBUG\"\n", + "\n", + " # Summarization Node\n", + " simple_summarization_node = simple_summarization_component(\n", + " # Bind the pipeline input data to the port 'data' of the flow component\n", + " # If you don't have pipeline input, you can directly pass the 'data_input' object to the 'data'\n", + " # But with this approach, you can't provide different data when you trigger different pipeline runs.\n", + " data=pipeline_input_data,\n", + " # run_outputs connects the output of a previous node\n", + " run_outputs=chat_flow_node.outputs.flow_outputs,\n", + " # Declare which column of previous node output should be mapped to flow input\n", + " # the value pattern follows ${run.outputs.}\n", + " answer=\"${run.outputs.answer}\",\n", + " # Provide the connection values of the flow component\n", + " # The value of connection and deployment_name should align with your workspace connection settings.\n", + " connections={\n", + " \"summarize_text_content\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " \"deployment_name\": \"gpt-35-turbo\",\n", + " }\n", + " },\n", + " )\n", + "\n", + " # Provide run settings of your flow component\n", + " # Only 'compute' is required and other setting will keep default value if not provided.\n", + " # If the workspace has been created with Azure AI Studio is inside a hub,\n", + " # a Compute Cluster cannot be used, use a Serverless instance instead.\n", + " simple_summarization_node.environment_variables = {\n", + " \"PF_INPUT_FORMAT\": \"jsonl\",\n", + " }\n", + " simple_summarization_node.compute = \"serverless\"\n", + " simple_summarization_node.resources = {\"instance_count\": parallel_node_count}\n", + " simple_summarization_node.mini_batch_size = 5\n", + " simple_summarization_node.max_concurrency_per_instance = 2\n", + " simple_summarization_node.retry_settings = {\n", + " \"max_retries\": 1,\n", + " \"timeout\": 1200,\n", + " }\n", + " simple_summarization_node.error_threshold = -1\n", + " simple_summarization_node.mini_batch_error_threshold = -1\n", + " simple_summarization_node.logging_level = \"DEBUG\"\n", + "\n", + " # Summarization Evaluation Node\n", + " eval_summarization_node = eval_summarization_component(\n", + " # Bind the pipeline input data to the port 'data' of the flow component\n", + " # If you don't have pipeline input, you can directly pass the 'data_input' object to the 'data'\n", + " # But with this approach, you can't provide different data when you trigger different pipeline runs.\n", + " data=simple_summarization_node.outputs.flow_outputs,\n", + " # run_outputs connects the output of a previous node\n", + " run_outputs=chat_flow_node.outputs.flow_outputs,\n", + " # Declare which column of input data should be mapped to flow input\n", + " # the value pattern follows ${data.}\n", + " summary=\"${data.summary}\",\n", + " # Declare which column of previous node output should be mapped to flow input\n", + " # the value pattern follows ${run.outputs.}\n", + " document=\"${run.outputs.answer}\",\n", + " # Provide the connection values of the flow component\n", + " # The value of connection and deployment_name should align with your workspace connection settings.\n", + " connections={\n", + " \"score_fluency\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " },\n", + " \"score_consistency\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " },\n", + " \"score_relevance\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " },\n", + " \"score_coherence\": {\n", + " \"connection\": \"azure_open_ai_connection\",\n", + " },\n", + " },\n", + " )\n", + "\n", + " # Provide run settings of your flow component\n", + " # Only 'compute' is required and other setting will keep default value if not provided.\n", + " # If the workspace has been created with Azure AI Studio is inside a hub,\n", + " # a Compute Cluster cannot be used, use a Serverless instance instead.\n", + " eval_summarization_node.environment_variables = {\n", + " \"PF_INPUT_FORMAT\": \"jsonl\",\n", + " }\n", + " eval_summarization_node.compute = \"serverless\"\n", + " eval_summarization_node.resources = {\"instance_count\": parallel_node_count}\n", + " eval_summarization_node.mini_batch_size = 5\n", + " eval_summarization_node.max_concurrency_per_instance = 2\n", + " eval_summarization_node.retry_settings = {\n", + " \"max_retries\": 1,\n", + " \"timeout\": 1200,\n", + " }\n", + " eval_summarization_node.error_threshold = -1\n", + " eval_summarization_node.mini_batch_error_threshold = -1\n", + " eval_summarization_node.logging_level = \"DEBUG\"\n", + "\n", + " convert_parquet_node = convert_parquet_component(\n", + " # Bind the evaluation nodes outputs to the command component's input\n", + " eval_qna_rag_metrics_output_folder=eval_qna_rag_metrics_node.outputs.flow_outputs,\n", + " eval_perceived_intelligence_output_folder=eval_perceived_intelligence_node.outputs.flow_outputs,\n", + " eval_summarization_output_folder=eval_summarization_node.outputs.flow_outputs,\n", + " )\n", + "\n", + " # Provide run settings of your flow component\n", + " # Only 'compute' is required and other setting will keep default value if not provided.\n", + " # If the workspace has been created with Azure AI Studio is inside a hub,\n", + " # a Compute Cluster cannot be used, use a Serverless instance instead.\n", + " convert_parquet_node.compute = \"serverless\"\n", + " # Function return will be treated as pipeline output. This is not required.\n", + " return {\n", + " \"eval_results_output_folder\": convert_parquet_node.outputs.eval_results_output\n", + " }\n", + "\n", + "\n", + "# create pipeline instance\n", + "pipeline_job_def = continuous_monitoring(pipeline_input_data=data_input)\n", + "pipeline_job_def.outputs.eval_results_output_folder = eval_results_output" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 3.2.2 Submit the job" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Submit the pipeline job to your workspace\n", + "pipeline_job_run = ml_client.jobs.create_or_update(\n", + " pipeline_job_def, experiment_name=\"Continuous Monitoring\"\n", + ")\n", + "pipeline_job_run\n", + "\n", + "ml_client.jobs.stream(pipeline_job_run.name)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 4.1 Next step - Setup scheduler for your pipeline" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from datetime import datetime\n", + "from azure.ai.ml.entities import JobSchedule, RecurrenceTrigger, RecurrencePattern\n", + "from azure.ai.ml.constants import TimeZone\n", + "\n", + "schedule_name = \"simple_sdk_create_schedule_recurrence\"\n", + "schedule_start_time = datetime.utcnow()\n", + "\n", + "recurrence_trigger = RecurrenceTrigger(\n", + " frequency=\"day\", # could accept \"hour\", \"minute\", \"day\", \"week\", \"month\"\n", + " interval=1,\n", + " schedule=RecurrencePattern(hours=10, minutes=[0, 1]),\n", + " start_time=schedule_start_time,\n", + " time_zone=TimeZone.UTC,\n", + ")\n", + "\n", + "job_schedule = JobSchedule(\n", + " name=schedule_name,\n", + " trigger=recurrence_trigger,\n", + " # Declare the pipeline job to be scheduled. Here we uses the pipeline job created in previous example.\n", + " create_job=pipeline_job_def,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Initiate the scheduler" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "job_schedule = ml_client.schedules.begin_create_or_update(\n", + " schedule=job_schedule\n", + ").result()\n", + "print(job_schedule)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# To disable or shut down a running scheduler" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "job_schedule = ml_client.schedules.begin_disable(name=schedule_name).result()\n", + "job_schedule.is_enabled" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# 4.2 Next step - Deploy pipeline to an endpoint" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from azure.ai.ml.entities import BatchEndpoint, PipelineComponentBatchDeployment\n", + "\n", + "# from azure.ai.ml.entities import ModelBatchDeployment, ModelBatchDeploymentSettings, Model, AmlCompute, Data, BatchRetrySettings, CodeConfiguration, Environment, Data\n", + "# from azure.ai.ml.constants import BatchDeploymentOutputAction\n", + "\n", + "\n", + "endpoint_name = \"hello-my-pipeline-endpoint\"\n", + "endpoint = BatchEndpoint(\n", + " name=endpoint_name,\n", + " description=\"A hello world endpoint for pipeline\",\n", + ")\n", + "\n", + "ml_client.batch_endpoints.begin_create_or_update(endpoint).result()" + ] + } + ], + "metadata": { + "build_doc": { + "author": [ + "loubigard@microsoft.com" + ], + "category": "azure", + "section": "Flow", + "weight": 40 + }, + "description": "A tutorial to create a pipeline using components to do continuous monitoring on a Gen AI application", + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.19" + }, + "resources": "examples/flows/chat/chat-basic,examples/flows/evaluation/eval-qna-rag-metrics,examples/flows/evaluation/eval-perceived-intelligence,examples/flows/evaluation/eval-summarization" + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/data/monitoring_dataset.jsonl b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/data/monitoring_dataset.jsonl new file mode 100644 index 00000000000..4161ad073cf --- /dev/null +++ b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/data/monitoring_dataset.jsonl @@ -0,0 +1,5 @@ +{"question":"What's Azure Machine Learning?", "chat_history":[],"answer":"Azure Machine Learning is a cloud-based platform provided by Microsoft for building, training, and deploying machine learning models. It offers a variety of tools and services that facilitate the entire machine learning lifecycle, including data preparation, model training, model evaluation, and model deployment. With Azure Machine Learning, users can develop and deploy models using various programming languages like Python and R, and take advantage of automated machine learning capabilities to streamline the model development process. Additionally, Azure Machine Learning integrates with other Azure services, allowing users to easily scale their machine learning workflows and leverage features like automatic model deployment and monitoring.[doc1]","documents":[{"[doc1]":{"title": "Take advantage of key features for the full ML lifecycle", "content": "Azure Machine Learning is a comprehensive cloud service designed to help you build, deploy, and manage machine learning models at scale. It provides a robust platform for data scientists, machine learning professionals, and engineers to work on the entire machine learning lifecycle, from data preparation to model training and deployment."}}],"context":["Azure Machine Learning is a fully managed cloud service used to train, deploy, and manage machine learning models at scale. It fully supports open-source technologies, so you can use tens of thousands of open-source Python packages such as TensorFlow, PyTorch, and scikit-learn. Rich tools are also available, such as Compute instances, Jupyter notebooks, or the Azure Machine Learning for Visual Studio Code extension, a free extension that allows you to manage your resources, model training workflows and deployments in Visual Studio Code. Azure Machine Learning includes features that automate model generation and tuning with ease, efficiency, and accuracy."]} +{"question":"How can I use Azure ML Pipeline ?", "chat_history":[],"answer":"To use Azure ML Pipeline, follow these steps:\n\n1. Create an Azure ML workspace: Start by creating an Azure Machine Learning workspace in your Azure portal. This workspace will serve as the central hub for your machine learning experiments and pipelines.\n\n2. Create an Azure ML pipeline: Once you have created your workspace, you can create an Azure ML pipeline. A pipeline is a collection of steps or activities that are executed in a specific order to complete a machine learning workflow.\n\n3. Define pipeline steps: Each step in the pipeline represents a specific task or activity, such as data preprocessing, model training, or model evaluation. You can use Azure ML SDK or Python SDK to define these steps.\n\n4. Define dependencies between steps: If there are dependencies between the steps, such as one step requiring the output of another step, you can define these dependencies in the pipeline. This ensures that the steps are executed in the correct order.\n\n5. Publish the pipeline: After defining the pipeline, you can publish it to make it available for execution. Published pipelines can be triggered manually or scheduled to run at specific intervals.\n\n6. Monitor and manage pipeline runs: Once the pipeline is published, you can monitor its runs and manage its execution. Azure ML provides tools and APIs to track the progress, view [doc1]","documents":[{"[doc1]":{"title": "Tutorial: Create production machine learning pipelines", "content": "Using Azure Machine Learning Pipelines allows you to create and manage workflows for your machine learning tasks. Here’s a step-by-step guide to get you started: 1. Set Up Your Environment 2. Create a Pipeline Run and Monitor the Pipeline 4. Deploy the Model"}}],"context":["The core of a machine learning pipeline is to split a complete machine learning task into a multistep workflow. Each step is a manageable component that can be developed, optimized, configured, and automated individually. Steps are connected through well-defined interfaces. The Azure Machine Learning pipeline service automatically orchestrates all the dependencies between pipeline steps. The benefits of using a pipeline are standardized the MLOps practice, scalable team collaboration, training efficiency and cost reduction. To learn more about the benefits of pipelines, see What are Azure Machine Learning pipelines."]} +{"question":"What are the pros and cons of prompt flow ?", "chat_history":[],"answer":"Prompt flow, also known as prompt chaining or prompt dependency, is a teaching strategy often used in special education to help individuals acquire new skills. It involves providing a series of prompts to guide the person through the steps of a task until they can complete it independently. Here are some pros and cons of using prompt flow:\n\nPros:\n1. Skill acquisition: Prompt flow can be effective in teaching new skills to individuals who have difficulty learning through other methods. It breaks down complex tasks into smaller steps, making it easier for the learner to understand and follow.\n\n2. Individualized instruction: Prompt flow can be tailored to meet the specific needs of each learner. Different types of prompts, such as visual, verbal, or physical prompts, can be used depending on the individual's abilities and learning style.\n\n3. Gradual fading of prompts: Prompt flow allows for the gradual fading of prompts over time. As the learner becomes more proficient in the skill, the prompts can be reduced or removed, promoting independence.\n\n4. Increased success and confidence: By providing the necessary support, prompt flow increases the chances of success for the learner. As they experience success, their confidence and motivation to learn can also increase.\n\nCons:\n1. Dependence on prompts: There is a risk that the learner may become overly [doc1]","documents":[{"[doc1]":{"title": "What is Azure Machine Learning prompt flow", "content": "Azure Machine Learning Prompt Flow is a tool designed to streamline the development of AI applications powered by Large Language Models (LLMs). Here are some of the pros and cons: Pros: 1.Prompt Engineering Agility 2.Interactive Authoring Experience 3.Collaboration 4.Comprehensive Resources 5.Enterprise Readiness Cons: 1.Complexity 2.Documentation Variability 3.Frequent Updates"}}],"context":["Recently, I looked into the options within Azure, and as always there is more than one option. I started my journey in the world of LLM app development using LangChain. I then came across Prompt Flow and Semantic Kernel (SK). These are pretty much the three big players when it comes to LLM app development. Think of them as the superheroes of the LLM app development world, each with their own special powers. LangChain is a framework that simplifies integrating LLM capabilities into your application. It supports models by OpenAI, Cohere, HuggingFace, Anthropic and many more. The essence of LangChain is captured in its name, which combines \"Lang\" (short for language) with \"Chain,\" reflecting the framework's core functionality of linking various components to harness LLMs for creating sophisticated language-based applications. Semantic Kernel is an open-source SDK from Microsoft that makes it easy to build AI agents that can interact with a wide range of LLMs as well as call existing application functions. It is available in Python, C#, and Java. Prompt Flow is a toolkit designed to make the development of AI apps powered by Large Language Models (LLMs) more streamlined from prototyping, testing, evaluation, deployment and monitoring. It makes prompt engineering much easier and enables developers to build high-quality LLM applications."]} +{"question":"What is a Azure ML pipeline parallel component ?", "chat_history":[],"answer":"In Azure Machine Learning, a pipeline parallel component is a type of component that allows you to execute multiple steps in parallel within a pipeline. This is especially useful when you have independent steps that do not have any dependencies on each other and can be executed simultaneously to improve overall pipeline performance.\n\nBy using pipeline parallel components, you can distribute the workload across multiple compute targets, such as multiple nodes in a cluster or different GPU instances, to speed up the execution of your pipeline. This can significantly reduce the time required for processing large datasets or performing computationally intensive tasks.\n\nTo create a pipeline parallel component in Azure ML, you can use the ParallelRunStep class, which allows you to define multiple parallel steps and configure their execution parameters. Each parallel step can have its own input data, code, and dependencies, and the outputs of all steps can be consolidated at the end of the parallel execution.\n\nOverall, using pipeline parallel components in Azure ML can help you optimize the performance of your machine learning pipelines by leveraging parallel processing capabilities and distributing the workload across multiple resources. [doc1]","documents":[{"[doc1]":{"title": "How to use parallel job in pipeline (V2)", "content": "An Azure Machine Learning Pipeline Parallel Component allows you to execute multiple tasks simultaneously, significantly speeding up the processing time for large-scale machine learning jobs. Here are some key points about it: Key Features: 1.Parallel Execution 2.Efficiency 3.Scalability 4.Error Handling"}}],"context":["In the real world, ML engineers always have scale requirements on their training or inferencing tasks. For example, when a data scientist provides a single script to train a sales prediction model, ML engineers need to apply this training task to each individual store. During this scale out process, some challenges are: Delay pressure caused by long execution time. Manual intervention to handle unexpected issues to keep the task proceeding. The core value of Azure Machine Learning parallel job is to split a single serial task into mini-batches and dispatch those mini-batches to multiple computes to execute in parallel. By using parallel jobs, we can: Significantly reduce end-to-end execution time. Use Azure Machine Learning parallel job's automatic error handling settings. You should consider using Azure Machine Learning Parallel job if: You plan to train many models on top of your partitioned data. You want to accelerate your large scale batch inferencing task."]} +{"question": "What is the purpose of the LLM Grounding Score, and what does a higher score mean in this context?", "chat_history":[], "answer": "The LLM Grounding Score is a metric used in the context of in-context learning with large-scale pretrained language models (LLMs) [doc1]. It measures the ability of the LLM to understand and connect with the provided context or demonstrations during the learning process.\n\nA higher LLM Grounding Score indicates that the LLM has a better understanding and connection with the context or demonstrations [doc1]. It means that the LLM is able to effectively utilize the provided examples and generate more accurate and relevant responses based on that context.\n\nPlease note that the LLM Grounding Score is specific to the in-context learning framework and may not be applicable in other contexts or tasks.\n\n[doc1]: In-Context Learning with Large-Scale Pretrained Language Models: How Far Are We? (2022) - zelin, English.", "documents": [{"[doc1]":{"title": "In-Context Learning with Large-Scale Pretrained Language Models", "content": "In-Context Learning with Large-Scale Pretrained Language Models\nConcepts\nFew-Shot Learning : the model learns a new task from a small amount of training examples.\nIn-Context Learning : large-scale pretrained language models learn a new task simply by conditioning on a few training examples and predicting which tokens best complete a test input. In-context learning is entirely different from few-shot learning: the language models does receive only a few training examples, but the overall system may still require a large number of training examples.\nFew-shot learning with Large Language Models (LLMs)\nGPT-3 ( paper link (https://arxiv.org/pdf/2005.14165.pdf) paper link ) introduced the idea of adapting models to new task without fine-tuning or gradient update. The approach is to elicit the LLM with text interaction to work with new tasks with accuracy close to many finetuned models. This ability of LLMs to work with few demonstrations or even just a task description is demonstrated when scale of the model crosses a threshold. Image below shows the difference between different ways to do in-context learning.\nUsually, giving few demonstrations (Few-shot) shows better performance than giving just an instruction (Zero-shot). Varying the number of in-context examples also affects performance.\nHow to actually prompt the language models\nDifferent prompt orders have different performance\nRelevant Paper: Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity (https://arxiv.org/abs/2104.08786) Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity\nThis paper gives few major insights:\n1. Even with a fixed set of few-shot prompts, different orders lead to different performance. This means the models are sensitive to permutations of demonstrations we provide them. \n2. There are certain ordering that are \"fantastic\" and we might need to discover them through \"probing\". You can find the details in the paper above.\nFinding similar train examples for prompt improves performance\nRelevant Paper: Making pre-trained language models better few-shot learners (https://arxiv.org/pdf/2012.15723.pdf) Making pre-trained language models better few-shot learners\nIntuitively, we can guess that if we give model demonstrations that look like the final inference task, the model might show better performance. This paper demonstrates improvement in performance by using pre-trained sentence embeddings to select the closest prompt examples to the given input instance (for text classification tasks).\nMain Takeaway : Based on above two papers, we can conclude that, for improving in-context performance for any task:\n1. We need to find \"great\" few-shot examples. This can be done using models pretrained on other tasks. We call such models \"Retrievers\".\n2. Order in which we select the samples matter.\nIn-Context Learning with Retrievers\nRelevant Paper: What Makes Good In-Context Examples for GPT-3? (https://arxiv.org/pdf/2101.06804.pdf) What Makes Good In-Context Examples for GPT-3? 1. This paper demonstrates how using nearest neighbor samples based on semantic similarity to test samples improves performance by benchmarking on various NLU and NLG tasks.\n2. They also show that using models finetuned on semantic similarity tasks (like Roberta finetuned on NLI) shows even better performance. This implies that certain models are better retrievers than others for in-context learning.\nAdvanced: Improve retrievers by finetuning on downstream tasks\nDTE-finetuned (01_dte.md) DTE-finetuned\nTarget Similarity Tuning (02_tst.md) Target Similarity Tuning\nSynchromesh: Reliable Code Generation from Pre-Trained Language Models (https://arxiv.org/pdf/2201.11227.pdf) Synchromesh: Reliable Code Generation from Pre-Trained Language Models\nSuppose that we already have the assumptions that the training examples with the same SQL template as the test input are \"fantastic\" examples, we can extract SQL templates for the training examples and fine-tune the retriever to make the training examples sharing the same SQL template together.\nGenerally, we can finetune the retrievers to learn to embed examples with similar output closer. This seem to work well for code-generation task but we need to benchmark this across tasks.\nLimitation: the assumption that the training examples with the same SQL template as the test input are \"fantastic\" examples . We need different implementations (or even different assumptions) for different downstream tasks."}}], "context": ["Grounding, particularly with retrieval-augmented generation, has a wide range of applications. One obvious use case is search and question-answering (Q&A) systems. For instance, when you interact with Bing chat, it transparently retrieves search results and uses them to ground its responses. Similarly, many users are now building systems that leverage large language models (LLMs) to distil and make sense of information from their repositories, enabling Q&A over documents. Another use case is generating content with information from an app or user interface. For example, Microsoft's Copilot, integrated with Visual Studio Code, uses an LLM to provide context-aware suggestions based on the document you are working on. This approach is also being implemented in Microsoft 365 and Power Platform, and developers are now being invited to create their own context-aware applications. Retrieving information from APIs or external sources, such as the weather forecast or stock quotes, is another area where grounding can be beneficial. Additionally, grounding can be used for memory and state management in multi-step generation processes. Since LLMs are stateless, incorporating previous interactions or generated content can help produce more contextually relevant responses."]} \ No newline at end of file diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/compare_button.png b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/compare_button.png new file mode 100644 index 0000000000000000000000000000000000000000..ce7ec166ed9261c0b17e14f4b91bd4db736492de GIT binary patch literal 22556 zcmeFZWmr^g`}Pfjba%IefOMCPq<{zpA`Mc~-QCh9-K_#D-3`(yLwCb4^f<)utns?; z|NDM<-{-l%KW>}jV%BnItuv1EIDY$n67}+hGCmFs4hjkizN(6XHVO(_5^%hVg#o;V z8Xd_1|Dd{PD}zv~#_0BeAJ8r3HRMrHe#GP6nmh!4#&%TEbwNQP=t2HP9dIcBfP$j) zLsdckwWlE@2RoarZ$2Vg?d9nwnm#jB@;>erG4621P3rMp&ho$pcJ_}XtTei>iDILa zV!>s&!Jcd%H;7$+6*jV+std0$6YCO67bz(un6>+VYz|*2^|A1FcX$7MCwFtWI&z_9 z0Q|W`9Qw7?tFd=tF^3X?kS~f;BDZ9JkARm?AdvRIXA~Z@v4#C}38?u0;Oz^j+*`+7 zzboAu*ox00G+qwTIB#8+a|F9u+fFkQO=i~+; z`071^5+PV>1Fo{T{H-Il+~Pq9E;ierD%qLVloL_=%!3WCjk8~F&JdY>*R#X@{A=h| zmP5OBRr~cv7qEdxhR2H6K-J-7QN+B{0R1Srj((!hC&e2r-i;3{tzJ88A%vnr79;#_ zZf-k>8>raVs^*W4ZvsP$k9HF>AD44zKhJ@t15F{L&x-7~$s_~!`MxC|Xxq+| z8ed(mTx;U?+7NfOu%`5>!X)kuD-EOfUIE47S>k(@jm}&6c!0hCNeJV+zLCEBb;1Dt zJin`qyV&ctv<{ldquj&_Sbjk_KkdbO?vQH4z^l|Egr7vSDQ*hEZS^9`=o*(NggNGF$PE=Er%o{P{;1gbcQb%j$r6Kb44! zcKE2n!YUsBH=q&RS1}oVC#ueee@F1EnRa2PT1er7yUW#Uom*}pHRG~ZU7LtR%2|5n zU(CbH!{WD+O0pTZ5u5N(5NwDz&7uAN*rTInar(1Mis?|wSwXgbjs;<3s}P4vp+`)*HB` zxEOr9K;N02Z%^&s0+Z45xWwJ#NYnp}^v36GYtUV878}{2;=aU9+n@Cobh?z{ z8SLyzokJ=W6=;aQ-@sh?^kw zi|;q-L{~h`Y(kmII)LWMqIGDUtb)mZ%k@1QoTb|IleECO9uGj6g`q)Ri62LuJeca@ zbT?)qvB&L+AZ$48Cz~mVVEFi?^#qh9x(YT5@H;;=x?jMf7VI6#5E?j!o_FK&&24&N zlCcsyHC!#&WXbwnhx$;}025{5(8P~qNgQ23Nw6SPU#R~^jj!#))v||?%c>u&q&=6` zzOjqu^8OAQFly)nF~?tltp=>#oLsqvmPPzA z(<>v^5PpvE8P>x2y{-*aCGl8tity7(Q2Jz3aO63#5U^X5E=lp|7+&nRVArrb(nBa` z3oB^N^*NUCk>g661?eeWZ$+#2HA1&?t`g+w4hTVB(;8x<*vjloMZUTm$Pj?4MU0In z-C)r_>&wJ2DL<#cGr$8g;<&XI3TJDKk5H&Z2QLP3)bFx*fpc78*Me^CBN z43$84mE8G=5X7H|?3=*=30CqFBCgYM6dj{|lmpJ&i;Pw3Ob?C(QX)31T8@X%$@>k_ zz}iQboxmcHYi(;gvYj!W?>uLQww{cQUTOdxnjZhEr|jQIk%vvnJ}1nHh6$$;v)5-Q zqFT>Xx0wB=o6)(}aFaxdc^T{$f=M{$do^#lzQ9wZZ?J&yTVIjAzdd)CIaywY3YhhW z+N5pAo8SI9&B&ixbbau{bXd^yL=YGw>no>!++qaH19am0R=8n2j_p9;iSRMErGEHe z#$Qu!%n62*G<1=u%<#htMp&>A;Z-#zv4PdxOF9m=sI8CHI0$ zZ797JvwN~i^VL97uI56DgYPD-d)un}A3OYLn-)&Bp&6sy$QF+(UbDMA#05tSL9XWp zcFS0REb}n~E^;OtD>DEC=&EMP_oNFmgEdaPFkfb*2~9o&89jxfDR_V0xX=wqZOp=p z&s0b-A~H453|aKcV>a?!#v~-5WOV!KnX- zrNG@&nTh}Cn-~g%&ym1k^Y*gF6d|5GtvwZ$&yKR(wHB@>Dg!;!w>UFjEW2K`>C6P- zvruJr{(Q(K3CYo#k<41fA0PnRJt%70I}9@E3D?#9XdJV+>WiUN%PS_WYz=)S`-{@v zJ%EA>!~k1z(Y$F(cbpPm@*j zPNwX&!;QPCLEltaN81kOYsxOAon(&*3-X^*ulrS_hsrtLDd4OzyA-6-+SMjrzDxtz z9IKMZAx`5V#rjVr7`?UF9AOVb0$L#TxZo!tU6=`fedRCLJ4|6~@NsdR_9;ePOC5z9cnfaFY95->q%qG^yDduS&6V@oR zJ}m-eUd!M7rqS0EwVq<5wKzJ!*dDw2`tl(E6EcB&tL_Nmuf}tKB@(H!_M361+bT9P z6ljv3v9??I;wnOPH;fA-rm`NzEGb|qju?&%@i0!}cT|N=7(CE-XUFEQ^2Lf_HcckM zag8oD;>528@p;pQ9kE5jQyqS9;;^zP{*w!`=sjFR#1V1tPxU$Vd_a98f)g1`$n*jflx98Fu1+0X}20PkZV3~q! zA_FbMfV;~Rz0c0G0vK2)Gxg-27MlkTqoG~o$i8;uXEIPFj-pIa`BMB)Z&0JXryC*8 zksi=zlok&>Sv7951xh&M^VufrYBy42WAV=q3}48p$3=bQBttJc zfBGl)wFPH0d?zbSk6k_8ILx`Ji!}HElbt9^N^JOQ(ZK-*^e+(sFa_f0KO}2>n_!zs zO8Y=8YK>?0I1;M6ldaCb*|Rq&0@gQcYlB7}mrt)u^7ILcq3fh#y)AuTic1C443)8Q63sC_pXMv0{y2KQlmY(d zO;GGhcF9C?!(Ws}zLUUI)2)(JYNzHg-HK1ZXB>m~(>jE9d>-(krQgc9{nJ<<^9$>Y zLPv~1d<~@~9@+JwXnHx+H{APCY^Gxx-(HP$Bh*VHLwf&drISzrsYgUg5tK2o;H8)} zsu52`{x5;Bs}mw7>FW0DgC!q*sv@fRk9Q#BMnOXIrh7tY?8d&=P}v8;!5)*oI0l5~$~2KLcK6Mfc^j6rUt4i+ zckyIzG0Gnt53i-D*gA+@7fh1HowoWbCG6aTF82mX14U`ge^Gt>B zW&f504%F(Lha|RZEy$={pK;gtDKpOYVy2@RGN`-So*W63kAABdAwRvC){M{_0+Nw$ zW<{Hh(y3<5yi+lc_^Dozmj6~~V*im(N@>9pTB4R_Q3a&jOVw7V&?7s7=fk(fQe`(=|DkKrT zmBQ18s?U-nr}G4-fl7)CTZm2YE9aWQn&J!gA}0L=Enk@2x>5|42?7D@&?D5hVd{;8 zRRkaY$enzO#RF^K+{tpV&&F)+z6-Us5XtzR-ar3L!f| zYohfe&#-s;zxK8Ulhd+_D|orC47J8jNzUa|GrCe>y4Y_TZ;o5+ytMkQha?zP%|F3n zsV%vxpVrPr`Fr+z&X(`Of9XUF`)Qs$B{^2(lrn!hFCg`#U@k>#LGm-9g$c5{vWqYJ zcJkz`L($8Svx8F_wiPD*L(Yqh@785yucp_bD#S;<>GViT?BrnTvz7y=4yH+!!c#}+ zd_&61e<#d$m+wDL4w}NS%E*JBlnVK-&Ia?R)D-01FsMZ&>l>UMl+Od%a(_^NTfJ zc(XrNC_|lHfAKV3%prH)GSAO8*A?nh9 z)haz66n9 z!XHz@4-B~+ZGX6x%P_|yL~35@ynp5+Lz>;-u-aZZ7Qj^tZmO`{_N{8N6cG`5vzXf_ zQ`UEpW4&;3mEU%t*#z~d12g6&g>ES``hB=s4G3`QBiBiPMj+9<dYc%TvtO(`$(pi2DHbCTP02?9tHMKO@Lt=gIwJu6Gj8^PG61d9dOjvK+tUKP zM%esJ#SUUS>p{Z=R=tCRTLQtffGQAhBPZ9O93o#sG$1oa`a6 zX5xY)>lH4`{8V_^b6K|ET~k+=g#q0_1#6yP{X-KCbGc~sgQW~GGhm`4x(G>H30_nsNyLkca%SyitoCPZyU(vU<}IwrZNA@{kz@HHMrYNDejcF-pU~+n|hdS&|+rJPgsE5$T&A2%m-bWf&CEpCZ z!&*%%ySBS>hPS&S&RZZ`S$*3TNnX6+03cmn(Y&W=?C-dPIMfp;oK}zihJs+%-+KsZ zZqBndxd@eOrQu`Ejpwgv76^JTzyL&Y)DZO4)c3zTj5c&2yO(ToH{@1f$^5*Ut1y0O z{B-npgOx#LxZstOTaf6?c81N_p~?kKp?RkW{0~^=Jr6<_aXMK$54Ll;{HaZ*2fIBq zi!$s8@Jq42aLf2;6OSX0xb^9}R9ws|Zh!Q=y}8Fj$7rLTgMQfC7JcEhEm54uWtttq zRlF3#lAAGOBcynf^D0*(OLWw>VcC6_73axv@fowhwV{Wh$BX;-cb1`$NtYJ4WBn&~ zIv7vrb|0>)aYPPHL`DadAlB-|Dj zy^0haiwN_UNhu0B2X3x!wm{wDmVD zKG2LbL+?QxqN)n*m)6HEP}#%vSXuq2eb)fJw87^^f5O@^=BU&vgcITvNmQzi0I1vI za}~@yvY$bYkLas1PFV3!@oXYCnsmplq~Oa zYvllr7|6Ow7|fUv7>4VRp%M?fYZEr+PrkK*a7)^PHoy& z^6p^u*9)Oz8Z2&Bx|TUhfeONW!+H-Nef8zd>qf)009iQX!0_UFkC2vz=14R@n@YSL zAe_zvuQ;+#XICcb#d{_%*F7!aFJ|yyQQlpC+$9c!Iv#<|pd6<~zV<&G`txk`YP17F zQI(xnjooE5~u^g4$ z8NibwQb*rIngB(rQl&W99$?RJ^^buvsh(fpV2#@d{>cvIt)Z9sP1aDzoHJDU6tM;% z!%FY?B)&U0umRXxGtd9p`mmrFRUE#R3~_U+v1U3+XA$Sd&%p(2K5{P#AvhcB*;3M@ zf|r2m`ym?XO!Txla=6odf)DLEOb^miavye`Ic62JKdRU z8${Y{7Wz4B)4r{^POgsSQodW4dN6H3m^Z^nXOTjL)cHk2O)CfboijB%0a3H!~}e0Ms3bs6E@mHlQS1z=?VH z;U~is8-0GKH40tK*&65`P{wShpQV$SO7U-m1^Vu9sy;GW;cS#Ojf5g1Iap;q(LU|tO$DcK`%XqoMMz6_(c=I3!j^(|$yPzPbj=oT z&Xw@lNNjpT$?vXTihzdn7;PN=GADeH`fHl;NjL(l={&Pj*kbn)J?uL0%03?VcU$qH%K$MssgAO* zCnCF{Clz(O>-uAOb9Agc?J?m&pFEuHl2_HOjOw{e{e&NPX(g8>V^Xy*jtpx%r?HV? z#ED{Ph~E!pi8Qqc|6T(=a2#hR!J`N9r~&m_rNx6!EtP{L_ib?~n`Se;KNB8kx64XY z;D3F67vyWFE^$a0oO%7MdCVk~_gNdr(}PVLS9Rsl5}=>hMKIu77Yd#rn1vcxhKL** zJ+R>gKhNxoxS3*}O2_>bqzYOdYO;`X64+zxYt5;D0f|$KG1)ukfc8lH8O)nv!cEw?_nEf5k;Y=Je7A z-ss1NA;dk2UDk51B?yK%svNjpdz?*Xn&B>s9heJ{Ek^X#5zDG>qxDI%MoM4fUq5tY zE|!a+S$}z3xcJ=P!R+2~i5-Md8h+nZ2|jsiPgg`? z^X*5thEN=6U&xw}9K(T;1&!}9Ir?N*OfA;n^B%OGVtfJyGD*fW>`2EL4#|>wj31+D zhXeh|az@`0bzljvPKl2p4^7CZ^!+*c8$G0!IQikd?_6S0;JY$OAb!wxH%${?D3|zY zlBh=9?zTOcsQG4*qS4b=X@-t!RoF_{Vk?hW7WB-(zwt#C<>L%+t)rcAKi+L45eXxW zA&up6>TrNf!3dSsl-BF`R zvqUS07qj~-Aa5d72qFP~h3sH{Q%{CLxe1HLNd#z|W$O@ap+)A&YgnLc-?&`ftvlLC zEg$@vZmJY=HeK!B=sy(J{KC`bB4^CkFIuy``BmgB-EfWWUi)^>a)3_R`6aV$G)|D| z@PO8@7tZ1Lzjch33n8aYe$lIAH{Wwm`4MV?=d#X*{?kg=UPXwvuR3%i|90;|^5(*K zK)~0jMUZ_o8`2687i$i=?&kJ*D&xHCx*HhxG60x;Yw`uo#1Uh*DobRKQIw>Muxvf} zv&aR8uo{V3m9|zq6i^UIO}SrgNC9pOOywVEqvC~A9{0%9H&eIBxPcv8!-kWZlFqipTdMsSI+4ipLoa?qU&cg z_K$5xgb&9<_#RcK21>u2oze2()ml=Zd~(l&Fn+ewHio%daK!I!;2rZ~%7Q-;uS0@z zypk5RT3BBQQZpPsOd8@?+2$}K*CBn^tc^7kBCia#XmoZSC!TmBWRf*3wtaL!Yo}h9 z;#9#D*_FyzILQbZhw*^gW%FqkHbHLj5Pt-`7~%>OXqH*{e0JM}y{~`WC)<+RNm%j0 zglW2syT6U&eu)y}OZ2{;FV62AUoy$h-oYknP0E;N1rvV5n6S1v*x|@pRJb2`)tSfk3~?9Q-ZTe36n{)=!0fwYE0Kw9$gMP*-;V4 zrGAqOTS#}FVa(5E5^XYE^M*sBr{lB(nm?R0;ApW)pTxv%`4H&{d;YOjFS;*sRt)Qp zo*<5CjI$_@J}4Bbr`%Q;-}t0qd8qfP%Rxxe8@`8l%kVpI=v9)kw(E=XnV1BR;z^N2 zhE|qPhvq~<=I23LCQ1>%4O8NB6({G{M98Ei=4k1urQdlL5kDYFOnr7)cBZ28Nx_rI zuR*%XAC$Jp?fa${Khy{g%TSsl^do1%$5QEz!1~QxYY0(1-P0w@p+AtUHg$C+l!h?e zGroivnQPruNaNW36;%MiAQ8vwzcMxbcNVDs{ndqP>`b?Y39Ze>4DW z^;gP}PC^naY@Pcw_(<*Z)<0en;KFi$q6D7hX15xapFXhTCO2y?PPX|NPyJe}MDipV?+bqJ z%35tz4$^<0bkaW`3RDSdU!g=vdr}JK@BPwrR3wR`0=GW2`M^F0{tIFw2?*Xh0JgDN z)yRBxZgyOZKIaF!Q7MuCIUPs1KoA=-JG`l<&Zec3lzy(V0SKlwmW_)4xp!qu;QX_5 zi8~#hhcZd_&h%ca|61ux`%wR$DtT0$aMAWZ-T9X1`Kb~}>t3AAuJAwI@+D8Z&QOa% z^?#u7yHn0L=j7Y}!;YEh{=0Vm|GQcNTOZ()JE5d6umX@yot)?|WfGS&uSNlvn{7kXP+*Xt*VzY`^)SlJH*a+x>2oWb zwgOhN9i#i}v`V@Ao7MXx$y~RXl~J+PrN0iM!_ac26BT5UxC7`xxZqmC+CHQpZ-!}( z^8;zW8-CKFZr2EiV5R_mkf{OeW*Mn(-w(PT&W^h7Pd!>n%KLJu!a_C>Lv4mXWOR3+ zV*#lH6Eu9-Uv6oT0JxzY=9ui!umC9XAUvo75TLsKS;%exa!JAS&e(m)Nx9`D$IbQ} zMh@x~a$I|gG`^%Gom@E2SUhQHX(6EdrbYXfdAr-VB0fHI}CR; z)27N=CXvV2(tyj6t~;3%qd^)N2T0DX@h0y?(20)Kl|EH$#k0pyF7G!A8oH+3U9H!C zNg7%Gl2R*BVqLk#17P)ulW|?T-e<)8j$HS@>kU|A#TTdtX{54}d;p%H6IL0v15^48cBWV8rZ<2#pH^G^s+#RZn$1 zS7!8ABCB15*PKa&uD}W4*(~Jg9|UNn9VMvbT)^X4|2%~8wliEUy1ob?V4}9cXNJwL z6@LE2w(+vp<-jy>um3@~D#4B)YB4i2lkR_aol)((%{@!y1B}HHAP)4%D3q*=VCP!O zHZ`08qct7LBG1IrqB!z=r+{%UTdEw-5QT=-ca)3{I|O9bh7cDYNL?C4DqiYXTq=L~ zD?c|&WP{XkV-JBKYi=|?>@np9;5oefO#g>lhZ?XHT)$N@9Z0ALl&0rza0acXOEap0 z;%63xhvI6RDvp;znh{BRsR6eJ958!Yg&iPMF-IGJPs%VC0vh38{bbRzLFsdV766gi z^3#bSi$OS3r4P`##54h8W`sh|mbfZv7l5gCMVEk+2M3_eC}^L3f+8lo?nIYeU)}Kh zx6tg8oS+?+qevD?!v%|w?IiYG{|pPYW2tN{h+4*lr^5Q*dVFlbFdS8YSCi2WxBKE zL3z=U%$lL2X6G_3dofBkMN{69=DS%S^w905slq%}*rNigF6R@RJW0m2arWAHH^sVKzBZ;~_#2S& zBT=NOTPdSpL{42l*TZQS+^9-hxd+=T!{k}<1TnCeie*!%$J5V4WdVbm=KkE zt{)Jt-$sJ;=Q!7EW>`J=8sm@ONsD_b5D(0Bh+DPH<+VEEtr{3w+lYS_MvB9Wek#*> zb(yV+%4uNV5PnG6C0rWWBy3}%O|jPp;nIq0n?t)x-LF;~>UpzlQr8wdcU%w0uWiV+-O zxdjG9Z7k^Ai#%-o82US&UTU`9%eD-13pg`BQZ6`vOPhR~Uiu!87z$67RJvC`ekle- zJLxP6$fKJ^(S@9NM6!$XlZ6jw&)UVq`{ayL07Y-LUej4L%4XUcW(}6et)B}J2?vO- zOKfe&1-Ysp;B@viEplqapbXlaO|w`c?y9GbQ%Y$B)jSy9eXH-R=VvLQMdl;xo+SFNFR9uK&x$(w(vI+SN-0tilZ z^dcK!U8ucmhChqFJK;>;vT%Z07UItLJ|0-9KZThyUDbB>eoh(N_0 zfH6M|hqaDoig2ef=C3e+!^ER9es;9td#;h?fM+q3?B?kxf!K_fBZz+#9!(cOwsQC}8fE(%lhV6iZ(_jsd3DStS?#cM9HA2TPViu~&hD zo9|UZY|bCD?h_LSW46dxkX4J8H{3=O_A>ZUh2;2_X*l1R2;6L;tt+)emC0d;%0Oht zXD-1;4n4-!!vpb*s~dE9=tQWYfkhhTv+pq8?(4x89pthI#@o==m1*4pqi2@OxC|Jn z7VaR02#sXu3~8%7-T1j`%YI(`6|ut~hkzSB?qF-PidV(*da4Y`JMY#qjGx+6*3II8 zKNJM)P#4@U`rkkxaOw+3qBrt}RQ$a*Su<%9>(vL#a%+fLZLZbB2IptbwtZuH4kfjq^s>aAIB9Bj){lM>S0YbRSZqcKJRKl zrP^~f4FqEp#fPW92u{?J)Snf=QA3O6z;+7J`m`Ph8O%-FDMj?JKGa*A&%Dlp?%Bm^3y9!`>fE!2cd=fhVXm zCXn0I5s85pP=#soJXf3#gPQq6WQH<>S1%1hUo2ei5-kL!=L4|o0DC;8qxH7y(-n$I zpO8XM86M|3+g!iwQMAdMu>>;(>=00-Ym9=d01lU!(S@<_t8l8kBV~q9#C25dN6~W| zo-ozB=MR1ifegCMLXBDRJ1UK7VcEA6?8ZFW9Q^MEs08qqn}{~T*64}{+??-a8jv8BfZ#JsII_j1nZ80^JIENIZhpxH_$s2Y1VN%+48F{RPPRqu6a6IQ zPvyGc?=)Lt8r{%BWwS%B8F*P2T!)fGY517N;~5{T-&{f*0v2V3;I=r|B-hNjLSiS^ zX&twH1h0jtHaCaTeP0<$GtLn_MOg28S#Z%_lVYTE@}aYB2j%Z9*|%foYF@>Yej>QC zWa`W_3xwiPZPC64(r1D%FI1-9nMn)Mq*27J&zVXoN@~%VD_LU?6uR^!6q14z)*i9& zgq{#ft5Uha4c*InPFB;{@FHDbNNkg3^#@@xSSg>VgkUf-p8{F@iZ>lnaH8m^ps%nw zQWcYYtjo1`L@elpOV5nKm|N2b;Tgl?L05WI^y^lN67RWEM2>u3HQlMZP89Ui1rfv*@S3e>17SYLZoKJ9?lvc$ zedEQl&xxn0H6uX&m?}+k?ILWZgKC>FFy!gJx%fRRXh-jA1U6>8k7tLVqnM(EyQ80~NU$VP?-f*(9sMzR)@89g}vC&GGKmcBsi_Z>m-v8#-@XqS1CwXwdejtBg zfc(-QvKmbiJpZYylgHtFIP}tDc-KU4_gCelqiu_1c*%2xz@_JS%p>LTGvtPvOrOw9 zem!L)t=|sm(#)kqz-m`zrwEyVMAz^Sq_QVoKi_Xh*TYUZ1>}bf@8XUEQwXTS*SX?? zpAMWq$Kt*#$|I;An6OxbJfcSzL2HV9oom#~0S?6cE#LBs=_Rv?>S_o$RS;?5fOv>a z@P_U!y*9VvW_54nrlhn;Lcejw8z_BPL{!a$&zj4RrwC$GjVB_NeA2Dw>9j1ygBv6e ziQ4N!ycIu|a6+M^a7uLg0t|*G6|GRa;-`_kGG=uE!lmT)KmNn z%6YfA`8Wa>a4}w<1RJ{w(B!dw3$!~5IqebX^Yo|J7lNhRuxCd*idNAVecH{kjJwga zqE-^vEsQyx73ZTn^yDDwNq5z^;!t9S1-ZT(5Lm*Llg49?4|>6{gf8ltm0Shp!#z{Y zY{%ja6$%{W0Efa*&06#)WVI})&SzxsY{t4S&071Z*ZMl^&Ug|D%oXN;#vSIjF4dv1 zDAqEDJ}3PWrNXSA>Yh^X1ng?MdV{_|*J`!I*Nb6C2_fhvb?V92sF;YGtV$NuQpis8 z7`DFg^7pdv%02DMN|-U@-?H!_jgp024^x?oV@=?c{C5Ti zf^ZX>E@gkJyT1I*I|7jW{x|+#{Ov?Z#TpzVO`hNvA9`jvDJ3-&NKA57wUu~~=|Sksk)38_ zdBg!jI8;xWQt*0+pCts*pKbZH%_&(?7jzf*$duVuq#c&u=6^fIWBnw~ zAm5899&F|;9uV+kA!^rmVCD-VwB8(Lmy3LlQ9i`m%HJu0i6Iis6V9*n7(+(Mmx~A^ zifPx-gve}SD*V-p6k70uZ`}QJ;G&jkMU*YbdKsfwc%CRlx42 z?PA+~pgg4&F1~BD6>>L|fm)G~tLGisM@;lyNF4dc6&tT^j@kLGkaoCyII|L(wyM>Z z)sevK$UaRR>H^d@!1wZP(rZ$Yvd&ZNSuHQsM2wQqp+P2pYduDgVopBd2(Z7czt`fO z$X)vll3r(4ic~lx?8f6`*}%~X4NC22)s0;n>{?SaHa&`pBSd?lph2XqtHDnsZ^fGX z_<%0=3C2?*MLY%KOX6L>uS)4AsaL@nzZcE;QS~Bq-wvX&WXcm_@STn|N7+lVs1Nmj zTU9R(StA3v+Eb|yjF~z{2X!^dL&d+;bI213nnnbE0Ff!lP-_)wS*}YRV%2nFnHxp@ z8p8|@A^ynD+2st}PmJ>+b)A6ZkKOF z`#qxQw23kaqomLxaJs1<(8dWpJ%>L3qoYHT2#TD2S;wHt2$G@FeYBNu!oh@|@KU5( zTDW;t_H~FdNl4DTA+~_2HUXX+J0H`x=g`-0_#zeY1al3#0izAO(>enQNOewdT=rz_ zrE)OQt8TfB(2&F4n-ETmGu-<&{mCEJHt`>53ZD9m$koNS52q1ICwmru;VNXqVNe}P zAME+U+&Z5sRIjXjr9yFNhs?2_U#hrc{gnV-)7mk!O-Cv-4+SZd5SfPI{uMjkx)Q}9 zSQJXM=*d7SpZ*$!wM?FiEck^OMSnK^T$|#ZO;mXP5BPtbU>fUcIdKzBWX)B_K;DuE zoMHLrvm-(MUd(2zt5u3>0%~+&^-?$hCUdp_D|AHSB#d*$3mq&=21K#i%kk#8L={c= zJ01F-*^vU*2-|!Gv<$5W))saeg8fB@a0rF+)Q>REOL48c6v^o?0d%X&8_6l|Q{K!9 z1IqV^XGosU)Q2<3CQoTFeot4j>XwJNDi+4kI&doV=6W*$A2_8qOT2eFl=Vv*(*i8l zM89MIc`$P}Y>5DwRb2``L)$duK7=uSr4}pha)&t&0AC3Lo5>q+FGs6gR4+9zwIj(1 z?--Fj>%vc*iV9T#b1JnQ&lQ|i%mUpN=4!3#FT1!+t_Q$W-}8m;`TpapHZM&e9xJ=g zaeOPCNRm*P^m4V_W2U3HnOYfn_q+e>j^|_zi*X2$TGjPT4NYKSevE%P)|O0LnT4=S zDZT_njy=-?B@uIrPyu{j!MgXdpuHeYS!G3KlHnz^kYSM1=`fUt8$)`vh+wjaV6ZilP^rsxP2dxcC0u zQh&N7RKFQ>u_wvA5aR15XtX@ipn;`zmk*inv})7l9AX`ep7RAV%9bmmAeEO$H+h(< z;1E3K2KexV*Nn-sVMO$_wRD*TcDpuhwG4x;_sU6OqyoIc_YPKF!4Izj2!5QH)-P%z z#g1WE)Nw<{iqXIZ!txwhwu5oCtbtzwFi&wWH%6V@u0`EIP}l+VRJ4zw*DP^7Pn-AU z&vChX&Dp3y)4W3kW^7xKilo3k`iF~q{B2dF^sz^#a09yc(9heSovePM-#B3lxVGHe zh*Ece-Q_>-bp%yalk5oN5`dRSU*jg|}8 z%DFzPq!~M2W87?AZm#xDEK|U6#sZ`azrs}=rH-3u(j;J~Z^81wPtPA2hq!9YLn?U^ zE|GS)0o3Qfb}Bv&wjw3-ix$m`2BJdm)4a{{K-a}jCHV>H4Qf+mz>Jv8h`NeEIJCDs zx?WVgVPZpWLT*{!2xZHWTw!G%LNz?jUr`KWBefT2rrynnL}2{;sHCuO?0%&MoXq+!mAHzm&}{^=2u=qCGtwkz8!R`+ zF$7l3&z?*X=`V@w6Z()@>6VIr_7HsLSV(PNMOeBzRexGbyajgSvau;5h7oD~8yMDNoK~6m;+g-Y9K6V+IbvQ2rPyBl(2uz-=U;j^>8;XY@0;-W!}Qqlrqg`g@k)HQ zcCHj%W!B_1pk2!Tm8}%0ox(ks(OT#3&O$n=taoCpgkxjZ1)VsrL9esl3rTX`Plj~o zsS0a~u(~`Dh+1op%P?2cprbFQp%W|_yvW%*K2;F<=snTna(N{F;n+aML?=-GFMxBaW zRx3T}v@A0%g~K)<)|I-e)MSVjxjxdZzvc|x`*$k_MByv`RVIy4r*LNJgY^`b(wG?R z=#i--2i(v4>08dZHCB5TqTd7sQl=RglkGj1e!6%UG@^>i$ung2H$mp5|G+PvNh~9H ztBU$Wl*V@rUQuOAib(4mzqfr?vDjgbNQ(NbJnZ;7%W{f-p6F=a>OhyViBdOH;}sLm zA;xBOOqtgekJ&XCJxNdvVr=y8a~ygx5exo5ZON@-Pm8GM!kAGioO^8OBGGM4L913P2Kr}yN^aC^};G?g;3T5xS6^c)4QJ|0xlNH ztpx-MTfeL=tbpZp1#i?It9MpqdZI&CpOME5Gd(;pu@%$HmL_}IP>zkexxdkdiZ-%T zwSli8z}_c%Vmxq%UC8pJ*L*S9=Hp2y=e7`sbQ->z;!&)?@KETjK{i5Wy42~?VixPZ zwQESes0P+KV?N!%^nypbN~F|b42`;Ea7mVnMcC##GjqPnX?ta`8NNnaN=A<^oGIk^Lrs*1r1IL z8+FK9rsJwHbiPJ?Ir1s`!wv{d=yt+;=$yH*hBd!M?>j!anpr$n!l3djipX(qqo?>Z z_I(Dp)dvaJOD*A^i|S0RN=-IYuW4^H6<2i#OcQO{y%115 z(DIQlsvtzz(rl+ofWsF>=TtpDM_Dk3TFL#}w1u}Tc~tRwEJ5Ojc}xjQHy)n%7MR^m z7Red2j8N8nS|NO=3@ZN&WvZS3)ySELL%IHc{G3)MM1#)QG9gN38Ae$%A&x=P$-YdZ zK}p$n^Cinjsp!ZMvNYMnSjHZOFcU+`zGvU!YaR5vAI|Sw*YCgIU#<(!!*k!yeD3%C zx!>;>d)*VhTmjH?)-7d`P*)|P%^C5lgl3 zJIV7Rb}J$BDO^qF!Ao%&!q3lwXG3)#?bRx#juz)ik>Cu>5Ut*yK#RYarSUi{(pt zir-#`ZBr$%deUE-4>}8p|e*xrZR^^9*5!#1yu{D+M5!_LISVRTWTzwUisl0OaEJAZhXoA zgM4Ze%g0CXAdc+AU++9GuH4#@gl`b~wJs|dl-Np5>TV<&Q4fV58eJQ`GikBLgnDUsqASvxVra zuAE*0tbjjxB9eSdeQ(xtWD4`0`8E>xIUqGRdUQPrP)g+VGye9y3&G~iqfQp8lwa8t5Om6qSYo%DO<_&L}Yze`mzgF9vYc~ldVnz(_ zUrwu#4%PjKDU{axp1Q;j+myIzMy0`!y+`-`4jyJcZSD=QdwdvDbnyNn1(x*x4Z6)g z8PtT&N6rAL%DXQ#nXt5hOcukT<3v)4Cy*(GN1W*QP3HZ%dEnnur#eJPhdBzD8XjJ~ zsC;i0a3b)2t22Bm_wIEZSm@;0xe#hrxF5u`&@2}8tlh*~Jm!zV6#otiCdo~7PYf4_ zI4aEqt&od(d1LCf%6C{18lh4$Ys6T;G$Zg97{mT4t|YJY1*H@{eNE(9t;5hwUt%XP9NC`DvED786y<+e6Ha7Ev6woS*;JImn;m%?w0Dd zL)=@d9s9KB9>V?5Vg9+C+!K2`G8=;RBqpo*PjHNs*008ruGFS@44g~R1Is;M@3~&V zuu>SaT*_1f!6J(9eBs)AY!BY4vhASU3rn4*8o+XtCT5%;B!5fyKxD;pK}bcXU#b0} z8gxH$3)0lWFWgpRBQYnwY64w3F;EK2aeV%-^vTw`*49*7#X_s|A&zmX5AV5zCJuwXSJm zATg%|1abh4qNqToOnXw+LM%*_sx?Jqz{uSWX8%b9wgo%t74+FvGL8Fi^#QmjIITa8;pt%;O!oOLmt zz$M?;{h47GF}>;)=$hIBs*-Q-%{_HF0dNcoN<#ru->K9H`pdH>^oUbIjtd~~_3+uu z1UM*_?W7n`Lm(_!=B)&RLPAAd$(sNCuyUtrdNVwPl>CCwI~++VEM#6xPY^tSpm7{x zd`4Vle0u0VM!HPF`O)NB-Y_Yyqf#dL8LS%1=7TSF(s9Z*MdjL9_wN*(ee#m&r@lX8 z4IWA8QXjbNoJZZgO{)R3s933fIw;ZLqd2}Xyos#GR&>T>c5QLUQ6P==%%ks-L`)e4 zVWZx7I@!Mk6$WG-OCRWH8Pk-)g9xY;T z3f`=5#!PSVvVJ#xD69eHOj~y$V^ZFCH+sCBS6_Bc{#}TqR{50M41;4obteS^k6D3m z0Rh0pyjcC^?gm3Ofedxl9G+s7;_}R)7K9&j_Iy%gY~#Ywj1rJ|nZ!Vj6~`6ep6b1S zF$v;Xy>@Lxc&XsZkE?II>9KpniFJ3UlI8H><{)!4R}R^pSj}=mv!A`WmfgUXAtCtS z(TP$Jfterwf3 zrVv{e!9?k{nJCxQ1BGT9;@dVHBbR6LGC2?m{jds(Suh`|P%n5xSfzJ7Iuk(%Km7ea zQuzqEF8^om8#x4d_?i?6U$t>T_wF*~6;6^?y1CwI9X@!-|pwj#7H{vX2U@gZU%vrj5JwBmgnMz5pp+oL(d-ouc}0(|fo z5bQ%kWK>qGpi2= z{Qx&85pt@#jH&6VzQ7ibnb0qDSB}TtCs{2&mTfpF4c7aYhEoPYg5lRH!rD8h4SJU7 zmw3ARoe5#K2BN>V)*hhUEi&KazL8E)#mLwUHzdKANo`fPFI^a1MV`L)MK0a#$E5!~ z{ICVavPA#Bvp(3XToUMHTSoQ&4wjS>O=2%I~MM)Oqwv>Q} zn@`kR2&4N^90R#oatE=k-bXe~k*z^@8Z5`eAA}^4%M!zHV&wg6{D>W7@|Z#ODYw25 z=MAT^KNf{suCyC)F5esa`C6edk>gztwwhKKUsp06X2I2VRj^(4&J!oubo#?({#A;+ ztMP^+f4<@+KWvn@tiGsJ%AI;1oNy3OSzMSzm|p0KWL3-!K<_hT)gT8eN_nqCq(IJ8 z330wfeas66)NMUrll|GF%O;Rw5eS7I6ol9lzIbcQFO@QiJvY8EDA7EmYgtO5e7GMc2FVCVGAW1Gx`7zdk zQ1y9uhC3A`YF+y_#+7D!Z0MKyL29v;_W+U%8Cy}o@n2Mdjnpd%$e(4+KI|`Tjfvht zx3TqDC}X=b9|t-)-C=9Vfslo{@A!GM_aD{}LMY%=rWZDeB$|VaI;!)XQJE8|bo3{k zh7^-8Yhx!q{_$ov5PbT`z@quJ_wrYxnX{{Z`|0nz6){Q4`ko)nT20nwW6A+}X3lK( zdYlS8j}RcBluCtil6!PX_)l&w)-&Sv$JfBMiKc(kjGyJXCWKob8Te?Nu0y_Jl`!%# zqOW)(q754|)-f8`0;)pW_S7{En2B!q7`wNGxOYFzb>d|0MpzQ}=Xde$R>1Dqb<7tE zz6UH5rNCEfDnOO>%>x;vtmOAivsf~lh#}p`R7SFsbk1b=+|9|O8xglmSS}*~6ofl0 zFcf%%Q^5Z49vJelAhkwbe}<5FpHK$Jv}CA}^gFh?%4+UNj*nQkU~)^kEG9rh3w^Jk zx|A%73-JX!hxytC(v2x005i!7M?5xrb0=f= z7T89_U9?VMjSu+ldj6%jR+8ra0rjy~3GWgK*_9j`1J7I-Ru^Ut0n=~>knDf?{hLm> zDYL(w5FGP&4V}`>T!0~C_L(5P&o7Df+e(UA-vk(mjF~L%(|<9bJA4sZ>Bc{s z-q8gm1rp7fqWiRv`&wV4#0sLpU3Ou}<^86AFBJR=az)OKw$iJFSBwfq;LS0`-Wc^T5!pO}hpC z#>#9eH-RS9(97P9)q*0Mjyl$f7)QEf$nAbDBidti_nRIW#SnlQpp-$7%=O z)u#MHSWl96`K8sFTl2@%epRxxWGm{9txqSzZL!NqM!&jgM%VTlx@V0voany}7L?x@ V;to;F1w`N9^t27N@-cQH{{!lIX^H>< literal 0 HcmV?d00001 diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/compare_metrics.png b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/compare_metrics.png new file mode 100644 index 0000000000000000000000000000000000000000..6567cc100ee9da1136f1f263e662d8e79511f752 GIT binary patch literal 53695 zcmd3OcT`i+*CnD6MFkNMQ7JY`kRnB-1&}7403oyhibyZgTL8gEuK@vRp@mRF?+~%k zM0#&ZCzQ}aNhtI1H{Y7IX3f9f%&-=Vm%@AR-goak`|Pv#4b{<7xk7uBmWqn%is~~( zT`H<`P%0|w$qVO!cjg)nA;6zAZn`Q@sY-s|UIt#AvzFJCr=lv0zI6B;1iZfJ^32GM zii+;r>F=3VXUt0~s$P9nMR|R1v(-tOOznA&)=eCW|Ecm(s&1PakI`CXm0nIC`+Zhh zyL-7!qC8n?i8(z6st=49+Ra8O$Hc|g>Em^>ct7I7kl)HM`0xBv_>R_J#yv6X}bSl^HL&YYJ0~qM;YDM>0y~j#$*}n_UnUCI7G%B@t zBeVY}WT@W5UTQ8zVKp;^uKfI!n!n{YgV@dkyhOt8|{2txiGw$j);A+6oR|Ek$e;G5@^32+Kh3CL=e0unbi@niRruWerSXU8|XPx+U&Zx&2$Jn;p3WmD>1nw%D-jg#kwA4j-bv zQfgCnr$-^=LCghE-ucTnED8;Z%g3D>x~#l8k>qt-Z{X%9y-KY*6W0a*UDKjKH#aO? z9z*)c?zj3}%zo%{Ue4~{4;*7(-kjCE6H`=~*rv-BR<1MUzpZIf@hWoVtWp<`S^^J) zhUWCIWRW#OA1c)ZcyY}Xp}K(d*>lWx7ief+ig`=8 z*OX7|?YfTwWrXWrC$x9o`nu9=J#bDo+gHYbLplrM>StG$-o9kHBrS6WKX;*d%%#TadS-S_52yZI$P~F@w#;strR}- z)GJ4g@xY6lsq5B3{(o^GNsaWUQL|tw3~G{mP~c{EfNN@t}~R0 zYW0goFLZ@H1VP_9_R{r7# z?aEMbDsrb+MW9fD^^#t`RuvzvQFySud1$i9&nsym`&pva#2~aFMF5xpNU3vjOhGcY{0P?r+3z6qT$FLDtUBq`+~m)81DNN7SeV=F&gf7HVmvbX%Km95Qqf*d zGr^*MF=|Wy{oN@eGgMcfXC)SN*N7JV(7bsfjVfI0W3~ zxp46&t|-=LdjX{GkneCY?MC3ie3Fw+kK&oDAhm=m*%}d>v#sO6BX|Kk3w&o3RQ7IJ zCDKa6#xLb~5N=*4u6qUqa`xYSpK_;wHFCjkrMSjnt}RwTC`;~LHOKgSwJygz&D4dV ztY!;QTww8!a2BsObZd)1U(f8q-f%{CV!G*MWs3LU<1ntDOTF0$&-c6VGZ)wLsjovs z?rGqRh#A2ugsX<~`8JhRqxa-xt)-P&Gm$Bm`r>3lr!vssXXzd>G)`Oyt_&E5{rQQ?7waUCH;o`kqB1Z zGJBsdr7xN4MKjB7E|wdiayi4aUZC#8T}boi`<5ASL>5^2p5d}WRANEQjS2GxT^nz3 zYL9z>h*4r2zy^kAM@e>c@pS}4MXrt7}|Xd##i!ebbY;n=p;Vj5@K?a zz^i8}C!g}zk&tq*^m0OKwe*`Qr+k-Bf(L_oGM{AZ-jVbEFgA~C%v@X3FyTQKRRT5Z zGF0dp#h}Er&O*bEmS@6yHgeq~e0MMSl7LHBVCAuQ6H}pTKX3l$xUDhtdcIlG!lwM? z>FmvkIyc16h;n@r|Oa*BgJmm=UUv z&e&Pr)U2oPXxGciuYYCx06u$%-#E~r;$GUJ{b%Z@kOui=dSD4;>o)b8^zZ~GT!Teq zO8NI}3k?UF$H!IRc%^=(hdr58aVhv`jdI#WBg5=@TJA4u3WLG}=?Bj7CZ+>JL>wi5bozE{y-oC_E%sjIiBLRDJQDyPQ3zPps)k;u@MnoKcM8Px_o zX8O_Zmov`7x0G|4(JG&(jfpb#lhH~F!h?RLN1KT;ik|Z0FN|^{)ej1Wk)@z=O0$C; zmrw7_VmP-MJ}8TELZjuQ2T3ZP?YdW$j{OUJ>-NW>2f6=YUKgFGNdkee_tL@cFQsyaSA< z2Ex)~wY>?qZaM`Z=1oZ3_Ck1cvW~)4h(->c5#>)4scOZNfO2{-GL$o9BS>KT z?3!bvP2{mb-)c4W-~8d!W0_|E3eNdJU7ksTG@yrg71TYY)6Zh%v~|94CcSn`r@%wsHOBREgl=MNAk;xe>sg)_0&@ikqjk`!%F34NQ62khN4$8%{UgZ ztls9ZJo~!JlCn<=#B53fo8gWi(`Kli$_;>p0`FeGEpqjcoPQz6g&tDZSltK4sQHQBXr zY&9?5Ws=VD3?s=6%^9QhlleyN;Z`T&aR@c9Bouyf+dADRZ(Ozur)A98Zl{wezpT)a zS4$c)nY78t)$v#c3TyrtN};Zb^1UH-gSVXtJB6R1Er;>KUIe96y)D{arvlmd@J zH$zbB&VvDpN1rM;NUuL}$%9_{KBkLrkwKa@R|?Ur$@b9=y}FVJf-` z;uUDx(Y+=7@?+;?=pUsc};`@0Un}&|S^Z+s%6VAQyK( z%d{Q73WenBDT_eBc4He;&K<_`d>YDSM1taWjF~U@e!$T8Ovj1yH#=VyLxg%IvGU{c zlxa?8BxO~0XV!GVRO?u-kS^Z)?1732&Th+MB9im+>{E?5-ise+XtEgew}cp~oNdK6$&ww69zG z3V_{>;-{Uc6U1*?+e0eGx*hQhP(S5o7ak`+FcF5vR1DZ+Q*f9H6RR9fB;kS~ud(BJ zb>sl|zXgIbu&?Vm`n--JBJ5=@zbCMWDazAHb%PGl&WO z{~>KUy}|#N$z1SLRUn~|up6&-thVaBE>6{kR0_ZDq44+XcP(I;Y*oy6CY`ST5;fE> zmi+VW2C>zx?0)Ag%gsy9L7C)6OP?!jT>kP~AFX(inipuw!3YrZ9q>oH6}q`q6%U-eli&E#HL5RzEmD3JsyX{K; zPs00u%5Gif*QE(}SmdT+;gWZ69nY&#b{1i~ovT->U4Fb8ujl_1qZ!8ZxX?K3)8I7z zZCX$1=d@|GzYwTx90H`86d=$$5(WyYy*FQYB{i((NV%`oEv9?6-T7b`ew}|cK~L*q z9@W8c+yixoA7RYZo@>VAm9~09kNwtwNor8mg4J?+p{oMu$-1+{XJ~F)6dPBS15-!p(KdSf|mT}q6 z$269%t!B*Oqx{xuszM;2#9(e9OKD_t?~(UhEo-f(l|zAHMbBdIw%i`I1Tg%%H&*R^ z6i(#TuTs8}+F%RXtp^}&eU*W-;uk-{45}TBSXO!TF*-sjyE-VNmSg}Yj+a|?njRUD zbb@(DSOH$Zc6>gB=M)Svcg#jf{U!_*uTb-JCJENG$8$Gyy!t@LRwN&CzGZ?6J(4V+ zFr{#PBs@EbrrR^BGiv(Vy1hBS_xZUi^s#8i~md;Q=0DRT{96Rs)v+S?cz~XvCHe01_kkkx_QYA3bdut0uoaPX!>^ay`m? zVmH9!M;e5OSj)9!aY;+C$rTu9VJ5weAKV&Tq@o=E4VlPny_s+Oe2is{v9yA>P?wd5 z9MHuvuf`-5T#Odhu^TE%?sHIS`FXIt z*wFkoVE-@8|MtbL_v!+5-;t01J_izyYKs=?Q|e;U(q}2_!$czp)tn?5utdsm?^CS? zP`pKlr(KN8a&A)na*Fo0b&`bJ!bqGg2_e53pXh%;d&te;2yy5(t-dB(| zMf#Q^)Nc()lRF5*IF5)=?S9nz&oDlCNu>%0av(~QzykzE%0y<~QwfhAkf#{o>;rB# z1TI3x_lMcg$^_WJ86RixqUI|Ag(CY!yBHG@lsWp_o8T8W7x}nOjw!e_kRfS={k+p|PvPda;iwcG%U?w|g9RElz9vAxJ6>%u zcbtp^N?yL&BdXxfp7`}RD|*DmYePm!O?Nu3NfH~kC#0J-LdBinSwxgT-f~sAvvD@3 zd39#Gg4?{351)gUsz%ozqAWc3)zRjUoi0pG583JQk2gng8*avn^(MD;RK z7~)#aXk?Xa>Y=M-?_opWRQ&{i8yDrUU0(4DsD3gWq|EGydOga|J%po8EV;k@R-oee za9f*lv^R}N7N{5YU#l9f-y%iJ9=pI>6W-+i`EttDl(54AT;65U!#k^x)<))%mVE8| z_Z`cG(%VX#l!f+!2&T30U0$hjsJrzYWHSeHCsry8noEe1S|9mq$D`X%^ct(QeGEXu zggwGBg;-NF<=Y9AxRZ*TrwqfOY%YHsppWKU}>oX`{5_|G}RErFS%=lL?5=TEjp zE%hu5!aa7<<=Yj{?+hv<3te^NtbRalzX7aM-Hd0lf;EheXyYiCZ)?vtrZf&>p!aUX zKSHTa2T&Z}Bh3hg=x#B=y^35{0~T-B@=dPTdkeX4%_jF;U6Ma6FK=)_rOV||^hyfcpD}B1I&hjlyY(i^xO-dlp8M0k`L>gC0eRy$ z7IU^!lC;HxW^CEJZYB0puIDv8;jzhd2u3T-?hVH=OOyp5Kp8^MpJKOop;)hs0 zm=x(n_8EPb<`i@5aUG_UkJP)yfIRG(my&q)>gwS$k`pr8P2HF~DB$o_55H3-8%AHufaVm0I+PWPo-wD_8HZ+ zY>54!>M0?dFh8@I{2@Qx7u)$2+M~asy@CwmF7_GZSNi+)-5LPvuPlg)Yi(b?_4wE1 zOC{0+nPG|^kZC6P1_wscDtZzI+BM;9gxP3W&weFVZ{H06$2LI3HJ#MDc%g%KH{-+c z`=m~=)stmqp(OJFpBzcs*st$ae_tvb6Ra4{XMMJLhUtF}rF@Da^xZhky*@O@CrMJ@ zhYV({|5EtTlpnEW7km zC7L3RTFhDeS$gT7F5KT91$vKJB^1}>9r;;Q-sz<7f6ls9W&Uc>p`uprrV9R^!fK;L zlk@Yx z0Y-@*e>rZu7hM}txGT?m2`2P{5B0o0EL*L)IxEC$wfRG;!}k(B3ffh`GJe3+2b^QGq8xAj>Hr<1-n4pR9dyu+RS{HbJyMU9+-IT zZFhg$F*UkPJ7?Quj3=CdkS0#j^`GzJ>6_R=;xC4Cr?bJtzYv-oc=s$xfwK&MG{R*} zjd1R%EwB-gQw zK3*nnir8yiYuk9{nnwg14tL$cy}=BJ6@X?k?BAIjJ_O^LO8cKQ z(fhp7>LdBsm?Mp{AD#i#0VJX+P(dM@Z`(nT5J_?JpAdS}BgIAToB6^j@~@i&TR8$7 zO|+)*0H@mEL3ZWTU@ay?(!FOwOCEY3ZnjSD3pVe672B>S6ABTdJpjl`Ai5gwOcDHw zJeCgC7j4N=Kcoa82XpZT1-cKs6Oh0#^)B zf?4;<_1Tdq2we|4b9TbI#4;#=0wB}sac+>!(QD=DK(^=#H5@#n=66!|%Pj97Vb@)~ zx+7Sg^;&Ys^-8CBS$FiyGbv4zvUQ%{j;7^>8FxU(8!EaTv9c>`roo|c2 zUeZ>_^m6h|{wvZNK8@$aV%KkEf9QKz-i?r-xk@x(wh8yRg0K%-1Nst?x$*bes64*( zzkx2-VB4$|xl6DQmpBWAHE8C%;jE{`2=xWY(iJ%*Q2VNPRb##fd#K-{qohEyF*5$6 zpzf!Ub=rJ~Elf@n;~HZC;_*zgtIv+_b5T-ezJ(%6_JpVK6rcxcoXz)Aq#gA}KY}@6 z%YCZB!H=+1_N<7OKKo&a?${1Pwx1(S|&qZ!!2cBkmR-=9kTOR`6IR36T zEq|0J>AAYpJpLRSNI`#z41lMYa|?{YVVRb#2!NiA1^)<5Kex2qOVq|+K|*|SBoijs z5-HWdMS0c(Wr9=f%p>^?>={t7TZ-J|v1)4mGVamk-q31-1zPk_kyq-G9gmeay=vza$2I zoa>jlSTH?a3ivZ1(@Kt;Oq}qm#K03uMrYy$=X+q}$zhq{bnnKG5KN^bf(m91M#mec zU0Ra4hV&w_zUBS+M`@zIY4q8|6g<`j;kvwVeC3lDZJtYrn zR12*;VXq`Jw)qyBy}+T5GGlUX326{-icBwa{|qbD;vVd%?eU4Cen}AI*B}-{#8IQm zVl;i5s+8<-A=*5iXNAOY>T6}9(NQFFNx^F0o-v-+Q9UDa*JRO{_Dd@}_ltth!}dM? z%T3y4t=&5dgJtF`gPZEDl1*@O)4o(cp{eJffIfKTv&bPEBPhQTf}GxI#zN4u`5n}V zxBZ^5#3f{(b&f?W!EEe(dZ9*7ndZE$F+bjO^cYPOc@hBZ#QcNnghNRtO~sHW?U@-{!rMrDpvQ zrtkg~eHJI_+TV_$Yg_JxiNcQEJrv}hvps=9mgjgAgK0Z@ddRQ^Y#*9Y2o|AVrXx`p z_8!bG`Q}PLvw9k5!3*@!@Ji1`xG_!FhzB+}GfchqZgyCt)>RnmuwS-XdWkoL%xO#Xk&Le zFpy=b&pMkkG?+nBV$fNR1RX3u8lFH^YhdG;oZoV zpMOLDRxx)tYyNA?jiX&AhkB{GNyZ0)b(YIRtF?7#8A2z7Rlry-9yL>+|0CgSd1yj*E z_`7n;*tyYFar$1<`CbTkWVynaF;_B-{^o-%h?OWVK%2bAyPr85J;jGQlcJp*h+SDi zzYT~7J6pap&+dR}j}O`#oQHmjk|j5-jAVA3cb-;HkAC<0I_t|0?3r~gyXgyQb$c8R zK5FYHYO=-Vv?wN;s9aOY5;1~iRX92t7kfys!*!!5iY`%Ex z1#50*f@IJV4YqZx`EGJp!Uq}{ySrTPAT`^ns1{`o7xO)vkrV0Tk$Kh=1mnwrSiH*r zapF|Ayq$fL_%DX(-N<}x3*C^Pm#?M9mUAW2jbUR#aa1oAQL-Rkm=A1TU%&Sw z9EeaO@Wfe1Ss6**K1jhmT|ojQPh}q4x9;(0xCOB?fVu~J;vRH;tpaR{R(lVINkTAB zORU^yGCM*OH_IE)vB@3gpTO+RL_s$D%0TD?VP|xu_Fpxq!LlFb;YAo7i5;A|C_=r{ z3?AVuL;dl_Gl1xKWCpEEx;5l!twe0hkOuWzxmKcD@CD{)5l3u4Daeljm&na;DOEkH zMb-DpP$xn#7v@YIt`ET%B;_~Z*DRHe*0}H5wzowp*8O2~;X`q|4VMca&W<6R%$Zue{Ga8JoMea^bv#B_*d z;8!d$S2kAvdSx&_dCHDmJ0pvbMw`CmS2CR(96|eOsR0vVJD3Z@8%)CGzZsGdHa?S~ zZGrTalNO)#_H;;xJu(rC!yNXytp(OX zay@vS;N;X_;RR>T^2&CZF1~MMl8cq*P~3FSRpkwptX&C0U#I~Wsmbg0Je8G1UKCcR zp+m~cu-Uj$rrsG--`i_bZqqf9W58b)P9_yet@7_&B-gE91LeRZ8Nd zek`C;s4Lu8zWA8;LZ+p|{?kz0Is25QIqhi)tAK*L`C<7bm4!Guu>)uxY&En3970>P z!}8)3iL9c}@r4{C8(r5!6i4=ZCo@b~RUgD%XF%4#N(kq4-&_^j z?7i4bs3re$cP%V>gyfz9*^=1$omDfkWnaUDME&Eik|hgz@j_NF_vKmSv(J6nLFh|h ze<@g5^NP4w_gyU7yQ(#PzQj z8q9}N%biA%{gvFcm)$XD`WOn%-s63k?L%sS zxhQrSs6^df7Je_Pg%>>X6pId!^~H(GfI$i#MfwJr!DECan3qTY%D^Hn&(rQz?zREw zgt<~0V*m>>2_ZbxZrCbZkt<%#|#UgvsRxoR@WaKKBUO z!wp4*X9`h|om>yu2fqOVp(~xsNcHHF%)&nGIGhkxm%@u#ZKqR{7=pP}nZOc1>n?d% zX~XxgWrfAjO{x7b5K)_cHPzpa1@8f4Lr*aT%xjdprwkWhmL7hCpz+R)!j@krW){Vt zht&__-j>W^g3)1N;L0+1(@eXA(!JsibA9Z&_Y~S5>pV+M?F!K3Wz>FGxY;;(4lChj&rOI4*iyWa4H3#i(hfWH`Yl_epGg zkfG$AJz^Fa(QBRUa1X(+N!-H(g{QDC%M7=F(t3X_ckjtWr4qL>cv~3lbExVF@U#c3 zF-1LI2zeurrW*Nh2im%fA!N~ zEkX!U{UiNt@YQQXwWhok3cA~LlvR=)V@fAp)k}n0)db%m7m-F+p5*vCHI|!D08ovp zlVsRk>08lRP)8f78RZ^py5Alm0_c}95G70d*ycnG6B>tq!l2%&fI6F^-HG8fZ7fPp zQ=$N3n}AEUARlHOJx;@I%;A8SvSzlwh2|LsgmjOyVKP6IT4O9K_yxfR-b#&%#HKF! ztB&`I7Mojzb$F99Q0#1H6M8*&N39#0+w^yry|Xx*^4^&z1v5&Xw@szJy7D5T{JvuG zBziV&KjKc)u$?mYPh*ric$#z+uGU$2c}SS-RQL=2-U80U{!#b)gVmc19i79I`#oIwwOwnzhBuFlIFt7#+#4Twn7bcT|qx0 z`vbT*ZUcI`IW@1oJ0EV(EU1=5+;0LX2jC=zs_it^;)p=fAwX5Qcay}W5ib@QipA@m zt`DsY%UE?ViSzFw8Jea5UEO_VK7(If8K!@=>-VPo;vYLt-L_3_=f0(ypAk^+hp#+T zQx$Xy5^EGG{K#wZj|0m9Za1em}WqJM%rXr6MP1PR$Is zL$Lxtf%BRMh`9hGi$-?P`8vn*Ir0(S;m_wz0;LmIRH>>=3M7DuALVH zibwBGsczMF)I!c2qrjf@!JGi^!JIVj!AmaQgPdQH;Z=aLGbc;oLmv5X$4?cq>DDho zMt)Bf0D^da>LMU2rT@9g(=^?Dvi2jObr!j3*_ykD(6kSPdC}tWfZETc@6#j8iOL8F zN7ILPZVE3ktGaObf$b4{)dV1gOWR88aW=)16qo%7vV6UkP1ldOX0N1NPnT`5o%}of z+Q24|H1!~3jtk-a_nj(GH9_=!;0Rdi5bpI-pFRL?t!@!e;~h)Yc&rTNoDxLymHw?! zZ1HA*iblWnq|R-T3*j*cQw1cd=6ydSZ%x+DF=~pR0Mg2dZSNgF-~GQICYzVtG#~|T z(*Q#>n?qU|Ztw}9tknSBfzJW-O?3xO4Qr2UnnsQi099t)FxDp{@YrXs5um!AWaG4R z;^VYyCuc7*a-CEELYYr84_1j;?$2)!CIR(fukKm4y|R_IM5-%@!t z&xNRM@a$baan1Gcfw(2VfWnG(27cR{O-=qBnSQ5&%*31)@vfIVON-rDV)h$7;CB_M zl7fPxJU}LJ1@~MzI9o-0NgSc$P04pc*fVALtr)3<=E6K4HOF1}iHC<_7c@>g&s42~ zXGI16D}k4Rs0R1!fmwRg7Mhrm`JR8&!aqkKU<*M3-d%4*uXI`ACT5 z%XYp^k~Ztve0ZQ}?=^3R25xZK4c>plDw{X^Y}=Lk861~S1K7%bSseF}$x5k!gI(;A z?sc>Ja_vEjFMnG#3oMAictAq?zyDtuK{$T4i=|tkAf6GZm@fb-qB|(4bSZlwPM0P!?@u9C3R1g5xjNS&M7+W*;6~has-pt+-(pHC8RS$afC{#*6B{-_i>dT}Kc5#*40J<8RoKy{9AdNkGYosZO<<7>Y-!!Wkm|Ecf`du1H|8Y_J9sA0ZHW?P?;ol`^8 z4?yCQLHey_SSn1YoF`3Ff(KfI2PQlaTC}c~vwAUQgHu+wpMBE6sP(ajT_Kk=;mMNjeainDJu1(n-IAakXl8}~HDz%L?oeGk9(-{$ z=h<^oyzZ~bpN3w4?|26vK5b#L5?eC(q6Ra_)Aj5_KnLvcPJ)Hiv*)yRuJU|BOVt-4 z8iqhpuED_~9m0JG?+fV`)^~vteIzGskV6vupN-(zZFwe6M~Nc}&+Wti_zUDDT_1eZ zt+eU=!w{4fIeqU`#PO0&p2>pc8<7F99|YB>Rnd7HQ z8n@K!dotxadH`od1K<>J1Y8y^rmyUWiprcO>qi02y7Q^kVdPgGpaxkT(9^C0=HPkD ze?9!Agi|Gir$0}#A{s*G$@=ncVNlrkG4`~Q1W0@#r>=}Nk0GPS0f*iV8(%N20Vnh7 zA1*=I2Og1<_{|zesvXA{exB-hrU2c?Dj=!$3OqyQ(|KwX0R%WsJ+o=Nt-(CB@0UQ$ z05aPhGI3U-%>l~Tl&_ZrM*tH3y2QS2*$X8pK&81_I~!H;J56dFP!zs`6&h3l^1z>o zg62*@Zj3JYhPzyxhdKPBxGPe zz>hKM#>rjaZy-^;byM!=t6!gD-64C^&Q`vR!&oyco)z?sDdKvC8(Th+w_SA5h&p z9RPwU$xq`GmX0FQT8GD^)zN-2iLG(c$~6!QV6<%_Z;H7p&ZI0Rjm3>aWuT?jS{&2>e znd1ONN>zXme>Cd@JK>lhahxq|Y~>-Zz#7?7G4n$(m~TmaF7`W{)-ANbs}N1nzA@L`0A z<>WaoWfrduVW8jAk-J92msJNEmm@@eyHLgp;ex5sKIJli(3Y2&zpM1-7Kz3Uc)H`K z%KA(t)EgCcpYK+Gc;vE=1~W55clBn;yKDB-{@am$Y?}~{Q(>zHDJNb_vfT_xf=L0& zqjuBtLDH{)!d4{!fO-H6!x^splUXFM-;>ycIbL zALtAkJROInpKZUbcp@DMfBYes`_w;B0?{6uhbW*PCT&&6%pzxYFjOWsHRoGjfoIFe^p$#$1-es0+Df6_9OPdaS zGg!MIcq$|M5#>DM;e?I2n8Ai_>7K`%I#neZ6e=H zMZ^u+5AMX1f7RHYIFwg#7{`=1c^^uik4B^|tn_&viC*!&-Ryd^fkWGW$607z)|bte zpWb4z*BOELIic)WY5C$Q5gm~u>oi6x@XADBwZxNPt)MLD$NQP^bU$#$r!V3 z-D;NT>8L5^6Qg&s^r14x-%A8NWg*F*$lfd_QTa8pRGuZb>Cw-O-z0p#ocSwVW*kuZ zvWTnYa=s0|boTsZ7q4w-Ui(5|$Ba-H4PQ(a@mFK`PIxRB0l%m%G zGL+SdXbK-tY<8yvOnMCGP7E#qmJTnr9YCncbQ>Fk*xi3>Z`6>}YGTImgiTF)#fKwl3l6{K_h8R{+xfN1pW`)dPeXKDOgD>1 zOS{(n-0C8i48+z0v{RaTn*%I}2}t(Mt`vQo?M%^?-NCab6@MqOb&F-h=po!u$C9 zi-wDr%P3iQc0HDIo*yQVoH&lQ=(#dJQtd|*V-#=$1*Zs zMsFEo&?umS)ve+jH3Cr5-Lf}KAP&(vMJlDyTXtsItKk5`Ja2iD{pTxx&BP0H|IK%u z`#KrimaS1&^0-*yvYO@Hg$6&s03c!UD{z1%lv=2R9eVX?*Nd+SS?C2(<2?*CyXf`o zPXWsZ(38XM=+4C602@C$r-2;RcniCk0sUvhH^I*U$ZFJ)$)yFkdY1EqLXKXeNselD zqIY`sVaUp{Wvi>?+W2NrDVR(8!j??Sa-K}5St)KU=Ku@ zs$K`_RYE}ULrd6z(wnm&I3&GxNQMQrBZ7UKpQ{o2b=NSz!-?a*C%&PkI7V2Ll6}C! zMXbn2Zni{>Aeo!Eu^TE3^Hp3=QTBKuTQu+DP}8d;gT1kxo8L0Bs|%LLK17&(053`= z)8ak53Tf-XyHvjr5A_gQEMI{5dVO!kNXz?R8|HZowbJcSM4ho_gK~YFZb~w!W4N_y z;8W~svQ73VJYi{vqr2x!{1oJur2S4J7+pd!qZzYX66ox0!zYX4C4nKjnhhcV(TOHZYZ zByDtnVnV?+spw#)dn!@xM=?A$S?ptjT8|T*k{@;)>ii5H>wRzANH#eD)Qt%1a)R_@ z5l08+f!=02qbsz>hvXGp%fR=7pF*_{=Rrq2Y8+c@!8g6>DJi zP~%OE@B><-o!tg9Ma|Kib3_WBLO$vj&ovz$#om}qSe)$ zKW}!&FNym>-BNbS)%ql(lu);DV#VEMYFjJ2XV4L5D!#3ZRO5g#pbOOx&c3wDAJq?# z{RI4Md8a^U;^hR{lf(YWcX=w&>;hL^W+TPxi)cye*P_dX26!W#LW`s}F-GC+egqE1 zv`PYY=q%o%bIiQYv01cHBcj(SXXCxbLbruj^~{U~`HFbQ}G6u|0A) zWopr|_-X8#mCJeDy{6fVL2Co(*{WePH-%TmUN7@a>Cqp9xjH&@rtceF%;dSS0r;nK z-4n=wxVG-{H7oRfvr-Nq>0ghMMv`}4cQfCIE`%lanu{%8w9e*f+n9^jt|#$YDzV|n zM+1_T0>ygr3}UK%z6mflz+&0ZO0fQ+_gy0-Sk!hhjMwu-ab;truem^?8<-M{=RuB4 zJ3aQpewij5E59LxX9hFH+S~g_iO9XlVKeW)puL-780VBCQ3mRmJ)SfZ)NoSk;cz=f z@<(B{$|EnB&;<8;a5bZ~BawKY7IyR2y7`0|?a&$@z%a${Z9J}s2Y*%z+&lD4xXYFr zo3&%hVG*hVSE*_i#~<`1#ap$9O=4`Fy}-|j<6H*%HlCPYPM2oPqoHqo6J9W-B~ z`kwD#I+^xJyux+RE0pN0HZZ4@R{m{3S6fwBRg>N7}>q#F*me%u8ruv zUijR4oL4>p@n)<~D+4k~Cao;8a{vZ$z>D6-hAqW@llLeg^oIndSGDXugpZgtW??U# z-`{MmFyLoImAGQd{imQ-df@NmLWug;@qrjKkb&bDSbLWZq?fr9+fG>;E?l$m8!K_P zU+=Gm$!p3oTD9)w|2^|Smw%RWZ)4@McK1>h*_|A*{yRR_8}Oh;LB6e_wj<12J-Cy+ zy1((RO&pvVIT#=&NVXYBWKHL$xM4nbE_rdhTj*Cki~x;GC6FqePzb_>L7Y@}!#ajH zCiom{7dL^?>iclvq%Uj%vkGOs|98tQMA)jYQxRJIS#pv@8EYJ#E(vkRN0?=kulF|F z)=N6W%pZx7y_VcTY56+3*KZ=R8BxKCUOjcE zA~HaDCSLwj(|Le9T=MlAV1B6gnm-eySRi9lpu0TXo|oCn{IfIN&|%Nqv={cw?WvOk zkfCSqk~#>3UFRj;mn`^=E8BFHJ4y_d!?noFU$pLz#PfjU2HdxQtn9hVU!!U`PYNM_ zXVk8}eitw|G&NK>8hmeft)c%C{T3q;L6>QCCG(?mRQ3YooDLP|O;{ANwe;^DytO9J zlYT<=iR9%S1p{Gys+SyASF}F~wQy;DUOHYFn)KPaH(4!S{jfh~x`K~m|FP7xJ^ydrIzm1x=Vi<6TYjVR``-Vy zY5+VNr*0YVgj&E^OGPF1A$Y`y_wWpr#0TMSlw3v6LTqw4}RYAUw6sLLpg<^Oiz{pwnGNURh^yyY(y|` zoi(63d~84TFjY3N*>R-w&W+Rtoku<&1B&RH_rZRRk`|`3ut|;etVScvA5K6 zPW*(+pEoq;^<1@M?<%bVEJ`&HTY3PZ>t;k3;*>V&N|PD}xl?^Hsd%26_L}HWIP`Zf z;cxF=;yJyF=_>7RH-&(m-sr*!N;ZD;p=sIZO|P_cM_e~fuz1bNAmI3N5TducVraYt zuC2+NO){KyeX50$s#CH`-8IMEOONQ-R9tp@a+tR~O7@xC>!unRWFCP)PZ4q%UfrQ(9 zHBZc@=ga9P6DHLTulp+x0h5X2Y1B;3<5CPrQ%`*OI)r`Nyvy)Qfov5e#nSm4R~J3UdGsRbC~@GdCe z%ofF(YE_MCpgEIx`_>S0I>FUAn)^6ehI_4SUd^E{o&&`T9YM90={w0%+ROhK?b>+D zsZus$Q`T`y+Z}c>^_>3azmycoor5+X?V1zs0chy!9e^}lG zP)1Ilf2E3dpE6<7$i1|cOApddRkc|Cf*0KoO3@ksZUit}ed>avfZt`AT9U(mPKg3 z=qlqL)UfPQd1_*X%n!a#?>@^X49!P$w2Z8wy2k}EGskzE3$`DRMI77;`5(-^XH--P z*R5MXBP}WzKtv?8N)}WUkPIzBQF4+{fJ%{b&LEhHN|Ky&&Y=h*3Q{DIoI!FZKr$5J zu1&w+`QAIuz31QijZ?pRv`TGu?b>^-x#oP9(a`*XyVn;Ep)$`%{nJS7n^WMi|Li8LB?nT<6Nq zPD84Sk}v2niOvX%V@Ly2Y8+pZLoBML5XpndQB3?>J@Ic+{<+)U=!m>$>SM{DN`>zp zgm|LW*4d{nf06T0t=CXVaC^-q8@ZhSd49n#Ti0ul(aWkJWF=~PM&%Wsdl+}kH1pQ` zxbVLH2`VFztrnK5YGOv&$gP&wdfHl{?`);JM)12znfZnb2f_v`zZ{(dlhi^8Zvs@UZKV~XhV~;~+O_qa&8D>7MPBs&>J|Q=`0(#WOcUGv9dtsJ z-b>r{l-Wy3>eM=he`_U$67AKPuCDJCaqCu|*OpH!uQqnysl=7{pO%!{Mh@k0fp-^k z;F&u6$rUrAM=JMa!&7%8Eh&sEa*v8tr_tnxsEenUsyyM~&xlDl23yLS(*Duitua+X zmJ5#{MTSn#-llUWj_ahyiSwR82i1k^P`i*`eS3b0^*xW6WaQ;oK?hs8XWa%fa9NgG z>nGaJ|K#o;598AA-DFLg5d-UJL1LOH`Ohn^bBWy9l3iiES8*GPt5-!}1pD}8k?{c* zU#5+-!+cDX9E5|cBX`66C zWrLl6wQDH=$?FG4H-@fM{_1C~{F!xus&47&ZKB{3o&K!q7U|z%?rJ_^S3ON9yD_yK z&FLd|gfMiMH+HgE%i}YgJKU?grAKS)qVpbyV_$PW=0%70=JsW7tHk?8O&Z7f?RD71 zC+pSm_=P^5;+$J+@ZWUQz-IB`BkwDt+qQlZPH$|tPaXUoBg&mc40?A_0y1@0VMe^$ z91Q*gO5Hr-teYJ1cRhXt92KXF>CWKtou_4m;n<}@Cm(2S?5Q#!A1U>I$6H)+wtUtc z4x<-t>ZNAM=%~NIa|>rJ|YBp3eyi*=Yk2ynf! zV>2UI?n%%2s7XLQZJX9uFU%#`qszW6eV1tI3VgaBFY>myK;lbB{SA+}bl{c9^QaGP z(fsMd4`n99dJ>u5_*`4ClZ<4+Q9wM@O8Iq+DNAy@@ir*TW@EJ;S$Gr0F>X}wh-#%u z>eBL5h@ls?s+ZbsSC4T1!#3sK`LFMn3*~=MLt2<6E9XZ3K}DL}x6setK6kV3H9eY? zZ`nxD3=Fr>OfPYZIafo86Qqr(rvI%W^t{;By)4Mf!&ahYjXPF&_zK%TgWd<>{KXKK z%AaxPa__H?+SG6PktBKi;A=Qwo)Zb65&jM5`&s@=v#L=$l#t%>$*uJ|Bd4)js=wXz6O%v0+@iSxvc2sYfa6eNYjo>}RcBd}zyTfRo+c@b^zMErA>>+#sZ5zPB+S zexc1E^qd_$FsnQihIwa3BP2*(u!)gwf zo{yYfPieU#v`Y@0HB2D7xWXp!r&o{Ke1~kT{1h*Hf1;?OeV9_1wL?3dHc?;%UWKKS zXSqo9{Jj?s{osjg5*t^RYQb{;yX+&amD}DiG}GlVT^Mn7%;vqMzBR3K4|MQnewBB! z_2)_DsDzaSa&LsHH&5F?al9hK^RcVxMak%Ml#l{cN^|nURV{62_=m?&I1CDcP#*Eh z7Otr(QRQ~Q#T$K8Y48MPcWUUg8{w7l2+6oVfQ(k^s?at7oWetqVmH9P;3nxbsAPR# z&f3D^HFblx<-yHj?sA-WD=u*I?Tpd|ny{{q516E;i*VyaE8I5I4(8Ij zj`T7=L%N!Xl=_QroSrI3jL?~Ee<)ZG|IJI?-SceKE)jsTwnFZH{ z-29sqaJg_BC$^s1e37JmX)*9#TyG-guADFN1=Gvdar(4(4E@d4CSNs~K>l{!o-7`^ z$Ii|CtDX#AMvOIa+Ez6^kd2wreY2$(+rsfV2~%Re!!2I8s1lL@m)i&z0Y-o}P7L}q zv-v+WRqCyW?A2topp5)N;<54j%n*r_eq`EtaIwLXa`I-rr}vO92c9t^U^TH~>U z9ocpoYJpwaqjmCng>9`T-;^D7%RHJ)d+SH6wjRP+Hkg;*BsedS;o7TUT|%MuwZUS2 zQ*(j0sf^HZy5`LuMaB{F5yL`_dOSPIHZIEIduMt3M+M-A!-tW;113xqt6R zvd3@M)!Ne!SMGfrU5nZtoj9uDYm5FftcCnz_4WZrgvccY>81}kxcNzM-RG0OvZ)J4 zQgV+Qnq*Ox+2RI3#{1Qk8_*LOC)l%Iv&U#Z&DfhE{ZPZD+rAXtho-wL=hXuRW=s2n zdx{~(cC?+2ssbBB8mTJqdq(9^kd=DYtz;DtwE5D@t98eGrpT+pJMXG3#MfA^yyR#T ztXDevtC0OyoNk?AQ%TmITa##BYUYqC2 ze|(no<-VE9?!flX1xqbe^0>RA83s{Ae(Ww4}eydz!` ze)aK-Nbd6A-S*O&S8&`XjY3f3o&~_R!*NetiZz~pZ*8c+cmh2BP8+Ai3C>Cp1C=O1c-oEA$G^6?+Q1i;F zQdx>Eg7~oP=RjoGpG<;a7zn0Rl8|v!`Z2ZEbVw0rlSFJlhBG!u?;k%J4J+gS>#8^L z(iLb)nHZCxS#_a-W|d0E{>DG(!~ZbUP?3#YbM@X+Xi!cv{L>vzc|h++!NAN72p`nv z_CzK}b#gD1czdG%32z_28R37L)&CSl@%X3z8{3iXtVAe)8HNDEX9>N6`z;m^IM%}F zK_b1b8*@-DAOC{~jFD{T4q*K5*vuT!545rK|M~;$c^|fOzy$mP)-k7Jip2@kS{Ycb zG=cMgKp16dH*_YY2HYlQHGf0YHa+Ljz>B6UcK(D@s0Yy1LO_6I zcF}HN38TT9Xjy*tCIUuDxIPEUAjl}*qZiuNiPu0R8Gt(@9#;iGj4vlB8O#CiNIuib z0P=eVC;ujY`JANWw7t0iCCX*w=x|$%0ObeF66=5z@#tiVFg#f;zgr5Z9j3Eh`AX!y zpCf)k2_nxb*y(WX96%^+LHpHjQ9fd*=ee88S(dCCg6q>> zJi_|mOr*7ypP90HpJh#R+Y@|6$r=$AuB;t;mCB`3s z>NzMv!C$X*;tlSv0F{fe$~ZO0SJ6|(qHQZG?Ojz#bA8hhn;ML~Ums}%Rq4Ky^XNO= zGnbG9vx+8eZ9cbeCFHO%ngS)%0xdMHuxxD2M9b2LYcgq}+O?;*_uDS)Z&+*6UF1IU z(9GPk)0+0`-juD-0ndthzQn*V66&PO$3N0q+$5y*Wz~6B=~18@(}<}h*|kbCuvLOy zhwccIMA?T{xyIFjCgw@0soJf^90aTav)=6rKx=ZR+JNlCoqhcpc?O$)(igoKwDiWR z-P>%7J)#etU0qpEpn|(pN3%|=yU=z33)2iSe{h!H&3Ko_ z6DQS@HDAG_=}wz@W8V7Kb&;FZtAESx=6`;T(+W#{m!mOxS8m;S6u3UE0UDLV0E84; z%qbpr4jWSH&F)K^zKfpT2Y3zL?j}lHo*nK;;HMmbq2-b{35E5GQ>2f(?Hlzhs7ciw z$2^>d>^ekzZx0o#Q~fwgf1RpZQ85+pHI_Qkq0`Fwx*~K)eeyB+@p(*`b2PtI*%>;% ziY9XZ44+y+lWHP}K`}PLcStWTWW4iq{PPP-L6kv6c3u_7G)G)YPp$5N9fz;Gw-ja& z#$OoV3`KJqsikpb%Tw7e130!N&hKw(Bef{0#auMDtYy5V)%8G43vUE1=kVkc_;oy# z>rlVav**(T$xy-Chld2K2&r@)E$B8@r!hI#f`otap4lEOx3cO=S8yUa8P>y5tk(D* z_kdSSW1N}XqA%h|D<7wbj(Kcfvv6NgGbo8?F#nEv=q5+-B%$x;7&ujBJVj8!Dr**J zrNV|0lM*D*nZYXMCCm4Q`0V@9BlyZ8+}I)bN zyCcpMjd{2Yb)nZ|pOEKJaqINwYSafJ{t)jLn2kdfH3vW&5y)k#0?(}Bhnv4i9FsUCT#{jgZzPnqsAl_}clFxO+I6{4n zz-0m&VT*OA#A6Nh2K%){p`8-5zIHUP@jy{3AzR#UrbjgU@BA{*@w#6~ALOFMbrbY` zacoN76E82wTes=DOX=NLUs(>Gb45rlK`GP^np+P!R?oNV|KVNDJ=_sil;~j>xf=Xn zRX9$vFI)Wm-bq#c2Xc~Kmcso|zuE zKzxnEpB2h!WvU+b95%F~oHc{X)VaRG?A}E0R!jtXW>4*%%p|ozXSJzc7IyV*6bXB> zTqS%VzJaxZOdWvE~ms_sP7M}#CL$+TG>Qy&uMpG zPG7aW=amrY&7UoY-~ANz&vjUnUWZo3t%_U_%fKq#OM7bn<^**w+-qwxDN|p7ib^ZGhjXQaCnZlMuiW}80-o4AHdXO_ zxlc0{^`WU=zATpG_XKwBEyr&`#SCJFr<6=scDt)WEf;fF5Hb&3LQSOC$ptYy99+WL zCd;U^3XTz-Nmc1Qce?piB|1K-Y`t49NLv#38A7dH=27B#NLm#|KA$ICNJORdO(*jg zOmL7F+ok_zY1=WNzQ3`(s&)kxEW(ydlrWA=W{}r^n690ms9ZE|^977#=Y{`a)_o^% zXyjcHmemP|^4dH0zQj@v)gj6|=05Ei6e&%c%kfeDr^(>t%*>}4NKk9-%@qE+1Lc+;z+*WaR9g?DV0t=8(EOt}0y;~8ZR z@AGDl1-M_z`cdnTqwY2%wUhABGF-=nD|JVr&L>dVi^Oawo(3a>l>MH%Pd4)>JLs@F zRV`#8JXrT7ThG<>?(}}p2AoY8IN^!J;*g?0v@NY@v2|IF6q0Jo0H-NeKh1o(E0L2Ka4U{T* zJjMf%IO#yJZm$t|BNCmsYX#@g5iv$00hr!9`kzsKM{tRVds4Mm{b;~@IbGQoUbbQ@ zz9!671HQm-f$QtvC_Imk>af3F!H-&~tz!j_cAo}Su*h*%f(nxuT(kK-c_kJ4@dgTjn%kLfkb*)vbjuki~ z?`w(OhzR1}F)7?<&!7s-IB!W%DV>!vq*7TL`Ig|kc8~~%L%+k!^ef$MflO-qkP(VCvx`%2UXYM2pDlr_ z{X69sO`7>Lo(xC)xi%(eJmc_=S7v7)nXRrn7f$ZG7y8Cuw_w4NZmbD#$z!@)`n(nakF(%v-= zmV+rF@3xW{n=T9%(Z`yo1q4Y-;W<`WtCpq^-%brn6M}EPD&|r?LzWOol zvv2$fxL7in-cwjN(L2Q;W=vAaU-YZp=Ccq&D+c$4cD`d%s{Us_nN7Pq(z6rzvH!BAb_y^|E$iSpjh6vu#?g<`UEx0dkQv?h$vlZgcp=*Sbm zhR46-Qr&OkS^HmKUq&JL_@AZk?j9hwMq-pB`q%_8UUXw$Ph|Lz-*t>>!sW~TwNI*D z&%Vibjs5!_^uNAhaI}c4fIQ>NEB0SR*SU4Kq^?F@`v1M!+f|NS;C4pCYmQ<|cu&nK z%|t4GeZt?bf9aGMckW*6b%7snKV35T@a2(^q0U3K?T$RLE}5SHh%m7@zxoAF3l1SvVP zf5IK&5*pxZz=d;}!AMwQbxDlewt8#OtiIToib+YIMTB0+v1B&Rtz9&M^yHz1{^fND77(~B8Odi6h3pv!4-KXAZ~pX8Vgy{%*gEp4 zmjSeQ8j;=#g9|;`Xg(XgV~mvXV3+t7pm72RdV?uEggs?1F!5eM=CUH-lDCYR>D=j` z5x%_AHy=%dkh>TGvhn`-c*K`+g-q{#{(_3Zq3!u72p$h!AmlVXu#6$krK@CmvJu=C z2o}(+YLbMGT)#C&!0ycrDpDT!YJ-Bb6$v7H^K#`Z1I@ioTky3Z3l#dHZ&^-*bSjRk z7$U0Bm4Bd1cX_ruBNml6QMH(7x7g)c>KV|rwKZ&~zz-<`czyT!S89X?(zDNDB?IjV zWp@;2?nff``s8t5tmqBuGe|jX=}=zV*crt zgP;&FTl+Np%0iTkO8@z(RxQ5+F?YyQ;HO2&tg~De{|%;NpXvm^#wlg2%;~#(kNgF0^}z=U-45!AtxwY*K>iA!qQ%Sg5P{-uql< z^xSzG)n*kavh}2o^?i|AQ3JHCD?CcDCZbT^e*kXt=1A)u+v!Sn#aevz&*Nj7lBql| z`JrDN2N|9D*GuFH&Y*Z(4)>Gt|59pvw7NzZY=b0KKW;d&J;gL;udD-sq#-gk}eS#`P+GQynfjQq|qWS zZ(u?iZE2@L^Gcs*|N>O8P|zcq$@MOUgzKU8`<{L*SHO4S_GDZ zlJUpCZd;y`XPF1g?&}r?;)l-79Ul_P1jf(igWY0(CZO1U^yX_DY3O>sYYFs|!6X#T zT)yIJjayywyZMI-c+To0+9rzlb8Ro9g87kKW=x$S_Qpr)O|zksO(i8pajwH!3_aIV z*YfLJ4+qkdbGlOTBkr>$B5j^~Hh<$@EFNr+7P@$Klul~#D{Oz}o!)M?L~oC|m7@nI zl0ppy2g&M|4dp0{SsC2EmYENnzY!KSvocVVO>g7u!9U!ajm_1p860(gUcqbWoZ)da zQeGc3(@5_ic?}ZC@o>xXeCQH;xuga%Syj9HW>6o$e~Aravg~ez;cb27f747aGT}gsSqhKF_Wxw|%N=VV!6zP) zEVcZeH1DZ9I$>v?uHT$M6Ic3(8*dOKdRSrst5ufqb`iPzPN|z7J4!+mDSYZRd%Ar4 zZBF=bhVm1SQssyuxzSc-l%fn4(R>y~(Smv_jdi>24;BY=!K<^;K4ev8^Ay>iu+|jH zzP10zH>6zTNA2z&r`uj;6aM?P=z%yl^}<3uf17oV&$m67W_t@Jo;t49m%^u==_ox^ zyp*gUR<2tq!Yv*^gZmRFWE`CYB5a=PB3z*g&Bk9 zymG29JmiNJ5p;F2Y}hz0iYdRPPa>4P?#QzEFvg^B%sGTU#NGDCYC%J;_I8|lkwLS5 z;E{3MMK0~iiLC7W;!3;2A~B=(lC^2hC+~hS=U9KZs$Qtf%&DEHQy@7gZ*EuAV`crx zZGYD&*<$)Vb?}1Sz^+|nY%tCtP%TrrM_yq1Znj3< z3$7=(#`yjiV$-Q8Kzodwv8*wQaS87cDA8S8jGmB>qk6JuG&<&J6i--62~j8-bM0L91&6(1lsVz@NT+jI27Sbv*c!zBA zg0A}e862@H3)~owzQZZ2Gwdi*PJJ{aN?4}zhlzak%To_thTTxZaj0gT6~4oXPV;a1 z@*|j*+`zdjUCu$1&h%=EV~!kMPx0@`#z9Dxw))&SXEozL4?@8S?DB#1M%i`@2 zC8>nG2>x;Oo$Z^1^$yxCa7VDzh5zcU#&DMB_(++W+Bash4iw7pkNw3Zwtx8)HHU%J z`~B@s`N;ifwH(!fEwOQL{Z>c6i`*sDoZ&+SiTPR$Rc|T|Dzj8%%7Y?3tWZ;2rmuJA zGlKAq(UugADRR@e4|11(vM{1-JF8v1uq@LYZau9!G@nr@7Y)eZBC*iI8@ko^EZJb#<^HZMii ze38M4)9ySy)xIGSxr$FIg;LbH`cjOB`)}%~#9hqQ!uIWSh^;TZHp?$^SnL~kt{`k- zRkxkUYvL+d(Cbfct$&8*MyD(I0dp71+$6AmD)^w+*8eAblEI_Ua}+xd0wGl8YJ0Bi~Z70`+nxJ*sd z<~6^6{e~b?A#&=vx&D#1uv;`I^Gd~)LEgvzBci%{a z#D?DcB1bC!-MNMj+}6nlot`zf_Z^0$S;&J|M6C4~HKaMjKmLg+bC`GaYl{)^6^bAd z=T`1bh}wquA173`-_6z>&SLUww#r1eMaGL{tA?!u&bS_GF4}+V$*T+`z7`k9=&@we zi%+A&her!(_n7u9y{J9NeLPTPx#vPtQY^BSVO+kw{1m>L?E!uZ!CIxcW8N|rP?Gg7 z?-%1s_?ocZl5h?9LhRO1oKkGvq*QZtT?6h_IyX+{LyL z{KlQZs`0LOO=U&V*x^t6qtwMC?`bH*PXsfv@7%WerLesGymmLCNO@y;O!LMy`@5%5 zl|g}qL`+tkp;J~XA5WU`r?u~VKV|P@y-i#f7Cwz|FOjJ|7cPOg?)!7pq{HV0g(zhqsW8lCgN78&V_+w zee&1N|M|*|JOjx1PXJGkjISvguN#7x)|((fvQ2Ybu$`5(f$1GDvgz@-RsF{+AADIN z)BqHmYu>~O;>vtcMYw%yQ0J6KiNqPEir;qOh_^Jqcrs~ULcCIpl94cQZnBk%R5yiZMW)z5D&H+ zy_U?%XpIF-;w@d3l|O=6f@7L@A%Cn6w9%Am#wf>LO{2;K_)N3iVD-S|{5nK0)IK4; z^$BD`BxjGO;w`W4OMl8YR%HauA3P}en(OCNUh`SJ3^nj@Nz&Xm#)v5dmsrf}8 z3S{RT>xPz|7P1XsaiBPz`ljwMGTg+1Y=Cc|6g!=EIEx7B;L`1f|5z@|ocVh&&xg~h zb~nvo?nnIr{VjcOcuv3$Co4@jS_GtC3R>tOC~7mh_oY{SiLVU40sR7pFD^qv=}L2{ zvq(v=&xiEX5(EPZkTgGc?ang-WX*<|Tyufo8{rlp_^``!6oO)wBd(&(tNKn?w?RhK zgjFdcM#qVJePsj>?VQ_17LGEAhbuq~QYqj4s7`BESSKE<7a>s#9snqcC2&Tq6f{s2 zLpr-+7An0EI9Y<@a~z9c#9S%5WdCavoGqawk3Y?rbSAx`Ls$~gLQZ)?^*8lutHVIA zBB$XthhM$~ED|RkEj{QwLi$G}l3!LLM|$PiJcku*O&W2fPH_fx0zEzs3p2~0qprZm zyUav#|495SsY<@ZMmGgQetDO?P#eAZY)5?j1Yvx80)PhvAWjd(Dxz(^Q7?=9Ku!#@ zXAd?;1kk&$*2Cja2w5$PI6AmjAs1j(u@s7YvY=;MMQkRI2lUdXZSTCMvpf#HTq=(B z5sc7v&Ey4F=_L}T?2RE+0V0stRMi)sMRRk6v8%9LiWb$Yy>Ok(dlM?Go&45Dz}hjc=dORLH5F)4wftumwZ8#NT-p^1eS@A zGyCU6Jlm5*EFr_h&noM}BRSn}vbTHZm6{5w*4vO(opEN>X~+g0%OCcer<7rHgQ}bv z&-oP(NZGDjkRtS!nrR?(7Y)p*|xz1NN^uiSHIT~H(#6=k?IXinJh8hY| z#qK{I5Pkz0zY5F99-I=`xdBXlU1P7%cS#2-GSw*CixzW4`m@K|RqTw%ra+_%vX`(F z(Wto*-bio#nTWgLN8fLFyyq(_U&Tv5oBDRhX^(w2LTw1%7;X~D@=KmD*c)kfbCg-y z?9KEFLZ)paw`FMmLIu;g2lI$j;dbHP`~=$XVLMC``{d*4tL)>bU^^Hg6n3B1CJOcA zuKJFl;PBOoIthYH|;%4tqN*-%jMw&Ha?x%>7lI=o$($lhX& zXZEagBTuEB>Q7QukrwdqhAJf*vf0hF+LWJGFQyq$n)ysZdAb+cpe7`ElxJ7@sIgWjq^i=f34q+ADE^e8Uis4cC?L!jJB7-!a0Y8kiu&{wQt#A{{5(d3Qa@QcS} zck#Y@PQ__&1}+4NJVT5yWc7}-@fO*K2Yn3mKRURw9gOMHW1MqDM6JD}^$2s@DqT`7 zlrGcELTQ*g<)iEEq77+>x4+dMEo++%uPVB&2TWLsc2-=Y$hSf|(YEOA@)73|wdG2R z0@qwE?sJG*WdV?G*EsimC4Nt+RxIgQ~@U!p4s!Uud_Z<%o7euyx)y$OW zUh`h!fBI#cZ-TJmVRyLs-1 zp!?-BOGiet08w=9+)0?$+&Oy#{l>N)&!R;ay8aG)yoGR76x|;+b%yWE>He{LDJ22t~B{&Xw2;i4UN^eWOzQpc*>nRj%pPUb1La-5mH(RKVD|f1)?nO##?H2Dh5L?kf0&=MG!< z^z;o33wRC0-!anKU8ByC{C!@%;=#G>Cz&5lIU7uUkDc*~3kau9-Wf5O? zL_`Kt@OYDe=)0x2lVxDN_Rz;Zy-{*dv_LE`gmJ4TQdiju?qr@+cz2$8c>NaXxlzx> zGeXJu0mm(Y*~(2_(CsYaXImqm%kY^Qg3wmX{OBS!p-DmPQ2!0TrOqe@n-sJ| zn*%pncafTz*=SN~TfEp9z59h!4+}?CJM7*d-LNClRhOE3l#DnU!I(K%t9L8MZrCkv zTOZ2g-Wm-np)8kE{L9O3%T9|sH+DT%zvK(l^ksBVi*M12nrFajs$R3R{sSLyIevpq zB_EI1MxrVuURYvol#49tZ~{W@79VdE= zd2EgFUwW=y?R=6GwlvlkxtHC(JK$=@@gW-*6(sy>_6qSyY=X0T-Gz!F~8QYt`vbdaQCq zCMxNJzoh(f+St>p^aJvqv8_)n-BBo2m*I7g3p<~fpd6%IV}T#!bsTJF?XNV7we)(t zJAw}>Q@Oc&y_L*In6!vPwZ8$nrWO6iYpx&1tUUO%8c)^D_#iHS{ z)k$Ngad6afcb-a9?Y?bJ%f`JDOSI6=xK^3rrs-m5 zZuY@`mxAMw@4-?HrI=0R?RTd3&KKM!@!J;tbSo8U3I`0XpE27@y0z7E(OXRv5`iQX zITOTJP4gz%g^Wf9OD(m8X=|iWnzQ+qx(;odc`>pb3sY;)H!^}Oy5|b>Y7RuUMGQu; zBcab@ElE&Gq*YSWm3z5@TRRuffCAH+#;%Z@g|8#r+os30N_D0VKfEDS)hgZ<9$u-( zYcl|1cq3)DxV*20md(d zW1g`k4d3pAqKoU3pyFkmw;Lnj&Q57L)7PIc8(%7~p@WZFd>I`xx7*poU^HS`)oEE- zS*(_;-o`LDG=>Z1T1<;~k`U^P|BkP5>(|f=w3|%&$@PwK#<}|-q1tTGo}m0e94BfI^TDCE9A)GCjVtm-Lmqa9P;M@O0x;pK7Zk zr|xi@O5M}@oam|&w_)KwRp)qhh7Zq!87j?8cjLfm?1^tf+4ji!9?Nfz#k5`Nm3NhL zlb;kO_0EnG(05|!1f3iX!rg!GUJ^Qrxr5&|iYrQLUa8TYEWTMiRitMBCa#6M^4=~l zmE+`Ht~j>*bK*0ZwxXbV<~@v8uH&7{N{u53`MG5IbN0h`M}%B?@t4*t%!n`>pzBv* zGv)=#KaqJ?XDfcRBn^4^ze%ALttV3jYVXYnxXo13nDCP)e11}H8X3k>F7kvf40nI? zX^P`!mF>XpYqOjIC9YxmFI4;Z?H>nQYr%=j?~Rx2P~!HH?8zRLIXU8(vmCxvgf_I! zh2J9Cs>pb=68peZ%bb=1_2BXRugDJ8=DtWaLGO$UYHC~5&5Vb8X1#SQBbmU83c{;z;@4Uiwd+&%T@=RBD(TebEevyox+IcBKFUEbiH zg*9_Td-T5N`SWA~N_@Em-hqza(q4jU#8o5q(}LwK!E`m>5>g*ab{ zfjW$kS?i_#ryV+HC7W4*rddX3Sy3P7y1@QOR+hZcFzEjk?=qAboPiefnTC$dE0ErJ zE=AXrc;Me`$w1ocDFiff0ITx#aCepwJN5C=612`vUB}Gx#mpl1N4`7w-5Tsmv3g;0w=`pbt3^@9aXOK7+AM9DxhMmT@P9Q(W+vgRNEMF$* zNQmLv5#Za}a3l07tdtq*Mi$jZupRC#-UQXlvg{8J#9;~BGjq{ESOCbX%Zm;J#K1ee z3ZWx1XwqF?@OXI8;_tWv^A(we%`qkW$^qC)XQ_2Lz9 zS+Kq#hkF27@Wa{3^1yvp;cyoqBV5TQrdLBotM+R%-n9Vub<&3%v_!iu z$;lEc#K}~(_yJ;LLkOVc%@y0XAZG}#JyfW02^foywybLAz5TOn^YAqAH7xYjx9~%K znBWrdasvEkeDn@iI_7TOYeXF7Rl_|n4@u_IF6#&Bajq=0FR^0|^g8(gv4>XLho#kE zvRyv-y|~)7d6}w20AJEIe3$D8oSRsaNS8BUVl>Ng5P@Fa(p`F}640y4wXA~*9(h{Q z8Gmz_Pf6=!LhRH&$FYKoY<7 z8P&+1$XH`D*ZH5Rcn}Anc^5S4J81elr-aLCpyPCHJq3@Cu4jUP12FNKD;Qojk1 zx9IWvD-ECuE&C`B9>TWE`NO@>;;L5FzhA>7yF*@d*Pu@U2N!`EY+-h|<>1E1Lu>Kq zAjchK#x0QUhf*VS7l;-2ge6z-VH&z7kVmk4@fRrvJUR+BL8ze8g>Cs6TDmGFm_veStUmK{>pb!JfH7aB5FD9W3mqgD=j-V-sFBQPK9tyh& z9{Lp!=O`X8<8R(;(FG{^cau?Lt64O^y+F`l76@)KR!w7l%!SXSpx$FL_skc!8$puB#P4H zc7W;m+3O(=5UWqI!Fg#7XxnoiP4DTvkII0)*jHJlTcY4v!WSbi?5`* zy9+5sWu8C64zgU|P-cPQD317?WF|blg&u}MwNNSkYHwGuI0o5_9V@<=BGJYwFF1dl z=7D1GE1Ko|1`unqbfN$JB`L3^s|3+9Lk`6bcN);=)I-jqH6K`s2sSJ6wF2~?CZ~DW z47S&f_!zxAS;tR)HbsWMAxmWt5y0^gCotP=Ou_0GWrSS=7RDhA$ZDM=s{Q3f0^86} zLc(~U@$Q#burY_Qs%yy$xC9d5cuWBj~$7jP41#E^(HB>LULh2vqu#hLE#mmG~H zVhBJ82jq2iYx#UBxO@_RjRWS#3^`O?#4Sh_}I2}Nv5 z=DV;lD^k8MkxZHc;XX9T@FxBPG5ym>bbitOZk-33*CBo9=<$nf?qjdxFAs%%CxsSc z#%IiX^u$H7M1mgWq36^q zKj!ykvxvchIPx+c*7KE|oroVX}u({ACl1Yn&SlnPX=i{IfW_k%0nzhpF}gr8Hp+;OG?z z@=1aY3kLY1a;MnBh0TR2Iii!KJXHFM7qW897pvFX?VQx2crfo{0UPBA-u8;$k&w!g z&7_SOrSy2SF$%zL-#m6@t)b3s$+oK-hndMwMkXFOsr{@#(B&?_9&QL7v)jH4ft>ks zK#Cz*RVFo!^O*vnvsx~8k)t5#^UEU1U%XC4XJc`7{(F2%GOre{YvS=?0l!)vlSVm| zUO)xk0gZVLfJ#_@P4wL^Qfl56rI2Un;rB(tppKlAY&NoYqM^5b&K`Ze=Jea)v^1fr zmJfm^YK%9tu4FU^i?WW4IvIVxxk7(ds++uum|d27yZoWX5nI&=AJlM?T**=n7dHK_ z!p3#1*{{jyr^Ki=1|PQX=aAj$Bkt>M2;p;Er!BPGUrNsK{Jb8ixcfWRA4I51KZHA% z2igJDuGIfkyC#ftoTYF1Ctfvm_y4Ne_5VI^Cf58cX8tF$qiOUc{#da+E<#Zz*Zw^a zS8(b{on&TvljiEqoQ9e`OEw%8A)JY zGf@Il%eDkKA$p(Gy;&{?1E$@p?Noa1BCjK~39L8ZGu%jb+=@)sO5Q8Qzin95?pQ`U z-7_{HtZt6;+A`rsH-MqkrsX}j)jbC-YO02!da_6Es_i%=IM09&aqgl_70LT>y3OA* zj%z2IDm*&J@@ucG%xse}<#t%*OBFY6HLmull))|=b;49KrAm0;Mm+1`Tw2?%S(!ss zNsSwqYp#Ou)>|YVCr9G#|M=5tjoi*5c6=c4;!Q0B$j%jaGn(Qq$ znd~@z=|Ch)MWjNogO*?+ zimYZv(bhnE{DYHEGW<0nms*;^-#tLGy=l`W+=E6M#+i_~ zIt{nZcrWvy=XNwUzEGOojPj_PUM^h`V7+JRH9T|_j#%}IJtsZF~= z|Lbgg04Q{LOQ`yb!Gc<(?YbqOm*O?eu{iO)nwSf!(b8GH%jTt~jHWlwxQ-$-1=2@Q zN<4>cOt~u!BqM7IVw=vgKkk!7#)l|e93_ljShO4SwTyNevB{3L9h_gjo;dL|U};&% zbqEC2ie1cpMWK3$v69>Gu)ZAWfT(7&&#WU`26NfIBch8_20PI!T78xx!#lg{M}>$( zm0^+BwAx2xrTQR^jtn)g6uEtfwHhdFqG$Afmk)zx^(xi*am_(h%?PkyRMpMKIlbt~ zv&(SH8UOyY88fz|=~Xp5dPsKVlCZ2C^=f1@X6k@)Q{iOH%gn#M2BtO6-y>2?$h@Iy z{Usu`&@*0m0CH7v%F9_fA`@Tv^8Bg7K1|rI`8d*Gi+afJP;Dp{?7fP3UE%E_Y=XaE z;jbZ(s~^yt`rOZF{SB|6FsACTH^U~_WAzX=3`2ts+TR+GDf%g!!*(r4U;z=0t^v3R zO=9z^2SE|XSou4Tm@??J$RRo{A_EUN^f@_71OadF53^e|a}{e@X1SIld&^9TzHP&m zb{0QV;w^r5T*!6n7`4=DQP0JL7>#ZH2n&WPF@DsXh7k@`I2^tfi^x0{#hnuY_%+VF zGQ_PmL_?}mdoW|hUpF`P>xQPuc%QOo$DI0yN)L;7W( zJDeIh1##u0%kq-)LI;n{V%*Qt-*F8c;zp$y?Gz-!CKXx`IwWox${+}o9*H!xtE!2o z#rGC=O}0XJ#i{PD4(kryI*w@*B~l$$;)zx&h3Vw*%|&8wp9j-Z$s~KseT4Q>9t`e} z=XcTs6LAl?7$0G(dZ$SyacrvLdYSj8%dQtul7HmvVjjIMPI0f`b&L2P-aZ^~F&jRP zpCPxFs%{9?EP}oZNf9C@zxpS|Sv^_Uga$A;#3g77%$S4QaQRsDFFyu$1cBGVMpkdeQJ_wUKMN?(J z=i1CFfKxpgEozZu&jITEIf3e>k-EnrrQU*b#jWr0U+W?{0FOl;bCR{xi+H z5fQCyS<2e5G|{iFX1No}cCublx|)2akoeHvr-x5#DvQvN(#BX-J05aeP-~>{Iw%b_ zF)3Xq^WK+zVc8?2&FJwMPOt-oNH=f8Z%hGBg0y7vcUKgB&($_@nH)9(Q&wN?Zlkpj zsIVh;7zKs>`KD0Y5d|`DpBr$VwuX*S+U1V9ZLxP)FBPL97=_l_i<$aeOL;-FLcg}< za;A<;6K7muXpeG{!9%)Rs}K8@yD5G3=nSbjmd>?qGNPwP0Lqms)uO)DZ1h}(%kGdA z?^F3xo!uqg@uuErnjE2l)bxufgZo>dzg(DIh9)+*z9gX9}Ug%nv@ZK0HXO(c|k>`NutmtyRLtRYKTvnT7wlB_e# zR6-H5j%|!3yRnA#NY^tpG#S%7bhRfROf)qjrIQ`IZfptfWhDDQ(i zz=k>!|AU?4jSuVq_Ha)+Qbb|!)DK$YIj;on6!^xZ5EuA)M-ND2GelE(cCpQHGndpe zQ~uC5))~N&ES;cgUZ|N3+{5e0&F|F*DsG{TWy(qVYfW>AV#R4 zF=<}NzVX#5g4WVtTmIs2BlhRKzk4$FN;WbN9)%y{dp#m6in~Hh8Tbz%A7*rdKf}S0|7564y*O?Zj*_ z+8Jq%m!A!}lqE`-SV?IPBFDt%CjFKUJXlG*h$W;rA0+ zs1Y!kT=Kh(RRl9h1EXq_8-QHxvLrRp^E)v$T(ma`bhU9`s2bcay^w@-p;n|;7nmX^ z_OjhI74$1Fyr~y*@*{UA!sbr$3q2w7xpnSN-~=}gv;kpz3_zZ#YP;BonJc{IZRnvn zV7@J1b~QOWqRRtH*KrM{F62ElS3jHIkq({`9Q~z33SG?+Dh()% z<_+MVWK4>bJIVn*l%D%7)l$rMR)`08nM-OE3RM@c&`kFy;R$;`%2E z4DcC&jOx$jYfUvpP><@+KWR&7bpdP;WmmI5(*fffR7r%^AGWcurTk{>b9+EHre6Px z_kB!LQ?u$_1#>M|c~dXtz}xovnLoF{=6OW#H0X&+W+?clh?rDFEUK$)uPPKmQy|r@g zRg_L2*LlB6uCJSDPy+}~oAV5OE8JIS^q@ z|9qg+;MR3xd+U+Oi`bucJ6&5}QTn?SHw~Q(F||}7*3xj(5;FsRtUpkA;kM3na*lE} zTYGnaJhXub3?-eruj;LN1%)ENu3q;+`{X{&-C==vmh%Z!2cNp6Tj+H>*L)6}yq(AK z9)$YAP6KVB*?oyoS<8`eLSl;V)JJuY>?PuwZoSHI3I8ckN4mAM4>c!3ota8mk|h9v zW4ZBD_dVYStLV*?Z6N2$0w+75oH(=xu#5-kAf7?ITv)FLthUEj?>r3uzcdeH(>gCg zdeS;+jFbRd^)k@(UF?ujvJ9Yanzqf-Z*}@F08c9w@<8wa2%_m3 zl;NR@D~4ed~!~3Q&KX@cW+@*N)1b@;A`vlQsXRiIXW)MHKw`c`eY-li?zBvAj})z_8a* z484&)P^{ON4d2rsI}xvufl`1v=o%z0d;F1m#mIRA9kaJH`h9TQM5=YQs1z_~s8QTI zi~U`++8hrWs5KuK2S`4#Aw13>$a> z`t@gq(1+8W1M;nT2IGBwWX+1a^Fa*nn%{9%CP}=09)-W6A0| z{P`8D_vOr3Ad>@9I~C8z8Hr`j`U-1Fhs2L!ZSgQu0fWFp9QW@yIf`nXHUuSU>y;#G z+4?Vn$-81>W_Wlbm5rCyPyC?Ya$wnL|C+pHYq)E)TRIWLk}hx$uuaa51|8CkdGu%m zYagGYw}*f*&S9!jipp!Te9I9}%PZk{9t@YiF}#I8yle!NfjA^E=XnDhaUmYg#a;7r z@u^K|RtcB?Y)sJr)V#d`PZ!RI{)#NqDxWBw#!`b#z+9c{l|-po8IsB9gIP_J?<){> zsV1zK7v^t~Z?8u(=Hwn9a$SD@DfTP{S>rg~?@Ac%l;DnrfmpaefVT>8PJrz# zd5_;!@kdDwk2N%0vw_;qe3y}Z%w`*-U%W<)x9?0J zkhJSMFE4F(D3h1Z;H^o4}QbBF#!yehL3kJX2urM5t2<3AS$$0rHK1|5#3*T19H+nj2=9 z;)QHq6*n51mTf4E9VXGIH#khI8p=kCA(JH?t6wuL}z_2V_(-;G4sii3^|9tjir zbmuJ(id~H7zQzM?6i63n6K!gGKBHi{*3vf(Ev3{P!^7>Uicw~_v2V6n$2G^>Y_(EH z3Lj)R0}{;DRnl?5)7Pyu3;=TR`@Rpn#|HZX-`?txVBy?eCsOn&b=_#MH4p^Rrz1aG zW!nFZ;d6PXcl#pw{EvI}8>O06t0dKcD;5XI#fe|sb?qUxwUZ ztnQb&MG($0v%^#3^M`Yyi-N242Wa?|3}lKP!jhD|ol_#)asQB4ez)7L(>R=Haz))N z7`8=r9PRuut=G4C-L|&a)V`5VBZ+$cQ-96$3^LVugW5THywVOZP5w0T2qadC(s3&0 z^7oI)0tMBFN=ebr1vwi25>mY_5M4~qq$0v0;@G4N)ouJ`&FBmQfh1mS*Q1mKm0ut}`ezpVKZE*6H39 zt)!t7J2o4rO@~@cb>Cu=Pvot~pFj^4EPV+qSe$(iZ`_{cS`MGf`NXQ@=xnok&U)*@ zClCy@U_0lf)N$ernEA9JXZm`adL#3WCUCD!$1D?jFn`UaLBWyJrC3x1U;9FKb-%74 z7F;2k^{d=;Gn$-y-TFeF8_%}%{$_F>>*H{dlKCU>`2pklgqGAW4s}8~awklTk*#%J zrq@y+Y~i;x2MuG1f{Xby>atuELoh;>kQaFN;XE`IC20n|>^QwObhv2haD~f2=p0<5 zsCJk|TzD_wSDN5ty9=CIz*9rwsFuTs52P4T#cY4MS6^tyC&-;8`HKM)N4@eKg`vn~ zb5S_DV6XoXe$d+w|R+r_C zhTwLgBEN+t4TR?vM&{ngcSvHfu+Gw;%Ll#pi6Jte?V$Yu7E*?ggQQ3L1|icHOcd+^1{b))L@G>pL&7i(G^69*(CN40p( z1a^o7_v*6yb>$;VlTzc$hob&ir?hTG22={xj&bQJGyi!h z7NqE%B5!*ek2!fq2IlkbOMz6fMUk4$@UUgKCvRXnW+R+ka-xl67;3!VuedjVsVfs(c2iG+=1_1wbY7n zjmuT@XQIp}y9P|&uer#%tVEN$b=tmmpy8S8sD(XB3lYi@;?A4DGgp*q`A8j$uw~hQ zitx7pLcb;(C0@T|l+r4T&PWXw|L`X7+ebk@0ofT>g|p3F0qNnB!_B8=%@O=VZakS! z4#E7VGh>^K9~O~{b?2b>m{ph@#D?DnZR%)k3U+ib$n9Cj0__vrQiXZEM|U07;Teo? zHu4ju+e;(q<9`Ax0y{9qfbkRb9N)`k`XQY#`J1LVi{jIiyVZfAt4B%9HGbP#E34l# z)Eh7@>m82PsxMx$sBMjnFc4yjR?E?-k+~^6F@;6mC)UWEa5~T_Y&l5ajh^O10%)sS*FE+JRVY>x1->7Nv9Z=oVa+ zZ47nCC;krM|6tP-i5J0~b&1uM{~@~b$jxaOQ3Df1LS7qF_NvIJy#kC0Va&p*Z)GB! z4q{jx@|AWJU;i!nZiS=4_s_UczlK3FFKfXE%E!qEpq1-gMxviXu1FKKDdq~&k6J!fo3bh_>%w!75yk0Z;I=whyq^@qx_?q{q_3qT>Ao78Eod{${ z{F=p@I+bTMP=V)5+mL~?A{^18zC{hSm0!HMkmb^k?o=ShI+^0j@*b*MMtH4^O%|%p zMXHtR%EK!9{piG2>{$69c)h48hAV*FBbToCH*o>DvtkhCLX83HiYj`=qQ~YMu!lL! zC>H8DdzFDL;fS`qXP&Tc&JB-R5lf27l;qPfpVI8Y(O4m^Ehl9z$$gxQ)G|o)cDMAV zG(3ARLf-sDtmw?3p7a(2$03HdVtw;86XjMZuwJ2)oyKZWyxyt-58urc-OqZRR`d}n z$}28?DhU`uwWVAO962Lqnw^&8yqs*vF3hQwB7#gXs&Df~UI#Eb)2)`~m1==AjLW>E);HKdYlNnpJlY@8Yw$>Uu5N7`s2v>DF6avh;Xs;((P9b~l!K zyp4H6iUk;Nvqcog!+ov~RX(}$8@@;MfbwO8uIGF;ta+2_UAR;@AKv`>iA}{$xr`%+ zK@U21ce}~eIsjr&FOv?!=UVKyY0utQLKYb^OcYnf}e4FIqOc|4r zv{w$219rUxb3B6Mr-a;keyb!gdtHs6G>Q}3&m29tM%sMa#*={_ja5l0jkI!7YFf4{ zAIaBQ-6XvG!D3Pa6my&LGP6rwPv=BeLf5w9e2)t-<*EoOpYz*h7oAwcq-?6`*QwF- zBbYVRUqq*9otesiJ_f8ESE!vjW!u-V&xb~Cf@}f%zIA8MYey1akokx5!gWTH9Bwin zbA#kwWAw&24cA1o__P1z#D_gbQt&b1;N-2eUo(k?PX!Tiks?C;SQBkCW4d1Sey>)! zR-QeIdYB|<3$tmEOsWr_OnHth*mrr+%BNaC9*K03{l?w zUShT9_oib1bJdI*W^8k3zhR}0ve~ul3|k5mPNRdct6RE`)VaJpn+BEQRSnx^^}^iu z+)vUV>D>DnbM$i0ZHM=put&LH?6wWo^f*_)RTLe(vXWONY1P+55(InyJo`w^q~{YD zjh#^^r!s6fd>K(1$xRfw4oOlQa|8Nc_IOE7h)`XXVIDvP%Ix2c$wCaRs(T>14Lw|PdbhumWBGG1EB{BK&cAQJ*;rKdts>#kl8>Xgzs&*i;Xgpj z&+z|!X4W)hMkbfjH)^EbU~ZVErHo8Uf~rTCunG{w2hLO&5>)vg36vEoH)!L`hRF^D(>sAzMrb~?#_N#b_Wi^7>M6r z;YZ$b%MJbLwebm<>os3krVmk*3o(Ev8me8?VoYPubaYS0fIv)Z0k6@2xgCMFH@T5KcJcfRDvA2 zp{XK{KUkx0TdI)hvy0S=G-k!5BFh?@UDcF5Z3l73=|^8@4gZFFTk6@R4A1mC~mKe*~b zisU^S&s-q6SIAZFO~@Hhsi~1w-*$1i{pbZjM7Azd!Y(kcY0Je234@Kx5{?EcrMg&Z!Ip_*paO%BDTFiB43yB#kzlQ1959PBcCT2d4+?dZ*{!#R~i z3}!t-VQUmWoc4^>c_v0|D3oPvjyC$bLiXRT)zN9DGsE9BFds?daO}$NHZF?g z{97f`4JY5{PUOz!K3vP!Z|l>`p)BLx9mg@Q#R?4*$GBvdWl4=i*9bHS4?%aq@ON}?lL~`D2*N10L;Ya*7p%O z!su$7YxX?lW=xJ^{z(2GKn3C3v{ikX#}eLnQ_kGTD5cDjj3|S{Vq@#qp8`{}2H0|v z9S)g8K*#_eU);mk(5;>IT@OMdlb>nDxYBc82=(x&ogoAb1D9PFf}Ce%kn_;P-Ca4$ zo%MEW>)Jt=kr65Mvkx_L1x=P-Ey+f8D|KTY)H4&=4bn$=otB5@o|A^CJe#q~$j)QHu$NQ_e;8xpyx7q{6HKo~(qWZ}I_EJcVu! zD6ge1sZ5A?^hoPE=~WF=pwBxdl{b1QY~>`$LQX(&Z0*bUE^axCwb%EbEiE<}qjwr% zilrGD{DHfM12KxlJsytmEIjRJV~x9E0&V9vVWn10NEBwhENBQAemnt|c@BAQv+r;6 zK7;ZAG6MH#-6+rp%qbe0Ny_zz_zxG*;r7hXT0IQy!1d&DkU45lvT?VNTsa@8&B8G_ zPj@Gozo~2WQI@ug9TR0I@?mLCsli-^t7(Z*SvpzR?Z74!75t0A(B$?Pt5M);!W0l7 z^^VCeo1nPdPxuR>srfDyX-L|@+XhExNk#Od#rB=yULm2NiEp$&;5AC zH@D_r8ol#DR~&qfA%ow6#Kb^Wa|>xY2=pzjW7f^ttY=TJUy;Eps1s&by5-tj&uw?y z8PS4CtiJJ1&*0^;*6c0I(7YA=I-H3zpDpZ6?P0j6E6G9lSZ&8XS0NUQ_1;I|Za#Xm zw7K9$t?tm2ZMP#@+@1khvaadq?c-vlqk;svdd&m?xra8p#eg;88S}oUoqyocj*yf^ zP+8MnyRg?>Me7J}&R$~at*E6#tlPtIdU~+x>Out>7P!X!wikyNxaEd(QqCyY*$3@y z5jd{VL)Pbyn7wPYHMU5;)&JM1L4Uc3>3+3{Rn^$G1BMWXa{0Ecl{p>vXWChAAU0&5S4AfBKvsX&V(e*2S?_)sg|hFBsfeR!A&(gLSGd(8Jd@Q4cDbI zsv0%ck977uc=z*HX@&bzfw>enY_mT#l2OBR^rMY`&+cL!uF;)bk)bFD&}kYUJ=odm zPtN%BXFkLWISv&m-WsMf&A1rr<=xOT#61HV0q|_{8P6n(i~22Ucz><1t?3$*i65cl zk*PfQEV#Cg`Sw&y?MC<<0l+fXk+z(GWh80exQB=))>I5@l^PAlsBKQ`U+(l9Ms1-Ea)l57vGtpjm3L|}Ak`k@Rzp>GE z&5O?ipjdHHRT;{TtmMjgR5|BN?`@@L7}uMir3VH)Nh--s6r zSdb&5K7tKdx!q}ATb-GC-utjRE^Ev>Hfm!!hGD4i0T-@fqj4)Wc`y$uEG5QVrbIzA ze6a12CB<>s!2NRED#|Qv)XylaQVYEtONoA-422E?IhW?3ta85j2I}tOGvKSI$v?N3 zd>$sTh38q!P*;ef?B{DIB-XsH@+YkfjS9wnJ*%MZAf(}Wif?;K$~}mNkG0rxY;RR0 zs{A@(wAOPnlt(@2Zv8F&6SueuH*IS=Cutl9%1g;lTwOPaX5?Z~#g>}65L{l6w3Z zcroaKE9y$GO9*~QK3#e%pY-{n35n0O+0Cno+Vk*3_%S73h#`G}!+SVaF+v&YS*-Q* zfNLoOD4y(CT^3HvWyzVFkM`b$V);8_oH4U83USHuXNH2wJ z$Pcv#8(LxwH|PrLyO|-y_n#wD(b<=$g+LJXl_FNBtcT>=$ELH7EFCcq^<3gIEGyc6 zPZO;@B;oE-wIa>Q)UH%id+LO)Dvjn&>LtiT>0OtPBTvB7A@v)}f5AZqWT@#U>@>on0)3zTC~yGH~dXvSi|HWn(rg$K?RLP!5}y8ZtCz(Z$nl=XF+3$@}m2fC8 z0^{u8OpkyY;(4cShbkWdtZe4JghSPxI4@n<%OJ=Y1f1V_{t3)TwrqvXE%X(|14sGq zpgVBxK22~%A)HJc2HCN1jOM9H%D}k-kh>B&S`|+IgM{ zxc3K7B{{vU0;ZExS7>f&cfPbY*(n3EW_7n185x(A4tL{G6)|*{@;l{ro3ODIA!%Z1cW@z;Kj! zIEfw7NX_3mummnuWF{dRupDpdZ9QOqd+_+)BdXEbl@WgGYA!Hvy`TxO8WSVcH$NUYKacRW zFbFVSV$%k0%pES5n8Qz?FWLaMj3jUsF9jzs(_PO*VD|(f56YZ3XFFFuQZ;(GM(K`H zR075t#7HfT)>Zg__*Dov8!0(Ct31LMJv-M%;_Q9lm(Gt{nPkquu5Bab!{rf;uYsnJ zSDXdX2u52-;uX%H5xdb+_>_p!$v5kkyj$1`*7DZ(r`J4y@toE*u5lfVi&D2Ys@Fpu zWu{gX9*?rFknTto)49fr!_R+YFqtE!+I@C znxp7)#~+!|tscC9i#qh;qh7qmHMzhg$5v^#>$eNisB=?r_-OnxuKNPn_~R;)}4_!CXM zZkR44IQTYUWWCJpdLCyR6!7T@Ts%}6_2#-;H@glJJVU{2v`r}3!LD*bx%GADJ%_6lzT;ARoIR(2rlBw%uVUovgUW6ZwNdf@oO*Im9r{gK4=HO=MTAJYQ% zoY8WC=!cvzyb>Wg2)~l4LVx1tLg*A#hdC{(4_C`XEL6Lc-MR8`-Sa2A^zO=4py1Z> z4Bud5D;`mqejkQs;O2$uhoL4^gNCo(e-VaMcaSN2534!R<0VmyIO@ASy3wXa_5n%N z$jUF>eIlk4`uei7oS(70>V#CQBD@&4PA3a5?}@KVbix??q_v{H{ZyOk>#4ZJ*>Jqa z0WbNZ6S(o{L>QlZyVJ2N!UsnoNtmx+RXV!5n0cV`^w$ZB80#n|6KoB824RP_j0}U z{eGdEzwlj=ZZ2E73dpU2-V=K_P2li}doB)nt0&H~(3{{;jKdA5L}-2@YO zz?@>Gw z*upPRWv)y&sro44dp+W2M4))C1xeaXs*_r2DyLx*KroaHvM2|EB;+ae7rGB-fX!j% zmaX?iE;Q7-;?hmL0saHlhjk6U)X<{oIJvG)x?sdp5dJy{qBub_jE9~2dCKRK&0oGr zPPSF>CwfjN$rFcc975T6bAP5wbih%=rN&})vAxvxg{ALI@=662F%egAOCia?&kIEs zawC;D%yLSdbxBKMJ)!O2tHIwuhUl^b|IdCu(h7n0F@4Jqd%t=WodKT`6%Od!?qU-N zL#e-wKJdh#T^W0t=7}BpH1!^H+SmxEhWWrkFv7I)3A!x*fL$UmV{%)+C+*fGmKA4R zwmPSVG6<)_Bs4iDE>`Kx59H&5({l>PS5vKmw(7%34@QAwr2f=F@>HC`)-HgeoK%$b z{rt>G7Qbu46djugh$ssdrx&o=QPt%yKzXlv^=)o$mVt($Aj2EIwLjT24bMUMB4Fb$ z2owrpp~mns9<%kBN5~y$xD_z~1*+e0_Z*!|s$2-vP(aa!XQ&pv6%AGnd1Eb}BkYswB|nA`DQ4 zF^?^`_F^2njps{*x((=aL??h@Kkt1F9rj>K(M$$#>Kw7Gsj&8>&lR0I*xVwFO6fTW zln`%|YBGc3fJ)pPm}%gu{9^se{5ta-EKy#tj{Lk}l?!j?-2fOL+hAm6Go0W2X?$l9 z=7vY(xJlF?=5V1v-tA1zy4KZHqVa0t;J|~?9;D(EjVn0+G-nI|K{v56D;ZYUi7%lz zH~_~j@ObdT5yS>}At=v@)ffKZI7kH5oy49yjd+}#lhi~az)U&AuV6iCmA2-AaH@{3I$ zN5lQ=vRx6(!_^Ln8N88)9{{|`wSt}tGzZ}4RIP2t8IG6sqk-Ak29x1&dZzjQEH&1Q zGS59cLK`A`Z}gxCKuMVPydJDmNGxO;m8`u@zeU*r*ZK8^(CArMc*a0BD_ zwOXqE1p>=>P+sD=7%9Cne zZ$~4PJjoYn*@89*Ix6f*#l7EhCO{rRSJtlxtt6Pr(j0k_ZTkyL&iAE4a{5uoTV+Bn zvYDEa9mQl}dA>B7&*EL4>uW~y0CJjk_37KU!z*Y@iWxso5pM^itpWWza&4@RFrw6C z)HGYaFNTg%GGbh-dEg=^-=&e(dI5MM_J~4llUEe_$>k^c8}SICo(I)<&hBsV7cMat zq=;FstY|%J)5yv3Q!;331KZx)Je50a>$ls%8nCclsyZrQTac)cwY-O3ha@cI**u!3 zH5)jE&|HK1Ab(+%^Gin5@Dkq-W#f7Io4vPxwIC~KdnCd0tgMHh&tD7R0c zrejCJ?-4sBAA#)=n9?8{Y-_RyNo|^;6^N=>Px# literal 0 HcmV?d00001 diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/continuous_monitoring_pipeline.png b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/continuous_monitoring_pipeline.png new file mode 100644 index 0000000000000000000000000000000000000000..ea0a3bdc0a68d491258f78e6ccc782a9333dee40 GIT binary patch literal 84437 zcmeFZWmJ`4^es$xBPt;!NOuX6f^-NXUD6!~P&y7sh?E>aKpF(38>B_0yBnpu8|3bX z-+jmbzMtL?@7H_AkO6o&=Xv&Cd#$KuLVb>f z0e*wj{UQ~7MRwJYdyZ7nPqhJlK(mrkkwQW$kH)=zgARVic9z$5MM5HIhyEdVIOUro zA>HOG$Vh2=7;Vp?``nwj6xwIfp(R2^!6sp^vr}?;m*){*Q_ERfd$~6=s=UWr);1Co%tsx@* z?~#O)R4enpA0OIK=>K~t!vF8L&`SJ&ufvbK)vQ;QS=M-KzwjiK5L0RNJ+37-vHGg# z)w%rK_mdpmqx;}z;@B;-vVbsy;pDNxp92GGYP+}(IPDXrgG15~Wv}vniA=0+8P8^H6f!HJ ztQz{>=H|JKj3yDcj-(9dTJHvTfqrS4XLJ~|)#x`v{jCS|^jYu#u%|F#A*XqS>8}q- zXJKJ5*`cf#)qZsJ^sHD9*=hgV`YZqmIQ$-%({1dxm8->x~4kS=;`{QXLGDD*JbH9>~veV-uch-5N}pmqCm;_)tW^* z>dlv%Z1r9z-|8{%-sqoeL}a9eu>n1dMZeZQfkH}3qaK!_`$^M!T|tf=Ah&bI!ON9cB+}`2QKa<9O7rE?uRr1{kmd8qo zobfHUH+#d)zTQcK4#@~+^&hiGj;^b{v1$@GnqO5@S&@;EEm??H;v*@?m7YI`SJ}Hq z>w*ng{yte55XMYKO-+3`uJ6oLJvyP~Eblm%$$pUuf z?IHN9)GW9J1PO6WYRO=Wzyn5jx9GX?m9&fvZ+AX{#Yj4XX$4{6=e*sSuJFBj8y*#v z@TNUP;m?ovCjwa=QaU=HPvt#sP9_-n`88QJ^J$OP1`ZGmzx7oq@mTF8-oxsF&mYu{BZJGhyt{fln!jpceD8rll(+dMLSziiXcE zgWO>|Rgv3Y6Of7NQG%c?I&%|a(<;)TTN_AE+?{La3C1Sn5C5~= z6$fJdc_?|#>Q3##xDf#sCv76K*Q2{Exq^m)8y5D-^RSC)WiV43242K=H$dCS^Tlrc z8gaqMTH<(ml=Z8I9c7({_ zwhG~;rKQ`G<CT{6r9 z?7Dhjun|hqSf!O(60Zdg>RqN|G{G3`vNHVv2tk z+_)T48cEE{sS3@%!qr?9TWVV3gK{Tp{(tYv_W!&q4tUpm$?nqs@?PV~xW32=y%Fl} z5{}S#xyhWVu~CJ2*H5-Bt!QK_rfP?+@!~jeg*~(NuD?b`_zvne9+4`=v%Yp)#VlI- z=BGChx-;%AC48!2rz3TE1Jby?W4ZfnhpT=mVLJZl-b`(2bM$=Oy5r14fL~Bm=7L1=Le~wwcL8@7LT37AQd3CcLZF6fdZq{=9@2a5&T4rHVXqgTt ztJrselS~(zIcx5s(U2SIpVtD%DdyAd~ z8zQ8cdA}=%3lQVgv%^(3uY7w0kWrTpS9*p^ja%>@k)ElgiL!uaj`1ql4xIJ!4Op8| zti;B~R`}g`aj2>2U1eWMq*2c7Q#{}@Dk}ADgQH)H*Eq~PX#rR8;p&fAkzO_H!KBW9 z+EBj#+1$I6`j@9C6UIGTr7a~h`<@c#i$Nq09wctYprVnc`CZ!h7FE+k30n`+-dq<= z?BD)dxK+Yqppx)25OkQ%nc2Sm3LUWiV5s1ThWr+!NVWFKR=>SI8nE+xk3;seqeE_j zkm_j$zCN23(y7yIUAf@rO#NEN>HUxr!v;lVW##Yq!ylo16}*M@2@HUvMAU{6g}QIJ>f?}=r^&INqCXc z4#o`q_HObz;jc4Uc!t#R)9MZS!+p;vDq)QTF5{B3!q7xjhB z2r+9UIXOAI$6}AmBq%luq#n~C<6k?JaiQE}QE-sppLaQ-*>1SaK0`XKP8ciHw#%S@ zJcx=MEk#6s;fxNQAno5K$i!sw*2*Hh21CDNknO^dC<+?h?Ws0uq>>5yD*OBU@0>=L zTm}@Lti#eN38g3K*U|EM*vBNkwIwBL~alGx>`d2-lDnp)`NI9a}od@3V z9C*@m!3Ed@q01qw`}O;G1s5lAJ)&4&P$@Elw!lDmkX?b5|2XFh4t~6#!%Pw*;#E<| z(~{k1Y{&N`2~dvjcdF99HyHShzr{#LCj}r`2Rpz~FvD$Kkkf>JG(T&;A@j5=N5PIF zz^36bfCr+0^=9pa;D7;vH^^I#CIjGCtn$!r{(F3avfTguHthe`I^g6C4!%qhuuHfG zXW1*_3OeTI?I6sF$)3{qoo9;vJ6|H9qAEHIB1GTvMUX;?4*;NLIL-r3c5H0yRk5rb zBkvu7de5V;z3CFh^Bzn1p4;Ue%aKA0gICNQZ?g~vEq;doC!!`})%p~KN$~kBY`4`p z$Zn#aufE!Z zc=PP^am5$V0VIv&Dg085ulf(1&zhQrmqXd=mGg6-U2NoIB zZmZ;3)64ZR+AV4%1D!ejTX(FE3bHoiL)VIB%y42=2JSLLAs++vH?0AvU%x!UzB zhSjcc)VZ<95@2F(i9q`e6?GYc^@9$YL^3w5^}``Knl84|M*0*%2j1GR?Be1^^PX#| z%INn?44UX4Myvr?*^|f-@9TXwDQJ#lG!RPTXFcB|LiM=mOctyVVgiC`^r2dXS{(Qf z*$uyIqM``12y&{+HP8`p-(ae^GvN% z`n;g^prli#dDkbXBx2*?jb5pV9t8Uw^$GF5`GMy0?pK zvD%jc)j&?Dwb~7>0M%`Mi9)UHP^XcOa*#671FJkYs`qLv3>+9AR8+qxu zpaatQs!WcrdA%la8;e86YQ2Xpx-gV2XE9!+dug!rJ2DQmr=1OZ&4r~7d-F|D>r?Nt zB>Q*i6AdP)Am*TV3BP*4qVdzm9g32RPyqMBItwz|ZaQfE*NNWSjC}0^<(q+%Q@p1$n=f}Qww=b8M);YzhtYh%5u`RnE~+NDNg zHyVck!7wz-QmuiGKv>)jDr?MKP+5~9a9Zn6h1%6eQ2~-t*4Eah(Scu8Qj)zH-%GXu z6z@*69l>^f3J)S=6x>F=Mm_dBltfA^9TPk24vWopw|&u@TxQ-yuUl_g4-Sf!O&|)}JLK_LAzDTzwY1SpBo92Vd)rg4R$qf2hY~7n zuk?`I)08eTs%K?V{eoa~ae+NOk0E%J|0*x){QTV7Ddf?!w_%_aIge-KnkZqEKeFWD z&o6VmdMivpLXZ2PH*sgC#+KZROGiV5Am=1G9NZ+;dwqL*cFRbxz%Zv;Zo|A0YM)FAo~$bHfwctbv@ z5f3n9$~f-F(}CpVlQ)=%!3R-4@3Je)4`1}UNNa5F7YW8Z!Yo8 zr$@Vn3)PHwbjvJHSU1MA^G1tolyp9m-(D@;=BTDl|J_=8Y^54%YJk+HuuR|P67!JZrPWo673mI9+kgT3VqwuW*%gRR5=XLH%?iD&qjrGy` zh>w1Ybe{Wv={wmY6}>jscI3LC=~If`T7hBKt3ZHB9oFi1a8tjI@&t%*6Cw5axZtS?Go z>kH*t$7WFkfb2!zO-L02ZWR6Zr(Y4rw*Ok&f`0Y%=yQIN;Chv( z{3)&JO8#V1z%Bg+?pn{Ud}g!sObY%^l?R+(5|7sUw>U!NzDW41f=DxU{N)zQ_^V?= zTQQNdIHI()9C11Z1FxSo`}3{}V}%k>53&&Q@FQK|AVnS*#4#xh82RRS;p>;t^71}k z?N4KL-x`CCp0%#g^RmTjPqqMOrq3pXO{9Z%$d&x(wnS}pCmbB}^Yg3J(ArJiRrh*P zWxJ<#t8NEJM{&DiUP-z1>gwve7Q-~d+1gG8>W@3w$sSScPkfCWuDwOV#w#*SSS?+T zc)+-6ikLUzPgu8J8uz$?>OVC-D9yYiyKN1WnSYr#XeKxIFyE|rHZQUe0Azj@3!&D;6wOO0NjPsggvpE zy`s%I~f59 z_GGHstYKEdCm6xMt$AtN@ox?Z^A3OD7bJeGRd`hVGToOKNx8vk zlrApiFvi^v^S=#w9E&r4#6L9?Q|0V1lGVmIA{Km)`(UV5N+J4_MmH%ky1t1NDxM`5 zRAx>!!2Nd1{dd1pE&l06y44z&Cp66QsHlCz4Z3ack~werVtzO}w?nsiHV+k3wyXs0 z?(VK7iv^n3;|wO05D^oxejV;n@RH?qG;9;4#zJS!sGcN>9v;@*$Ze?Xr)_A4AFsY_ zfK-8aYilwO4#T>2}Yt9*aEMOST`_lUg?U|Hq!od+xi z`7xsB``MN(D~RA=JCcZ2Df=P}(h`Gee9u95*K^cy(C9wPUtHDdyf4rEzBMougV2o! z6{qIvT{q7!2C%HTk#|8uuB8@;flsM$MewL^cvSc5-Nv+v?L>iH!dC+PsRh4BTt-Io zysK@Yy&Bz0SodkI16Ei=#B{c_9k#I8e;P+lT-7?x@zTU2ybu0b64BY1AlHVA-S6Fz z-D#X*-gPOeZSCccUooBoK^=3BJgZpJqh=hL?;(D z&(~FqS_g~-#mhR_$u|tb5+fd`$SW+?#c|e_w;c06ni?-Y+$SSMBNOFD`*B}#P!a_N zM@Ojpp$^zF?YqPz+yu@}(s2hGi)LM!=*1|XrOhR$aWg~9?J0QEVU^9%KAkcodioz0 z;~zKvZt3gEMp2t3eBzVl3O+wCq;Zx$nP&oCw-ATk-5t1k9x_okd8S&%~C68^D zDz-d2Dhlp28*)2siOEwm6g;go^kbbmQk}Z{>9A<@W!*9qtY1p^YSHiYz9GV5&_TwS zr7w7F6#{T_Bh=j!3m4BdyzL4(v8(iXFL7Eiw0OJoU}H45aJ0|73~8{>cN3xD^`%XE z_rYxdk}Ga(B*uO2<+kl=lZK|O5mj41^re4fb{@el61lo`NNp%rNY716&<*X8G4&Ub zkwZyD#25j7eR~IdR5wahD4_=B?;Z_pQ0Tl5{At zq_BZfs}OC}lq(xa>x!$sI@jna9fbLC(}eQSL}}c*r__X_MiduIbXK0RF5nuwCMWVC z&Kt_hm`OY2mEXSjqU*(&R$<^48~%G^tDWSr3a35l0=?qo%FTirTOyS(u z(JI#A#3ZDS0bRVp(i4?FjjY-CX;>oL$jBI#6F`S*|wz^-ZHR5px5)9uQlVtMdvgav~ zYbr;bVwy~lb)-JZswB(L)a>+W(09bf#Kh#hgU6St$BEDnny@E_wlOQ!z3l@*xq*J) zUYc&OFcZy{gNRs|=Pqfnm?XF3Q;v6)XUm;*iT7gk^NL6!bv4^*JihkIa9zeA7=Mz~yagd>H!(pY(d<@6ef_H39koL+kzOK^Y z!KctcMrX@6Nq0i7WnRN02J!g*?m&CwJ(sq)`Cbh>Fhx{3j%XzdI#@b(QM2g`Hk&co z#7MWHOjf(*+0WJ)HG(1%dEiy}y7Y}B9ZW@p3Pmco@Er=ybWq!O0dKwx(J^v*95O{b z`biXA#*_P%Qj>5zM$!o$T%drhaNgZt=FU`;5-PgU{QPi@p?85^b#gu9)u6|P#q7Xv zvow>(Av?WIL&_QqmHu)pqR=DV+kMrz+g_PHuHY~As$0<^b{c^p5%B9wTn zk}Mc^^-=b%V;LNMN5l;gQcuKQw!KiR4RAzhWE|+(6ym?V?&FX_yoxbv6aQ@TWl51L zp-v~f$$mCVRFLx*Io;qhWMninlA=t4^(2$6Op~qLfIndS(6R3T$x8xEnqkSdKK)v^ zNhu&X_HqdCO;>x6pN&Khm-x!rOfsLr(Sephmm*j+fz;VlyccKLr~g?Wv@CS;4=KHM zQ8CM1v?P%OgLU)++H~aRVhl2$zu!e8=zT+&JE+t)^6DsV(u=KZjfl9O;BF7!*#EPO zvtOur-F=0*UWN*YLRY2TR_;I-iJJ_maM{R|&$;z$pf{~>Y4@6Hu&Ny`dOhP2?NCk0 z4@&c~Mj5_)*@;G30-KhJRHBZ`w)H??pNc5YY^ElQK2jzk591z)Gl4yv?lMB+SiU=6 z>~QYl2&fk093mX1sozhk4$>Y>^%t8+Q7&CjtVxx%!qyj&r>b1C-?}>veTyH>CpLGa zqql;KVz1{qUsRMEd@n*qGi>ouK*eEJO3NUw*A{I8l>E$$Tw`MspnK~Cx?}2M!#dUP zV&=m3Y}c>?$RPE;TQybR&*nuPs+PVjs!90dH9e#dvawBAjRc}v&uOv#^7;@_fp$wE zztj8=Ye|nJ{qWA$Cf2A4N{OZnPei6`9&Qai$ljya4KTtMZ*Km(w8WOHlsN8gY&Mt8 z;7v=cK7W@{4&VDaq4d5Tta#0KaKvWi*hhIm&R%BnHF#&Hc0&XO1B;4Cy)9r~W#|Ey z?LLxJaK!J+UlHms9#+M?&s7H)2UP8?e_B4UO<{Tf7t#Z7N9D89db{+vX?#G;j%a?H-+ z>^w1lC6mDU>UTa@w<2gRazB~o4>@X~+qP~)X|?k4O|N|D_bGR+06r7J`IP5yX4g?p;k> zVZEA1i)Y2$EJKB=p54Fai+@A+TH?L;>>l?~G3b~GKf3<}FwR5z$J~-?s77QAS-sDZ zNZKFvr#@qgB_<{zCeM^Ylk5-rK*nZ+lnF04`6fqC|KEz+0W~>OyNeXG(}mubI;%+a$X%@3aQV7I$2L`Tcf!%7jNBa0pUi-49b$)QVaJDc}H{v8tF`hNN z#XkUXcBaPXK}18twbydn{EcMFL4IAKUdR=u6E&G9|Xat$24^ZIf+3TV@3K zqN2!s(4vLq zWMp*L0R5wCOnFc9l~q1y!{}dpxEBkZqftmS()vml+#3Yr>ktt5?=as7PO$=rj-=Ocq)M~=cqdr1I=IRzf$$MC%hh=FAEkj zGV8~NgyonTfX_g&Jpkjk?Or%={iT;QaSzczF6N zJ?UrS*Zb1C-`xsIgdC`<%SKVL06zv#?JPe2yE5^1AcS@%8v7Mk6ew)$A?f%7WuD~rKA<&64o4Xz z2vBgr>5c;&+73@}DmbY;EYrdF1z-~90&m)E2fiJS+9z@y>O02U5b3&q5QYCEm2ax=N=2U3Zg zhB<2K;uVf_B6AXq%rtzKA9o|zhU+SJKRExQQIa)j2}H%*^s2P%|-V}PgIig*{$Y9NlDJB6jm=%;?T~*UWZA!w=FB1~R zi)ZA4){tYbD{*0Q?E&;l<0@_B67HpLSH(f17ec4C3x4`%?XPWNMq1G%$Epv-eLPG{ zF=0k`%l4fc{cpO`gPtGgN6UKG0uZ%!ygsF0dGzdAFwPg3htF-f8Q!$EN)4C4vPzZq z{awY99LCPddZe8c7N!xzHg7jxq|kWyvE=CLbSDn<2=Yl47AwCax2AFNXU?{#vacAd z#KybT1E{gkk>{Pv&CN}@a6TT%afbCiU!vL$#Wi^-2e`Up;h3}5OXZQvXFA-$t((L# zsY@B}ew-gHRnE2tQeocJLRRIZYCtiht}M<@xkAvBT*SU^v}i-o)w)4;0KV+}BdX;! z<&x9=`)?I3l~Y)^CW_+$MFjEn6$JP{2p)-amc40r!$Kz6)}7m?@mQ1!k7%GTZ8`WV zda)e$cWg-Je&^iWwBoMPqTaXtOJ%Ry*CT;j(urF+~i%Q%zqsXNR&pxs4#IU zu9;Es7+b9(VGOU8clH8&NCGATsF8KW`p1j)Nkb8fP9o-=F+U}VgqivwUhZ9Zcz7&O zARhJ@v8);?QwZ8K;8O~`-Ti}6aK1ZJDzNuM>}=L9p)PG}!a@7|bZ;SVgdGS9FY~IE zxw`Iy7LDDoqBGz1uf<|Un++e0@!en?%fIa7oBLhJjDaZ*6kz+tdrKjMG_3>Zh0(T0pnK_?N@f zC-^^sTH6`9HIeI{QnPA=Ah+(-e?_84?SFkhWjD*GBgygx>Y=LsSlh|$4jG|a%Yf^Y zS9}n#8zkX#R?P1>oFj>b%bv((IBbN0h4%(W_Lw){xFrzLYbvBePwx#;8-VKK*_%_Z zvR3!Jc2Y|^>x=~l;_h|x17=#B(zj=7N& zgop=R({P!q)6Lfs%fb*L?&0a;fkJFC2c(7y>pv*%@>q3Wb&Kz}0d1QpaA)SmyoWJv+06d!m4v81W|rJZH?07a#5{ zfKlb#JY>WcW7|o545HT`J=K0@n=5#p$Cj^((ZewAUYq+=3T``QC*O2=lB*d14ZP@D z2Tk(`S`@mlj!E0ODHrDiU*hrxR{LO>Z&&ewpqo5biZpO z;OZMNk)m*%t)l}(>gIkhpu;KbzuEa%0!yGEsLP6*MF%sL?ps=r(+rY$D1;o@$G*Mh ztJQF<_q+MtFGB2fm@_m?3URrPblr{rBc}-qD+hY)w<7CSdgW40whTN~zyh7#Go{B% z41G{oM{Ms*Hmh_eoW!-Vie@URfL7;Y+|A)MYJ1O&KFE)SNBN<^lnLA(5TEWwLr;JH zCzNIxc$i4IxVTghA6nbETCVBo=;($kER<3!;$h?Aq#YK!+f#qnv{fO$oz>&V=4<_r z{yL0KrKK4G+>_Lv2GBU?A|UEXsC?w%gw%-qd|a>G#*(2jHsj{Tdd%F_HJX+Q20w?H zY9S@&$G`yft&2#FN}_}*d%X#cz9pmS>uMy3d7I z4s(@YH^1=k$)!w507LZ&<|Fr5os!-=_|L!3d4?5319%M#p!yhzP@hd?P&{$@!`tNZCUEo&wE+PryGo z&7Qd}xyOly!uncK4B4SG_Hl*Hk{7wC+h^V92%c>|d70&4h$=hz7SHS6%`nyCR61Iu zs}vIXkmZrcThk2MxY>^@rw+gSDshdCY4yj;C8?(wvq-qTW^x zjELksroo# z2GuhPdPYSL`8ltXj~IHs2ZyE;dqottM`~X5W6jcXmBAHD!;H!e9xn-KM9hmitoL!m z@HZbu074S%*7AoCLK;p07L?v$835y#1;9`mz}f@;3a88JKsYN+1sDGJVTC+`96Gwf zPp?~-%#oz=fen|O2|Y~%QwqhbO~hj0+(km+<;$3=y}>T_%rDM2UT^|6$weeS7=mBb z(r%0B{d6)>EOX23ch6TJ5}7m28hP-#cukkB3#cA4$1(lOY_~I6&gTKQXpzK!@UGau z+Wr`II@0ei4e4_4#)zJvZ7-@uxhdnt(%AyTbuT#wFsXfXu2E6Uf0VY!o0|tGh@B5U)|g)~_PerZ{31qKU_mH$osZz7 z_|pb0p7{zg62Jl)+C7E6(5-hY0e zb9lRLFKK+W*M&%jgPz0QuEa++aqH_HL;r_ z#Njc*spX%>-@kvWwl}QOM!qvBOjQmjb$w!7NhxiXx5tw>n#KM5=MSB+|3$rlK#nXlP0;2AcM>?r|ACZ@#`B`8jcT9l;ge;O%npw^EG(czJq^JP!6W zdglzqg}O`hEx4N4e|;c}8<4osapiA~@V^c1m)W0UqS;{Q-5>5LT_-A5llb>Cj;MX% z_Tbw+P6HkBi*?jTvPfJlhsjSq+?=vD{Ki@uu0Gij`>;2S#j)~cG zd$VwB z{FBudK^t*6Kqy@fOD%#s({`$G%f4j3moZB*)d4V468Td~?w6OxJ5FB)F>KV5E?n0- z8|y1q7TZw6wP(U$GyRWdR-TtTUs%iuv0&gZdwZR3(*Yx(z+SHXaE=1hsvJ>UXlFim z1VF+^*!A~#52HXGO24Gk?)qeXgv)%*P~(~&PRqq=%HC&S>On*gs<_Y86jxaG+3xIa zZY{9^LS~q4e!b>HNg(0~n7nO##>D}7DNq>CT+2gMYI~^hq9(|kt@A5i+=v5SP}im z9HpG~bs?`#%s>>4#1lD!?8G)=v8m^KL3-7;Kz2XYz|p zWztBujcS8`Atonh@xM7qguDoG?Q}0+K2CKxLHKqOyDnY2r*J02xw2|Cnga16PRq@|xfKkWNq zRB7pO6Gba6p#HWuN$l15PY#43Q*9FCJBSG6wF7FT7xf?8Hk2n%8R)xFKYzo~uKKE) z#O@F1=@$|A;%j!EbplCjIn7%iL~OKk-}IQ@Y$|y(Zt-rtc3#oeaCL)+mseLpUyAwW za(D%BJ75#mswl|;rMw>UWQW0q0K_Hn=^T$k!q-@R>@50uk7{Y@*}DzfgqZB`k70wH zX`N017e$4d1>f390aCQsl|t~W7a#7u=jw@C(g%mH+zOf4;(jO{aFm5);-36z4;82< zIwF2rv%qxyZA8I+W4e!>PVW7(dZ$f52C^j~i{aOgXk$gW3dF)PfxK{Rf^g>AI#akb zWasLpOXM<6fJ{D!E>erBbS1{kAD%YzS-E(WJQ{0@78$KPWTuCrd4E7g2OUoWkg5{J zZ!WhqO`g?|FVriLTC99nfdS&Eo=Lr1t5mN5+p_W-Xrhyy`S^NDTmIRKCx;o3dEKw0 zhcZ+J3oxN{XqYLV7^es-mYePlXPPmQi^KTA)g#ZmKzdu@D){ijV7-zM)YDg4Pige0 zOK^em7Yi_QcXN|Bqqz9|Y}`6QuTc6Nn}0_%&4k$`7opg34Dcs!mK^#6Vr4ZutTUdD zr<4t1NSPzJN0FwStOmYtUY+fX`Jh2eQ=4>%;td9Vbpx9)vXHVxmq}pd;SYp;pMGcA zpT{{?d*T3XDCpKYMxO3ZmHDL#5&-87^AUv1yMqxS6d9RtBec2iXM+LA!ViQ}2?GtJ zfH&jg-WxS?8y}jj2`5@e&iP$cFuerH9@`F*!v(I|ESjC3s{Vkg;$&1xKmam5g;5x@ ziH^I79{8`=yOImFOS{+7{YyN(!f;r%RqA?xdZ>=UKx6xIw#nn~%OF5qGFfyZc+ME| zvdd9U*_{~ypFLbLCdO+zYP~U^op_H!Z}j|M3E0Nl!9ADBf>*0)nuB1%#)cYMTga*( z>I$0vyND!<~=9{?YWi0aHtuk-CsaBD!0uqUrN4{IJKnswzZAdd7p?tz?s8xc@tHqmS4Lmi;tfB8Gn+lT!FCo^WH<`!q?e1jBs}1$G*g3S&etG~6AzEX;#}K$; zc>HLjmt*Cqd7$@z4ihWwCqe8jv+<{&iYX5)tr)n4jPHa4W$6Gj7oPR7nko9AsEuwO9X06Sf~rH3}>gx@>}G( zVR^xapI+uU8g9%u1>gxb&#cqAiEJa-bsObTgMr!OjT5Tk2=GdngMy)ukS6B6S$DI% z0t{p^!1p-mW4;Lt@PNbnS4Ti8SsEa5d;N;jsL|R{fP!8wsuTR*vj6Q>`t|uzG;FaI z8MqcFrtS|eN;q8my?NSNWi!J6dH>(lo|1`41~gv)7RQQu#qrs@T`(8+ece^;=6K|i zf_p|fTQ<@h7^kLjmzvJzVuC(aiPJ)5)E}Nl>_$YT-`l8lW*44D-?x#4JCWmK)gGSPi zWg}xjEGUHkWi;Iz;LukD3=n^305xXP(06ycjiKyNXeMm!K*I8`l8DZQLm6Y(jyu5m*`tQ#j&sogtGHBTV z)6Q=_`4R(Fdk#w!C6H4KF4P+J}c$KxrTQe@tQEdsN_YWNSD@@7p-l9QGgL7dJqql^}jF{NRLHZ%P=H~ek?2)PlOwglTwio3)kxlIVn2t)9F?XS|cf^ zKM9RI0cxwodKvL>C2(nJHHFkOkuh5facfC_4ZtC@7gm&w94Z{NzUQD*s};&r4A}S| z17AP}vO>&Y6Y$(aHav+fL~6$8VLQg^YS#_rZ*a~@ zjE#*S@shbtTb}@9wD}A4K1gaS8~W@CwtZbl|Lg?SN-8 z-46K;05jPuR?rG-Ws5|#aF1YHs;vWC3*IBaS)Ed@yyt4&+q!&ig6~nU&KQmGpRC1_ zViUNclAzH)Kq=|WmO}>L$w}(PSQAH{1q2{9{a7e_BMmNJn!3fodQx8j_Z6HD#n|WI z{cc1+JBZ9Dz|K0Bi3z!eM@tO#S+xPZ!*RG>Ik4%P)qx}~hxUAX&U>#ZS?JC6`2mH8 z$7+&QFNaJ?-)mB;?_dJr&I z*}?qb@Ne-NRM`EDcT$#?p!%=9x2bvNzWX<(t#ccac@4UAm3SyZTFQmIxWl)%$h&Hd_jW8LYHQ zLS7ivw#m!p8i;PZ&!tBTxk$mJ&UzILAknBI$nTou)=vsLGT6Ua;i)F?#3I8cJ>ZJo zK&Swm_X#j|itQnhKPPId0J0mW60I`^`IQ|ox}c!LRKetOuEE8jMxLsn3OtOQEQ*Q% z0+BW(<0(Q;;XoQvO_HnU%6R@lm$E|nN48oD%uDyvyFA?~ zytoe=zP>n$^iOo>Ctm^`)m`~!dGP49y?qfLrJ!I>$tl2jbC3FZ+*> zcMWu~jBE+FED*mV{jL9tLKCu*a66a>Z#gQ#i`)cikZlH?z@!y0bXGXb2;$#gcB6&o zXB50Htf^nUbAX2ivJ}^t`RsvoaxsS^xI zB4?u`-VM$miQ&g+9{mH*Qfp_L1kk3vNdm7d`#(PcF2`#eybsRKgw7bc(d#-pb&yv`C0by~8ScyL zJj2D_Y6ttwER++aab0o$Vk8ndf%WOJ{p>2OoT_tJ;SF$`kERjoa4WYS0tjG7{?)VE z3(usXy7YCFnMvmlF!-1Y=12|)r3p8hPe1J)0hIJvf46n1$ZSdA56@F^FLsl;=?*V@ z50AR}+M20$L8>L-gjW;&dgHCM;j6P<8%fUho4vZtZ<1!_n`S}QfHcDRGYDm7%r~6V zV%~3iddWvD{#<*)j;+IhAw{1rjF5s?lY-x#0pd)~o^-1?!XlQ4nA6EvH9PZB?G#Gq zea(M%4h=VY;)Z|tPV3L9%O3aSA7_f7M1<#YI%Ty2<_W$# z01=rNcqXtmQcIMtK#COW;@2>%%nE>#%VJ)$GKP0^zYW}SXjbrM* zthQTTRhg*@ZI}D_Nj0^>8p9>4(VViW*92LZVA^i zcghazJnueeJG1zL?sly8V>V>D5NrnXCsLKmQ6Ri+vdO={-T5Z1xqsGSdLpyF(9$9< z-f+uV)RD~;gkFl>fnm@lxAX}26CJ%U6C4whfGQ?aI&@QB&abJj^j~2Co2tQ{RpQuW&EBvC7!}%44=B9F!dhII z`ge1c3;3hy5oAwX;HV<;#x%!(<39{BS)pv*hXV|_b+kAMK59U9HY6ih01tVO@@(gb zl4ynLj(k9pIg6$eW{-HlF-WST?7a@G@Tm=MBb?P|vom_&gAwd@RhF?FmuO1rRhQ4d zX%#*({QfpBHk&2}I78N}-lYgRjIumTys4J~YKHBuLYJ^Dyd7wR*6rsrPAy{~6KuyJ zJqi8U@>nivR94{hg*Z~76*9#&s1oB2ER0W;2~qB^4?8tQfz!!^!9zqOx%ds7qf1e& zhv%1*ogDv5BHx3odqS!{aJcQ}!;%1`nRr5axnQ2f+Q?hr;5SVm z5~g905cRwM{3N339%;5Gq3tNvF#47`pVg_dE@|z{x2aW(%Sv7EqE!&9V|xO7z$5Jh zD4b;P9^#RTuNi$bpENYFE>LL z76Vo!8((8ea)C?ajrqz%&=ARnKA}z8{`&&=uwYq!DaC0TT_-tw3=A^BnU$$sLZpEIO;@Rdn?yL<)77%4PhZGl&;$dw(>+z#RS&hX zBl=O?s7KAiKEunECU&RE{8tVw zz{rP;gJkxal+%oBrD2(|q%~d2 zM`k30bJ*Y4*Qe9#POlwmgjT&ZI^;za7&Ps^r3gMhBuUsTnfpZ8Y?NCFwB zS1c71q)IRm4VR{e*)pk7C0h7%js?G;sj*McU{q4Lib^&Ana`6<_l&mHv-BRZMMMfE zC83PX{^4N)Anto^+GXX_n{S|Qq;ya%?YyB#ul~omukagaZmuL4__hol^6JqDQ?a+T zwUH3}$<9VRNCZ@~;>$igk;wbR;b3F6NlF-I5qV#lNB+a?N z_wn`ZYoa3jpR1&Jz-cf@i<|()sFXbY{ULbvgyVRBADR==;dUx9P^$Y4R$bQ(K#jL; zFEqeMrJ(obyaaMNi$+dZf-uhiY|J3{=wg}6j}Thqufj6m<2CYGF)@<8iPT}A_5T4A zFNocW4S-KcU}z(I)UirDJR&~G3dH44yL&~Ho#kcf0|BYqNykWIyQp3AWi&sKR!ZgI z_>lMPf3bCzQCaO#x2L;ALAp_pkd#nB1W7?fq)SO@L8My+q+1#UM5IBwrBejymhSGl zYoGhxJMM?;r!&-X+|R?_`@hy)bN&{vgv<&V!~qcW+W_^sa>d)ZPcVM$W7yB&N|Ag# z$3oR~!zXC^8;DI7njR1BS_R+Y8^c5d1o~YL4KJBbZ}qrg_SyB!7rpjgpLQSFx1TI_ z?~za-lnNC(gM!Mw=9&i4cOd_Wbf_zRl{I0JxJ2c2`fLdy-4Sz3j*9089RHRyHK|^1vgoc!aEjAi zF6vSG7WHZ-eeMT0a%T*e#T4_`^9JOBwBaH8OU@$9#QWzDGI9R<%l_Ng$P;z}M=~P#IK*D95c`C}kg0>FgoRrwY z?^1A8Di#JvrWt&+Y@po65Q`r(3=Stx9zpWXHyJY@$*W=e=HMg5rT?k=W5C<+pD^q; zZ>E%}k_{G;r(m?S=5}QDYNuqF(JfKqS{cnL526y2(V864$~tSYnMQh0kXl4*nwYG+);#`JqHVM*-6S@&gp)q5ba`v;o-%YT&aRU2#-rbsg@xdO2_B4 zZGc!sDx>{rFu#TVrznf~LzY~#)b8UgPU*r_qYjn+N=0?h^?zdF+x~heUHPF1?W^0u z+!_)uh&uBN4THIYB1kqyZDm%-Ir3iwmv4_|zC7vnmB8@kP}UBAzo=PQ6t{80ExV&^ zGCNjAnc#EpZWv$)jSWhDR-czLVEQ#D5%b^i`4azC8j-5;Pf)#IP=Wv)y;gCmYZ^!Txmwl z^!%soTjz{PKT_4Y%BceFwBY?W=Rb1^lQ|PYdk6S%LdDJVjeCWr-I%S$e#%GeZ$ozx zsg!VU=O1b~`lGw-NY3fGcp=OW5?J%U=+wCdND=z);<~WduSzqu%^3@PP7t(VgcGqZ z+Z(_$?BERjGu=xFW75%fIs4y1dwY8Wqv*wJWfPQ-JvYV}U3S<_w?O7oeOm6kLt>67ju zTZ1DsCRw#w7eb{^!u`j!FbMQUup5*ES>~h$L1evVbUy31gmY)KU)b18JP9j^5xcd- zof2VMbZrp9rFPClL1Uah0CO zLHfaBf)~=WA50E&DM{Uy8T_b7L`Z_k;8Y8L@z5F_V^}tbDzDKOq{_}>iP-iiDBO|M zL9MM;al%aDo9NBO`1ZjMUAreK#9fyIYxCpLPvs{{Gt|$h)i;^wTKGJ#86aI6fN-V- zbzrulX5)Ia$pxy!P+E}mJOy$Ad<%=g)Wx1x;&FoZRmzxTl$4B034*b+%rbAk!_R7# z2ZL$p1Me@nPr_zkWd`wKeWFO$k1bcD?g`kB<6ReWsTShp$F-WeUCX>0hr3cLEaIiR zgZj!lN)~IoMI2V6IJL{3h@UPq_h@=CV%yXn#O$vQDt6`WQnxx|U3+eQ=l}S+pA$9w ze0p}c87>HXn7cj@f1+Yp2J9T1+048-yw^lNfyidJF(&w?FEtH_X1&b=11}(+`K|VF zVlt3Acq6jdbeNm=hwa>=674_r47qEQLjPBEd6D8#Xj8(d(kOJ`<&oZ*G(WNMOvIgc zfg_+LFea*P%=1p^xpo}uiWIHU!40#a+_<_dB+2YUhlR}59$;OZ9Ss8u;q;AHvriiC zsQiSS^yt#Uf=g4tBmgm-MLx1&BU4&mbdBwF{j}GtpxuOkonA;bg8ka_O^srwv}WnH z_U-U=U|iUlP?*?4`khAHO+)!(vH(>-E9-X?C0#W1oc6tXF_!UAt9wUv4nuXk!{JQc%tmC_$mx0%{LxiQ6Zw=9z|GTCuKK|0?a-W2b) z-g>a6f+~ie&mEC;yDZ*yf-4;c(eHQuwQv%0J-qv2zT_6k^QlL)FS1pPWAVug>q(!J zGlqBgoY*^TnzHbKuR5Hd((}u^tR0WRL81uzEwm!p8s%jS|kM zDWZ>u@-;J!H^v;pUM)uBNgoEZYgVpYNzb1`TD4W$OFC+f=YAl^Yb*IEtZqw$I|l7; zvE^8Wq(aOccHQcOu(YO6icK=XnE+IQ0)4()$0~^NS|)+|JpWivia%RABo(akE0MIX ziY^w z?!pXfdA~nukBII8!?gQe={jWLtS4%+rOd7g9B;(#h}z#=;8s_%ZO^_QKmc#bc5n1E z*5f`&S}hpai5X^d7vKbqwx!H}0!N6S&=RkR|J*5L6> z4x$m%$~sJ+@UAg&KRs})G+=Qhf96dNsv$BVH2EFRXsVntWc0B~CMXIfsX$**eTWp! zjepWj@2*UwW;uu2je%rTdb1G*fzvl4CF-|-;l~U80|`n{k*yku^Khth#@1b&O_Q;C zRXME5Z%w%yAKcOQvF@Y=FQ8+Go+1JKlkfgDzt#L{>x~rWSrRn`O4@{^BPRE@k%|kg z@z@F)z^^`OW<{_2E3NNRPMyEXR^x8x%uEER8Y!i?TLqW~VTc$3XvHr_Y}N36v%N&i z$r!ar1vii#c=2HT2sXwnxNLjJB=)~J8ppJlLv@1-y@_mvh2ynU0MsLuuh=!Uw?M-^3m}M;uzhV}@sZ#-x z5;40?qG4txn4fk*Zd6DTbuiv-o@V9be9|QwLYE#qCGUC8o%M9QlSexDb*2K}&Jk`v zu_F`hhbo-hT+y##wJI1n;Q z=Z9-fJufkd#n1hB;x87;^;ky}^6T-%Dxhdj&N=W5CxlvgE*5{}AU zbRPS>8(h2|6Ho0x_T_Rl0*hi9Vo}@+C3Tqfi4i*;xsG6PPENAu=U=Nv#RR*bRyyr`jS)L+o<=4R|J$}`6)pBA>ZCJt+7q$YY8@U=Cxf4m{$#c5_rlcBk$hK7 z-p>budrOp8%EWYi9`BxtU*v)pr_*uTja5ebS-FI9R5V{_9Tb^B8yzUlN^w?(*BtA# z+rdZ3-n6J4uJ5Vzn^AaLJ=g-xi&fd{ghd`i8w@EMi>D*RQwca9uH`y>kxE z#SVxG|6FgWzmkHm-5qdar|s0%WrNZ290n{Z;HLb4oQ%HB7+HvuG5N~#eD*9r4DE7f zbxl;eQu5H5c&cxF-q?vC#P3iVGAzMciiGw#79cX?VVohp`-iW?y!_3;)U%KX z>=|i^357FO(X?WKu(BI99P?QQoX(F_Bn9*{pdABm8i$z?S0*7DK+&wHt9&2T3 zO<#`F3v7bF>@mFEuTF|H`Rq>xHnI(LM>*NdhjI~+D&JNn0WMK(Z7n5RJ0lodqWMjJ z2mMS{$8xzxVvOaBIUlj^i?Q7QH&6B{rGU|Qn5eda#l8Ilfgf}}AFR(P^7{ZK!oHxX zFqGe_+8t>;+)ImTQq=$!h<30~bi1-kh57)`6nXnVv&f#7_2{~vRfZ9qwJI5M&uKj* zqbNU41K!CI6sIgdhQ?MxYNNH79?{Jo4hi?{rroV5b?YfJkmCbay;sa;cA(joP?Qm_H`*KsRjD0 zV?WJDqaWWM$>o=ux7hVmd+AXhhCM~U*;jS2oaT(@&*v+G-_DvSN_NNfX#GzjhWE=K z=E`U+Ym7_^zjq~K6{L+m1Ub(zmV5y2gf!IwK9wb8r3dnldA7OP#PklAqG?rxA5#(IO${Vp1lB1IVB?Zf9yUxJK7p-@Xm^j zg`46T%y7O_PU&7VA)`3jdun&O|AvqP=`Mfl&~OdI*h$sY;9;n5v4;mKg+|N7qbO$r z@~{_=Sp69qq43G9sjnBX{Hp0c@@U8YFc2FDxOjeh=gJ(bjm@0Nd+!w8aKl z8+3J=WS+|3sXGMbzj5ML zpLOc36Ns;NGbw5_t7o@%A%+ltr1BoPX22(N6xfbP*naR->>vv2fJRPvNfov~Oo;2;45q(>7||&mMCxcr##PcU5?8EFCOq+QQUm*A~!&7`GrCMnEB~p%BZ? zwKkL={ze9!y7rdMbnj?Lz1aj8bp3T_oA*z^%abV=&a@bl=m2Au(OQcv#Hl;17_;BE zTTvm55Sa6`FoPG@?}_i|{V9}bisj3KS@m;J*YI9I&^~Z3$>Zj#JVwVzs(v5)wdHdM)OAR{rc&sd2tNEPuV}?-awUBZ)`k=S ziydUDs?~OXLu;)``8h zA&M}yJ|u~9@6Zs32aNUJgJv zMK`OVz1yBvLl?`-%Pd=DZO!6pvCf<)lU7rg2Lu#6521oFgZagkHp`W^QoO)Q{!zFa zh4011{BJ>z^JA8?Tc%f)|0;{>51b~x(zLai2GtLK;WgZ!Mx*naga7aRIW*sXzRUZ- zcX}tv|AgajmcHl8SgqeJKJb~oS>8w((DexLFaBXZXx?A_R38~Zu(0UFV(u+qDajD30&&_F*GzR^+QR%4?N{EtMNNs zF&FF-7u&vx;KYND>j8D<1eEH;JzTOy~44)}99Qa4og zR&xpvtM`8C>+cvYVgE#kX+<;I5c>ApAi*&nC8gb?&xT`;acvRPlZ!i}S_P)7f3p$C znSJ1fjrVYf72xas9$DAw*T4x%Jo)M9-m709-d!mHSd5Yvf~C%o2;DD5(L)!PaBDr* zW<0{wYXaIG2fh1#xH$2pvXj-Di==jEh?NncOUNz%x@Xb)PQ-IF@!j%>JR-A!wDDhj z*xB0J+UE+a@3^1zh$vm^A?)$IdYhTJA%-jAVAWc=C`1rOv$Y>VVO#m>2W5kdZ|#@v zc)=H~p=a{>E&78rDy%_RmHH=kyW35na^Tmx)@8Y*F%^sUzd0>Pu?I9^+vG$?SuXKLOR1M9ZEsFc{z9P z2jr401ibP@1;hnpXOe1H%JW~4+sCU0qr|kg&)+`%mXgtSSej@sSb(jYcs`6j(RwIN z^h2^qhHw9J7F_WBqvfXMda6kBr&7>VCm=A=P&Tv^Ot+o=tbq#R*`u>?P`bMt*j$uQ#L4mS_C$$q}U z%?WDkVoZ&<3}=3rl!s%u>|H_(43!rU)Y%XOtKLw=U^0mlw~oK!T}{U4V4X;Ha{z-2 zAt}RMNywNQ^@XLuedoHvrrD1~*Bp^yj)>^H4jYHxVpIYTt$jff>6j2HnQso5Tclf& z67!ov!@)DaY5WsLf+Un^MC>#!PkO;AxWb3H1FnUKt5hq~I?rat8t;F>&iLTxw}^%# zkX#N}`CafStZjQ8SPWCN)v#V*5&V619hL0P@?Mlc?`$7KGQVlaRNKoHbZlaI-%o^@ zN(t{tJ-PH;IZ(beiksIY|DAEp%~`7O34Vp+LE%6H$hK55a@P(*bBH4Eu`iKY2qkA)HM}e}q z#8&Hfg?h98ixV^2vEOPnQ%#Pl1&_s;V2+?XlL>nRW<1g_flX%hQpjd93Lp^+Fi9u3 zv7uLzKJLg$>5Ai@-5s+iSSjiwkwzyc)TFUFE6cO%F?%KpFyHE479C@nFvIav6f*kJQlS(VJGV^`FI~_~eDfeMo!H z^2-C@LI)5Uy>pNym9lIa@z9CJe&7jh3P2ve-bOua+(qETly>3t+vh{3{BFm#+3E$! z;NU%3N*R6_c;J$U;p6pt1&G%Jy9n@I&t--6f@a&aYN@poNYSt0RG~o!Ne|1?a49Z2qxp{%=v}xuwX39MK)w7;2*g=qsCxg=lQ2NwZfH1`hN=U&^~qRt zDr3rVrVCbb$@39xg3xCti$Gis(HFU89?Jvz_|4eQ%BYgSTOj3NKY98djP(zWNno%s zFAO^afak4~<`dUIY_!r%Ot+N*+Qpa9h^(oxz2 z$PhMJ3^U^4kV?zy-}2m2!}z+iv5UTA%p4co^%1>z8uQFav^<{H^%4VamHn7Bj_~jX!(j>)9JCy zv9B#BNE!xApmJMxfvNi5CnYOb`RnrG+!wz2>LN>-TO!CJ1YaRzSbDXd@US6A>6*WEp z_XoVdc(`PrV$44VVQ*De3gGpU{`##~oO$Cv1*tOR+luB+09cogN|H$b`aUTWnlI+4 zUowWo#eSso*5*T^=PvNb>3n1n)xLL?iEGc(C7GjkBv6Y97?|t6=#PVzAQ1ex;FMWDEk(c8kHYBh)^(5Q-GxrYqvYD|i2Ecg zlD`LF=3r)nQ!`an$y1i>&llQ=)=73O`6vvf61OiQ^>LS)3P|(-MGBvL_}d3eBkfX% z!H8v z4ynqG7vy@MdCUEy*tOL_gEjk&I#2kpfkEh%XsbQs1HNF>^C@m3suQ5Qb$w}(xHXJ= z*DPQn2fW9H3Rsx}7+fV%%4>~N7}E3G0vfjeDjw|1U7EDt`dZ`5eQdU!K5K)Iq9KW* z&A{O7Y_39xF;n;4+Iac5)$n!Fckz-0><_n40hac&7T7l#))!dt@dW1%7JAl-Xa4;G zu=hfEQK%xW{TE4du+U_S5t6t`-UV6|)zavaoU}aHqR9EV(bio%+hK(;hCqhGqwQFsn)(&%aS}$=InImD4_9Y}v)_15WJpm4n`-;J-NsroWS-zbu>C=U8E3T>M&-TnJX!X7$c{p^IJwN@VvQBd-e zRdZG4x2`|m1U8_iFT=F_whT5r$i7mo1x4oz|UYC{NbN-x4<7l~yts0hf9S%N&ozFNi$N z_ci)bQT%#B*lzJ`Qe)fk_ScC<@CIx8?D*DsOlqb76K_E7%%9M@;TFH z=-=8x`>a6^uC0DTS>;sDYiMpr0d7I(wEe1tWpBF@6PrGQxv^1nlI5}oL~LD%*OSi; zfeB_J(%%l=Sy8NVUrSzHhz5I=TMg^2Evuvl*BDaTYtKrieZ~eSM;eb6T*9{zh@QY<%QQ@ zrKF8`P|N3QYF&@os|0h5yM2k*1j0wX8E*L0w1s?Z-!W|4FKJr}n99SUG7!ePW(?rg zs5d|X?4!fryoLk*f15SkmEc}p9W-->_8ppa{wP=A(N9bYIM}5?zofB>&xbHB@ z%@)=V;P1!RsSnC>&JTkxBsRu8kqFuO2QD4ro%lL9X>6k|evNkk>g zLq6Y!_M}oHQ{m;t+hiaLoTdl_(rMy~JuaR80T4VH@LW>R3o`aA>7gi%V23pO7S^*MB9BjI}IExIv^DF{b zFxDK+mRb8XM7sv$VDRNxyZmvvg!RNzSoXnRximFo@}~FK{-m6S>+XC&^ArsfZVJ)n zg91)7;D9iej^c${fiTgN(TE>X}C;rJZS3DU00dKuq%yRpL8O3Yf&CK$(`J;)~9R(}b?Ylr4 z5qKj=$~NVwKTeb4U;ysc*7{eDE0E97jJUr&DHV!~L0HH@;nP6nLd9VAtW>L%rUzAW z{s)}g?~;Vy>iE&3xLjAw(^49S=#_V#Y2!5_DO(k@8)oY6_C>lK<*W+>SmM^fG7qXk z$A>l!D@`Wb4)3?g-K08^VWuT~Z`F{_sNY>-lRHke3%A!{8Q&cF*%&6KZDtDL#l*)% zf_6r)R?ncffgC*=3=T+?#?0BjZ#jkDrBnl0ZY#7`C#UzmImlaaXr4 zw-Sxy#>2UxGKSCw){FAj#Gd;5c-#B?{mEz)_q;NEaPXXLPitBF(%*rHY^=~`A|pE) zeenacEKyf+rXpC|FUK@~TYMK%rYXlHgdw=X`C!f|t!x8|H#%oKucY)nxV?Ua(caXw zOAxU7_F@%Cg-jUuwrKYci*9$5Ds&}TpUU`7)(mbQwaPn1wAPLG4d)!=w%)U!bkUK? zpSgKQXJqJM8<}(ZthH|3aH}}E;bhrlU6D<-UFf`>QxF-B0s8cbjv$pCl2#qj;4Gpd zX|PHxVqma^l!yIO>hs`m4wSe<-U|k9+nKtoq7VjDpNCN*FzFg*FAmT70rWAGH$$_o z5<8)|q85_>k#onKre&YBO8iR|HGUQDf=DU=WrEnBlwobG2*DGjB1`2ri=f#mn#mdN`3!PhG?8 zI}GC0y3Wz>$fZ&~VEa8<{~ZO+?akMZ!|h$nsR73S!Ju(=#(?sy!VnWbx9E8JbLc0r z{J+dOb?$Sz<*(=kR(m*313C`7-T9^(ZdxUdw~i&KZxqdR)|6UvE*A%WdO&5ZmQH`h zvqCpI&?@1BV#3ovj^Qt8ZH4M>v9q(7><9ry6$R%plWPxAQ1FMIe{z_2PCb8haF6i& zv*%cZbgyGf2tT26Y6RZbxqCqS3#jEEZp{qG4-vrwx!|rmcVKrhO1ioRN0I|U0zq+ zXuD9kX@$3bh5v>^jY!>8*(Uen&m-1<8iuiIMbxIpo>CtG8p9h!wyIM$1plfo?~_Dv z_xMl`9h!+P!%0-=v5=|7`7|!ta-OcO*VQEr0vj+MXeCsE^8m3fX~qj)_e^1kzM~#P zM0-DuH3YxJB(<$E`^9ZRC$Bx@eIv<+hrZa}(WM@&c95zKd(=7uuB^VwkA8&o*}7*y zut(GW?IR^hjYhRn#ℑdqG9DVYWWET34A*cy0mdERI91Z)>=up>33em%zY@msQp^ zCv-cLagX3QF=^8CMd>)at5Z?vtok+4+jIO2*GvQ@L4$_%1XG+r?(-k?)cpC(l-LQdM<#C;^YrW0B5#LKU3e|td`pgyRe5j( z$>;AsB&Ph^qLz$~LA-5~^53SQc#-2!+M z^&{T4e{*=^Xz0t_?sKB8GFeyl{+}xEE%p>0l&v3|{T-*;6qOb4{F!DmUQkg)PklTu zw_V#f=$WLzGxR>t9(5k?p#j4Q2`Wp#{;o_2ij6i(gBii73%Uj>>QnotcX0lPg5N)- z`1HX^lxDoh@|Xl9Ay!7KLvP)GCv$vJNXqiM@d&%t@Xkx3zi(R6b8Jm8?Eoux!xmab zJ_Jh-&H5(&y$McMrcSYd;P>ykWRF}8bqgjeq^7EhJ+s0?wY8t2s2}w3?U{S*=J#ed zM&~rI>;$#ePEiFr2ldpGkY!Fj0Ta<1fx}=vi-b*6YTwX{XRR{gBKU6VU(@fJ&qZ3z6w$Rsrr_ns{dAvvoWk{8b(_2P+;m;I4L%yhg2qzhj(^T)!*1$YT8;cR` zfmu!U-M5ds#$w#)wHGfk&N2kaU4S|q|M#HicCvt?euC#to874+<1B!LJ;r!}g z1}bVd6O<5!_-J3eQ8iG!pAFwZasH^bO$4@{b2S^*kIrAr+5w5^V{jaxm~>K*tmijK=NT+9l`mE!zlYDO>Uf*rNd>D1S^*0bhzpIcl1 z?&3J69x90RZN=qSdt@h#)6fD4RkUT)^ztfcMIFoB%2*|c5dPbk;{r|O--ear+S;5- zFs~iWak=o$(KJFQ0SbX>iGX3Rj7(?H%8N|d-+C&{-@j?DYHg1PG&BZWFT)J5*m3o5 zKsE9Cu2w7Rnj0=`1Twyc$wWb1=gk3t^hlXyW7bl>Mh)$Rx6!>f>7}JBm zLhqqYEjQsoAuO7g!V0_D)YjHEr1%gW)e#N7_mq}pi5zgkfOw2lJNgOYzjk~kCBN$; z0Ze2!`+GRw>b8gaCS#@DRyLbK2^&a-m`m01nLRn5-2-?i9wOolcr4^Wd|+vR%nSjM z)*-R0_>fXrL%rd*f%zvUOyUBof088V1~%f(G$c`T(=80m`$mR8xNf$*BgoFDCd7ui z>A37uyO{Yd~0TOG5lw@#V^ZqSy!1laoj=$9UE9n{7}dZX`66>xx54n z3B4(Csv8aK(k^Qod8lY^jsY};RP}-6-LU}HkORX=PyhbLM5WaGjgkDD1$g~ieUH<* zg~1MgDB2LU@CYdsFxRL30HxZ&3)0$?H~O>SPmGEzqm|p}xs` zi=tA(a?PtB==1N0srUt$mDvVui;(2Yd;B%sj; z4P#XECij=y!3hHYtvSn-#=J>&X*Q#+dzj~-yY7op8;h9)4nGZzxetEN(TQc2kP_1s)MlQt;9I z&E^?ijaRm}SHmgCj#&K>vIb8>+f7uTewUX~{+7u44-wd43n+2E2U5RsDF+0Xhca{4 zZi{vkVBnGPH2Baxfd(K`vxGc;+6!4$^b7?i>@f>nG3J;$)yldhQEXeU2qdUZ)%#}3 zsZK>(rwOpWEZ~sLbb2cA=zD;STl>RCcSY8){zGL*WuT%<)ZcMZ*1o}%js`X>$V7%g zH(qEeiTdFcM)G3STJKrEK2fLp+VHModaDw$IXRvnnoWpTW9DBz1z`z%JOvotIM8W- zo$&eOf$TS;zL_(CzT#q;YsEV>7qt7g>Ox!ny!1~&up99OYI8&l%^B$e3%u{Kx^3vl z$!|gWWRWy(gt?<9FD&ZlZh_Q6=h>O!$?j#hjB=EcC$}!=wb^TYwtWUlQ*q=3tcmxG zi}gfoV3@C|(82hu(D3a2sOh;0Bd>YMVo#>t1=gb4*Tgmh??9=jL*N>~Z2!9+En_F% zlX1V3`tUQRbZ-=v^`>Z%&dDXZ3=ppN_}3P01XZU%@>=zWs%F-j@^5(7z@whW2kQL86 zRJ1f@3w%G14jl<39?Bd^`uZ&k+uRx9akLRwJP#@AIZHC?2>uDMHmpp)hFnn^Q3-D2;NXNQM5Bvjc(KvH8ruwn zbw76d=qraT(!WN-m}QMd{KMldzx4aDt4Z#<6S^V-S_fxo5_B6SP94^E)FYtykcXyaR{8woYuHpF`h|6 z?4jCmBdT8PAV~932zRHM6T?H`EaT!z*lR^5tL4+M!CJW+5a-)D;Ie`Y(#b}H_eHVG z0^Z65?A^#tIE>x`Kw8(&>=)pOILnY9Ffn1z#1^54XSB}{1D`|yxKsq>72jd>{ziQ_ zJiYY`>%0OF*B^)^0UQZ)&hOySYulM`JBF_S$svgf6Sbs0>KCN5;%I2n@7jQF){&ot zLtbd=%`6py?5h%ZNb>%UszUTGGK6E{;d$mt%qCh7NM!h5KH9uG#jtxnw`jRQFseTMIGq*OK+qZr2j$vDJ3}F7( zDBij*TZiaj0y9QPIrUCL=6n!O6H4)Wdf2#%MCQiJts)^wKt-59JElZzze&!mqR~ku z(I9A!0Ta*Q4C5}>46EwH{tEdQ*)OX@RI^oHMmjpXy4v4Dy|b!_r||3 zBsi0$7Foz<9aH$ae8}^(-iWEgU}A?$H=YoW3HyWn7nt5CB4N;b5zKzR4tQJhOH9b> zN21s~%jA-(n~t%N+F#pFvs?J$Cctky725mJS}#bMMI$Lq@<)qU;P)_d;P&?}+mp(%d0 zf1TE@FV|U-QXH0WfQI)V-w?ZxqACrY#qfr@?AdbK+S=c!rARjI2usn7QKRPN=6DlgWg~ zd57t4kK?pOCC9I`EAVv~b${UvXnZoEwZ3II4Z-%cC7<6!<*sR-rIhoI1$(U z^9@+HeQ`AXl+zQuPCt{v&OLVP(aX|+#o~NQwUHvk-&S;)JXDX1Wt9XQ(v?|C@hH^w zbMNv2eS3KIW$b=kp&IMFxm`7y@(O>qS<49kn>v#xs~1`Tl|HQR`m!~I)*`BwrDUml zpjl#>y>P2R5*M@ru#w~pxE7(^>P+5W?G?J(;))ZukJ-dtaUagzs?B-(p~`-Gs#fOW zxQ6C|)5Bj)%)#SQyOT%Lx9R9Z=Q=pT4SzYlf#t<~R$~GP(Hu@`2qAv3=6LZf$nCfh zo6o2Z1zC~|Stlzk?E2HyoPJnnc=Frxl9}G~yl7GF7Z=@Cc^iEs7>#dv6zKNA)3pG| z_rBi@V(?qhx09V7k#p;OKXTlh$S&_l+mzBB{`wgIq~&}1jx%#x^ubw7{JbYN2Tx!< zs~+Y&$)Wb$+Q=ZcwgA|_*5M0U{|jKd|!O;b~~%~LK1WD7u;rP2aka9!RpKS z3X*`z%ZttG2hOZ^Ot8T)$~4(oW=#zl2t?@RuL)W9q6HUHGpxYViAKpM^oSm6rm1>2cJ#lJP-LbhU|aw*vm^GQ?} zeDJT@>pqVI`Tyr*|F{oJuz_A!@b}{Mh(64EiY&Q?ROB11s4$WDL5V!YO!qg@^XTdSW~#|=olSOgEjqpapuOTH7-`fE6x=&p7mx3TaO zrCrE)%F>aoyiiOl72+5-eahLVz^~eSB&-tN5S+l z;L=K!OI>uF!3&*T$Pb(EvTNsg@cUr8|Bm|dR`vxg=V1YQTQ1nhV?y{LGO;lkJ$Ur^ z#IF(UcDNp2lCP?(-(OdiSMu)T?z90HE}~2*z^*XBW4S-oUDs`Aai{t5HpL%pFaZX`#%vu_n`C z`nztQybNcBDJ-hr49IBKzlsboZ8Z(OZWOenbm`d8!5(qubzxE5iet$6^Wf;onlR$2 z@w`~mKczSq`sx1NZ7MmZjfRsm!+)gpgdx5(?i!CjX+Ob{1|fUO$wnJZEkp`Rr7s%}_9 zi4%tST;HgEgQQ6;?CFSyuuY$!HOQTaSM?t>IhyhJ)EdANu{OkotmAP$ja%v_?ZgLQ z$G~*nb3-~F#s0Rc$aaxD{c5#|!0N8-zGngcIKwMx;eM%*xXr$H!)P;FbpO%V`S7je zxv?VRMJ*E5($zL3)pmIInzZc!$QsZnQAObOFl|NX7gGi}2KU`>al-^WHeH;{K4T zk1scS6GM_2j7yY6qnxW%X36?E_-?(JY6Eax?rl9+ZZg##IS$%LLu;L8s*TibpwLcU zCtaW)N;>g{mqO!TDAnXW9QX7$8e$n9X`0xJm$ffN~Zq#cJ@!V7KeN@*oMc zOY;A^0Q`33>iyvA_u%S}{=0e{ta~<7kNmEo_X~3LU#e5Rig!iSrvS_oLKKFhA-mSA#gJ@1H6W%%Xfp0xV z(^F!|U`10QyNI}tk4I6n5n7P{-#?CbkC*Bcq%qNbwjA3am;n@9v=Nnj==Xu8@@bE# zj9wYU@CCxH5dMOU$C!x2fN80onW&>aH>`v3PjHJ~P)2V{KgSgA!9(1H*TUJQclyS&))>;mN9JQR() zP-i8$ni56Qej}<~0F2+)l(!HjfCP$ZayZzXkWlpe-|QFi--=iif?)x$<*rfRJj*jw ziK7YoelP>%Ic+?amTm`^5n5%=r>Qi{WhkU{Em8|?T@%yDQ) zVNI88q~y0n@Y_WL!E4Vkq`xH@QyJFMz11qZB{7PbXzRY+hz;ghnbAu7EKP;ck5Cgf z07A3w#?$w&02y{#&dAi(T*bdt1V6$r09U+|E*pApYqDo@QdcI3f)Q{ILW+EC3hyfn z$M2PVfPLp|aA@D#%3KE))nA7Hd|S4!D|Jx^CgJoHkr+yH4n-aYdq^DY?CvTEJ~V@u zMfisgAHE#%VT*-_@f^T=VYJL4M^mdX8esXGl}mJcdYu3l;6Mr+=TJ5Wnolzs&rs`k zTB$`ZIk@C#4&Zt_i6zbf$^Sm}K5eT2HrBVv!3{`JjFpD?)8IqP&{!2oqv_^W^L~5^ zmR%s5Jvhmn7fOYG<{ZM#2%DIZPOuxZrnv2UPsw_%(HGQlBYQiZ0;jwS`Ue&y&~xc} zxgHt%GZ!p8yGNciKjxYxuJ#SNnw~H~4oD$_PryU}{7m9q#owDL?yosG)Sl?Mtr%Ze zyFpY+OAX-Y2qgVwdVrzGMOFLw=@Pk zmK3qXo$zZiqc&y5y9*Z!U^)!<-?l|K{922!43qf5ej%dMd!Ac&eY{TcOcF`CCGvU% z-n!(K?7E}+oedQHXRtn?BO{>RLbK&_w9(%$*eu;_36{|>`8s8&jkA~T)38MSRUCiUp$&{Ep5Jr&P9sjk3Ffo@Vi24m5+fDlN$&@& zgHbKDocC6gJhp`x^rCD?q_KF+VO8!5)fb#o=n%5d;mUbYp1;2)@eG+f;4f+$9|RSn zf`{KjQ`j>(W-w1(|MYO}E(CodALB1tXx5$*26#hWt_tYoOIZtGm_M>QZP%!sE{stzlI<#N--e+nq|Do%w{--70Dgxn7#6 z=o~XXx~N&02cgxnyrbow1-aTqt(?Y_9aWJIw+lCoCgWc8CW~=Inq1wXD-G<3(6yg` ztrJzb5W;nJ-#q`h2?Sb{JgzQ|1YB=2tkA7aAUnK})foRY&a16Q3ez<0gwF2~QSwOx z64oTW+XL-l^D9!d29Y2#+bKiLL{FU}G-ble z;GTF3fzReF?SGZnO&3ik;~aAD@6o~Y{~~O)H}_$x@zFs~u~7Jx#m&JO?|U4 z(L4-43i3~s*R_6{y^2wJZ#iD3bo~IR$eO+v*^&)(-Y%WauU>(7@|#kBi6ebUv-5&^ zuMaIiAU8ha@>z6c+(5^S5`)fa@NH_gB<2H$Rn?Wzk}M=VmThu*P~f@pa`>--8#Li% z9I!>owvwJTx zmDWc~WIZlNhkQHfOj4Wyk({Ah{hr{)JcC;dP6LBnBxO$D|&G|bRZ>k-)UQO9sSBes#8Y*hnYBtrx^1a{aOa=zh$#xTQDJi^e zOFd5BVq#AULxAMZjV!G6L~lHb7dc=SmIc~;CfRm4v+Pd}+E$?#o4UF{p}d?upE+}i{Gn|lp`%#^7IBNjb+o7N_j0)l^WSBHXsq%fFh}!5dkqhMy8-WuznyQ6YD4y` zNqOnY#=*7T-0w6LX2{=3yk|8YXVJ^0?u(i01w5$=1Lkq9b2NGU-o6>+!pGnrb03b- za)tt_Tb4AO!#e-(s;B^^bHaFgH{W<<$?f1HG89n?j# z=(QrK1Fi4ZN1oHUEHjogJ0t;GLi}uO26;V+YwgfZ(#<67Kf$3CqBb z=D4G&S{CIXqUjyEX6PfewDjEp*?S6AmPa`JtP)r?Z_?4_*NWKMrNk$S@4!V?BTkjj zgqmw%1HT{l`r#67Ej|%>jch_1$a)$JpSTkQEZCg=9ZHi zc!&ZGdNdHmwIJn_dK3_0=+i`C5W($peC*^5$bp6ItYolG$0Xq;BG~hc31r~|f)Sqw z&CW!Av!goz>@yIQi-8Xq&l=Yxg9RYtDXwaf*SjiUJ~aaL^QkT%y(=^oC>Nen1 z7Ge+;a4iS)hA*s@xA|{hn)W7!10ALv!7oH~zhd?Uh}2fhBn|KKwE(&Y9kG8q9XG~| z#rdpwd3kpm(20??%YwKY@Lk4bl!ukVVxMQu`+sBsiSvjLL?48`sG6(soPN*q>d4cD z$mBrybV=fh9l;=5G0Oz?Ey)oTP^50_?N2*dV8<;>Vf3qVuVT~VP+@Y>B zlcHLB_rKVB%eW}JsEv1Mkp>BoQVA(RkWQsLL|Q=UknR?h5Cx#TX+_x+vo;piuknR{m6``&x6wXXHQOakZ|Q?`Z0!H95YZ;#Bu!9lOa@w4dJdim+` zr%w!H^&X|-gj+k%4;cd>gJ;#1FDJCYxRPZjE|pnoStuAl8DM)gwMYT4qjbrbs9k$B#egRtxE`S=vI6G z`CGelj6?6MUOne$u|I=`;oejMBO50{VS@ZQ!w8hfCF%Q;0Yf5gm#Kc+d zmnG^R`UmqOTMaR~c@2l-w)!DCS|cX>A+f52^MC&?Ykhe&YlS34_3LvU=m#L1dBpj8iSv~1SZ&Tm(eDf&{NKrw-~;S`+vi~^NC z41Dbwcw5aLThn1Zynuz8in!1esXFip#14H*X%IjrUohRhzLq`t6Ej4p2k_*#l29|x z$zP|^=ILHv0rX9z9x@68OUgeoJ>b&IMM?A8ou=Imb{?-?GwALOK3_AwiUw?QJE~uB zCg7?O(vv?XpDaLj6=*fy7x%N8WH0Q94u)nf4+Wr|reN4huvscC3}K_FE|GaDEfzj9 zT5hT4qO__W$;8O`yUFhil>AGm20w!9uuaKF?m&c-)qQ&>`3N}ok3ND!rWEAc z8F?+cW1BBel>oQGcP+@M__)~7UJ!E4d)KO-m*l_pKid!}|AFmBnNAcC2}J$hWNM6LQz77pT<;_T_P;q<$sTyPo!(&HJvA*c`eN=qVnQt4(3!2V1vnb@u78wpjj?( zBofbz&t|aK&9_9*gLwp=y_>W7@f;3Plt+{GzJ7%~yR$T>(;~st&-ME1%zkODN@$m@BT_AKL{BhUOo+ZM z6)X>o3|Fzrj{A)xG1O~(gwQ)(O8U1}z3=YdaIydzdY0u-$1KfXXOC}CX4u-yo9;}j zKW7fP#*>a$+6)~O(1%O3_F(7rtC)pI@l$y<*@D>@LEls7e!H>Eo<)+G4?evn_!pBB zt!;mmdToLD0F(OL@1!91UBR2Qd^TFaYN1Yd!X@vkj`s0UJW|1=()786YVf}^>Be_4 zu7e4X9^sFSe{fUmbElo3El~#Ew8XFRzu6Rc)F-G3vH9oU>b;>-smKVpgj|7dC-Ua&I>s!iEcHCb5Thcn6D~L`ZW%i7#J5ZXBc@@E zn$Y0%viw^ljIyOX=fe#0DH?|vrC7~eRCLj#Ym-Tal&IbQF3ZYeGg1~oZ5gX)JHHg@ zN*Ly{DW@tN7x7u*LkA;n^2(|?6Z>`Ed$`>fDEAi4>PsH^=O4H?GF!$9&if13dCt5w zgNmLNTu^;7V87Eiy!@_*grzI#)%6=JasLjntnSSgveZ4yI=flIoms?A&8fxxp!NA> zdr=)3w=06>bKr7Wb6NbyI_$G7RhcyNN898)&X3>67#mJ~_y70dtvC52P5G4*CrvA3 zkGYd4_0)+P)@kaw`_$X#E05A5O~(W7w;Nl^Mt(q`A1W10%c@bxujn53sCc0cVeu(K zFC5XXJ<{Vi_J~Z6AHH}m`DI7q>Le2TpZnG{awe(uzevpCLc3TwpzU-F}7fwvJ zzFXHCHFm&qQ-Fu3fATIXCf-g;;NDZrXMIKZ&kn%M_&H(z5t7cb7VDAGAWKcuTe`}B zb8#fs(C^lS;h}6|`+4@cTW0XlCs9rc=66ZayYPdL@7+1m*LxT#pD@4dOG zKJfkoe%nF=Vgx*RuWmG8VyG8k(ua~T=$&zXs zCml~}rFX6K(wAa+EnjNq=Iovf9TlcV@n?enrsRM308=xxVf=!>S(i|vAK8qacD^u)vb7qd|hr;N*!%DKxL7u$!5OSD>&z!_N)M63nPPxdg0~;zGe{OK04Din%h=EBeupF7K@{-LlaSH zBML&1u=CReB~wXd_&yS%PSob6FsJHI>YL?glG1jdVZ@MDV;L;EwtI)zx|vYWSXzyX zJwI#?k4pBf0bV-25SpVo%}McgnzBEQ;cdB6geVvFP`=bxHAP~YDx{I`iH4rMOh z@eI!VTXi^){BA4dIP9%LrT=cZDk5^kq>*q^f1-FK^rH2$G1Fd_;bu?BvWyAVnO@}A z2%O=)c~Md{<+Vx!Jks_%thdGK2hXk`u}x+y(k=0^g5vBpC7%D4BgfjFEHQNkTGjkc z4OPet^_KuH)+rWeGgS5M<}J}=ECAEyC`X=*H6)b%m)V&jV0kErra(yk>cUe_tL`}Z z#ry0msUyQw^{nW!(w03uHc?lyUoRC;FhXhifQ7`OCXwQ7VWrbjHxN%lL7=o?4?2PV zq1yw(=&Vq+S2_fLrc9W(yC*uiNtEIsGnkSU)% zFNPN`cLhG0ES_^b!x-lpUS{lP7M+;fu1=qBb_kiVKBHHeRov;a|1!H3a8>W_R2jl9 zbhEJ|Vk9ia!EmK*$lKQ5$dI$bg8d7Fet746!*x*CK4qI7{Ic1gRw_U5R;&1h0Vr## zDS}R!ghbzm9-R9;GQw4!w|!Q7cVuDJ%~E@o204_%2LL0bq9!-&F#>=#4MU^}*#eVC zkMaX~*_(YROz>IjuuR`kws=m0S-s*dd_Y5z|92I$o}kHK{8=jhKn8+f9|rkAyER>w zk);lJgHt~g zv8=q9^0KsJ#lWh?7|^>@%ju89jX|_TZTYF?pw@dTi990tZg6NSfc7mAQ1s{^Kn3=( zUjK#nNgp9!0Nz~Cd%_2vsxDnSZ$m=Ximy@iCqEHCp2c`(cyoG2ibxFXKUb;nSzPhKf(@+gwNF zB4`wR@$H@PC3{KEo@Xpftmo{f7PRu)O00R_`|hH*zLga+VBpXg=jl6HEfIWRSo1Ej z=wx(6n14FqU>rHm6o5SJ33n5v^H8&J?+=L)~m+?ustKd+QGK==cSV;VTS>1 zDrga@*?*e8Nn*X?5^s_}jQen49{K*g@;T3ouQM!Shvc{s=5CgXM(w6w**ap`kb}By zUQkLQWxu=bAU5pnFx)ikbO_4B_jR%K4rp*D)o*Sd-{aXy9^f+5%-pIfa z^Ot8cSC^LY<7S7vZ1Rb-y1vJ=a~9I&VHxTRD5}*Y@ES3uul8`~aBE2|*oC zb?zpLYhn6#bjJS-3BN8yF9D8ToYkxEz!wPAn3jKSVeS?+xr^F^*TN}n+uoelSE6YL zWo6}P;TIRD2dB5pX3DRB$Kn1HagYjJ=uiG7R7Q>fW#Vje;V^2@jjT0kMZd{y64=zU z@SRFsLJYDgbrN(T*8UJ?rW|{XHHUx`ib2!)t`eu(hl4T3s|mek96`c(Fy3nC?Geo~ zuc^HJFx&W7%wW3)+3R`G)3Q1xKhS#b$w6`M zIcrANMl0l;JuN51_Ds~$yBH=CPNx^4XM^9*1pT*f-jQdprsHJLt4>EZYvWDRi_>dW z2@+0Au#I(|Q{`wsD9>c-xasxOBgpR})EkDO{d9ED==IM_%)aY{CK4R?l(u%F#?Ju3 zsk-x34{5xi=Z3s!cia_zSh?JDjfSgo=5E7f1vPF#+K<1U`D!~<9D*yqa4UuoEkPUY zx0nt!&uEp`q-iF4d%sl831GFodVlU0wk~87kq9-te;)0HZJL@8aCt#F!mSZT(7QXx zg*}yhZt@ph6P{r68}N2CYW6rT(c8Y5hU%!1jNB)Jo@OvayGUmWnfC$TK}cYbt%bdB zsl91L-#wJf|FWIdWm&IBZ3ZbSb6}dPSt}{ zlG2+QNWUpFEQSH^nL|Pv!Kek^pJz6Z2^#)taz#hH-m;Z(v7DBAFgDAVHT`1;d4X(x zpxdlggTNW)7bV=fxh=T#g1|m^976niGVJGB;dhO?3 zmB^!3)isN`o* z9d}L->)lc@K2Hw+*P5qdilrZ;`V}}<%|tzRPiZEIKY=_*xA{?l!iEQNq{4=@=m~!V zHVsbM-c52$Wk*i|xh|5|0edxH(2uL%*FCJB^LevHjQ&zL+IaM9hzT=#D#&#mlEp%u z@x+PPL&HXr8}2I-7%e|h6U%8P$R9u~Pv+}5re4iVKMRy|LK4sf-1c;2sqt~Lb*lbL znDdxm{ISp4#rX8}#B?2=R{rd!cCjZoG?E5k$@d~=npk*ONTZ3zrtJ8y=zGNLT`b%! zJluuG_PT|(`Ywd7w2!n;#VhQ?!?2Ajnkpoe8f8XAur{TFz6)+IF+S5T5WK=EE7iBv z($4yD7)N&TAiDQDE3oaqDk$gVNX+iHO@e};Yj2~-K>x({=VysehF#?3(M`r8h`G+R zfcN=UFeq_#Bz%pLN6-hVlU$dPVYTP%_tS)s;rLPyTUuW zD#JeetGM-yT~E&00Qly2Yi^h)&p!~CMg*wi(Ft4Ldrq6QIT?{n4z$jhM~BNf&DIB7$o}P(pH*iB zet|w8XSoR*D$$2Klmif}0qgnuHsSr7TZiNYZ?xsT-3;ox$H~L!2E3>qE0bzkkL}~6 zmIW|OkfS`Lx`-^#LQ0#MH-07QmJ~+q8(MpO`6nKeAMpOKREYRJHvO5$LEp3I3hJI% zS3467v-*_`=jf~bIHXRsZv5O9F^(lx(i6FP@nhf6W2jW)=K+jZRkIwfHRI@(&?+&; z#YE25l(_2SpNCS07()6w6RoCxQ-|Ddexzo_24aRB?GDpCa4+R(T)Dzpri;ck-%=Bb zT-4E5v!(rf6|k}KqE0@`yiT>c?^FtdX#99M{h_Df2q$R6r&&i!ye7TIOq? zOX0!WL&36{ETc~Y!j~v%i{7y0W5~^P9(t?pH(EEE5?NQS{JgvI$f9`!0@vhbN_Ug7 zrxg)Hcm{`SL!aB-8;fRe&-VJC2OO+j4Y=G}-P_-K{$kDi2N5pee2W3Q39@rv;^Wk! zrh~rTJtWn>+`G$b{q$07x5=aX=dDKc>hJU^GTv{*d{J2O_fGCR{h5Tzn*Iii!?WIw ze)1PZQTuXfbaGeuqf;xHWCy%`extY~Zeu5r8on}qep(gn^tgmE!Fj8yITJQ%IXy$C z%59(5c1(Q;G-NI$iYI;^ob)wjB&M%*pAOX$4vmnE9MQ!!9VTAVt*k}b-fg9ra@h~@ z9Hj4c_wq#z^oX@zaR8-j8gooVv%$`Zf3`HEptAVBkNr_ZpSN=JnfHJ&->RvP&$5x| zsqP;;{e;momuc14*zU8EM>%a})dLr6R|okO$|GYFLO67;Gw!wGRgsInG=7%=uI{?h z$vLVkUhIy1s@&t^kMu4rZS!x|n5l0hT6VFSsW3dsYu0e{0xGD5@uBU&fqu!^d++WI zH*c5C599}8($69~{QdZs&HWRtQDg*~9*Q6HUX-!(-G3~`t_ELrK&1ja5@YkCj1*JKbkP29HZHy@HzT zeH1y)+?&}=w#66)_X!>h*}A{bG@eDm%1rZz>=nA&`3LFAsM&3!>D0Two@N?ZM-sjK zJ+70taqGjxn>oW=j;83lzmQ^h6&gBc$mn6seND|CbVLc+#FzMmh&r;s`eWy*Bw}cJ zOyKfpu;rAAJf)EI?JH*>Y!Lg-5x5YB8G;}VU=(afs9p+1i)uhU{`rAuH5j;rXbhL~ zj@J;OMAg`M2>9a=r|J9yvCU!JP9hZ~g=i=I-(T@8jg6Wm4qQG_43dIhMZNlUsaDH_ z&zqRDgiS^s%$Pzi5b&dg(A%xI1I8&A*K0+M$V1!K>hFsOzth22r9&;BrM;b(6MI;83i!|kMFk5HU1{SXrYyNq{7i^ zK}^2-ziT82EQdk|SOS!yQQ&2i@!Z#G*IQHC85j&J|DW#~L+Hj5Gei=L;-sE zK1kKSDw^g_{Hr>pSv5qIMBgE!(L}?LJlF6*-{vdnBq?`nE8Xo|x8MDikJG6_Z>#qh z1Vc=!2Xm7iyV%W`6Se@Wah-}|Bzlt_pcGqX}V2>`!6dn7g)E>jMd16&zsqNYt1Vvh) z*)}r1di9FOep0YKL!!BIiM~KHCmQ?<4U`bkin_Xq2lq~PPccLE(QM7!SAX@WFwyv* z<<_p4at^rFsrZ6k?Yp&-#3%!oVS--fi=au%1bz3Zy(6b= z_X<5{%{ApXv_Pj7aPBNz4ckG!fQj7&wTEtwhmw3GCZYs>`TuTWgT$&TLgOP$^ROqUh?qU%zwvxoA!24I64tuKk+U}xOTs=A+X5D&IEPg`Ip1rr38)!+MEBl5 zy#wicY*aPGY%pI#8yDg5mmQaIikjoU3EHHI*7gRTLy=>M{E16`lVoWMUB)>NmU2C$ zw~zi^RdW9Ry!OI!Eb(_>F=)KfQY8Y6gx%Mz;?A?Y4_0}gl3HL}J*+w-ox}FxCBrY${9N=9A7+RmxsKYy=9x_-fBwghlJxhNqh4x7nZ!Em^+Z{TW zGgq6nx!JUCUORuh-03xOd<-z(g%1MLU*a!GNJ(F88vCD^|L0O-ikOC{&Pfsr|G&pj z1ny_w?bgq0mH7`P-z^kcaayO<>oni`_gsd&1khl#`*bP&+b|&39dDu zQq6LUXayZV`GQAeu3_F6EQHS9DrN{g(Zm68V~GV&lOx@3fAN`{y%y}{VUZxA(2JDL zRZU|fpnezyt_n%!DvSmo&to-iYCusGmzG5L|D4n!qJv@mQ6Q(LUae{`?E%T$vjDpn z|Lq1O3RN#Mr>smm87 zhdb@Ep1lN8s~w#!77#@2_oyWMdrK;S_EsJ5OXS9a@Ysf2%57kr<{TO@A#?ssl{Qtb zXjKS*f1f6(9pYuZg?xi)&h@O0air9^I{vTRYux{R(&M1azoxOfIBfFYqPO$*x*#)r z^U0qHYu#@W`AqA4Yv-TYvU)oq9iFj@icaFMm0?sk<-8gLb&?V&V2eo+3avUp<@N*= z|J_*QW?|VW+!oq@dXA-^X9Cz`?*9_^QPLc&R$QB8*C|OXu-VetvJopPA2jQT4XdNr zM$hou{fO`F{{{9hFSj5mk3L1MfST2~`$)qrx$3@`CtvYAVY)CAtg142-EDoon|Y9X z_`CnMp%?Iz4p58bU?sw^9UiDUGFY~M(8FjL|9yvAkK;-i1zl|cvJR{Lu_6`?8BCGQ z*)=!I$2abi9x;|8h;@jD98AO%rzih^pPl)(h_7_5Q&8661cwWO0nu*IzkpA+1M*e9 z)#CxM(M`E6e~Z63;nOEcTY^aymmkOI|LsR&A=gLu@&jWc{%2R(r&(3)$=XyFM(s^+ zVB6ba(l2cSAuGz|BGKA?2r7x$ASKJ)eCuOHE3zK4h4MRLeq{0cEgni>2!w04@=VV_ z)P(woOX!D7?%4kcT~+^op7eo>e32MzH_QHA)D$Smy&gw%DTHLIN$y_QhF&g3SaQuE z$VM6M`~l;1)+=bDGJ`C8nai>&n1DMJBj?zZF7@p$tHdtW8LY1yaR!R=y~5*n2M2nR zQy<5$sKgq~%?%Gb@?D%Qv7aPQ3cLO7ksyqdC{>BYd;Y+%MkO?)~~cTjjk3NDB?$%nfhPQkw`?>)xOnkF%CGX zJ;C@_)I@(6rz~_;UwK1pwfhZ>81-P90z1=S8ZL?dUha9tIeSurMa5HRuGNV*XQykW zi_ieYEKi&U0wN_)45@5a0sBy!m2@6@dU*maMZM0$SWNA{Ksi+MT=Pr;lEcNMHRAw1 zW3((W9M*(cpN^%5@2NPhz4d~#bto7#iAik!L)R_sib|) z$FK^o^yej5CVQ5VXGWH@1D5 zZ{Jh%b)46HXwO!NG25DMMERl_Fv-97a#@y#6i*J&CA8ZcLG8d8aDG{YPq!QW6kVtH zx*+#pGsR2^By994Ol7fK}jMGI?=jf8)uItVo>|y(i+zyI0fa} zZHHg+5bTsTpFRdNmFFG74f|y$gjPf`fQOlWdN@Rxp-?l#dGRx#@s9OCZXB1y)!_bh zb&z!zTKoF%jIn&2`5e%=n+%!g0IeF^b!Cj z=C!+Gxp;o!vb)A5As4&Y$*LxP@E&#^RsUOb8o4^%ID(fwR;6EO0xo{WE_S3%7;+C^ z=>(ipJSbMX)jd8i*O_D@H&&Xf93myUbPtbEob#^#Rafj{SBegupKl;A&Kd7sh5mb< zBw#G=wf~hx**`_&?j^VR2~9xrnZ#*L!Gvri&GU|3%6klv`vIK~huhv$TX$T?#FTFa zACo66wB}iNJJ8XuE6&KB{%I#;F)9-=*Yuq^a>!`M!>|()2c8|^EfB()N zXKiNx?_X)>T+=1+p-@TUF@N`zR1E0=H3PqsmAufaHP$I`3pML5PSR}K38<3o($}`0 zR7vGl9j~$tm^nRM&IndG3lT}4A(d2~q~i$mQAcATU^8Lx`ky6EvI8i$LvW>4}@#8pbt(f?K!WWGVt4X=^>6-b-#Q%i7aWac~K7L=P|D+ zNDl*W!|<}4GvpIE8|F!fdDk4cXhnh5S6eFfCy?K#i8k=R0v^GOW+}$@RC(AO8Ihn zu*L$4l>ADjRBQXxf+x&LqWrWXKLgxr!Qh|+V4vA`wY$v5VM}Ewq%*8D24Kl*w>yY9 zlKR4sl!dNJt<@sAD?63+Fkj?VtddJBhatD|;rbZZ1y;EYHWDTM$O#>Lb#Z3idhXSq zZOzMS{5kU}wMX)(Hd&fclW)5+nf@n9wDuH%j7o|ci6AWTZ(1#Ql&GHSY}i(p-rXes z;seL(_Ke}QfbJ9)(Czc$9)n6rZfeD1FioV+grRpwY;q;wj?i-Y*Oj!_sXC_vXgfC`6|sYBAOB@ za=~jJTcqK^rQL*k^()sB=jV)) zc`c##fU=}4PZh4ru$w9y+i&S#dGU1cp-KO7!H!qS}RGs%Ac;Duq86V_S1lLU-N=RV}kcKd!k17J|G=Fx!DaYs*2 z&t#jdrJNTCZ#!*O@co)L<&F|);@F<)?=EvOfRkXAqNrT+t>ibm{<&%y_Z=E`g5T4L z>4U?DdE7TjVh+dG*@b}{>@O0s+I^|X;Xsvzj=dSL@K7R z{q?cE^w~V@2gY8Dk6wmp^9W#;lbkoKv&ACzPfn9v zQ?eJ%z4tF+PaaMV=-!*X&D9ZW&1(h%wBh8?bFRbXg~`{7Gu3`Zv3t&grYn{d29ey$ zbf$F~&?+V?J|jNkEfH|(yXQ4#?kzec>oDyY@zD_z&#<#rbJrEC`0UK@hx{1mb(j#`_I!qnY?FUz zlLb{f%uKJF{(67YqcK2jz@=c|V7ye$az(G+{pCVkq{O9ey8o$l`R2@4(_IvS0R!hS z<=}LD!|n9kIN)L^wvDmM^ykozFtV<_{Y&WiTjCw;$$eU+y|mZgray8CqBo0MViD2kx#YhpJ|%51O^`u*!(doC>I{N#(3T zdC4SbfO_G)CtxNaDk{9}=Z=8f!RD06u7eE7N*M(O1&4)Ij`Ca$7JZ9rydk~OxwoIY zEM7Mcap*&5k+@#?g0DEh5aqrk&F4zYD6b8jc*m-8gfyhUDr^gEt4k;PKg+0aMEb z3CA{RNq*o{>o%fKaA5s;I^UTV?Shq>SmXt(cE6*{l=U)4jo;cOBt+iU#nh2)W1j|S zy!K6I7|e~VOdmSceqBhN|CHKyfX*1C@7!+4lE?`AyvFjDWsGH=s0$6Od zl)%#dIzyk(pD4-l$BEU;3W1Nq5_R-`c2}1}b^HDmBOkE^$w(R9$;BMtAm;(CiBztEvYE8_)YNYdreo>QUd_Ps(iy4q z-!$qM1`X@X!z-c<8v2g(&%Z7uiVjNqucRRC<8iZ8c1?No8as(jf&P3;Dv8_r-Tu~$ zL|OXaZCrN;Cgh)uf|i5Ie@?DxX+Dg`O}-Y_KH3-68l&7B%88~D4k`MlRS#{e#!AP! zNwTCy*N<}D;$2Gl)CP|7rK(*$6PnUdGEz|Z-Wid(;J7@SIm(f4*L421BcKt|Y#l*= z=7&!wIx4E>V`EKgyQPNF!1ll2I3)3ZzcHky_9nm5eV5sn`wjNMU9l>s^iS9YX;TWX zy!J!hT`uF}<3F5OO;o&a%HPL{kN5+XcH4!jy5g(W7iW*M$TRFM2NpC#h&B^|ncmLa zM#Q}_UbxS9@665GykKpg&r0S5Oe<6Fe?C2}%p87@cIdQWrlDA^b6U4CQY^9WYM1YR zP!XB;%%xH}GEQs1(TXW2fbsf`dnRl$wBL&Z_A~t{pvEE_dik~gC>u%!^xZvu7uHyI z0Y+bG*@e9?<4Qa(BknCP=C7^ysK{KT1}<f9RRCvBTDdfDe%LJ^N zYP|@e)rkNOqYCx=!EfXf?5B{?#{+u>jSrz>C)T=dQn&&AFYg{$g~z^rey#nDcFWhZh-jIbTlhTNvNOm^E@jitstKf-VKg#7GVc z*-3w33mAgl-AnqXl8{!P0X-z)#x@eRxVSmBctJtvi?gpVv=t_;xQ9#B(n4^o)LTwm z25!AVzvoSHd{2de*ME`z4`~w3|}pn|DbPH=*#?>?G%#>uUGeA zrh}BZvvQKCM*nfn$B49gT_*W_FBi){P^APYQV5c?<=`5{F={vJ0PcLYY2JNkf%PuB zfgm1&gaD)3v?z4R+ZJIWW^GK{-bbejIV-MJIo)CoPu!TPR-?Fo{KWgnPa4&>8HZl- zlN*y>JIjI-x?E!1Y)5pN5P|=I7ceeHftVE+C^1gh-~Z+u24-4G1woNCLUwV861gue zAvvmP`Zk04Ou*3iaFoc5kXPu>R*~vPM>v*4!-RaD>q;>bV{p100Tqpq>u)wR!$^}? zk?h(HHClz|fh*5Ylw|n8v&yC2&?Y>jkB+EGjwl_WGsOSkM%z_MM&+p(q?S>oh#0w} z8-+et{B(j*)UY{X?izxaHsGq$IWCY5QPYKLj1v1!Y$)eLVYYi@c@KRu*yPdmDAp&; z1sX;1U}7S5xG`lKhmNqjjG#Bz=RhEi`h2n2P%n|qyBM0r!o-`skBO&wE$#7F*Cd`Y z`zife-<9a6*AeRD<&yQkNx9(_HSSZZdM(_T3H{SpmOZH-!RbPmDR7OLL;lgju{mH% zEp9dI=rJJ>IbwJ>T44Z)RP6Kn^DL=?*vVNnwRkMsU*sy`7=D0H7R{@3-z(oAL&(#l znM1#(g$)aFY#DKlBI+6iVsif1L8!?iVQ-2WEyy=^s=vt!iTvLSJ$Ds+JmpQ4Jc)%$ zV8x+JcEqzjk^V9}{xk>+F-thI@%-8Xhhyy>nedLE;kTU^xNGrd_2GMc0xM4-oer6S zT|$}fm?n(4=6zbJA|1`f=>> z204f*XJCOjQy>C;MuqVDb5(3N=Y%1O<8Q74Ua(<>AMf85nYtW4!)WcNfpqUgMOqDmuSj~a@< zTNz2|FbvEi4wEM1(gN_9RHq8L{L)7!jWj`fkESLLemmK5NxZp$K@&k;MX16Ot^77z z#jx33r9a1#8PlU94qhbU!^|Ur8Kn!zwUrf19K~0GL1nJ(#%1;!)Toie-e^*Q3ZP@; zErAlHQ&)T`N0tpeM<9{l5{kc!Lxf3vL3p5uZldZrg?75lsuY}tCUNM2Td?2G&CQub zV(pX^aasixJ*zY0X?30|p+ZF?#CO!1r{Ek4_bu?OBbhU$;^onoSn%5zF1|FN zxSz=N`Sa&teZ)2`L{(8Pn0SbmoI5@g*`}+~9)E0^gv?|yYtUQKe}uW;vYpCJ8!R~c zv_4a!&S6y{BY z3EE$z`sZ)VCh7&d7-i3=^piw<;td7W9&;nwjk{@ty2PYN5G`)5gg{Urq;s^Z&QVQe&H*PR(q6IQXryPy85Dp-E4yIAOi6~ zf|}w3-YBSdNa1R(U3R!h8g7C{ngfwLa5OC26Pg22jB;(H;ce*kD;q|a+>NAw3NY{%PIn|0b@YA3xQku1w>%t4tx12`YncSKk#H)c;i9nDQ13o|No(ylEe zbcyzOT?HCQ&0b<5rdJ~7z8X$rvi$YlK(%4Ve$v_LC#Y?Bsf8sdgH*PROl0R&+2cS+b zL`&Y;OF$Vf7Qmz_ekds^AOcNaepm=LL7Oar9jQOZ)`RtFVQjH+wQR0R633Zofh&T6 zb%i{JL$AC;D4GUcvUvT$4hA+p_tCiHI;c*)id`?*OB?QNHJmx<)vh{#Y%;=ERSW7> zC?6;VaE3Iyy1Wp-#nHQ$waj`y-DgvXOuo+mLUB`37X4|I39oFey>XMF5SdA6>G{c? zoPT;<2C5GdrRH)rBHd6OHKO55rj zLitqBHcH*PKQ}G~N=(s~V`7Th#k#k_pYeUAC3*wuf?DmGIp7*zqFJu}QpopTlyVA> z0&1m1c{RU8Lnv^`5fV@U-VyO4ODEm!T9CVl4mCp}q-ed0Utu<`@C&=WaNM@WB_QBi z%j}nY1fAP==;=ihd7@3%e1@?r`=`-B?aLcEuhjDuTW?*qc2g;YbfNR(RPjK#%V8Ae z1{3kX@+ViSpgn(As>g8e=NsvBq)5GwHfgugakuHTRwm^v$!Vt_}?~$ zihTa^09{949mq94iq)I-#B=Vt^tw??hpxD&pHe@ZHQmaCDBa?!J7f9c(3FU^FJ&pC zw}-O*Ng;6cn7_$IkAF3-$EA8>)MOWZN4rE*252TrFU39{3Sk>Zzcj?d&I&hN5qC~QNcM?2(Yt%EQSVt7#t zP*SraWf;!aQvwUuy8+#4clWMR*+CWpqcf{Fj?_*Id(LqmF?)G=YWoo;zIJ!I5XvA{ zAQD(n_dtbRStaRfW!eP&t{-(a4s!qK+oQ}ignHVLbLaq+=lx?)ZO8vd(ihHDMvH%c z5|1^#Oj&v9cs9&gpj%Luq|%z+fsYV1@BH_1$woP^`x%t1B@I!$Qj4zC=l(rHdrI>x zFxD}^eKBBi%rW+-W)^UmbgkGeMu+WkY;&RmYB#+3ohAZh zRKl@(k4UhwR4SRQ?TWQP%;M%@U@XSYx0M`vqa%K(K_;$iXrs6HkG8@aNQ3+r)=+{Gx90Ms{BkiGG#+OYxXdD2gipV$6CI4dM^aoXJ`jC9n|f+QuZHv7#)5O2=== z4k0-aEseViUF;}YQTXlKpa5YIw;4gQTMy)v$4jGhJe4*>Rq?Zg*U zg+{JMgxQ`_%&&(Qb4Jur3JJFxy^33t$01MLCh1GUAz7xLum0Q`Iy5S*N2Xmr=vTRh zDW^YtROK*X6MPR5So9HA1X8uUKf~Czr>^^K+cyU4TY7p#yVKaKZ*WMLqBFw5jgkFl zctVsrkq5Ww)I+N;-&U=^nE$b;1sYr>IfBt>yuzBv;BcK(8IClh%085EESK5Lu9GT7 zC+eTZgiHs2D>cJsdB;#QR=Ms}u&x(!37GqaHCE{oLL>NP24CfO5-(*KFFGhJzf+vw4`hCR|m<2xPvj;Mrj}UG`sD?DFz*Sf61N%$LlNS_SEt|eD8N>q|xUWG9WbxGV6~1IgXC_y9t#x-ob+B^URuIo8<^%77_RL zs01T#1r+g&mpr2!aJ)&oC>kmXbOV)@hia{GBFa&SVb)f_?AnThBNwW`t>t>okNx+Z z37T&|p!&`8=82xdve};aAS;&LxD}HSwlCCj>$JBQ`CZwgJ07WT0N-=80bq~)NCN@u zi7Mn%XMqj3ZQ#~+K(do5ZD8=7yx_(ob)W^1-3V<5wUWI!%H!OCNphVk7?GjX^jKzP zI4kd=?et^kr4G&%ftRtk^S|1`M0pYB{jfH|F(6q+czcl578gOB0_>Z1xLGgoZ{(w_ zq9MkWs)CNZ&kFWit#iW@iXJz-Icw&*fY z9CKcTxT-yk`E`BLYVf&GKt!rARi95jcxUz(sTm8o@8>I1NesuL;$iQ@&AW0iDZ|Kq zC6#5$geycEH7-hvWMJJTeG>!YiKo+%RlMdGo%Z(M!f3uI0c)VUn>xD)4 zBYlYE#?a)(7@uf=>-)Vnjtrp@{bHP}nq=3Z7|+44pH?Ju|J<~Bc__%}zG)*00eg{tJTU2?WWJTBq0!PK#GWxco8$-7*Kli6)HQ^#_P zU;I>b-Wz@i#e{lP%Wf_Q^Zc)fn5l?)i}uuZ4junb?<(vEh7CuPu9iig97@Kqp$-{i zb>`(9iC1~7C{9;}-H>dR-82Ut<*4~kr}|gOLbYdN?WyPu*$k9xp5*e*0wUHBus|%R#6gG6a3QBkGJgyzbJWj7 zf=j(o=4LyRhzDrt4Pg{(Td&^Y1BWq%M(zM5zs;mHEa;r<@tl0-;1Cf5PM3d@hXPUx zTAo5KLJ=t4%HQla2-XY_j|$_;Y!eIHtUuy*1XhDoAxhOy!!WY z_TgW>dj-o+QU6*r%w5*)sr08fO}cl%vCpfwaNJp8`GwWR*J6k^W9o66W&O{cn?*|z zJypr`yu({>{Ux*@D?yv#!zVW;DmO;)BzFey!oTy%rDVRoawxXaX6o0@R3RG_C)p>e zWnx!6)S6!dm+T`5o@8PxU$VrpOPDmxoQmnZ9Iq4*dcS}5>M}rGpVG?U@9BZ9MV+GY z<1YvxUx!n6jg~a}c%-y3(FXV)L>fKM-9sb(Kp%3!ut+0OcuT_?WFgxikvzE#g<+m@h965;ms?j0a6*rT6TBXiJU&71LVJwd z?sAVq+@$T^QrD78-x$fP{`i?Xh9~v@9uvNl@ zxmf&}lfpqF5ez|uGJa(um$(c}1DkItr5@*#A?QEtbtM#Ost(f|nxoW1{BMqCt|TcqLjPTEDc`Ex zYS@ctIQ|>n`DXU9a05fGA&K|U85FLffaR#P)R~0TtkDNh4Ju?Aj&?AiLMQ%-l0ZNv zgt_!lu5wbgYf{-#Jj6V zsakzQ0N{{vceHX8$zfOcus+>rKu9NQ6@&{oi;Ai@MSieDInVzh0KjGB4K^+LbnjJG zUW~Wt9yKHI{B_kHkz?WoQbCqzSTVR}!YqkXqbxR-(_om989izj|KidhhFBp+5e-#I zWe}+?9Mg$P`26jT&vG6&BuFPT|mnGdG27C*tuNNH?>id)XpplE_(WA2^a2VRhb>Z0=p59JtgTFKp<)}WcZ zi+gw7FHPbmA#D(NT=)0*!>2}K5w}g7IL3el4#cKZoBL3)o|5duZRqkA8-ZvUA>ly$mH&VEddsjX*RE}Jk_rl_Sfn7Kv`9&V zh_na@NTZS>UDBl>Dj^-x-AH$cN_S6E>F%y?+^+S!-}CNc@9hu%2y@YUc^e3-TRRR3R)Xz4L)9T z$ugsUj-nMoS{VSUQvjJ(EYalUwOMzE-pOAUm?GfrN}CYQrlEnkeho09(2FD-E#Pql zwQ93+7_(ZA1Zt9${`u34g5qg1Kt+9@=0Bwd3?*fIu~+qVU$%ZhT&{MpB5bcTt>G-8 zN}3+n&eFfXizA`z16SOqUjx!PkhHH+VUyA9bSX~b+smKXhFI&$cAjF$+iKwf$e>hA z#ZIzno={JjF-LPaTio(cja;gH%x&bI0sE6k<~mdmPmJec&(BZb3G9%CEn;QUy>a${ zozi7n0muj)TBQv>L;qnE@Fzru5?SK-jT_ApG6t(izKr7 zB3`&M;m34$tT_rX&rD)x~_+7sBvgJ^W*Ez*01c+5tJw%ymSApiz&bjCER<+8yK zs7Y1FMW1yZmHQCH1RG&=Cg5kB12=O(jB>S03bDIHEEGw4Q=66}IS_B$fiP5%7bXUV zRZta225)>Nh0cPCDb8E1^G469wYD#9eq7@&@Tjf0ps~# zp!TcLS*Gpml$ef817KiK4(@s2V`XF|`_Cz`e&Mw?aEO;yUg#;VY&LxwFX9d!FKWO1 zC}?0;V{QzfC@jIAG~JxJV~!CxO2=LQ8!S$YD-QT4fag84uTe7-YgPR-lwR!Xy%(O- z?a_m6iCy_G#CWZyKOArO%G1|p(knbeY+Xm@`*$e?9Y4gy=}hlh#*Ev7tD$1PuIN@3 zVp-<0WK!(I`iP<>eh<;MRspYdmKWfEo;-w4Zcq;XQnS3fTfqFGmlxQ02V{D5HbR)5 z-hdVTRQ`KIL%UNiZe`d45UOfs3%g@cP<9*e0@v@p$pEvp^H|=h&aIKo?r8e30c}ja zA}Ww*r^G`7Enj?*tEtYpufsCbh@$utC5FC zD{s!7!R9i>6>?0A@(-i9a>@2_HHxq`PXmd@0QWr*g<9bT6Lpo(U=8e@*YDw)bz<4UDsWOwXMr9~Xv*O~)PPD6ETJfWR7bN1oEev<(Qy zrljov(AjW)w2RV3rfd3O%>)t*}wt zh3A-OX8}rLxk(Y9^G25*PgKfP+QtEWkJ5GjMHcP%50`0Vb;UbP;Mtol9@Gwo>Rvia zB}X=LH4jSS)b5EQW2H#2)`xt>4PZssny_8oE(5GH%=P3ja?-sf05Ol49Zy|50`*fM z-0W5$tNgL?$IubUN3CBW;V=kl3B47_s#O^|ic3nObF#-svbK*^1+$i3Ekbwuu5l3$ zkF+)-8&q>Z+74l=24fciB68ovMqCx*x)VtUGS#fr$h6 zG;1L2r4{ykzvIeEd_!rGy@we@2#{GT>v&PazQ*I>4Qyp;7%sOL6`{LIL9NlhadbDS z*oH?i3|GKLvK|-!a*^5=h{q?d|L(6Ymof82>6MN65n<@^W^EUodQe`w()q@0_%0Kq z5nX#^ah+Eg&VgX*?GNE{fT!iY(ufFyARnPQ4-D+3d5}=F{FdD#3I0L&%ESCNsGkCQZgKQy4hF6YI!Z^Eth8@` z;#>K%|Lv7*p4Vh=S`G`wTSXBOQK_kY1&faiPd)USWAF&M%sseucT!d3q02Sd+J~nx zzvP1&)MxtthJ9EoYwo4^IRb%&tG8*;@__F+{`t3;R**9md``-V#tdE5Y!e*7WIAND z=QK4P{@w@hh}Hz#Nkazp zqI8B{YHZ+f@ElwyN_&SZ`o_@}_7+1=W(8w+aV!EL3nG$-;BybX?nljm)(n;d$@@QL zLIR<+kB+?B63WIVK_M(cFCRr^ZO|SfSZUuqf?B6q;Um)}5Jv?bzg-G_v0|5}`s3Wi zo4TZ1QJf0II}E1SCJidV8^)?mawTa?JKAIie7m()Gi6%+eWSSOXpy+A1{_fD=wV)i zqK%!vjG`w#k48xF+?79wuc`yK&GUD-{E160mqNMsZTIT|C~doR&V`8dP1mF}cnYKd zCxt%k9id15fs@FAiZWFbx zevXJd5a>;jPA6g-1<#Bb)CxkTZI9zoliFlF+2O?|Dl%&|C$tj&UsV`T=4pPd|F>w$ zdhE-0f4TYFK+l-XO?U1d+v5t3YCdyE6yLyz4Rh`8-7aku$~k)WcLb0-A;YSlQR)<` z5~0cO@3*XU&s>Q)AGn=tD?|JT0P6~vW&){6bwlLg6r8M zwTjx&_qrI%Ro~Qm1_oa06&L1<(hwp3JzPcoJf4GS{^Es>amOVV2Y9=4=uo-3L}GWf zQnQtVj-eiNzF19w>g>YSNP2wCih*+h#;PniiYQQn;#7K-G}rxh$q91$d7T;Q@9QR^ zGy=xILnnPdNc9yc<`-b@Mu^If$Pa9*D^0Se%-7P`4!6 zT(r$W+a{~V)zH7445#;FR;!YP9hL0-x8XuxonDyw*em17Z%-rI|`XJmck{j{xG77a%>WL$Q4>K9F z45iZ8?bQ+0C@ymt#vC;lq{zscV^s@V-ZtAJ!}<_^mS#2EMwj-ayAUP$#95HRSXVb* zDmQ(vc?`_tU#4=%G7H{_a)xr(nBgECLj|RYY4TGW>q~Mp5T^_%!(J*;QQbE8coL0E<G=zfAM^xy(kA zs5fjwk&S3#S;KGX3Q68-xSH+Q7Z{aG!kn?{F{$3S#?`v_nRh*f1YsIzU&EgHyG8Zt z9uIYI4*n6h+Md4*SFIcG{N201dCXfflrD;+f-WGtG8g<UE-adR}w35;=O$7>$u7MawH z7qvQqjXh22U_qIMYb^NcFGS@X*{`myRb?;n+b+_AwWrgcQJjhl7uOui?D=qh8!U(D2T{SZ{U! zb}mHyHw~5&+%wW-eo&+#>0;5jJ3z_f8hO1M0zR4I+}^K}{N@q**WCN}?|){PoI;@^ z=aSah{kW3S4G5XDWzgRGy`h2`)=-tu*pq++2h191Jew{M0<^}y$5$m%`mMBgP%j!v zf#K4vb*#C)iWd#I3vhG-kZ2GD^(S;g84QE3jHo1aJ_(f~a|WFXZL$q+t%-7LL$w;` za);O&A4K(x-iz-X9K5CRYV2SX8SrgFU_+D@zeU1~V{fODPGL`{<9t`b>`+B6xYo{= z?N+2uctILjeKwYWFSKws7a}?UpqCjm6O=Xg>VXFr$m4puZ6xa07c5 zu70eqPn&ru`9T@8F51DwvjJek$j5~);>i5-o<4+5a1!VfU_+g*3aNGE!*!*3PtIXu zHvwl{ZfeUFsTg)CMy_q816q1{xT}B81WSi(9WBYKk{!XI3SDYA>}W#*11nI@8fa&6 zFLE{MTCXzb*2+g$ugx1JX^~AqA;xj^Y*+cM3k!!qTiF_K*Uq8#vUPgn;#g%VvwBez zGk@(l!9B*3@z9RB$6j|y^SdaiTgC%=Cf}d%E*_mg5l1tTmfJg(8Bud?-3;``fiI2C zy?)6>Nv{c=$eMZexn}5xhK}ei1l2}JVK79^vOKjs_(Djv;9%`3W~zx+4T1f)saQ0F zz@$1cmtY2bx}!|Fp2qu0jy)kOL>u{i0=MKL0f{_qi52cOJoNK{tQm+D;>r`t8*dEd zZs9_`;3?pGs=KGz+i+Y%)23U)I~j2%O^bf- z92MQt|Iraa{Hy=2f`qpq?|?d{`%xvdp|a6@!`n4G*MJc=sd%tC$HQ^g$lS|-MO_+r zYkCJgbob4cjU379(iT3SXS^FYj^kdsHGo(Ss7@90SGm}#o69zY-}@>v8UTjO$%YuH zQhuOSkoSz^o+AV;;~cEqaa?9?M$NG_K1IeO9gG-oJYP!xFCHa$0Y&?cP`nHo{Dd6C z9_Ky+E1t&d!3?JAqupTGwWR^zx@|?)HG&eNkJoj_j^)#LWk!KnLDb|Wkxg-Eyd{%^ zQlzC?RdVDe<6)$3b!Zca3^7yXqZxs{6%L?)ng(Wb0DOzxN+$M~deZgR|NQaHCp<#D z>nO=9!-aE!q+BLa0kPF7rbj+<+>^DCj0tfN(X^JLqMRjx$3qSmo_r|<&j4D4phH{! z^5;jH!daGN!!;pAV8&Se*|&`3a|mjS5#=5L=5UxWq{<#(EH!IgUnpd?XH!U zU{=XP!Cnsh7%7|cb^KqzL?H`)pahtq(N(N<6iDPGo{AvTMGt6-tFFUxP(}(rsC*%)P=QhSsW_0R%&y#?ZHdv~e)F6SYLGAQ zFPw%z$*rz|DJ4@$eI24ml@Piis`)0Ic(XtKqc`0Jq1(&-jcw@`kPX8j@Z!(AjlZ0h zAJ=GTzd0Y^HiS&a}!t z`~jT6XzB~@cc$4c)qOti9*(q6sxaW zrAQx0_u0QP2m?BUql)yLqQTA{+~5Q@g3-ETY-rP zno_|lS8v`00bJ01{nkj$b2ZQEX%_q_dB0;{01U005hJkgzwu>Nzbm$SL&2{=kZvki z^8EDqK?R8M4MInA?0kJ@HgMzAWl)Nk6dx@BT{dAj0Ao}4t9M)ux9K%Z!*j<{HDc#})9&I!Ar0tdH$@L!xf}6cP_3XI#eHCJbdTont z;4k4!0gq-~<5l4Hf$o@Q%xOEQh3FZ3FH6T)q4u`qe77?p6DFz`Pz5jp3-QWL$waZb zXvB$uP_WE;uB6ze^!~w86Ah6om9b)4Xy_4rOZ}wW04b^lTo6!`Nyzwphm12Mw)(Ki zi>N<5pXo(`#cAAqIr_Vtqb%TplNi)=MHp*1I|!7kji|PG1L55M~bx8 z=Kgk`ht-zKqrihc9D@lAMNGLct#2>CFI_i(1Yh~`X@w1m@t(ywBpdxIgSspX?q7QT zg;)`ehKk5CBB=66gOTl9U?CN#2V9ZB!XgS=y?sJ?oM)4H%El0xS8-gmEsj@X(}%_s zk$&Nc6UwOZ7xzVt`gDX(zyOWoQG%milog~#L~`o~A%r1vA?GGE{s(P<1=@#Z)ZA4EP?cA-Ej^m}`D$_F)~$Jvfb zDyz}77wj&UN}nm@7|Fm&s)H;m?j%0i#jm-LNh{~;(I0x1I#m`PvQZDT0fJpr+FH83 zaeTsoa6+c90?#q7yJm5VUc$q4r9Se$!A|8skMLsn!iyjZOc5*O%3f&#mD*=HWE9J3 zJ}ljnCd0-Q&K3--dh?;}60}+}17VU2!2Qh#zLQOdl44=UBUDcKIx`*lYeL$)=oGZt zufyLyKO8s;D8{ZEJb39TRtp@!CBY?j6qN@k*;w>jL&HbQx0=8#yDg%)+Q}}ukuT{X z{aJ+D)9-s)m5y@IpWFq;5b+r@H9|>I13G54VDO=Bn6v-CR|Us|!qC79I{t{pghAG( zP_l5@^>~{a^`3;oOU9}*yoH1!E%7(L*x)8Zzo16ftb;%d`~$C0?w`Isu+-dLM$aj9?x=A+8J>Iul}#;Po_em|3xP&gmSrOc;^B0 zKTSP5(6~n9fnaK2U61gEhlvrfq~#CgH=eaLM<#)7C{J;u9;EC2*o zf?)$g99U2yRnx6;6S9>z#e*P7az!&K!eu%^J8NZ=C+!@rKv#%LIp(sJCfLMJ$}%`6PNz)hP@sS z(-P^ELjt)jk+XPU;VIF}f2D$bBMdk`11js{Fu>v1t2gv)7j#b!{(sr21F=DLO-&^9 zOs8URo-UM8l^f3zXfHss=kUUGmIrvf{UlNsAOXdZYLm2`^Q+)%K#CAAkD)mzuvq}Y z`l<~$Rt^^#CoNQ$*)Cl#xAD#RDp>!-VqOMt7?CSTpWK(BAz!h#Ymf#>1ki$&5?B-J!nL-b0%49z>D)gBq@~SZP=BTOa{d?Q3 zXg@NZrUa~WkLF6+gYV=;E?#J!Sf~@>wcUINJmgj=1dY|>a>q)f?Y>Uvg^n^&Nl+W> zRA&{x+eZWRcafA60=6ib*=qI^`*O83UJqn5AjvH74m2ZNT?1c)sBVCBo}) z8Uf4~PaNMZuv=&aJI+_0SPI4kg8(SEAVGgX`MC;IW)p%dui6)DYf!pO>)bzE-oNYFE{n+8CppJU1BmGZF+SRTrQ35+{BEgN5bGLk=Yylk#HpC?i zx+D>kb!Mns@@9AjWCJar*Vfzb6pp&TvU$?S(b@8TYfgWkO*pauu5?SVyfDpNXH2k; z&)J(61mePqbyV-*ZE-DZwc>4wbyi36#cIzEKQXj|Btd|Ek|8*LHDJO!x?t2{vuR3R zz<4v}^ZtpdQffmlrPE7y>^SfoINks2IUrBza*W4Iag~-v;@GT(bA9yeTO&}vpFZ~6 zr%7b4OK+t{kQ1=$Xl8~MI@^xQt#PQs^i9M`RCnd zGGc{PvH8PommdmtqYB*WgtvGXhjiQ#e8a#+7T&cx$l|l+c9Ri+nJ=*NM{a@`MY>bl zMK-TOiuhM_0SJhYK4WJ6w-`N1J|hT$7R=oC1C!n;K2qEGLzk=(5Sz-h4}4M@2nh@U zmysqYS7p1jLhq_cSJ*AFo+F>|fT==s0W>`_P&COVnIHBsq-H}O*i-%~=}lbH8}&z# zc2jGxyDyJasv5|~y>2Ngecsrir3d__`_{iWl~jN*x6qYQ2(_gKa3U!cRDd}(KO|+h zhNdetY`>>J)1n&a!*9b`NYft4)f*C;_f!E`RrS;CJ>=M#o=Mp;NwUQA?J*o9Aq&~k zrX#}4z~$n+aZTItKv{{Bcx8s$K%iM zoyjYyq`X!(9?tYzSSYiocUDpxr6%M@`8|hzrc}5 zh!$!<&SvVJqm1pf9kCs<2z?e+Fc>@(5F|Q0>2^tZvZ68g2bxNe`<5q%T#dKZS!M&u zfRily^^|(x)>lcPU!N%%PER9Kf~S1wQ2SBP);u8neQp}a{sq#BJZ8%ecpVm#FOom# zVibL-RqLwtIN(?ew|1W`i`J2p`6k!PcX*5#mLL^CNz(U#%H#IDR^?s@kg%F+fV_q7 zYo<^1n-**QxOhzhwJr|$6_42ERl_6y6~)9ycqVwh2~t5EdTmCIp+9DUw3nKOe@PV@ zTB^wANRBj4WAvzvV4pS)+T}^5EG54kRCnXD!iCU5gEz|yXvp^cVWb~{Z4R^lV(HbC zLOO^TRREFJ%TTIs@GERE#x4#^EPZj}82om@u?vNgDwldB`fUEv2PW8k17#ubWAt-M z-#6PEY~IMwz3qWOOvSlzrf{w|(?CT>B=vqB@77O`ZZ&!0T%H<^F%F!Ju^(q|AhN-w zD)aU~8hnAf8{uY%grbu4rb+pe^CmxINX>dQ`UUh=&UGm=;cXV!a?XEwKnh7=B$=qS zB&j}luU~iy5IxIy4KgJ2VW4UrgRDD<{ng_P_2SIP8guwOtMy4yVp$_dB`T)V6i<87 z_rne3R)Gk#kScpRYwk~|k^Ma5zOU_Nb3~%!^YwwNCQt$H!qlB(I5ulA{l$H~hRWmZ zsizDYCE4)6wSkCeq6l)2NCaOeA87QYTfWNBiNnpRT*xNlEBslUrAilETYD9lfPnhG z%)$8sVjI*xRm^LJG~NoZX~|U1Gpo3Bh_(@Gj(oC}vc+z{GSKk%GY$7xImBb19H|x= zv1eTsv8rQ(do2nhLC9^blmTgJ8~IPZfNg~GR4nb(^Dc)AX$6iuuT7d=wjaO@yfo}(YMywl z3}{^UP!mYw;tUdu$mxVb0@*IAq^XUna%;VI(!H~$x)XqkfTqmZoQ|qff(yh9$aE+? zhcbSoT;S#K02uvkk%BUabD44BtjfZPv6mWIhp(ygYcUnF>JPPu!wAp_^jXB^h2svk z=8$Z+X^`~taOl?UT>>cKRye|D9?7jnlu&Qb`Qv(zgk1P`%Oiek?|weNT9tSX>R!Mv z)N0;jv__kz#lO&rQu_(wOg}sFv>G zBNSz#DjKZ6O|7}gfP-#T-<@uss=5t4Tce85rS{d;)zKZT(ryN#6_GSM<4fj%De!_AxK)@}M!WQ7o$eOoB8)425Y+eNYVj-OHtqu!e%nWW-lQ~O z?HC&?kB`jJ<3>Ri!>k)9Fo55s(3d{w9z|ba|ch3}U;S5IEGhoZYA$AqX`Gv6u z=L|UJe1X`#s1*QiX<4mg7?jvK!EjQiyLjoc(-`Q*^$~}OG3!o2`@y6BtpgCkN7(Wu>m6<{3`EWU zw976#4_;(YfJMwL^2UHBgkiApkaxk=w#gId24-ocQCuc1Q1Y6z1Ok&C<$jgk6kXKw zA9WhEHi)VYGA8o}zAGX9)?5cMGb(Vq2qRB*v-n%CP5Lt+JIPwa!*xsjmOetxZzl_L z18NZZ1`5x0e1ZAkaG_J_eBbN#C?1etJ4OAdkjA=*?ke`tTB!MSgUyUIZ=)YVGr`HB zkqCr4T=SPgad9~VI^BTN{8q42D=~Hru^5c*v9;^OuYQ=Py1sLQR$YkcGzQm*Hbdf( zYtCajpw3DXRMDL7jMKq|nKb3d(fJD(IQF=Y_IK?VspE*NhyZJoOlS#XVgq3fx7MGz z&S;rUF{0K;jynLjz7?!ybUQh;RlDa_qj|BM3RW<7(RVBS_76_|Lqa7E_rhT^F7@pX zUAE!dJ96kW)oGuLBtg?!(<`y0?A=*syVo46=~C~!qm&UAqoPK-5BxAe2}0wsRF-C< zVE2v>7Dv0LjR9+yKzJSGt(`GocDAd07JTPPM8>t54%0{hOHWO_3!(2nU6+m(a7cDN z!BtF^RpWKI|CG;W{w~-&^K|-VLhQT2O-~nnwLKR|mbA824@e6DUtT&vZtxEQwE~-F zS(i=ipo2X@frQ4m>Y;qc#m&2F7j|%Eb@JlmXw) zd2f~XP@hNiZQy4Z9ZO95L*a0okP_fFUa%_oNYP|L$@=@4m-+y76R5H~2fv|5%;B;? zy6y-kkSa<_19+6WF`8=26rBB}vS!b1TY=%`(>iQC{0FPBM9=~OFj(SHn32n*pIhwe zHt#pBZv#~rQ9AKZ5E{fqm90aZ-+e!jgU<(IJH>E}rOozyBPM4QROyTrY%d{wa8uxI zKEq({Dk)W$)e3vO)~u=HMXfYn%;JrlYCIeMu{`el71kXo(SV$}F4wat`Fcvy_$8nk zTs=^mZ+}8%RM;9S_%ne}>L#T1F$|7gI@p}ynE6K08oJHV{o>rNiS_B%K<5;zW5d1| zk5Q(cKeEA@(P3-g0P^$Hpwwtsi0 zZ1wVoC1%6Sz`h899!D+3qWBAqYd9pX$z%ftA6vzWchSh+VTEf_pwM5teBjMNZiJ<~u7}COLEj7fs0LBPdtI@(9oyE0h6yNm zDp85W2phcc(7YKMh$54-e9&b4RL4?{WA?AEn0O+p9$0rU?7`Q?AOKu42ZJRYDc`o}0<(#HrJ7dJ{@71wI=2>G6u$x1Iz zir*%H3EyLMkqO)T3&9G*p<&0xO?sKG%NuT~(U>U1a8~V*3)DB-TmUH1v>9E>#wI22 zqQou%dr?rgsnG!`Bem3hR=j&sB#ar`5d#s)phU8_Gk<(uP?|wQj~B1-Dns)C1KLVR zpWs5=_(=9gFU?C`E?u9S`H{F|6ZQ zvZgFI7g$hK4pi`He+agGCD95US&HJ!xnzJuz(w&@qyI{Q^TL&dvms*o5gyb;A_oC1 z@X|&`5qQ{-uyM9e3l>N&T_=5AVrg&=yJ#a%>5uU}k#LZ>(Vw375u*frpn{?5m~D>) zKZ#rS8!l6>4mC51KcwKZd3ySlU!uZpjfA)Tn)|fqmY>n9vx20JTh~eHRde3~w+pVu z%+tmb+eU5${iXi_jg+=U%5_r9UCG+x69?q}9FzvAjgprK_d!vX)t9Rw54eMV=#BRD5lo5C&=)49heDgiQF|A#!895M{c*WPz%# zb^JMvZ@~)0vrIT)#iIjW(xcNLhTtiKRSxlO%q}O6XHSRSAxIolV7 zO@3Vjc~vz?&z`^rZJJYm0!ZN_@I9geU(XgsjH>Vjf_T+a3Z9i~@dR!wc<~+>y<&BY zpUhHG6no~RM`*G-%e)4x`-UZ0sLKbK2vf|I044bvxeBb7`|H$^JccD#nV(7W*)6^d z{o@MKu{%qHxnB%%g$%FcU`SC{P`OJHrq{bS(R#cvs7wcb%>&Y7|Hrkvs1H44b+M11 zwkoDQF%`r@1(GAxYFQhcQ+v+17m*ZbRI(L+80@WT73gi;c9;eb0$3L+43m6NFW+l; z$GTMH?(gLCX2M3T)iyd^^3r(}0|;VK?lSFP+2uUt&&i~*YSG!7dDDz`euJob;6n7I zXb>Sr+sgPq9aH*uJBX-7y%ZKD)gVpm~uu+g0Ij;xL&zzlllc_%aQNg4c>~(>vM3pk8w;chBD2b z0#|l`jH0}>Q4Ti=qR9#`t6I#J;`d;h+s|mn$Xq)%RdJGt%XuR&veHB`F1NhlmtlpX- zpKk12e>tWT0=(8JN=yagBpn|tz2+dUL>~94u>Oq)iK-$793$|g#sWbKHF@38(@Xzw zI|HdfjQh%b;joX7d++sTi4cq3cIYO-D%!YZMg3!Rq~0C)YUO}@6)R&?PzDM**)>OT zMT21z%k8^?%0B1yEJO4|5w~a!Oeh(Vb!yaLWPqi>>ZcVlZ2_iYg!0F9SBld@yi6Fm zHZgjXmb;!*eB~v6KB3BHBe7V; z=YrlBV3$h`yiud^!k8WZCR>9-x#uB^ibdA7*6?PqDxCfzam$dgAu}iksbFgJTQK8R5%?_|54NN`02+y^D`CWSqxx%s>zTCxg?j0@NdtW4G0Dlf8 z1eO{i{o1%fhj$f)xI8mPxbxo0)2-2ssYVTxHJ39-=aiAZ6k%J3&98UmzlP6`mxl4W8QT2y zj*tTFF+@k*zU90wJX~Q~DI3Wxrjo7B0tCM=k%b1WO7g-idDXk>(qo(_LN6OZ1WVg zd|>bQ+TCFd-b0Sg{$DGNx7?WJ`AI7)bhLPYKi_*v72qqbIT^C=v41kQ5+?{;YJ}v? z6}Cvpg;20r^G-h!Up@Zu1{-Dh#5;1fZvwX3&rXjIQh-)z#dI9zum(8E{Uajq3Iv;g zCP)=;cXisHrv`?9z^FM)-P(6!r{O%-Ou#(Dey89msR$Uj7Pm#QYav7pV8tSW#P)Sh z5SkzfW5c=7Ngh$KqP#$J=Y zg>IjnK)AXhN7cku!^KRKT+p#^SxE~p+R;YwhXHf=u@Bg|+tn5 zYACKxT}qRSHXwCC-|Kx`tDVx{h^86^kgnKvxrn;lZiNrY*Z6cpOg4PI%;rMC!rMbf z;3sPv4|pg0zaMPuc!efzP?{58xQAWMmsgMJl1nJowKcQ0Wg66-d4K?es`6 z5^OfXVkaUVDUiX~l74;W5+i`VhJf z(p_?kfizV%hxisOtL1qGc(EiAZ{Mj#`9SSzr|3sE45Bu3Db|ht1*VOJ#(nWMB(Vxv zb`<3fQ{Gouw7<8}244BG>%wclnoZ94_9>&Xt4PoOK}z}-bTNv^4Yy!woFE}NXfz*C z#HNU=b)BL&Mko6U?Cuk*7`G(O&7|G(Y4IkAPBv&?3}P31R8WI-M(h$vWpX2S2D~Z0 zt4E27hYoPCy-i@*Wx*8VR=bI%Ol!O7b~m=%(J8Y61a5_f!!I6~ zwsRd6#$EYeibFOZ?YSs0oK(RUYK#73bhpV!;X8R5 zJ;bD1<2hR1p#0MCKYs%jWGEEnHB}V5yzh2&YA5%x-%d zRhI+FBXmgYAY#4!dN7v-q)}G$UDRsDuR||FUN^!*HH|E=_~(!6`2SX3e%$*&zPt^R zy>KCy!{;fN0TxG=Z4a=Q7egMj(A10n9tz@L0_VO3>M2Q(w#|SiV1e|i9f&b{?;<%F zDNre?WUNh9Jqmz3=e|b;oW431o6iRz3Ze@T*!-Yul?S+AbXF5A@FmjDxBk6eRTq%! zr2z}YGnfP10ZMnibfRjxl|Bc(V9MmbpJxB!CfM6`uV|Fg!XkUcnOpLgKe=h>M^Nhm zIhWz~)UVHB0Ha11jJq64C47Ic4<_GMn?HO%6x0Mrg0;6EIC@wuPD?2bfay2r(Wn7g ze}@nPTMmVKbif_FCH;$=vhNZzX1%{_oB9HwnLU6=G1bfcdn|0#dKOaxo!-L%v zvg5XG*&Hrh3j;(R#I&r1(q$R183dlk8zXsEv$7VChNrM-&Bv=YkIWlpU{H~|{OwL` zhyY9;*J@8+R(!~6q|h&?(7)xf@3A~yC=Z^s#x1V^X(l~@H#b?uZ*g;4sZ-MbEALEr zoNEX;D;k`@z>^82c7btjdOlH3KRpM3~{ZHN}9zPvF1Q;=n`3KJZOf@Qy6 zODxF)N%%;5VZsw_R-KwREwL2)U-g=!l##;s!6|N3frnRYc$->5Q`cIpUt5LneEE&G zZ5R7cLE$0V(W8GY^;4J2Z+J^0VEvJKZ5R;8qYCEE)!rqch&!^MaHfOF2Zuh}MThS0 zOC5JX9pYv+J+$#QCY+kBni-T?!FjdEnr7ow#sjtcd8)L4mTn?8`fuNsF*q&EsM0}5 z&*scrTnkI1z5K~i=-`jePWhVsWn$#TVsJX>AEX1LmCb{+TVO_aSr}`%pi}6ei8)JE z+eu6@McP(%I4iH#I5w%X*f?@kBOqa$_fT{0{#Yrcm`-?&P)eSRPF&5Y{plJDFUifT zb6cCb)fJnGQ&nL9uT^t$C{xI)b+~X zG8m&Q3|9tdT2~pkmSLa-KnV*EJPdPZUe|ouY;1X9wPg3mt7AC-PNO7H<#OmI5(zdC zvv%m=UskIhx3bkn;7-7_!#s4KyF+IC0h=X<+qh=)7mn~}+0U9``co$eyeyPifkW?O zJZUFQ`rKP2rtx)y3!(kM`fdSnMs|>svH3fxV0tOg_hnz1Xms$UoTD=%s_KW|tV;ft9ITmye~;%B)aA08x)sZvDU?;|mm@aH{b-#DtS13x z6wqTnH)#cq7{!$TbE(2u4`j?HYD`&v>Hx=8E}vScRQPni!I2xKlLDkelm)Os;6p)) zKpf*PoA$Za<-r7^kp?Dcv)~Lu8_zKW65G+It@!doJx0cz<%WDRBE8*aI z9XT^iyXfTm38*|WAZu`8tP9Z`Cq_;&7MUlF(tZP%Wh8_JN!gNwiK8BDB0cedyf!8-W_B|C0*DtvaVr0u-%)2U^OkJwWiV{jZBJlld0 z<#WnQOdN0=xu53%VR&m+qigygnB3z7`HZDhrzJk~^h~%ZjmWpFDD51KX?+2}%dr%__wFZeZZD z4cQ6d`~kuF^Y_+xp~`S(TbVPGe?RZ7H#8?jY=xeasGW0^y*75BqR{yh2~_||Gv|b8 z^}^=gauf$Po)wTb50;-09H6Yh8elSpBYr?@T^tzmz^p?{y z`%)x(FRmfjHpzKC;6qCyp%HEDMIHx`ZYdw@va1FRAwRV|KWUd5?o7_fct73h+hgDq zLfq8#cmmo{IU)?x$pw?%$73MhH;vT*u7U%&T+@*4k{Wg4u1B!H(%W4g?nzgmS(vD; zE&r|oc?*Sprl9!RTb9AX0tPYr*cr@2on0s(ewjl|v`8Vdf$Ryi$)sKIY8CBX=YbLD z{W%pKb7=M2j$j|lK*C(GF>E9om-4_^jzY9@3+9@`Rk(sa@Y=A?DXf{ope_t)6gZz1 zOFWd#lkxe8NM{XjFP>aXjLOoGVq`X8l2ia9pdbhr zGakr}eE05M6HMFXXymcZqu2X!Um?~D+SmlxMmP3g3y7Zczsdexv$_@#t6?3u#_bhl zAB=Ti$>3T-t~-Iqyc0rja~Oja2)N=c5!&Y+ZH0x3Ffx6qxGXk$<%J`0Q@P_`B@w8N zokkkQvd`rs&USUWI&`XsH9NN9OsY>1F&1f4clTg!!=q=iYdU#9ZZJRZBA0iH81h?s z=Q!yBq0@c8YH;ao=B3}j4C)R534!4{uESN~u+~K4mUFM#3M(wwuOnmS5#zioC)qX{ z*G3NPgl>RBd;G#7MOzeC)8c^L-1JX35V84MBmY{}tBwR*oxh^Qe>ZL4Qu+sd zk6j_Vs6BK!g0uq^?wkY+hUjQ2KF5A-GIh8{k5gwwyRxd$y>%TPU9iLv$9FxB8|SP` z@SA`4!Au7&(x1ESs43_QB6B$!O&<>DgPHKzk4{gY1}+p|3a!a&i{#LdvG1MqBsA}$ z#6-vUF?Cv|_&KZutud!sr@ySgyrpcjR{M7250XjM>m;g^WY<6V*#2RsC8Dkd@}?pf zWEu|_q)72)G*-l=&nZl}^{svB*@~~HO*!g!cmLRVFtC=pwfD@abcpvxktWs^_U^S$ zxFO5dH?`QVoBt(1|GVDWdx}xe$7^-&NEdV%#0TtL@4faqnwx3@q_`x4jz#69m5_(s zd#{gm3`Pp~{S+t&T$$sH~Ka{CNtv0bpz9a)XJ(cKMk^2$9^@kP4wNq z!ORAx#3TYq0s|v+9Mek@*LrI>?4KMvFfxwvx6aL9!d$^f5GAU1%$@t0derxA6PO&Hwe{`ln& zJJua*@QU)U9{iP8!|I~5I|nGt)cm78ZD9Pr@s=_I(u;N<(;9k3N0{a$d;4RJtMKVo z+*!n0U@TL6LOd@n=e=v8v8r(OQejPfe?0fbaCkvI2klP)JE}7 zyij~?qOCSzSkb$WyRs(4?jW_@+ED z+fQCj^9F_Kq{%6UslaQoEl|Ag z%qyS$@b?N>I9Xc1z`p<^xCY=1E9FPTrN)7-Z9b@2dYb;&rpZ`v%_)Q^v}4Z zMK^qLdB8wF*i`83cxyAy?L?*}ton$tsz#pRnNhwR7VV+X^4{!m>{p-9Z^=N-gA$-d z-0|#BgK*sS%wV!q!2HJl4R1v~WBK5Jis_Wfg$HIUf6VO%OIC&p8#UzycBJ#(nXGx1 zb$Fv_>Ew*kt1-9vms@B(D|F--Gpbo__TRBZb2xkA4s0p`Cbi(ue>_c6se zat%>x>rNDX(Z92e6}X9{$4=t?tyDpR@)Qdb?L<`plaFyan#>a_t5zU)upeTMTFuzZ*L?&q^-}sxjkF0a2l*PA-=T}iLU*(5`^RYC+hb9 zeocaQ`yc}(RCtYVGR?xSK?Zx`^ zs|Q_%I(tv%H~4Mlsi1JGTGA9x$(sXA_On3lmec2j;;i(%9e{KPqMSfEeIfbc|GX`d z9*9*fuqXWR4Cdw1zV+hHK_A2hCkx6WumAzsWyIqw^)ibDfGB=uJAt*u3}CQeI~^qn zcR_z``~y=Ak}X4^F$Pw#u1JL%v0@u{IOL#HVIH$+2D%e)l9t#S1yOs%PPe5{joA?r z$9SUO@l_3&bFW+2bZVHPmiqjr3=TRDm$g*Bs4DvW&BO& zmBn1YS1cw&`3VSv?MuUzmq}9iE4dreWD^Pj33vyIOBHz|ud&qP$do+Al%cF8{`}z( zy`5Q$d%ma&nvK~^C!6gDXsiUVy@kM;K+I@DjQusYGn0V@%k8()UexIL(#n%1xh?#= zEWNK9!zS<1cgN7|z;`tss!v&){f_>-=)8O8f9uA0WF{FZsC)tZtP1jZ3ucUdlpW54 z=H_$XU4=pQhq6V~xl*{mLNX<7&3(ypSnNSad&AnGGl*U;+Cl%`71Sg@5{B^i=2l1A zl@3zl>|V33Uf2YTDu<;i9GCK|PU-lvF7~AKKQAp%$?70AK<18@ga9KMQ4TPI#bN!# zvxGQZo(N#GXLGNV0(GH_}$ey zZh*~8V0==EAvDihjrG=gU{Th6$*?twJ0?UOocfISrsH?cyR#um*MB#hGRFGf6y+=q z1=mAG<$R=49ugor#Fo=uGQIEUGc^rJMET!Ge={(tsGO+QR;)&2Kly99vu3QxZm?#p zgSe~uX4Pnob!&F%wI=NzN9?-r|Jp8L)6G}!4jZ8PlEuT?I(UIHIOdLOQ&%mn^w3V_ z7V$HNQQ7+`tp+Urhz0+-f$NHNTPop)`i{?KE@ODD-4Jx?r# zSj$p&vL{6LbqtEs$=H_&#iT+KDqCo??+#fqbt3z|o5A;UP3L*ed7jtz`v-jIhnnu0 z=DzRizLwAR`Mej$M9iG`zn3To==kNjQ}<_FT`Nlvo6DkSx}6iJWO4^vyqC~et=N(z zeD3}^mpw<%{alWbF}pyJ3Lvh|hZLF$#SFc`$0o#QwhngAZsUepL7$1LV0T^!nC+v@cT1g9`GiB>b2w>@z<_ zbBUlmDFt+fSsymB~M73bMCq_har9An^8FodK5FKpn8G`IBfjm6i;vI)pW4rofa zJhnA;koXw~&Y-&@wUdJ0b$m~vf+qQn#YZu3Fe4Fq@Pp6z#g$N{g?e6~PkM~DonlDHzX?Mwhx`zB$Kjvqi$=CItc{w3z0Zt{qngs9hejN~4 zI%&XZA2|gsJy5REm+*6=K8OChjw;rBwnOXA zK`~L*Is%F{QeJW^2WoZLe6)MFRsr;aR_5``c|E)w5y=DQhv(%UB+`-z2c?={j06S&M08ghUz9YM?9yHKw zlBmU%_=>D+iH#lrd@@-1%zELzi42asd|2LhC0{5osK!9x&D`W%<_M+C^YK)aU}h2| z1sdk8aB=$p|NN5Xg7QguPlZO@mY%5jn=8OD+hA{q<;|rmbP>R2K3RG4coWZ=Z8>dG zO9pZK3(Argkx%X!MvjTAFD$9m87x$1pcn(y8O2|VCQ?J&5OCA}bNehm5C{!crm{v6 zo=XDUypP4pGI=G~`y?=e7%F{nc{g*w%b_8*b^YBq9@X>fE8ukus1W3JET*xs#6Fx> z_%;RryD2Yu#J03Li5J9>xAtJAzvQDSVlt>ed;MLonYSv$!e|?5YCGdisgC}|a;dYa zP!s4oGs5Ivk-{iIuLL0*WO`jI+_Jaiy@}CG9A_Wz*TmkTQ^mxfl?HVOld6r4v>&Ih zrJe@&P4M!4YFQQWeOpVz8RFR)gaN@YLw+K9rnv&zBQu$riTfVDxqJ_|+P6rsHku~gE? z!$J4~%#F z@AUNa;_n%~%&I;v3%^y?{sy{#l69S%=X}kHbQm4QltRI zpj40wJz-gCj5cQ|K`+YJpzSl{Is=ILx0yiPUssj3s_R+DB29=xE;u!6}Scnqkx^9{~ z8H+kL$Kg5ybP)@;aoG?2nD6x=!Zcv^#(^;+K1~OY+X_9ShWj$DMqnla`8G+TccmTq zXLmfnUF0bLbx-@Jr6SupZOqwmpif-RgmUA~y{qwn&ZrOm;AK9|lLfoAY&k3IYHxB9 z{UJ5`3lIIOgF!+ob+!Mgfnc6m^LMC(@r?;r)1H>P_>iR-Kx9<@2!<~(;L^a`N6%%+ z6FyLs&<@8zYR*B?72}+B1wz)xMxod9+H{_bMLVyW`vIql2spnYCaxC%gVgHxaw;H( zj0N7F1u!1Q9F+=>nsi}m0A`uZrFjwL52oXt9F7cfkWeWkyqcJ zL5j5=G|djlvpoVg0|EDkkGPGHyO>=`(#%0`D?KOXT^6HC<|;=9-iyL_?INs?Pg~yr zj@zP~z{;dyR3G4gux}xn#>7dxVwn-&nu+sj#Llr6?z}=88rPgiU^N92-|vA9-2fxI z#0Q3x*xM=5dH4Jk<*$I^N)Di^1YYd#x*_ct)|eu0lU_Yz^coDlKskiqW2HAv4ubr> zJq+{M!gI(#XV5vz4&sT&6jT_2I#H?zu}bWbmiE}QO&ImVLW$q$y178-05kVO(lN}w z^X7c#%S#+Wm&!{n-S_f(5-vlL?mGsTT!!{2c-GMj8f0FmW>mp)2q4Xltu8aO>Ku!$ zg*tAn1`;ru-|T?x2?qNccvB}Jdu?9AX#pt5m1@Q_p%3tn0N8-hIDXESKt_0 zQMJHF1={ByjY?PK5a;53zaP&28U;ftEPw?}&(TcO z^cV(FZ@6`|X#pZV$_7Ar>n_a4m*DVMF6x=oKrTj0H;x02h#|EJJRkB7ECbKf;g&QM z-(Y@%JFq(hFQlJZrP;l+8OpSH*`KG9^Pb2-1cikkEfCbA$p!){3}0xCfgr!tTS}aQ zbjTeV0DOEq3{Al81Kd>vRNTdO;wtsqsA`Jw`wVVT&A}?v7W(=CUDa1>P-0H~Mz_8o z$%^hk7u~_!i<%mPI6*f{q||_^;XpxlO*rWOj%d5TU^`xu5C#7Z{CPBN$IcPx@htmM zeG%h_cp@0Tb{KUn!L`5LYN6i2&xPTEg1A{elfKxA>=nk z>F9|Qh7@p7`p-k!5{@gc|2{N01u=Hn>S`|P(^zim#X0onsUUR?4BuhY2=0{E-} z=Q&!LL#8Z?#!Yabl8bsVYJvNa<*=3wcUK;|A}Sliowvf5rV56HiU&%~Vq`lrc0mV} zmJK)v7cv}%p78Y&V9L^0c>FQA1oIcral4^^!^vrJ8{>lD3SVgW%&;M0`g4(L7*yJ! z_7w~C!Bh$OHL4yD=5UuvEhN>CBZ*}q=q+D%x(P}AMp)QIKo#l`4Zv%^mQo+84MBM& zg@IjSI71~=XD2OM3!xx+1c=$tC_L!ti{@1A^%^^?9lC=!>BZsk3pxq4>m3Ge-9@|T zu+L*M$6o?z296s!+UsO(Z#V=l2&q1yQ; z#z~@ipTuQ)eFzsju^o^2hif;OHK2;dk6>&&|B8ATCql<`9_3XPs@+N=s5EUT#T=6sdXrrP5!OiK~w%O=gq7ihTMQ%+p0Z zFt;wDcu_P*d9T#Pi?6)AcYr6S%vSgXx-`BB@ec2D#?Y_?oyk1~A|2dw+V+Ml!Uny zWkHQztbvr5Z0fvKmmaiDP@!Pq?!He!E1*8 zQ&GCUyBU_|Sugm+l5nZV$i7wyWyL*Mvd@?NQlT!Q7SF;`O~yjgTwwvrYFGM!2jwzx zUA3W?HQoWpOJ``CG`X;077mp30?R7x^U+%xk1tpXkZ{Y9;7yB(k%D2N!V>Vp(sq40 zS!m&dfd$d9EMNbXjV}9~m+bmxuag7Uv8!Ttb<%npmpOXj6pHTdh3Rts-KLsv0H)Gc zY5xEuEI&e@l&%Ye)c#!%#CnI;Qmg{ZW=Er>pOCbXj?=y$*JDh&j6jTVYoR?{C%=@w zIG{zssXZr*?Ra|spj%(LWm2IrAF;_Xo{_^S%{7|qj7xdmg-m%%+}4N)>y!3rkxMOt zAtzHEUzHtsKz!)u6yiV8$E{zaY1y@`TyE=}5pf_p;v9E&8?WogRNs}Z#Gx6FLW_w9 z$*v$P)a?Fyyda@r`*c8;3oDzKX2y|Dl2>2vCgb+Bne$}e_+qJ1*!5i5;C%ajTNTH& zHQbn>gW{-t6^fiMIkQR48o2Q{+9kDBQwZ|v<%P_rgWfY7dC}Wvu{}LfI(fOCm!Ry@ z+K|)0+p?8C`4lA1H`MCtzmLz`27J*QpAVs|?^~2-6Uwl}ac|uld>cGq%CYHNd%8Ek z(yZj^4U#p2l#ie=!}2Jg{bDtkrHia>roOy3unh5x30m)Sm%lOm*lT)FeQWkXEr?Z` z2pN=zov?US3AadDH<(2E+ zgr#vwwZN%mj%}HHQ%6F65ecnxYx=C`$#45fyNBB*S@r@hRnoJR?S!$~)@%#GZX$-q>}JxTc-5D^G$NrE;_n~Y7#kla4+-gSLMBn z;g+&>ownEe{Pay<`F&1sp2dU5ZPP>|*}nTou{Or8bx_E`!fQ!+!e@CxCT`}-MpJ~e zI-k|hj-LLIu`P9q4DW&;;mn5l?}cW8!g6MB+>?Sfrz01d2(0#_E0Qc1vtrCE{c7{? zj#s$LFU6UJ%nY)xJB|58pg1QGmPS0^tBG0u*an8_oJiB8i(1Jo$(wXOT9K{G#EV5G z3k{C5_w)L|#Wz;qo>^M@o_=HVa<5Q&w&B)9Bj7&8IMS}f2}|#=>mJM04mS+k`Qet# z?KBxDv>6oVR@C~D4_v&n{>%3q?ce2``IGbZZH;vg=K62EblTwon-SHv=_miJ?+AYx zJnFzsp`^u~vBI+nnVWB{2KMg~6TO-K`qiC^sXkAIUpEOUtJyU8#>8h$QLIr%p6$Qn zQb-7~{58&bnL2>3jhm@j81>AOnV>v0FdMlGs zrMumHP0=ENp_nEhvDW_Gf@i$aq``}pDlf(rw&xnx_D1YoicD0 zwJK^)?~2XF-Qff|-cguR_jAA!`PI6xa-~F@v!aGivRjnW!4H3q1RM<9N?(=;7f-Esuri8^}4G1`1b>#g|OFapyz>IYPNOXbJxT3t8FJnn$8?$ zemyB#wNVl`L@0^vBsQhFvL1VEJ3CaYEFttYCYd`~)-upaST348jIVWw2#@?VNm`j{ z@37d_>GA6jZ71t;1>RS~Rgp`RGI8sTQ(6^0Zi6M~w>DQCJ681DS65q%s?#EZO7Gvw zvE1wgmf-HLz+=umE(?QMz1r`p>7E^qId}Jmnh(jS)cFrtnmZ9X!si8ox))z>*KiA2 zh^A2{69q>|vq26Kya^`_^fLUW{QaSF+@yN1G1b*ofs(D2JLGWWUfQK3NyiIj&Nhbz zCVMvJyRK+d$XMjKZTsj~-|%NuD=*goE|^{EWGg4 z%`kfBi&hNuqJkdGv7F+{9pHmKKp<5 eLp4$L+dJbVLly&O;~ly!Xo literal 0 HcmV?d00001 diff --git a/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/convert_parquet.png b/examples/flows/integrations/continuous-monitoring-with-pipeline/monitoring/media/convert_parquet.png new file mode 100644 index 0000000000000000000000000000000000000000..ecf335d6cde2763fb104ccee480716512d5d71b1 GIT binary patch literal 77777 zcmcG#bySpL+b?P$(jYA$2-1yos|ZN9bj%+)5}HD zS44g!dUNzbeWQFAN%BTT34W;iI)UMmL(VGSX+}E|{K|L*F~&D~NrRHZ!TAYGM#Xr7 z)(9G!Gr34c^J4G?a_V_zK_BZ181i~TZj7fVIcTJqBvYT0a)~mMQrsR1?QO^dKVkLPY zAAF#4GmE}rAoHPy54oJWDS8s>|CIPEF#0ul!{m|kXF0AuWlANs@X!Cbw0fPI(x;q$ z?yOU?HRUR3Sj#d!obTTq-+3>l!K(B~vH&`etzW`Ok`?t|p9)d%ON@PKOPzI^wK;Z~ zbxWQ_R-@+RMSppZNffwdra)2RALSZ|`~lmr_(T}O663MiZKamhDEguQ+%FpmIqPt} zPI$JpsBex;q-Js>BWj2_tmJR=4uu+9#aq?pAoh^`UpC_G=;d1OTcgHz@U<4vprzHH zoYC#9+APLA9YMK%`s=5x&|i2j(Q?XPMY%C%+tT{^zpkW|82L`QVvdxObDCqKXL?9k zKQdeE-)BC=;&t)4j1S?Y!hT*cmj7?}Vbc?h<`f*JuTBBDl`As7Gori5Q8;6aNiO;+wE{ ziY0FaT!eR#RjnQ`)D@0*JGqcFSI2GJlTN)Pwb6&Gc||bE^BH`HCMije=^YD zn39ba)qy+F-w7Y5c>L-`yzk`+^5$JCg4Qrf?PXRaeUHQS`#wi?(}BFbVrAApR${b` zFnur9jOU6&2dMr%GiBC$v6k96Y|?ZCPmR-x%%xOc3Focf;?ApsaJ7Ltw=Mh4nk87O z&ncYy)!K{Bks7BZ`)ThZlG<1q+mB=Bsp8?tYK_j(5>tI|eL0&($4(Bi%dM-+qE|my z?j(}KAW5Cq!{M5)`6)-ZG1aj#^Sd=-u`7?%b=)#%JN4?E$-wvGVCIiMKAvH%Gv-}t zcN3xIH2-l$-L> z+$d?Io*92rgj-a9s0Ya%=AHK1Ye`nJ`@SQS-Z?LE{KwyZKuNI0C4rE}dtc3cF~OW2 zk4eeT>&M6Jmwt@o*K5$01p8DI3oT~Nm^Dpvb!+(X#r{B$P1Tfu*V#tZ3=&_G-PuMw z$+p&#o#Lp-_ablpu;y&DZf>ODrJiNhGcvgXmOIJHVmr;F=PN3{-{jITuP0_m(djl% zOlwl)4HehPb7!!1_z~(;`nO#O)aXM+K1B{Fi*m4k-)@>I>y==0h96D2O$Qzf>vOF) zoNWej(&8bDL-$VCih^8*b(x0m@=rMr1g5zC7WO(A(hj6$)3+{ZX%d*QCbF&DsEqe# zu^TOh%zWSL3AtIy5NA-@S5fjSLpJD9gxKRk{LV(M4QE-s-vyQ9tX@|aN01n;mmiTO zM)}68zN_#YnD5oubh>tt5w44r*$96~?~iHsxiH9>SzOq_*K|6E;YV*zS8}ATE-&U> zS%!`K=dWjHmAm6k)NVGt{owf1oX?MBVSBxjlKJ0vJ2aAsW_~iI;^^DeyWo(jaS`8y zuNBsDt|P84waBy{VU#U>@q`s2E=3kiIz0D($s+sGrA!GzAbUDAu`|g|Js(}vOQ;Z6 z7h9(HXoB&=@AEQtfd}PT8k`BDq4u{%iy_}Uo{PZe?GZAC$kR?bv^|))Jln$G`o5=( zqgfIbOsi7`)<~f7)#*qH`5*of1EQRxPUFi`TAd@XMq5otjpx6%(;-!Hw!x5xU1S}y z5B-WxRF8oP23^-IjnF>|PPe&ZTG@sW@I5v`uS(-AM5j_t*7x(>&XR@F#d7rV95ypu$R@BRjt5P&-YSC*_LVX>BG>K z6z{_^9*jGVFS^B#CV8|;pK`%!z`#Qn*3~^@i*Uq9;CWgeE z#C-7AQ3~PrkvU&HSYIw*Bu$?)DvL(^A@Jk7C3j-*;!Q!tOLn|AzC7s^)ldz*HR(UL zb76wvei_i3g5n>yu$v8((dznneiWnCWgdPT)Lyc>Z~yqZP)=D|ZJW!(xDMSkas;!O zAGeBOMp8NVJ>7wPqIzEOq@>uZI1p-vmLzd?zV|F{_0yWgM#Tu8wt<5*RVvh7+vu$s z&W(ba#l*mLiZif&nzm<$M6$kBbUp4=tdw}GnSTk^QrjO= z;cWS_H%0YLU}}AklRp$KhaRnd#7Q}DqL3b-+wH_vYnQCr0a6B)vp%$ z#w=q`>vW>FrYUl_6X~Eyb`5X69(sZM!tx%6C76E^Ua#9~nEA-`<%4jMyH5SZ;RKS} zb6u^Znnt)Pmlv)hb)ht5)zJX)jFswPTrKxFo(V$Ig|HK@Wvx!txsmf(s6sH{7vwz9 z=he$!L*&sDBZ=+l(z=DZAw7RFcSF9!K>Ea1BXW;f)E3-H)$lWADhUnc!gR1_9=&u< zkuKqL#4=O6=~hzIToG{dHhMD`LT#E5*Y|3c48gpgLRFK1p{^9qP5V+^xnAxBHYr^Z zy(qUM0fb8 zkHI3=!g{DRQNW*H+=Mmr?l#^BVf@Ydqud+IldcmsNbjR5?ftd4auNj6pQVDae0EyM zNc=VX%e(#oH+=kuvQM?OCqE!yUi<^`N0_69>?Sy-V`yyrh9?H4|E6sdJZg zCp&U8;54({(79iZsYSj4uAdFB35{o}iNbLF>=cYOl0b^va)Yb-&wC-NmZRM2VvR%RkM^cYK-yub<5N6!Oso=RiPhGH~H5r3-yQNh?JO zp~)?ABn&qs#;zCefIuF7>H`Wj`fvj!u;ZQ6-&hh?c+P*8ogG=fJSg&QS$I7uy7irR zy>_iIEO=LSOsM59OMmOgNiJD1*9l-BkzMDA$)@G zlz@bcn#CE;KECJtwI|6mLc9{nFyCLEuQ7E8Zezg00ARU7zx!=|E}wGA<%zA5W$&fZ z#gjDsRpqKd37_-hV+BEqRoU6N&6O>9g2s!;aU!Q{D0fo7VIlu2M{RWrs(uAU#;h^P zefvkk%kDW_D*rLF3ezZX9I*X)BCjs{>MuR-=X*pc1d5XURN}ZwB;UvD>b#*Yv8=(0 zp~)hoS%~mskh2o`M{lF`(A*P=OujKLM_RbTbV{pbCD&7tw9Nzgr&sa-^5X7-w<+r5<8yk)?RkQ-Ig)DoA+bI;-k zNOes7iUCpw6&lu&*1iBlWz`~aPe{!+S^Z*nsg7i|9Z3z*bf9S8BXWVC2gu}?P(Sz0-l~? zpxiqxJWI&@_@44Qn7;7WS(SGS-jIq8PqK>!OmXR^?2suXc~XS(9O_&t)L1^YLL_q zzE*^gKyODqxeLx%nubx0PfO-ZlE@c&B?T)&YdWF5E#E6q97=)wUt7`WHnlEmq@>1P zJH_*n3NV)uL&5dZPEH=TK@@%A_Sj!;Qmx?iI^HHtCDL!D%nvVJR0-ZGt>EYnXnk{sug0v}gMqFt)ltIEu0kR8~^GW2v-8 z52d$Uwa{*4L@{xn*Unw>JUqG2RhI!{z?iT@c(dCc0W^Z;r22T2QetkPC8Fa|kq;X< zbr*xu)asE2W~*U)Kt(c;qiXUoJKzX_{wRxX5Dy#sUZB`S&|MlYZuW{nbqKgwoV{*A zV{00K2as>z9wHOJ8vHPq_kTJRK0SGoxs&ZI%mB}LRsWZh+@ zp8;qK7WDFLK#I5f6}Vbf2$R2F<_>8aiz|oJW}xP?;(IL46D>-y$JO+5N&}nkm5jX) zw3f#!-Yf%J+68IWerSuhIMnG1TLjd}_Sotk)$!627ak!M15!1r0yvhfZXPQAxLa7|Qu zSa}PnegJ@(y7aNXR3?-Cy#{YLpN5WY%>o6kLRmXq2r(=%fX=n7N92u;MS5V9o$uL% z3{t}`8g&87h^LHl-}062CNZDN)XQm3UVw@j^ro0@)L^M6#q`I=hlB^*_dd)v zsoWE!2u7OSv(Pi~yE9v!JD&u(Z@&Gf5mz`SUO+ezhWCKNiSJ-@O4JJwX&ymhr0V?> zz?Ny&+)>jtSSx?WC)h=0Nr7TK|UIFK&xLU){tL_USqbqT?e4U(i z)>JZ)gQLU?*79WYj^FOfpgzlLx#}cckSgU>_&IZxPt=@9?E=LZfe6Aj1(6v^K92EO^mA8w`k(q}X1LmpT zi(2|%4Ar#~Dfd|fjY@n@Y=^>AFAMiYAx#0tNgK*iZa!@hG}C=y2n{J;pSCbcGE{95 zvXy<>lg&S@^C-XSx5o_FkL=Dd015ZO<{mOOMBzXD)_bmQHQVn{0CpV}ag>NbJAu(? z8}(wAGkw<4L4@y>P*~;b!CbYOQPT)qP&0gvS8`ID65kZw-vk1cyN${zK;O28fy#UT z*9=(XI4PoTc&M6q+$MQ>4*=U2m2o+$sL}5<=+NUp!Xfb~Ejd zB5-b_<9ujjeyIh5-xvIw>zq?I<*^!>{O1@V{gA0sXWY~lE1iFfV%;VW^ zF6HLz6tn0{D{dxK2nZzAW5!eYwBCfxg`*<2^IYi9yW35|OUXpHwIcPjb9hW!`5IPH z%s?IUe|vnZURR*l;M3N{EEoNK;{0&&J8}KfaBrY}!{#a(@WuC4M*(&s0o5eE&G!?4 z0us=X2mWgFICM;e?J>`vto)Lt!42W4)WNxAFKwrr0gs3U$gflN;rJUQXWcsC?UyLC z3D_Xg)c}2wy?|T`T|T}0fOsPH?P5ZF$VzTTcyMDTuHa-be|h^_XeBcWdcy{oAh?p}+fr0p zZ8W|(U@AP9f7g3LUdR=s4**)YR#b>8RGLq*dibpVIin#{9wifWXw-pN@B!+8=|!m(}SdIgrtcI{T3JizO$S z5Wy`QILFRQ5?aQ|R#*+K-l&{__ORFT+Qixb1D7+D1J3V4A!uuGhWLuwZ;7mll%M`p zb`%mDuErVgq=BvWh0&+Xmn^UBBkW-iy`|&gwaJ-IK>sa9Pgx-{o8%fRz^di_f4)+3 z$!JI=FmgW%>LZnhld!RB+WpplO+!TO zJM=D!-i1B|xYtW&hNzmLm8c&bx+8PNO)GYBdIZ_ND;e-Pqb*3cPKC@>&wSW^#AzDF zu(U-nY*^{JJZRa!tvlv=dAZX@g9PYC{@dqlW3=GSNa0MHKOPn~XSIntFh7yN)Feci z;w_sm=5nS{D$96$VER1!5NNG7+w*>-qshT-K?EOLe(^ID35mbfjamwgiR!r_E455A ztq=Rzr&h7mxjXWtd@5jT~!=%{#zw+K(r;xb>8!7T@cQB96S)W2?d3zIxB2_z;#EyIV?;nZxGS zMvhkJA~!2zYMEto?#@hNTLb1q{;x;uVk7&&&B~MRVa$Z6_hbW}%Z|dTk_pzld_vdX zZobCP5{p-|lVGSCuPh4ZqgkC4Ihu6p1cIRz2o7kPoYmaFkmjCbXbO1) zxiF_h7cdclQh37j0zh!MrT5OapFA(AM+qM)-}42066w==wYXwaNHX2xypD5p; zs-ZK74Or-5HO=3i61)LQi>Gfyb767se6KeMfJnYI!3w~Fu%`%6MymUy)%C9pCH0W8 zG`pkXv(qxzSgp$CatUNQ0a@G0FWDE}#f`)%aX09w(Zux21_Xj>G4yNt2VItVuSA>5 zG3Wy&w4`1 z(f=3a)!$0&o7bX6FRPUOlA&z*^Kut>KFtQN^NlI8`*-CNPyaKJgJb0#y|1xSLY_x0 zSIvIqg^^94(GKqo5)U>$Pm#kU@zEF|yqPIaeh5f`EqQFR!kCE+&-@o)^wl-FO%IJo z4h^dr6Smv}^s4V!+^#DaS#JA3KVYzT_$1C`*nj%Mf`>vDte_1nH=e6WIleSwJ1FF% zxh2OEr%M5=EyGyn!Ktax;aW|jGl@yT^LuitU`6@w-bkjy69~R|=B(VjwR&ro&B^O& z_#*J3LWGk9Chd_;1X-W4tx%rA!KG?Ic(Q!Vs2-u~O6l(2&r0pn zIs-wUGF(Q*++N)luBqb%Hs?QahqX)DFvLlgcMFMjD6*Vnew}t(My&!0+#@xumT^lF zv{8(Q3|X~t&s6cU8s5jHQs4Xzw3NVbW6UkH6G*U9l!40pClQb<=CKMe!xs|DP@7o2 zn}AOQS|^NTsq3Dq9;B&8NK4zWqR*^jehCS3I+b0ds#wHVXSDQ4O&q%(;iXUnU`^UCe>n(DRhERr zE{D#R%*Y)g^#fgVGtA3m734Z0Ex&_F(mPfBEj6wPUDz067NyK)6(wC{8@@UO58d;s z1)CWv`jqLGbqp(e9G}az{>`);Mdcx+`qsZxYXpUZ0C7wF>4>G~2SgIvOES$W+bIeq z|JnkSj*@8^XYCGJQ4a{;9D|g$tdhs3ksPT1N;M(v)cltFC3E=KPm#av30iDJPJ@~Y zt!%?wAfPu|*>apZ?iN^Pb^%_b{=bAGl71-k8kHZYchmqfQ+vJ6)bFDg(8DdSe*jYR z_5aN)5#auK5@GxP6=%9cOafaw>67k5ioBwVS%^JE`&9~sVmvpb^sh&Ingl~=oK-u_ zHN8%vnCZ@ z2q%`-B()DcA1E?L2!+zaOUSKPf#sF1h2jG83O5t8S+CXj*rB}dlJDn@SQ}LqAfHPc zvr$|bLN6Dl8>)=@WqLXOYq$FU$5(^aL`=#!V1>JmT9q;zJp~LB8n2%C^0W_wq6F(?+2f+CsMr(Weft6dk50~i5B8_hm_fw z!xofglxyQJU zf)88<)o}NWp&uEbpDaOb!qzr)(K-fzhxFR(CTnX1@ifUW@D7C7=l)s|z>5vO=yxt^ zI9>fqY#=Bq#qBuXKAO7V0rHh5;5JVbE2Gk>QQ#;_rAm0Cs65c{!kI%dz^lAN}axtDyO!(wab^+}ThB#Nm6u49565 zFubu*y!9A((o`sWc^fcIm_MEf^WV%6sDF$g8Xxz+UahIpvnMO%1Ii%>LOhe;D=1M7 zPPYx&Dg781uOfWUAK^R@;p3z+0?irrppFFmwFQ7(8#eXC{;&HF09{3BzfTGYoI*0& z>KWMR76^lIC~eBsa9)UmE{GY)s}mKJ;2j0{Yf7*NEQV0z=cHg4hH3%u;Wpbyz3}VV zUww!&uu-6H4@RveS#19J1aWnSAb&?2ghD~8HKo;wD!WmjmP}PYJS1#PeFYqNeBWav z&1^B68yNGvZE$Vk(_s)Bj5U3hrM`~>mS=IZ80*Q79V`w$;+}`0!dc9&BYE0ILb$h7 zK@cD<1PG{U;UmY#Hs5zzQLYeHhHOOmfJrrxiWrZw{{mgm0PqZmd_aI|KNCpSx`|ee z@%7#)T^gUsk9YAI!8eF%62E|+8x)^yHHpiNy>|Q}MLvL?Al$__pa`j1hy~3VvONr} z+eQG4Iq3FKqyHpiqTiM)Fz#UisG&fndmSTqpPRE(G&4=(YJp(_?X!#!KeWL3DrcpOow5i!OY6QuXdGDZ51{fcCt|gg zL!c>$Lb`dB8Hjr`-rM)D%u^*gwBE<|rpS12drJ*1X%Te0#GA6vn(#xLKj^@ZKsBcf zsZ&cGFa|1}8w5RIP{GOybx|^FV}qG!>M^aR(3wbkYg}C(Axy9SW*ewRmzbC%kec-< z^bhiAy;pkmGxKZZqi`WQqkRh>5F7<*sEE6ieyGc~k-R)M&EMYgyG#86QI0D0-|$>$ z)l}{o`(!XvN_QPYsf^!dgl7XIP~flJn%$bdJgh1Sr?1tDnhSb|d%(p3LL;8`y}IcO zPE@T5Y=|O2n@xzARQJ*i$3j$rAZ;|lBNO9^+TbQCjr)(vpHswgLFv&fDlGL=!UogTE|+u}V~9QI1A8s3{j?$l zMSqxa!*6szv@UX0KEUWqtR_$-f}A*md=d`?6@}h_e9)02}<7k4)9C@nl z+(c_IdNSMtLM>6C%E7d%bD%nrN-~=%-sIooY&eFpp~q#HlCsFiDrbY{7S@-2GQv|R z!??uDj$eJD4{UOeu55V|t6$P&xQOExF6bRQmK=pj=3jqX{7@SEEAxDFJP^>(&orzx zj|9<420_n_69nb?nR)yak5V;C_#kM^1B%@x0oSuZqUQn%tOaKv#+ws^TSPei1Zvh3 z(nY|DINwDPcK4f)6c{&9wj+C$Vj<`00?+yv&rMy6B&vyU`E%x+bKs4(*jh_TO#C#L zUzv_RfQ+$3$_Q_8H5}97Asf<`=}2y7_rgfo5@Tw2V(v<4Ctf_aN>p+6)~N)u@G!RC z2e_fLuCFfF1v}(5HWQ1Ut7?-;H|$o^vV%Jat9;|Cua>z6&-oHOaR{ux&?%ibpNrE# zU<>4{-3hcjL3sx5?5egGz>tsVY!_ROor-IDRfQSjP^S3T)c&G=6O!c}6Fvm(?9S2k z1-40xAiL63&xyD6pUQrLh3a{t(&Q+r0TW#oTkUV{ug3A;j+t3@NU$(^im66WE{Uym zmUNWJJ@EiRmsPwnf18f_=D7jL)}Sr1^WW*s?+q{Uo0ckS{I<}RCxbUX|9 zA*N;A?%5wxF_5Z`weQJx*b`37RYA3sv6O?D?at;$t!EYPZXvFG&OIxJUgnAP?F0T1K%^{0>+|*o+HH5{ zN1ygVyOucVy^cA$UTtFcCkwP;cU7o0KN=C;FgXW=o*%DM%J|SPIq1dvJ37F(G)?>1 z8-T(Zo_9aG11K@@KM^|#19O3rI=4*XWIl@g#wqYkC~OtQyk66YjiKk`Pbti_%CwA% zjjCIJe0Rt1hQ!6s94UZ$EsRC@evFm4XRk_HNJF0Dy~h=ta@Js{)>81Yd9RXIijyz8 zXZjY7_QmBZRAT(3+>%w|=jAMIvJEnhMIM6S4ny@5Y~HS9!d-jzWD(#Q>uHU~Ei5Wg zDhj7sw1(W`)>&1ecML2Gk2N;QyTx9Cqe4^crPV=2z5GP9jT=hl zZan9XHBJ11|_PR7a_osv*Ggx!V_@gVi+6@2&}qsjV5%x%)H*ERWkuKPAB& zj=wY4P%Il^4IpNN*`R;^0Ja&irpzPNF#H3_wnwLphMzf6I&qY&%v1Gu@#8Jy7_}v3f4?Tw@y^NHL497K#UDs9dFztHyy() z-|p+)?N2e_w>EdA-ic`fO}I;{j-Ptj5n`13%+qz_yDxx?KJYVyluQp z0Cs2M_msV};3BK&iEK26pP0?(#}qGE8#;uX6`p?k?$~DUbSTHOPU9$3;yob1=yq>O zVj|MJCC7mbtkvo-LsX{LY|$c;HJsZmB(V`@3L|*gZ64V-h}U{5LTu5RgkGyxWsWu| zj6AlV+_{H#C_E9i@*U_l9yj786!KAy1bHjIi#N*z+A^{B#l*btp07D&{{UYm2ZLPH zo6pHgVg$SIeiNgqoXQY`kshQLR)%<({@R zkfU;*6l7vc;104t(;tCBPqKUcAK~YU3GMz#=yv!-z5lbMMaqHHc=@Z?b~M_C4kPvR z69)1pe;&>}O}atUNS|mma#!pJq5UHvbf1-8*GG5dHI6id*+&5V9B0;f_VE2AlgSmo z>^RpbP?8kyG5^?wliA~xTOZ4L|Qr2uW1?D1CJ1c+@rk`W7bat&68Hg}&DKY5y%C_S;t z6KH$JvGYTYlZ=`Vjtulj!D#vvl0GWM*+>;ku|RoC_~g&a0yUg(qu(vt!`@pB2hqt7 z-jkZ*8ze#&&zGmETRDagQjl5R#mm5eh+r3(UJ4B)wTzQR2yivBkDT7LUa(r_5v)0Ragsm~8J%OnPI zrJndhWtdXE98rnqtadYo4&bf02+AT^n1sMqk;C@{a5RgMTK(pOu;+gsdQ8spOc73PrJUIlk2 zzX0AbwoS`DoENE@OW2MpPV1Nx;VL8MT6#2<4+b7`$2f-m3d2gsW07e)e5mtdm#AER zt{7L{TO9MEUrKN;%Y?M1Ks~Utg&;TW#Y(kwXr6nzyw-{QpJwU9J<44TNd-FSYyPnQZpW#9E>je32qIjt>%js2EM>rp)@8qC@f>k4l}!iyHnIyGY@OY*O*qr&oas1Ly7lZ)e)eXPVnFl`ym<~ zr0TV;%d)9V7kX6QdypAAt0lia)RV$vy=9NqRoZsfPqGl((tS9_Y{`A$G* Clh` z@u^2VyR-5qO%zxCzBNJCy=__!R0CO%X}EB9pNk`!wL9ropAoOqF!K-K+|FZEZxQar zE|Q3t4p&Oq=ZnINM{33^!1*Fu!{>md-E+~8)K7s}EG-zx_wp$9*JU~k@bwGLJ5K{Q ziRn$;pg_4nAcKqhyEByPQSo8DR2IJX!=KA|_iF=5Uqp}{C!<)j(>;|Ms?&otvb8ul zzwAXi4CHyWpMAz31)CE(31YX|y*Yy}_w6~q$b}azX{E+--vV|{LOoCQ?hkal)J$$G z-QGr^jsl5TxR1X6EjC$dnu z84f#sXRnmIM9}?4&flB_)h1%YF)j~}wep)>W5b?c9+IH5Xe<~GqlMT?&Nr*B%IhS(}zQ11p(+v+eS;JRfj8Q&Y*zV6;V0V^WNwm zSeI~X&CAu+)M!!In|c*y$JQOWP3qVbo2n5bPk!Pv>s6kYc+#qvM!eJ$!A}PC-V~7Z zl2sa0V=~4DnnB%TKbwsKQ+BrW)h}gexObQ9+~nNo!*{I}p>vF$`9#ID*q-a2Fot$! z5Q$);dAX;{X;Cf^46=||+?qPxmaC*Oe#3Yq#wCU+^LZkLEDY|9jVK>TxonvP9vqL2 zqOJRTsNQQuE7GR0!EG#lB`Za0;@!-ME`zHpm%(Y1qVhP{ zBS2%KS_kg`VeX%El}?r7ia>i9Z;5?>zO(@BN(qWOlFqSc*j$T?h9mmDb`cPhnobqZ ze^J@%Xy|%H-tRH5C_h9<&mybdF{G*JGbYdR##L9nSV4Ww3UyuJeP_ZWZs6Bcvn{-0 zY7=01cpR^&vHH{l1bx$Q*UZ}V;GrveIO+*d3&fJVWeD`mRl`nWkF&&w zem;K^^h2f6!pN#$d#wRL2yDkvm7PpepiR@L$f%w|=@G}>mOaAvny6lcd_h1W8CPK) z!{BxgTqyXh2AiqHJ&5)Gwl|i4Bw`}Hv`YLRP5}|4qQx=jtp}k<1S-lVwxUfcTr{M(WrvkRu>g?m~Cz{R?@ic9sL`g zUC~W&Oqe`d6@)xn7f9c9J9nqHUj5>_`af>ny5qf4h9s7X_^Bef0P-v3B<3_A&^$@l znW2==$C~`4TXv?K%m`1)H(4}aEX`P_-1CANwNS>3h|?^;{g1f>olo$}L>g;~lb7z> zzeb*oG*AY*kR{agIXF(&Up=qBI{5|J2@}BW?A2Qa1(qIeDF6Q!-I;(LYnAubML>!C zzR~4YXZo6aQXDQHWKnt;Q@?On{QF6l!b~1!U^wblz&Vjmi6qwG99qascfw1YH=KhN zeGDheVm=qV5Q}t(MDIdDeX zJ!dwXxQ9$cw8aHW`(Em;oGf_l8C0)^)~&^Uy;DA4{;1MWLD^UX*}hbWWj;C3dI$}n zsoQumiEe-t`R&T=`k>%65svrz)l|Lcl)ux_+_~`f=G4K_L0)PeW!`~(cRu5Tw9K$9 zl8!0k<&pfkH+stW{2UKBQ@(A_lysK>CD8+MTGFrBQ)1rp+POPAxVQ903;_@4Xn z33egmft@J(xZM=NZ+Eu6Cr3u)y9>4HSJ1$h{OX4uzb2jXm8`rZZN2aYr+mMhJkGDu zHasiLgT-33&yVerio{eOk!A^YI_qlK;OxKY!{cAMIe2U4cuHvh;DpwBGn^~M;mBts`WI=7P zFg%*?pN9DZs#5RYTTDFLu(yQNP}-EquW97?c#=+Vx#%o972)Hzx}1u1Atj)Rw+Bjx zxQ;@2$rgWZN1T(N3h3EJj~0zByeSvQYI~0K;MuA{)>%>{xFlGgP@eMiQX+G(&8r-| z&+sFX@7cZ;54dMHAH?Omo^^3Cz%&xHW?Q7s7+VCtXJHaQR?k8TNs`;Z?Lr~Y%<_BpKuMc`&`}n3SOjwYZL}og zT9DPJY1qk3>DqL$SB>|?N|I>P?G3-T|uKF&HPah)x1<3uD@dn@GHj2p!BMg=%43J(TY9B{2N zM?IVjrtZs-9JXYdF0DOogTgI`j(&1^M#S}SIMy@AM>^$?iu?P?Z?v(7U$hQ%6*jV)^~jT=95=Bnvl zJ3n4gnCULogiHk+%dVPLwT-wO4g1ZcH&s-YoHr~#khoCU^bs8q4I6p0@_DAm*$_$? zcwlSMyVaI(H^|AU4eRKu`jY1PS<(vf$666SSIorku;_78m_~F?dmDR{T4rA0yCatZu7oX2`4dbtBJAR@JlkJPSP*NbLr%c=U9UfpG`* zZLJgT@df zQZsw6G$#{q^!q;LqLIgCUq&DND`8_cIJ^!?kvhUiI!a>l@;=iDyc?iMw>%YubeS$9 z!khBm>nk$eNiy;#_~>)+F>EAmTy*`Y^I&d!@4L5U^U$4!cZRl1CR;T(4arS&(Lfa}bxmR(MTximpIFPjqpZmi_ln5G3GD0l{SG+A8 zY7O^oUk0)L;dt)FGisiv{JSwh$fPOlWNYgD={;V8%lZQsefj>HVDk~Tz~JAAR>szAi)3+x+uoNo6p)gzT% zCR}T3UD_ptc8;PVe06k2-mVhvjle_`>wK4V=)(6DseJ3JA3*8&GffM8zK4|grO-RA z!>r<1e8q-s+9KAdNf}d?R7fk^R$Vr-dZ&l9Q}Q;2sQQa9<{bE!^dJYDMZ}VfPMyEq zDTfOlwW&ZJ=w8t(lwOVYb_S$U*VIcqyic(;afhsz>QUZ^5@K>EQIa#!{=7%Xce;Ah z!FUk&$OJ+oU=-9k7cG|hO=y}j!F02BMS=w8wF^0RTnR&ZxFBOiB^V7o*JOLEn8ZFA zon0{E0&1&jWvF1}Pr=~};8tEvLoV(9U6h+?ojKJ`<QuTqh1BAC6?1oGUQishF-SG_TVA!zN#3z`T2k8l;N*$@=y-*_&IEnw(# zvM>{G8jguNuKo7_4+t8GQGIE}o3(2c){%T#d?0d8hV%$af>W(V76rpg$fj-KrCdtS zj)N0v)(nKja@qbi_j>2(W4T_V^pUkHP2HWPIEapppmhoy!D5X6GxLNbFN5B0m|M1Y zW~M|TQM++I-)SEQNrR~NdO+l)qW4>uY>UXl`^0Q&Hx$R%dyFlAgo|S(9}?r^jpeJI zp)x8kbX2C3+GjqxUCW^n=YIKZ^QzV+fz=9eyfXtiCOg43Dq8kADYZR@LX?8-n~pWWE?su?P- zfR8CfQ8}G?0G)tkosq3RGLdZQ!A-184S|csu2xQ%#t{!6gx20YtF<*I$;19?OY_pO zNa@#wp>B~&6=(jNUhbpG1XKULjLRa%jcGZd;m2(XDZ1JlG^gJsGH3r6YiAu*W!J8I zMHEm{)X|jj*u5-HOta+sV4{74LpxANvc=@qfBM_D$ z^794@@~LoiEL_K7W#t|PKaZ3AF9;t_RF({HX#F7_l}~nfi|9wzWLtlC@&55pJMn3e z5n{CJD~DTk@myfX_5+3^Bm4*tUyoR;4s{M#!1po&>HOv!CVj7C5M1kh+*o z3eQr49@1y~r>s;Ge}=^2=mJ0TJ+bn(eS+?OJNwcD`^*ym!XEki!HTK;@t8`P<55KK z=+J9Q@8aR7J#N{-@ojrQwKKUd_ChnV2MrxDvNGsv3JvO1HAR~#=iDDOrG0+c-4QgM z)zJRRNq0udI92_N*G&22#1avlxe_-~ucM+J%|UT29Knvrp8Q|W^;_-Ukfy#Q6D<+) zWFL+X8=$+^VzjrX+d7YGn{h9|U_P@;zFFx0CH3Lc5z_Sg$5<@q9Q`h{L zlb?sYX7cWRx5zqmG&dxLvzpZmP6DI&Ll4tX9xSg_+_Xv!U-fwYc3iG-(n9q?&1PaDc~;!1%dRhhM^F64-Z%1m1h!;0O1ws_emVI zz}7h~!WWu_EFA~=I_JVKza!<{C)tZkl0-x)tp}HU7kY(g)O)O)&fS}bkBjAWX&ZNH zile7o12d3}j;pZool^-i-16zq-?ZBlowfPsMaRl1>5y3_fdrS#HwRf*{9SkIZB6^Wpl4`eCLl>%62_bdIgrol{qVHGVZB1{wUx?yfC%# zT{x>3FVI4I9sjMw%0$By!Ljlzbem$K-g@<iGv~>JHThJNqA|Qd+C~x{$w=G(JHr3 zNbAU(;ugExLD6SyO`LqelMif4rJZ|DJz?-IZ1qvt@_#>Pslen|@ZnHDx?GpBxBg|(xKQxGgLqMIIl+&r^z~u%9HpBN`&0x7eQV|T`0-tMxJYk>bueF-_kQ);Y znriR&-GP_Xuo?26Y=U_xJQl`}0r}h(fp|;oPJ9L+?f}cp0r87`n1{aGCPGd}Pk}=2 z(Yj^vh*>B1k&bTL%$0|*yC~8(`)D(7pUvaw5`R9URU*K6;?*Pe$or$m4!4_tPgf^vv&Mt)QR0ufbcz-xcv~eWOs>< zWli>4cdjsukJ|^m<;vs|+E+I|g+~@_K1`e$tQ_R?a9$hNUT6Dro~x~Cy21lrk)}{V z*e_{sbOy&;sM*La8nbKpn0iuyy6e|&KgE24bigNgsd%<^(MDor-(^Ph*3llvd_sd! zxD{e3sS`9~ijSSt#MJFRzbSaOd}%xfN_Y;(jpzD3>koUMMRrwV5&GJfY*TF1_4M?G z3a1up<=4%vH#jzZu=e7h%V;Aa~_~(d9t{=`9WUQ|JzsrQf2g1(}2Pn&8!?l-pwcka=3wx4Mp-l;! zzxlN7uK4{ktVu6?DM75+qhiLo^TdrHg5h+GV|$oRx?4vi=A3syT+8xX%*M2-$V5-= zMC!S7@anI&%s{dtFVl^KHuuUgt3lOLoQbCN(s0KQgknTW${(%C`m`;lHV&m0Hy&GS z^Y4@QzNneC9b|UYOZ-e}68Cw9uTUh?)nxUXOLDp1s)m&FqEYQ`ld1dmOApR?A)DT< zGy9n@vnSkZ!fLgd#2Mn4UpSwyleblDIAQC*9{8C1tEup8;+Lm(VV&!6y@s$D_mrpd zMwG}4BHfZdLh*C_47Kk(ei!}`N|`E)-HNoISW|ZUF8qvXp*q0fc=5+a|A4@V0nF5A zEa5}HNQ%nSLv>O&s=NLSE$uVE)8(iab=(RcjI^Fd209LIA!&~?xm+KH4^Ol{2+f_J zX^7A;83}tdKmycn^*g}b?GF$l670a)#v_1tAebkd3c=4nkDAKA@o<_A7%S-9LAM-X zT>vD@LHg$l#0a%rFzpbS1&VJtkg{MvTmt9IlT{=baN*s00^y-Ft*XbaK_5g~rrF#; z)81u`nh0_CE_Z??{Hw>c*mNAnp2M3?R_?KVs#5F2;sY(@&SisSk}CkD!2aEMq<3vX z1=?F3&07i-QrBjI>u&8hBk9F6=XHchTX|_{;PT6Le4-4im7(~$v10^$Nq{OqK;Kyi zlR`va)1`3?ne%-^&WufsA^FqGKY7d1JB?zN6i&Tky*WLwdtEYTQqPKLvIn!18^X1G zRwJjsA_{9d&K_e$Tf`_<7CecTA0uc@SYiI}-U4o#gH&!BC%P!vXWTofR=)(SXX_?Q z8m7(dl4@3fK=~ffiyC4Iegcpao~Kn{KH?9Y;r!%w?e`INvye#~l66B$T#~Sy_cK5R zPVD#O57W~)x$=;#(K}?8L)`3;OsstDg?0cm^}mxjuIIL;xf$VNjOd39NIM4;`dw>R z>o1!_eAw;9UzNp;B6vTN==yEn3%pFVq2k|sq<;K%$Zviiwd|gNj{d}~yPWgeAtrdI zI9Y&|fQ_3EBPDeq`MZ#i(khoRW3xnhh3CbOzrExuY%)htsOE=QeZb($KN=f{Vx}H{1&kmUJ0jeL%ZZS&^3}@BS^v`7M$*+=v$0&s8E=#cu$8BjY6 zbyXpo!=`H9GQsB$cn%>G&*oyGHy`p1YA~nd=jNEep8rdOU5o9B0B|zw0MvoF4sg07 z47b75moe?#4;Ns1V;0bW2b zh4{h9N&t+6V~ab$5bs`65WL`kMs6#)$yj6rp?^T9+?+*nA*FUujr$X*8(|G7* zhSkL6{`KSYc1;@Q`d<90vdCtx$qoXs{=1jdyteJAjj4UB{8x|9(6^9pjev$U4{QLK zfKF&-x$5#8sx}}!r6t!2w3V!;ztcPaJh`JPcAx>-GJp{dDuZqRHp>1|48U}_1Y450O!JIZvcyq@gAo!Pan_~Jmo2m5Bu8-3{gHKZZb?~g1%dM z4`XWkC5vP{jv$|-cuc-{==B;@LzBM2M5ZzQb_BM_RqwOc8TJ4lDvip4ewa5*-SQaJ z_su&m?NCD?Dk$S^ZwLm*4d{)5fNm;O^XLkAqrdkY#HGJ;SfMjp|BElHP;3g>K9fHN zw|u*HIp{gh1B)X%X|$U;Kz=X_^&6%LQz?AYKUnhAi>}K+*4{S2?Yr+M(6Rn5XK%~r7Fy>DkYPZiQ*H z7Z5HMEQ1O8Q1C%_v6Fa=R^;C<9RqagXqh#@(l94T>Y-a50e3DkAVY7KPCQ+30B5Q8 z`Kb zL^J^}iL~~C8ij({DKHyH?DyR%*h6|-LzQxULdpk#*frPv`~T`ky(6sxtTh7@L2*&3 zW*@@&LpqiUvru^4o)+<5(4scn+LsTcrlGXdsqYc90V76-r2n=^WK?69A>GojY2jqkOi#`TG z1bEMf!yzXiAP4}BOgNbUCqDpSg(IsNP)#Avdrp1)e>{yq-P8#$>jxj&;V1PhN!ImqG|79TaRTq33;*?GAFt~)kr?nn z@A5C(*#Gpb5?BID4YZL#o9ln@h5m;>B@>lVp96y!I%Mr~23EQ2%QR4?Hk3nXC4iLR z>;vyBAUmJ3Dc5#6{$(J1U=0monvutQm-NSN$R z+_waa3%V~AIlI4}c(>FYJr5+yCJ?wp(ONYy6Krb>Rr@IwTgAyQwTGS!mT9rP+vG!A z{R2#S(<_6`=Z$FYTV^NAiCk5Dx)$m9eMh6}s#u`lSR~-^a;O^rA~XOY3{%o4+_Az` z`yqLO)t>8%O;OkyZ1`JV+~@p)uXUj;|2R_5xsE??PceCZw*==Z4scy5%XqP&C+1E= z`V(Lw!v@45iG}{;_7-p=5<*6_9pE%-SJlM^EIRI`QJI|$a0$Ezcu9^QQdQlgX*VDs zE##Z41N*8GGCWgOiccQ3Too_>9mI!}Rh34`{=_t$>SuLr8-OYR9=nX1IbbOytA#s} zX~fJ$R1e{eq!H~t_zu}PpocktGuHNt1i&=})U0b%>QB!IArS*SgNV$5TLT!S(PKJF z=JicmRx@T_mDdmxKkjQNZNMH?-{R}<@TB|H!bbps9|V`<)U}LHY(O1R4ovj00~dTu zv85;(b0SH`7Eu3*1c0?s1fx1bl^xl21uj7i%&_VMi&Jm}Q$m%TGZtXEr4eEAi-FF} zHNg+;imG4MN`O*^C)yFz-d*HRKowzv)b=k5nAKoPvz3N8WiqP+Ta^U!06c^buzlmd z-VDwy4LoyU>5lSZQ$VqYZroj#ZgLSJe_q&WbsXfm9c|A8g+ABYWPhD&l<2Rqk1w-( ztAc_$Al@|`l-bsCvHlK*z`4TD6rmN?m61;%RR-3pUUWVfYV%o z%JUe04dw+v!xz7370g*iAUkMRt?UBu>onbifJk40t-)c6IM!*K;kGA=CUhyvOSlVJ zVf(gZcZCCuW}zKt9+bWqkRg*q+}&9@P3?ROV$1-2v$`g5Vi=Xb4_b%p6cNZ7qL2vi$~htCNfNmH0DJ`9Pg_Vwoggb^>j4a@s-PRJQDH@&P+WKC6B@nKZ)5&5|9Q6M`cXP0YKD>?T<8z z`nc45bzy_#e=I|CewNa57R&BpXE-`FixunNjpi zNne4!h-<}G_aVbg2Y{JqKM>w+$7j-hiPsHFXPim<`3n+U3W`OnLhd`eSv5_kh_R|I zC3qc>$Z;{&X^Pk2^tQOPd>MWSA)Nzw))Nzrd@kvpgrP&H!zxp6yWQ}pu@%N^o2$92 zOJYbFSjGkp5o=Af@n|Er7Vn!p6ZorE>xsIxZ`~^|dbcdbuVR!qgC~S8RaP$!5L9|j?l8ySJ|>k6KfV^6e41_Zgq?@ev>}zGRPC7A9T*&ovAwI*q0f~ z?1UpfV%@)C$Ja$!vGLOKqH?1P%z9c6VOvHa(eM%Ms~sOA+L2B;AKxe4O9Gm(!{SK%-o^Gb7 zZHXJ{*7Fw_sHa#LZN(0Ae+_i`ig4l&GBPNM;cGqZHj@^>g_1M`cKT~H}>o$QcHrlgzTG*pU$pPG`3hgOMo z;T+xm0=QiB;5&QgetA4IcS)3Z0iK5NRs6vRn`{|0$IG6BVhDg2+tm>SmEOWm^XXXG zLKQH(S8I+_Rv^{GBFJ!v&)GTmMBj2_););56+(uEj%uf_50;q`F8x@u^fv^B(DO=8hgLuH0DlIf-rihn!%$juSs(~rQq{Thc{-}A({XU#!|UJ?mIqd(SU%J1OX z&y7I$eO59uDxKpB*!Nh#n_?#j>_iJ#+@WM=;EGCRlk(bS1?%CfDyc)88XTcMboX+Y z{17MLsmxyi9KPZk944O*tQ0bd4vOvv()>_TBEPVPf!y5mA+aJ*m;72uETRVF69j?YfIInXDEAxJUncgHoDiP zWiQ`&wQe3xPobYijQW1WbLmU}xY)z$gler#Rmwk=^<%*Pv0;|svcmGVC{bTv-0z^M zs_g9OH&xIe@v{8PBW>mP{u~pyc-w0_*%=YzlH4&Kx@(VbZFKX+8ZV!MLy4<>zdq*t zkiJE!rIb`T(}HZ@NGHGbz|V{^txLxtLUgZNQ29`EIlGj@a?#K^x%FYCrKNUAj9uev zA(yEbu^07&xsF-r_v#;h6eCvpeoK&3`^;6RAJKJ+XfiuF$rGb+J2_bNOoHK;r>l90 zo4H3|M1Z{J7fMc_DROQyd76cy`h=$1qvTm(Otb`p8d3b#bpB;CLunx7C6~zhh9V`&TNw9E6AMsWIdN|@bTyjwJn{>y{cE;?X&CS=^_Y>pg-rA|Y|JAA{XRDcY(|(VWARW366fKJx2xr|aG%N_ zRNKaW@XUuu)HNpYsp2+N^{+g@L2IX5cmG*uYBPpf^!qFlgL*|g?YkHcDBTs5!D%y_dkUbK$Gy4F2!N#x)bUn+bGs>-6)2W)DLg z7c~_j=3|mPNBoVCPZYf?`V|S|U!T&CJ3G^4cOYe5lX@8cD9uk&!<8i&=Idw*<)u?E zl-nXNG7c#)I^7pOrx>TFpraOS4GQnvNYq6m&=wdEUHm-jX><{ys~%M@7hPq&Y23n%>z z`Bz-~D~8IgrZUXLTv}ig8c%cCIRud9110KwGquBFSseP2Bd2**w>@7`jo*vk79vL@ z$5}gnEr>F>Ce>VRmkyJk{(R$&VH3>QT{211<$u|qke{sdJx4Qqz z4t%8tf?~ceQLFX@h~rYlTiFsq{n-S#D=uP0d;nG4M!N1>RG}3IJxQg_Wxc}wyTkr9 z2%K0e)u1q;mZC2{pV;v*KWh>=;U{+_1*>=OgfHffI>`s;#vzM{l3pB6S{DB~6ijVl z3q+;D;Sl|@!=KV_0~U*Td0ZPmT&Pm$O0dlN=lo<@92@2s=f1xYwaHc$y)aD z@Sxw8hABqz)DC<&uR6;6C!JVsrg7WYPe;i_ooL{rTVimW7^wFn)D(+0Fh&eh*5+xs zC_b99&D7!9KU{J!d2N@z|C5Lo8*b2_WGp*G({iq;RAuU46gtLPh1zOk4} z+lQ7}tZ+>fEz1I4ifuYG520oZQ66$3Bj;oVi5ixXC1ZIvsRr6S7By_=2*MgNpDPF? zKotD(b_9d30)g(LmWoWo&|6hKAYpC;_`;22-z{KmF+zr#H+7;X#K&N(+n?KF;u^39 zc*qZM4(n>9a4E2sf^U0?pEd|rC_jI{TS=j5lk5Bj*8kCgu7s)f^!OjKh*5L&Vg0{`HhH1y-$^HzH>)nWa9CKxd#msif^QUptXPV^x* zcUX&kQntm!4~zf302S-#2Ybo&^X!KqEzX|HD?j6*j=#PC;)4$V z_k=dsB`_5-SB*)SMoGeKdN4+Jy`Qsxduai_JA!;yik7`+hvM*8>b~LagsqL^(lsuW zUn@GWqt{d9ZDL=k;1!$IpXBLnyNl91WL+A%b|SDx{wYgA`E-|)LUFCO^-5uc)9AO`Ei-pGjMBYPz~Wrj2~ zRAKd>eyENX%OS75>~++xWchJp=OX)~DVG!}*y3$3zK+u%c_#&?XqnRFrcT3$omN~2 zPOAc}L>cN5?8Ph24YYH~8RG{Y$9+bt0=&Or1ll&s)beWL;UByZH{E(S>{Y56I+2T- zhy$iivfr}7%*UcaIA1eU?I@wdzDtEc{h_tuJy19==U!w+tl1yo9(r8{yr`aGDvDE4 zY-L!_?9+3r7;a>6OCQXUkg{KYs92wr#I>bESJ~UyNRu}|?Pu7f_{p9>&+VZjW8NZ7 zXh@g3ZLn20Ul`8eEI9hHkOdbnF}q%fYbq$*CgsM;e~b|Ko#`kq>o$A7On~Uvj5m&5ALp$-qCiod)hLvYuYKVD%D-%m zI~<^GZHreeY0;g)y^_}*5uj+oa+D!X2a5`K&2yf4;zerxp59*52~Q|&(3ybiwX=QI zj&UR}`G(R?xK%rUua_C?SsnxB&SV*?Z4&VX^Bo1+uw8|?9YV6xUGFN0LvN>bYh#cn z`2r6IYbLa0XRS5VDA=Q)tio`^|J>=d2%?nU(AhLgq_tZEM^e3@hezqzQ`YR)&Oa(* zj2nlMG-r=sWoet@OQcbFt!p2J`yvHGS)}?>r76A9$wxA|YH_`nMd4od5dw~{v{BAt zmgx~+YYZ~+XIH+cwC11lz&x--3v-tXXf!pf^$V=J{t!_cqsjkdvO_K;jl_)!1Zx!gYca`zIiHxj@NE6t!>Fh7H@pK#J_$*f0b$4;a zxS&JaX!32p{0xikToSR*?GWyQG7?0|n)9O8I8lwA(;1DMvajQlY09a2j40zlzEZZO zD%eczY<>5iB)KBD8wanduh*yC)#hGhTJD2k57D_ytHj0x`%Am6eiezU$4%pHLZA#) zE`ra+8;p9uX+c6;}7#41UHq$$&v0+x@g;}o0*%AfR)WOqB@e* z#ArV`Ja7=aB~4+prq!~CpioGyI^0c86cZob^Igsr(T_7qSFlMG9thbwvhPCK&zYxe zo7hMaUvn5f-E=tQBMd6wJ%ge^^ig8Tc0gY$R6oaPT0={3^BC@#W<#F1DT7k196PJB zG)~*n?Qqk~Q}0RTAah=;sHS*>9p;{4n#W)$PV*Bcg^p`OjVxz|8b>tAa=)spWExQv zGsAs_MtuAqxr>V59^)Z;(gG<;oE&W(i(gB=l8TAyn*=-m2Ef6o*&a!N=g{?|!n z>l~m6Ird%1VqhxX!R1~E^h?&;|y{)plR1 zx%}0AR|ro7*<;0Db@EjG}o)2mxA{@4cE(Imh;*HjK>cwM`S zaM4cH5h2jIB_$LvI=@Y%NlgwqbJ>9MYj+9t)dWH78VMRuR6%)X5G(3QB;|MR0dRqx zjj9X*px=sIT04M{SB8k}q~iw==JO(Y4*O)mh8+p|4@cJ{e0Tytb&$MP+S`COQ|U}s zvj-5@E${&9O(H>-h0zsoc?CkrAy^m3uwH>TeMx~^Qq2qrnzWpdd1}!~p||}v1bpv` zp2q)Y`h!U&E(&$B=fPiKKs`iDXoS22R0+g^g>vPKfOL6Rj?(4zHIW?P!HtGiKvN=Q z(GvU?&yeiSG9U)Y!Ap9elgn=Mfk_$gR)7O#Q+xl8tMdiFuknz(<92ZFq@=gZly z2{fO;oI#_$4GfkjE+Me342~Et$df7MeqoII20cj(v>3DHA}Q_8@6$cen8il*)x0x) z1+~W^a1j_43SD!4yFNKJqIZ7>6a0CgrU>3|0 zUeJhw0v~7XpL~G-JLHT%1!jgK4bWj=SP!K87(ynHXrJT{fcsGx)2nBH+ru5?tOsbM0Shr?F?9QMy2V%R>_V@F38;n-nw0q;Q(l& z333E75Z@oJf~=9{!1&*32AVFqc0%1Wyq1&YK(^mp4luzTI#K{0A3Sq$4Q0y)Lk8hA z+cu2w71h785AIk(sV_;D!PFz*!+HSP;%Q0uenCKT%)9r=X$>^2ML;i>((Zxg1n(vI zJ_z`yUdPsz0J?mUdE;?vn%q*zDVPW>KC}jCs(G-OaiAgs z1x@ua6YhtlfFa5P;&Hhdl!|s1EIA3-Rd}CO0oTXwLGEGOU#`7TDcm9Ias;3k6F2xz zs@Q#EZvc<3PCyN=AE~KgDQy&;Dd-PDd|%GD$-JLRprA3R%ewm;LIbbZYXJ?LA@?0r zH;6>e4KWxYX@>Qf0eR{lKA8+=?Aokk zCiomUgLjMs%bg0!;wfkb?dPsEBZ4_P)5qA1D3+)h=HtzloW*`~5W}ipQRG&aSyJRiZ8JJ5_FjM&;;+f9eZiu)#>84w&%t(LxDPPG9dFv--^XtTA4xFC$4Z5ACOTJu0XgLu z>>SFF7KjOwjN(IihQZJV&<^%B;u`rGdmv?q)wNgvV-CGv5A7Df1X=_EHb;Cgdmo(~ zumSubMrd}2pn&|Mx z^!+oTTsDylNz#q&w2Aihp`Ie2Bh#mNPVi^w1b62rv`@-2yliE7Q;v|G}rjB=77nFd9*5VCB!)azCzHc z=b>B$^nh>e0Gg&M3YPv|-MmBOKdYt$h~yEp=Ib&!^5j@Trt29OAbbh3u962Xn2vJ% zscpXCAhOT~Q*xs#@BxaCd;9^}5xm83MIsH<amd3DiGm&Vnn4<}YpXct>^2txhtV zu#RPD;ARW>a{gdQK@AC_5v9xtera-V3$7q>%Z7gO3&fV>w<`Zxd4iW8C>cZg!>J_1 z$3Ol7&kxc>8CrVR1)^LJly#dJ{5&G^%8S;;p_Rvcp#w5bf8ilQwqgJQH zCfKQAB-1s4S^CM15NeXg+vt?fKTbMjRak`#`xw(j9YBb`Yl@s8<&5t<#twmY8KlLC z(B|-UFTuabSi^Obx9L!hIs`(K2K(@>+tN;Ud1i?l*WB!txxLJ;ufEDweVao~3dAGD zMt;pG1IVA9j;;^`oMP10UZm*|6QT;7dKb2N%N=AB>kFC~)3U4G+K%krMn_aI66FSW`49ulmEwg(olKiJ=x zq0u2$KO}>B5=QcjL^?(*;UA&I3Q)M%f;5(ZYgGr>aIhiFeRdgrHT&PyfKQa)8jwx^ zwC(WdKitqHP+kat$gf6FH>oH7{aTVgysj>|!9Kxl&~jtLB4(8J|BEicktzMH8i}a0 zEi`lfRI4zo@i$y~y#*Tpz%@&@L2bnGHaa(9j8-Uhn9TPN8?*r%nOc)J|JUkNyGF{v zf&UZOCx9x@8bp=N14;coP`S8Hg2VHkE<@4sDDPw(=eUml$Yau?&(NAZ{8Vpt_TO#Q zjE*wuelP&-(<0CR^r0yqsD7Bp-FI5Ey{;|;N@GUt&51x9)W+MJ8P*MJh9e-;rwtd< zHB5gdNcuM>?KK7$@N!nwAO!PGj@pifL@Z!Pr%aoxnW#5BEeN%rDAkn#J2fURw7!Nl z_yH5N9KwBYeftJN2uFg|4!`r zzb<+yx!x&yg-=!?4*bSkLpts~-=N*AoeF*Vq74 zxES-Ik-}#m7!QGsa%lRk=&x^}z~np!^g4LLt2Zovl3l@PR&cGr75L#y2Z3?tUpAjU zPXfwHBMHhU`e1GdE%0){I1YgFij~2KjmXT$rZj}~qo4?>90JL;(l3i#A?+Ipzon+A zXWf7SI6cB}Izgn&yQCP4q|xE-sXyQZ3Iy=X%Aq_*TMjUvPfaC!ngZk@ZCswa_sZsD za#ost{R@bc8)RAq0>F(3NT8eypR<&a$nY@z2h1?O?FZ5!K)G8w^hvw6ypz=8I#$gR zyf{{CzQ6Y~sQwd_{yE(T*2;n-9|NBvr0~`Bel?5tQAKF6_wqha;l>5Ynho zf`=;!z2E`Z1vUl1VD$w#6(nB)TfaUua)*+Y7b_6#)wUod&>!f}zWh3^VIZ)0{TXOX z7SyY@34+c+VLejL1&-Hxh^c$btr2>#ItkauFQ*RDksFOO%CC;(DrH(Sr6m~XB+Vq+4zL0Af66cgV&U!1*b3@e-Pi&7^ERTvfgl*0oH^H}IeW zK8j!&rsVfvQ|@HRdb3jJS*-ii5~Y#Cm@Zv%zZ`JgVIzb&idDwtntNz9X?p-M;aCa}e{9ehmBTm88lZ&!;HZ|WVDlX9oQYvABsni?AXot8n zjx~~ea3+ruYza^DzV41nBH0Fl`D`U)%wEATZ4OFlm&%y})y#4zt-^rF#`|L{l5O#7 z(?p`uSrrTSDX_T(EQ;%Ze|pgm)5k|k)F0yJz(JKBezU_Kl6!`}-F=#!fRy31xz|IH zgR zU!Lmdk-Z(B#@CN;lGN(N#ex2ar(h0>tbs0Df)u_mAiRMGdCW_$-q^lQ( zBF>Ht+0zB8TP_mPnxHE%=xZpbIZj*g2vdqfrBl*-E0uJ0@UVci@HjWY<@rFxu2y%F z1b<$_i2b3@pmOg#U<~R7nwGpNlZiL?P1iJf>=xc7<7Q{Sy#r=RYoiXL{BWR%#gb!+osM(sb|a^ZP5shr@G+}o53IQ*!FDq}ZQ;kj)_>SD0Ow`e@(^DIbu)yh!50S0> zT?Ur4Gp3!pZMZAYlo=T$83ini>X^Jz@ugvZ$$Ho;gmg@|pzD>k;~&6@zzL*bVxN#f zky#^6U+LP}1hl0!VR;X2UY#zXf#LJLNj-9hQHma(ZhHsVs;l(7@%Rm%5#^4ZSCr$K zM#+~yQ}j3W+7KR)K7`zJlWv;DTv6bVY|R&xV8R25mzGB3(S{zmSalM%KZdxiYM379 zz9K#(?pQ`|aJc%Z)m}xCu;9=abdo=x7ZI}0c$Io3G82`Mg#glc!U=0XnCLHDQ_}%I z?qJbQhDi9=G{Nc4dH3^Ch(Dp$%5u!!wQ z6s)sFklYm3MN@xo@(nxwM!O5LhPk>?vuhliAK`8(2bvu$M9vDW;RJ8tQ?D`=6>q1u z2zAvgh#F~;6YI=$D|E=v_aeE;)i(t$SIA9-+3694mz>7f;)RAi}%Kk zN3V^fyAW!dbeedTi|X_yiTAerK_^8h_a*N88|O;Q#2Eifw7}iym7yPjds-q5TN%ew zx4v@C__^3N!c+SEvP60n1@e1%1+mR0iP`DS-p^0vm*6bMYd_&u?OaS!?(KiLN39|- zp4dV>>}2#HiSsPXZp6N_FKy0paJ*U5Uz0RJqbh3_WWwyuu;qzbv8rFIR=Ow3Oe z1RR~e+^ToJ;c>YkK$Ww)z*eWFZv;t-3E83B@9e$b;F?g5VDY7!6{sUpV-GjLJ`zX> zAXg*Dsj9pZ!cJz~=(4K;WmP9N{t?=E@bVk2G!vbBLX>qx6={Ez_804xwGE2`V- zFPyLZK;B~*cz%d6Vx(Eax6yKGME1~X7+a_LmB}zIyI1j}&`UF6ougN(1NR&5W3K3Z zh8la;CB(fgjk%*P?TzUcvf?Bi^}Y6ZburVzrEDj2=W|nm7I(wJwT!h?m-s`fmV9os z>4zMfr@Y?nJi#h>3#d5V!m0v4%=ftEa}s6AF}U#ZM7i_OobX&(?xW$^(RcGt8u$rf z;RVshDOAOxt2jH;l}j7)z*A7koS#xl4YYu31v-Szzw%6=TW-0ruPP{)>{-7h!~u6h=%W{s zJOA{7;?+X7bZmw%f{luYxz&9ae}Z!`WS{2!z;3H!><0d-=&Eg-dwmoBEX}WWaTCFY zf(7%Fi~fTM1H6e!D17fuRNdhttkOjCP-V78$?lABYLNuY729=2WX>5tSaa6-aeuk;PHJe&t*Ot_{xl|a@LY-0B0E6 z;D^qtkjWJPtg0Z_)C<(kNlHLcS80_VPz_vdczG8Y+sp6v`=^a>NSa>^z1j2#*jed+ zIscN!QH8aFy2+|78FdEax11$~GBheI!9&+}_|p$TwJZXSS$GQ2aXjcllxUtMV+rqc z#>>uv-YtHC9q28$3xq2B$t+3~sk-GY3s7lB7^EgiovEoOpOw%c`eYcS z!VNFX_JWtyCQX3!-KZqhw;57H1_lM#&qR~UZ0=j;)XpB#dP{MKnCaVf5F!L+6@RWi zcOZwBm#E-Q7SbN^<@_P1{7!*U&E)COfZx3t?;xUt&>+l+`?+R}%QeJZ?<^K*0+tQj zG(zDIn9j;a$T#<^J{_l8wcg^cXEt>fl$n0ASymPKlx0O!s`k{kJsDlI-iK?&Go$S}r^H>!-Gy^g8P(;g_1M z%^GSRnwSsj;ov8v+u?Q2*L{Fz5PbR!Zo7-F_bmq(Nz?OHe+Gaqm7WfjS$FHq=!Q{4 zzMoko_Q^i=cQV%RO6Y}q3qHgsI%w#3qeFwOq92p`T*#F&mu@Hr8QLkQuHn)%$}4%H zs94IOnCZ;ajU%?fr@#TC<|5rhTk{Bb0dp=l%=3b@uQ9?IP8623t)pr8 z2o|}m#j~}rgCEE8@lel9(g&O0H98mOz%zgXq)# z2723EcO;zaa!UiuD5Qu&)NpU)px+3-0^jTWh>S`b5*K4Zjd`Ip+!jppLNMM^YU~Md zJ^i?t#p3+*r%!6_8pf8Wa#T5!<`d*;zJcteRO#&D*E@If8P+%?ChkzCoGj0-ei9R+ zQ$r%6DdU7dT2c%Fx4}81gT3=apm%Nc zVeM&L?3VHn>+NqYQ!><{R*!FlICc}=TWIKZ`b6Nq@B?VSH`?78;@kJ%xjtYl@}Sj( zVJqB{j!F6O{Z6Xn3x^*n`Du5?3^Y68v-t}*V+`C${9T*r5&x$x6D+<=QOj6^vA_+wa*Kv?$c5E-9fKB!t$F zthVpk9vC@ z*+g^)9@RYN^PxjgA(&d6A9Z+UepPq*c4=LRzPbc@JmHGCK3p{#`%eeCp_}duIemt6 zZ?B9<@I!mDXxk+Vu0$Psg_5UnLA{DLH!3wLf6(q5bS4_hXT@eMOU0q_;=NZ+75 zq33ws%TEu%99ctktr3fE7o5lE+{V6cCwMTc@qG7@m3a_Jt_3*`d4Ayr|tS2!}h}gH-&oRp3_>h&Z zqNcFs-6RPLX%&iujI7SOt1`HGD9q$@CH7P-d^R<(JU5J0wk{#5 zm#kP^)s2N{A+(#Qd7QxV0wub4N~$+F*+o<|nhx{5&%(dSjH0(D+)*9aLfrb%{MHHk z2Oq|wgWY))l;*U#*;HT?I!=lZ1c_XRUY=%5;qof7fs3a|C6vH9QN^rg{=z=f-whscQMWfzsL6a_(_`;nqWQDM}d%-x!4? z1jiM$pO~=rl_nOdB1+JndoG>%HIKNjsbaqv_Ge#(L!Fe=_73C4tyWrdUI0TKbq9c)1rMj!%uNFz5|MSVn3ta#C zq@O;%iaKL8wqo;Eq&PBbH<`W8!ALk-QrcJGh(0hSrm~2G_T?QZLdqLG9NCAFS-f3) zps)Ap*D$~B93yUbn`>bRMreR|R2e3H$-#^$Vb@Ill3#7EcDd<$k+wc|7lmdkLAev7NS@Kyxx z?Z+Q564F3U{rr2ni1$88A0mWViN1>gKiYz`JBe@00Ax`bc>y|iHpn0sbn4x`0+MYy zR6F^CXU*(4nNZH$Z>A*PKn+6Ap|Wkci2%4`rJa!3LQ%RM+VIHBo#98fS26f`Fnjx} zTzXDH6Z26%&Vz&t$wsG91-Y#EiC=*QcLLxci1c$V!NBz%B+K|hlnL6LJlxl|HQEK` zP+t08dD0ko5n80ltyA@8s#cy(qswc9TTLum;?MNluhWhLjJUkCnk`)Mg8PSYcgi%h+J^KP}6=%2l;-c)Db9(t|1p4gTT^EM6N zA{|oT1@tdvEvTavrGs?tqPAi5*}GQ^tYAJ>T7dB#DSH;-rnQS=*y4?VR&L`LaN9LR z(TTUappop7A+4_;Mq0Ra%CO=ji{?HsjhQC)m2gi_x^Y)31&X!0;}Vh=i!Qt?owNcN z8DSk3ERC)AxO^qUWL)btvRFcI-1lL`wPiAu9SS5@M~&znQ&cHFZw#=&I!cGa*6;l4 zA!(^9w29=U>{aDt)C4Xf#G>sjV<(cGSwaWALPuF=v**h=x{$ zAoH)br^2glKCAhg-U8GFkLh>h5h6c^aNU=l1XWxAhR%?%Hh)$B9Mf3ev zR};faW4C+!(Ki;d;|%a|E<|Ue3u)yPw7=dijPUbj40fSJ5&Tg|^$P@9V?o|c!^h=+ z=xVu-U?JgpZ@)dhHYQ;s_y!!y zaC^HYKJ+e^L*W~)=rnMI06@>50D!FSCMaACWUP&+dmbP5^9E3ZB6-6uAxY4;AB4Qx zWT4qnTi342)2yxARbuC}|I20tIblF-qexb^yfbEsMX@&__Y{-3I-i_kgQ)ez|$w{Ief zB60riY1o^7)JV9ktYIU@iotI^O3H+jH26gSiBMGv;l=JT7CkVv$N7mVD(zbtp!cL5 zlZo<`7@R*0@d$zR4Q0S9j1$~>;MD@xtUO;t5Uz#fa&UBd`yWbL5qxiZ=wrloX`7VY zL4bCLffeGht75wVGS7h`QP_iDw(XDMJR}HIIl}1LPlG6su$_1geh--|4&gj`27AM9 zRrNra2hj5`GC2^1>EnWbUrsjP8V}OsBmX8;yiGrWJZ z`(RCjGNCwOU5W%FX}8C@YG@zb6Z35AdBGS3Z|n8)R_!#H_5(g4D;mgMg1j!W@Qpx* zIf)#D`^5iX_u&ryU(!rJn55p#PLtS8pzLPg(=qqzKMgio0C(a4&8GuQpvS<#ILLko zx*)JRIUNIZ`}BN=^+EmZDR>~u-QStu0AZ1LP9b09`G8}FJrvwE1qJbdw7;LjZ0IY} zN)BTJCZnJvFx{nrE@b`yGN9N8Xdv#4bTUv7w`;r1<3V%FWWZYFg{CrrP0?c5THD}kP4(|l6}Q2>hn>sXaOz2nZ8 z{Rjl&a$diLW|p_mhn}fk%h~GthYp&5P(DBUFG8qMn!2g!y85GyM4Btl_B|+@_yfX& zWj3pyAc?mPkc;r4KwKyww#yn~L|zZvMa)B_J}7~SZB+IJUY)ekk(^yA@^rXU-T8ag zVMpMf6Fqjz$!(?kUrb`|s2C4YwSmh?+0gTiy|H01>1)0M>f2zma3zp^o>1mP-~dGH z!Qdn)8CKQJ+69En3c{4E_l4@E^fx)-4ww8V&;y1p?e$+TeP{5VWle`~+50HTRH-!d zJpcp4+bcK`qaag|AdnUAC+CJHP(?@&`q_Z zWtMfHfys|@-F2;sXOLA$A7Mb z^xA*rK+n{9mO`>d7!s+g9uMGj8L)*?y3ATtYiAm`OiWXoNjtJ!L(x#tIl{m-2t(5n zE(R_`_sX zFlxNZ|E5Rh$`0osm<4tk_;uicRXzp@z>gvNX=_d#8c!n4NQK`@1Mh!uo&Nn+zP+Q# z0%lZB-e8WS-6H}`urR7aZx9xzHssiX&fFJ4!r949o5{fqUxQLwJ zKPfDcD}Ymi0-RmmpKq5Bp=g`0LjlxKs2D^J{w8AwxF^=GfJxT|h$E=l)RhEawaI+1Vv`=geWr*ApnnlFK$wNY07s5@f=V!=xxG-v7Zcms#j=lWnDirNMgPgcm~X9?`d zXn@(NI|UBC71z!z15-%+`xb&!D^S{R^Lz5WIco+WVAkL|wJ zqjKEZAU>O@4HPg05FR#_IuU!2v;#;H0jPR{}U<44bXz>jD9j zQ2J=tGz2FnR>K}Zg>u57()yqeUJr?445hg z#E+((1OXjMrkLtV#a!Wk*KI81ez=G(Fh0drE`(3O*9|u)A>bo>sM=dlEY zUAqe)&uZsksx5}y0~Uj-cjhS&`$>Ou+)y2v>9gj?3oVc*3Xw@7`zESw7WRQ}0++Qz z6#@b+poxlCL(^Y*|D8`pleKevIdj@d9{T?Z2!h?mL1e)P`I;i}n6vabv0~LAbGdLB zzYXGz`*efI7oX%X1EU@g!;K4Zs7ir5L1#MlQ5~SsE?4x8TS)BZE9-4Rq$1vEi1Mvw z6uA*C6UO@CxwgH*=fVxJY;5}V0nLD{lUM?+y9UuQ4)icQ7-2B3hpp^>;;VuN^$--p z?t)z#oiGH|R+wxuu=4@%WZ5TGL@=Xv`VXn6or)_Wl{^go{i*$ka1*I8qd)b97&xfo zI`uhI5y+#U$~WW_KY{e{C~t3Tv(x$sZ<(#k1V! zfsB_K^^}w0lrd+7PkJZxU(8JMb9I|Hrq2BH(SouH4mvqAQY#y$`b6S=QQm@J}*Q_)_wkO(AkLZw4%>A0nvyg>29u zhetJDrz(+cbIkoZ2z>x+f`HwMfj-xp=2Lm_BARe9WKvVs+ZRqD`TF51qUDc^@?`g2cZd$PQ}o45H2J&RU?{mqm=uv;AturVzHQc)b(z zRzCEETM+|iOPWOHSQiN0m*dxg2_F_T(*Gy-Je}w!EdM=8_G!jl1#N$y1L>#TQ_5m3 z@{vF(3h4A$@W6sBq7J zdp>HOrWsFr{n`$Gx(JFs7(j(g`hSZq`iHdw945mUxP#j3(@7Pis18kw;tQM4Dhz;~ zmZ&*yAJE$Hf=rtw$D*CS^v;75R(fei=)~s}OLP`}>-x3Zw45S)SZ)dgGsUBIuF%m; z5;weBw=hR#u6UFA9yry(06&ifZTwbIi@75-lN*j@Q8L>TrCfEyyZCZCe6fin%^AN8)3MjbARL3soX&Z24yAdP^10yZ4Psvb(mXpLS+31GH#u-cZ0t&5ciL1LZFbAa@ zGCHmtCFB02y06e_S@Cv5yP)3JOdC+q4rGv({UojcN0ZcnK6?Z5Llm#B{g7Mo^!lX4 zC&HIT(6}f5DKNkKp71)HsgMarU3IjIo*SczDs3dIm^SY7ZGSazrm+(Q-*V%JQ-3(~ z$U|loleZ&t99U-_P0FHm0TmcZm%&4N;7A%}f80J9njjcERwk(IDq~!cB4`fJo8d+ZeRjn-O7-ytLS)lLqo1bP_@28w^_qF4KInXc)AXcDk^QWWG z3jEq<{k7jC101;TcsrP1vzM-8sO5^~RB1#h8N8aFhYwlrf;xa5UJ1^aV93JIJLAic za5QI(x3aT*dXRnP94s1`f>J*!m;jw?DHL+(c1H+?Tv2XDEB?q2GL)r6!>IJ*!n&Yi zD5|gP`I{b6VP<&xSF(6(-mF?s;6Mi2{kGxuNl7k8ESgjMBN~sysX6MI10JfGeKsb zQ)@3b3k|Iu?7!^+u#!DQ4BZi2Q)x!4)C`mZ2QLy4!wz*_$lp`|?r>)_NF9E&Y(*rt zCwcFHx5poykFH6dd|j!k2_Z+~$sAC-2%-0_Qwz5-$%PLYDY^ZEstm3f(@b#S5(NN7 z-V4ZQ%TA3AZ6$Xo2%As?Fq2k@WSazD&%bT3>R%Tvcu$Jhp(cN+=tkBjJsipng1O9Q z#+rTp5?)hix3K0NbB9-8Y=nc^{0m6obx{6mHI=6NEoYuX;4zxnlM#+x%z@pB$ zzaP*+Y=f7~z-&|N#sok>yC?aU*I@ahK0reM7}^lmzsm}sgepLg&7xp#kzvsh`Mnhi zy#_=&0qjg>bUJzmloryT)+lmuSv&Jq0?qHynXkt?Ka*18wW7>^l@UTL3<~>@>2gQ{ zm9Ulj%%92goA0uv5gsi8HUgQSIwUt>+KjWxdtoO1d9jcK2LMRiobwm+Flk1j`$=r-1a0t@vmUBrQ#gXDowSGIMIW%hF3d(tSu zOVjJJUx3|C5Pt+1nW4UH_f5cu%>I-^F2TVYbso|K_zq-SO{@M4i~cmLN}#z(UPgK+ ziF#mM%QV7}#h<71;qt{Dj;o8af#0Q1daZPvwusj*V}6x~6VD(`r>%%ghM4}x()|1~ zi<(oSp2b$-XZc-fJ>l|;<}s%ZP5w=hjOPPPfN=HWdJ8pO=h(XdW_8+*|Kv=O%j=tElIlj& z09p!onXOLVftbJuGt?F)Tb4=5o=KXM@K^YEX2%WTE$jAL`?P$P8nA%L^l?M6vPs!cXxICP)$`Mgu zCTtHGjkQ4yl>e)du@?9*;{Om>(Xk{!9eb6jIgpT#>kh9rr_N7cdakL5#4k6r2(Bd< zqIf-}Lc-xC1!wKCnzAYYfa(YTfi#ndd7oTNVUp=Fyi@X9f^jG|hWOG*RsOryO@It} zA0;y21+J`!LHb2Cqt9XncC*U5Z#kU=q?G9YZT9*t!=|YV3P}c`dEyV3$2K~GXIPAf|>-;=;y4pKl`mTo3fCVqk-7+#3ZOX_^<^9+uFaKa7sz%`Gm4cL#?nJ1Ace+ zm%MfSYil6Jv*Sa|q|VR(7XVB0s=z1ot2DGEUw^KFoLjgcx5^@5RN8{ATi&0kmO{<> zZ@tpJA3dVki@1S(Tmb)cvxGDTrDIAmctORJK!9-P)C*bo4ri0@IOaA}bH@HtO(z9? zst?tY$E~G$RSCp!0fohN1r(DwAar(~bxg|qUlzI?k5|H-x=CEL08#s;0i564tmOC^ zzZ66Ee?dw)aA@T}4geMQoqH+c3aIbQWyB+ipFpZpNDwg$fZ(aIBr{4m8J>y%RPg*8 z-CKi4~SY1i6R17G2ft69pOBMNo~5xg#edZQqMa&AKH3*euf>VpD9~cy0ks~MGmsXS0?Z{SR#k)E5(qZ*Aw}m+ zGZ-k}B?8#3bakFe!;}EF?_qTq1QyL$90zC0vvKLlZqgwnL{S1Dg9aD(|By(XU-M6- zDX7gwEJkI&^00G3A+Nk)P*Sc)+9m+1O_qQgCI%`~>8t+nF|Z6o(k{1GA;<1-%mDz} zyHn-MCa_V=0jV*2UH}lX2wb05H$QgT!iGFI#eH)iderT6l3R${Z+P`PUY8WANg>oJ z2m*tES1u!DplhcMO+;4rQwX<+mqk}IEL@YbSI5+~bHLAvCP zXZUds1fh_EX&*Gg;K*`=B&TXc;f_Nz$(Goof@q__+7l`YgQ4n4umc)_+s$NIfLA*x z$nJ-jAlN4HD9qSUEK#{T{zVst8{<> zbQlW(kn(vj*!W_F>@Q)pFsaygNWZcVc}$vF0Wd}%G`?tH-*v2koC?T58pANqr0{Ix zC= z7T5CI@e*&mTQD6B!h5In^Q$6Qoa3KKDfY_%?h>T?_i~bb;C0foEl?JRV|0TSPgxBj z<7*Toxu;2Imxoc%@?lJjX^VdFP5%iMJy(kTp-A3cv~Rllgmh3qlN1FP0;>1KVE}cN zf<+k&wEf)#5s3V-MF3kpXD@xF7e;>1X93tzED|nnW(sn{)H0j~%WH@Y7=KI=0~Feh|9E$v|B)yxRk{9acIh=VH`Ka? z9O;|#Zn}kTv4AIb!{HlY3S@jPdKx&S53ufsHAj%P#SP^`|_&r*f?Vl5?0fp!hVG%kuqU*80E>!o&gH z65nCT=mDm{@s}$-PbV`x-KYZZf8@c;O4pT_$4gi;#OMGtseu8EuF6`QOvQ z2Mu8iH|=E9$I-*dxLBBY9YJbuL_3NL3u`vR3JZ(34CbA*`|36qDvONYX>Wx!o<%-n zG!TA+g^BwnDWd}OKC%>S)o0<02g?`z-9{-`bil%uzlwKRH8W(VJfkj}Q=Lo-%2bVix3OM42M(!_^xkEz?QI2zB&{JfGu+CH;|BP15UKcFx_=Tm?@Xj@#h5c!Jc@{mdX#lJ_kQPul8Ga7-uy zFhxR=12y}Nu}I<1inK5v z!-6qoH$5YHc?leo-B$m*+XV*-`Wto6ejQLA|Xm+r7u>@)50s?+Nn<}U<`@x!pHN8G&llQNK zS%?||6$k>KWB*e7MM=oecgtYd9*tZ0l;VRsJ6laN-(Tkd%#7ml=}sEYOh^X0ZZe?x zj45efySqP*+bv=2Fp9KX-Lu1us8h)6(DijtD2q@(>EDTu-0xxrh%min&*AM+$@Z|_ z^_vxhg#)QCG`04H*u!tZ_rf%A5KcEL(zAU*rGsN|-#6FoylWst?Ip24DjovdzfHt7 zwX@yDIGssrfzH1SY?F%LTorrtcDso`5h}#A_X!6))Y{4mcjXJvX+fbwJBY`80Hcvh zwjH2ip>+^m3l34MiDXCWNI|y?^_*+^aQYUBy^O(PI-t}~1%nb}f}eU>UL^+HEg`ND z8nsaNzbB%VIDNNr6d9u07`tzMUKZHzEu97}ik74+s)xu!v@*76_{PecTcDCbHdfncOJ<9NqM6}J0Hwb4# za+U7nC!^>y4?0-ezb9UO@j#1zd+UM>&xV8BOM`b!QP%d*#_@b<(F)Oq<{ZHwIzWLd zV+2e#=ZRQ__w+}@J@62+|kfIJBfP2Dq$NHVl#58!yo3GT}bgI#2(py)?i zL%17}1EYbpe460P@11T830Le$)H*nf<#2PhowaYIXZgzHSr95Q$QKB&0oFSy4{@xH zhkXK&+qjRnr$>KdIfSrh2*yp(PAD2P+sss1ZK>+FfB$$vyE5R(p`HA4oHtTn%AmEg z21_kP(4%cf6hDUBRD?KRQ~*0nS)?tYfH?2y2%oN zX<-iLt%Cy*4w#(1$kIJ6-wy$m%vpp0-}Y-bBWdA<-8?2n2tkkv`AP^O9t&$qWeOc? z>d;MpVUMn{%HNr}BJ+b5%M<`A}CcTo<1yR897t zi{hL~Au3aB2xkx$qJbk{Tesn~=o~N81*S^ilv2JOdotKRz*5OG)B9!zNXL8@g?g04 zqP9*gn~$nrkQ%k0V8iOmhUEv_pP^NAipc2w42-`wVJpBCyhX7z@s6nVW#sug*Vmsg zpGnpXjvMbkZH)*YCjv0(X~2{>n2XleSbs%_hjo-G_O@IKg`;d7D0#(i;y-n>JeWS{Fb8i zZV=(}!$W%ZL!KbGl*^77^TWQJf-=y@i~yg9ID7(v__ejQE{A{KfAoIEghDnG{oOM9 zDPo5VU@sL1aeuO*8cRT8jM{wpWa-uuRI4(OWO_vCdI@9q2r@4j=;;Y1Q)I7MD2zTQ zl8Td-mgbU#BOTd?o?a;{E5}F%1_p*!xSyEEqoAN@^$rawWT_OY)>hd6?o~?lHNcqJ z-HYuY{1nHctx;#x6FnjpjvvdSRWqFZ5EZ{>>jF&ZVh6wSmLG#&hdzbNN@HWS=gSI} zQhE9SaCTy}8p~7M5F^l5z6700ypC<_LxGKW9pJ6Jn*iG7DvlQ~%8xl z#?2#DR6+hIlDsBLJ7|ig;U4=BWv6 zq>Q)+{U1GMfG7p?oPGd2&#eF{Q`w;Sap<+rGl9a5c4jIqHb6=Jjt``_jiflwMAsG1Fq6BfQlTv=D||JNX0I;fA-?i8@(V}D2N+Xk5=mH&RVFfs1v%ymiB7!3F$|``pLFmepOl2oU zE-z5!WX|8X03)lq9RMNC#{LDl0kwyOhQ%a?CtxecVdh4;MXy@$_HY>>Oa}d8<1rA3 zI9!d4w>!Bnhtg{O?xw7gsCDM7p_4tQ%!Jv8G);cKD> zGJaQ$3_Ykaic$t|PnI6WLX&_4HiZSylX24`4s#3oUt!^3YC#De0k8ZpSD)}Z7vV@j zU9hYClaD6Qy~QMiW_xryQ}9e88P@jz#O>AoqWl}=zjlpBeOS5xty#(|MLamY8iok+ zayQ7h*{*6|#&V*HBd>BLBnluBEei{i^De@)x!yxOJ7Mo8{=y@utsN5~8Sv)eTryzr z*j>z>!T7=Wyu3cjKZ}t&EPaZ@VuI<6Th?QA9H&V>CMsQ0h&Qdu-40`9{&mJFB{v^b zkw80&AttODoW{HL816bsO3QOgY6Ho!*GXOTwQ3Ka)6#a=y1CgJA1=B}sRQ0v)Ow8K z&oTj~dvDa#{-|Pn6>uwKRx53(^t3Z&Fbe`mbl`wxLvuFDHNOohfHyFwwP$ zJc#M5ad-(!N-apu+Yohy0N+Eg`^V6bi-F(Y zCexpkXNyCdzd{gT7)R?vF7WT3AbBZ`bvOxuvgu1yb^h;V@h3!^wVGv`BPJ3BkADG)mHjZQ@8X>Wh6d`&N1%in%U zZH|Hcubmt-Bk}<7Tk7$!;9L~1_?GVpe}juJwR!^rDHK@5kEMUG-02LE2kQGF!TPbB zMRLGYP_`a4&XzM3X3hZqs9Rxti?7OBu>^)$4hT)ik#eMzq6PNOAOgq43`!rK0_04s zfEHwmgF!DbIqw<

*}87bC=c0FIx!f(I~i_apN zt`Dc>@}UK_s>zC02Wki5+!TW*{)DVr`8})SEC!^)1i+5zAMfUcY*TPJ8bE-iJ!sUX zK~~yQ0JCcT>s8Rb$A49;0DYwf@Nm-RzbImYQhs(ol4Ln(v&SD{kXizt8~YP|C*@{u zuWISpuCY!A+oDr$vk~f9K(YcvHF<{%=r$n3^T6I=xi|(Y8wjR`M6o{Ca<;&1zCUCkhj28H%b2$9$I9KKH;zO(O zeTflegABdhS&v@${OUM7o#-yevo7$Erxuk9@@Qnp;$t<}~{Oh2o7NKmcT9TFjF$9A}yEw_r1c}djLUxrvbAQ{9 z`?+Zqx>d}B+J-jXQt7`>_<3?Au7cpFRb}FjH>cmO#LceO8VhG39h+KqcV70hw<0RW z$)U#K&KSZi?5BzczQ4K6;wG zaNYF~Ie2%zeipczt!r&C;QW;enQ3`XkQ~x}yjJn^;QMNY>D-L-c3rdA=?!Dmd#lfV z{CNu-6IEgrY4rContIN^J?hK`!16Sh?H`7tu*YI0bEz^#Pf>&SB_@|ERBma(mh}}} zNkL2}?k1YGhAU`hWH92}bAvAg5gMFBrR5kF$(YGXvA&omr>n8fE%MA?^Za;@4^0?* zQ_?;TNvrSe+*(s`asO5rFDh=U+HvOx35;PpbY#{!Wu>>n^`JxfZ$F-%@ez zEuAQ~5+pM{dtk#_-a5-&zEjUtmMhqAoJ&%zFpa&X;6+$S%QzZzY_T?YfSl2hW754o zRiWn9z5TN-R1PMG(O5nfbZYZnvwFyO*5mQabPCuEG6a(6BdiCbOO_X+N7KDu6P4|R zPGnm7Y)!nsaHUYGXXRmPVh#j{QJ354V@GEZLc?r`f60ev!-vf>^pvvkmnJ zlcv;)ot9%v^yNbgDfivxyH1cV??gG&4rhR-$tEz|U1^3NHfC*sj=P z+r&?@lsnGmtDsli3tx#phuv2zt}D_pKC0oapyZvO6^x5MzOXDuqxLReat=B>SaZHO z{2hw-0K8Kn%qbZel+`hu&7;Y^Wj*EW4wT-bH@%9j3${&v0ibFv*t+L)UhUKB_s}}B z?K(Kt&)&ho>TtCu)H7~Bl25SLm`Z(`IDCrJ{n*q4oNx1jLPJt<7_GH&X)?c8S1pI- zk2Di#|*3eBQ!z_O$g{=L)ODMAO&%uA5iO_2X5qtGnW2+eSjX(YEKu zP{O3GF0Vyqjg!lc^bf)+xs2s68jOZ*tgVXNTh7il4r+;xJFBe4R}zU zmeg&hd2GE;*-5`O>G1B5vD_#i)6!}Fep)F=;!+(O$jicZUN>@LLy4JuX;Z1|Y4K9l z9iLgl2y%K*T6jkzBx~!cSpz7M{ET^ElHXTWUOtbw0obOtei(6YfV?q_G3vLVde{MY zFI-)T+MR zP|qlQssZJGtrXc8j<~Qak_w!6ul}~<&q=7CBfL<|+`FrxLZ)7uG7}`<^9@o7cSYv( zx%J$dE;ljIVY6EnwHVEbK>gD@=X+DL3_s>LN;m5E5T=tXGeoX!X?PEq8KGVusg9EJ zJavlMCzD>7L?&B2e)7EI`y8iX7aGAJ`JB97@mPUMp(Gsno6+j9s&mJ#x;(DCOD_Yh zH94i;Z~MhTW-X1Rcgr0qDehc$c9Dmulg~PGD!2NZuDvVErG>}T8ZtV5rB(H!&ocLQ zN*#*esEQMi&i_K6uFS@ZkEwH@d%(W3VsmIEcB{d^_rn<+RyZEFU+_-FEQN|Kkap)r z-ZRlTc2gAXt7E}?zPu~7egF?`h%#B7<%i!=&gj*a8x}DqJ~2LH%NQJBWwL7{m=I`w zTJg*4DEId>c{P43S0bX3l8Z-HnXJdLRWFKn%AW5eM%nj;il>72tgK|MxN2z-jJe7$ z4~+MS7-s9|Ql8GbZ2y|(GM&wd)2$fPuBN+ox*&IEww=E+b#Jo5*^+VXr~8|&;4pt8 zE{pP7ca2ru@M>>rE&?M?!GVBa*6&$o)oXzXT(8v1(jW6=`5ie??~BSR;Xyl3EUw1f z0+HeQ`r`RsTrU=C`K!!T7f`|{HR1=MCrKGL%^mHz$;;qDcy2)(^`{;T$_SVIl`qKH_==={p;WgmpgP(^WJwcFW@29NwYn-BX9{yDr~DEK5H&K+1R+C8u&4tXpD#AE*1?`u0R&{uSxN^)x3* z&8tcJ6z!*Qm#X=*iRZK$4~2;>iTUvrnZxprRQU^OdXI=JCE|W}b*+p@c-tda*EqHT zGTy}0a(3>H3YpPALL`D|2V-3F* z)7I_P`P^*n&t=a9FD)jS?#b3($#|#wjMI5wG<^GT%1g}sG*e5n*8Mxq#Oc<; zoCZfIel}laZ-qmY=~mIgt)uCzYcV*T*^;#_HxtHR+0rSk)?jS1sdqLk`}e3Y7Cr+#rs zy25T+rr#__0C-k8eL#h9uq~qY8L0Jcip#A$9hh`i*DI zJ=RUPw_@R~@}IYr2KTrviC&uL5b;6KXw^k{_;-bM(=D&v&qeP~Ms@y9qlOo~E|eu$ zO?FiYF@4uv%!(Y}Ocobc!|rN+ZC}3`nzo}m#L80tS8%X0o~9@yzKnE^3800dVAPTG zefDvPkLJ*|F;LI#=~jukIcUk|8i}T$%DN}evFlUnyuw^2IjD+AkV0%e_QOfDTw!^q z5@N9!NM^u-xj$0h=Z4vq_ZN`sqSmv5ke~d1qeB?L_zTnE0&z@^2<$pAMefD3zo6r; z0Cw>4<#VSK)BU;$wA7i;yL0*C_WHjG{OV;;7j%FFKXs_vj87oS_A@NaLm8U=Kqv)c z5{$&O))7`*t&MU>Y5B=Y8sF{j^qAeFL38?iqfm=D*NEH`a|n5fp?CSc!z)jU4)mcy zoFgq=ER`RBO~t)E6%y--;IWsEW?wNQ8v?OrYhaJ)waN!;8{7!VT2dm*;D6d8*ktR< zLg_iJHM}$CKMJu|M`Qi^>r_OhRb9L@5iX}@|9h<@{DlAV=R<-zbYGQ~jFxERLRQr* znMbW1C4BHqZ8s5#ddj%l)2&l^Q)2i#=UP{_RMuvNUnNN`fMc_)^;QZH4yu6(2ThX% zC{kYDA@InTYIv}835|N^?|8VlHRr-A`pah}GyP(*I8v96gr*NXNJ$sMf{-y0D%5v6 z+EJVYN$=r2)fy2<@V(-BxZVC78I^MJJ0(IZ^1VUCurjQ2*rUOYussMVD;(EX7oDRf zlN4|*T0pO6RO7ZmWY=7`qC9zf2B3|^QZcUgb`y4W{Msp-826I>2TYXg*j!^kW=95RQ&*1Vn|p+K)c1zx#sn zJQs9?au3dHA0y6Q05Ws04gei~I8(gG%>98X)}h`OYuT}ANh`W^iofP^RqectgU^GQ z7IbsTS?wkhrY%czN-?-dDgJ3IJy33+c&gR}hu=<;kv6Oph2+FthBi`6>p+A6p(|87 zX8T;32Q}%S*7?$R%aH14m)>7b%@2n*K34h6x+*j|Sfy7V{b#5*QBwPwNOy?TR;El! ztvWO)CHXLqlk}E!e$<@}7+2NRG+dDz4!MFhXqXf|X8z?Ht~b>S7_{9bNIq2hrhO!a z1~+QzYflqMEV}bv_suLtNqk6pZc|CptWY)ks5$CtB^1Nnatbt*mC-}%M~(T6DIU*; z2j2!&N)zMAw)EQHei0CSM-Vr)0=ntl#;2Mk@`+V-c^#?5)&2NP^kq&KGDQxBm(Bxu z6uS*ck3>epS({7&K_coX2xfyy(UFrQ9wOHw?SSgQUGXb4!@-%AsBZqp`NSHnsq^kb zs=Kos%{+XHTGYz-e?mzV4RCyhT1?pBQR8@04<3nnXjsatmgrbPu+O~oIF&@)%NJN% zC%~Mv0>lZls7s@Ozxlqr{-+yeRH>>oQui43H>o?+F~mr(ww=LF@J+RXOJpeeCuXRO zcxe7{3bCe>WNgoomafuru4{9$a1VQ-$55AVlJ_5U#T=!^Uz;!Eb#3}eR$j}jh%ypD z>g_1FA!3^7*Bk&>&wGJGa?%U>%Q;;&KkrZ=OO#b{TG5lXMjbq*6oN|DuLN`QXD_dj zTO0KJ>Q=?qp9pz3Kq7=)ez-ASi4^QWMk$l&-(aJNB{2knX+DmAb)5x){_;Fah_tVe zDLPnvJ81s-SOIxdi+q5Mv@or1l7crx?~uk1^nUYU@ralNA|Y}S|K;aO59fMg8CNt# zU~sojcbEfoL!1Ht>3-M&lVi$%uIiLDL&=5roxbv_i=~`5JZy#c{l9P_yafgYomLH^8`a1N!mn1^ zvs|`rQ9h+}P1j9K=X(<*PiM^9XVo$v8Q*A^I!4T2tv%P&ZWv7BD6%mj9;30-`Q*~X z5PtI9vgPt(slqMrbhw`PfVq5OXJO{j$%c4A8ht(D%3Y<_Svkn)L_8vPeNkH@#)5z4 z_|H|}PS%ZvdSA|ygE?mZJdZ5CMID8QXG^P7UNvvGZ|C6>Upih6v%TXBUUBEggb_X;0F&I$#(JC0_48Hy(&xw6wwakMId-d$YY*71Wk`=VG!7 z2$nV#f~%Y^1mVqZ3mi~ZQ(v*|wNS=DKN;$G ziQpjQEqp=#s@{swa+Eo)`|Ud=%#~CwL-rHO!gPD3zRWnk?OBW!*K)GTPQ1@g7>G)p zRsv1{{!%T3KTU>iZWuvi+IuAfr*nzG;k$7HH6h$M`VLZ=29ELP@5ful90!s(Cu;26 z9n7IuIz%@bU=|j`>AvG7x;|fOf4XtKm#=W`)JT4wJM282BegB;_1X^c=N}&%mh#;0 z^_i(2#_u&*W2&nq6Y zfG!VNdDZS3Ktuo-OXPCW`NnCZwczgjF_mm2QlY6QYJ9eXgv@#}Yx-d**PHH!XPpz| z<*7I8+0U&1Zl{HP#f(c-bSA+J+;#U#oR!#|tMg^#maX|kuJ+rLtGp$0io~wKa@$l$ zRsNw^HkX(8q1|JD(Bf)D%+$_&YZk6yq4fU zLLTduUEQ}$gL=N3hK3H31$t*@Uk|-_tE9T4lx)>)9oJ{S9=SAkY>Iq!H<+{6#3su9 z&L1;=up1M5x<r34npqcpKnfl1DyeiyW9Gtmoz<|x9`~=m$=UG%a3Z} zYBU!|{6;1;1_UP?BU5WE4DvMw2Q5*qig=(Fw?A>Vct=CMuNPS(nr!RVUeKv2iy$)D zPZo~7gI;&`6G;oHv5@#6&erho1|7SYTXL!(zks}$T-^1Y_>5Qz3~i(3aI-D5PwRq* zj9bWNer3f}#UxRm!9J>Ax2$WPYPzN^@(%UPgOm<0f7{NNkJRK& zkk$4Am%bt`^w^zERnofHZzQ}d-wsu~V0>FK>i}AKcyr>~gwVt;qx>IJuWAOIMij%} zO)uTDQv@nvZ*~$+oqrTFZ^sU)@*unebv?(wm+w6oD%61@fBS&+zT*LDy4#d^400D0 zpQoOT%SWUGgWkFE(4$sV)flhUbf)^9*k|$~ZN}(|8D*~Q?^lN>koAMh&VCgqUPZ!Q z)vC$5!IZuGb*2xP9q^T5Fc=0gnPcp0+|0C&5)^Eel3;sGIX}aLhh~qCsj}O^>?*3^H~69<&%F66ge>i?*GMXqC72?6Ua5P)`PL= z4$wP_M4H%c`=$mP+2Hy8SPaaQk#Mxd_^l?S0)nhd`RcZ7(2?1@+66_Klz3WP2@RBf z5@3i#3lXpvu~Bnm7Z5ajfWuYgJ|bnP1aBWYkt>TL>k&OLwzn$y;&=udmOzvg`t@s> z;R%f6uj{sni=`8yhOVqY?v&HvGUc$&Oho$XY!kV&oYY*}0}t+U`Xg1^-4$o`o;~%h z4`u#_EnH5QM}_`Di6=#!;=gP_V=?*ppMzQP^#MYACL#utEJ}DG%?EOMeeHl}iej(! zms*8-Bl`A^MDUAk^XV*H62&k?a72`BJ8SjW_LNWDOT{U7hr5(W{O+sW?SlsZ4V4k|!fXPq$XY**YqamOZ-46#}s8vm}gZ|L(VRvaAv0LiFXy2*COQW}3-4g03f#2j{ zZxyZ0^)1M&>@ZM8`e`m7ZPQLfM3vy`)P9UzYcPIW;QmsFPL z6`#EawY9c4A9f|&X8YfpYmy!G;#H-|80w07LNou;@18&5@9wb6Jnkp@GW1M8xF*Vw z-wE`d(@jWVEcl<%-^bZC8&e=XLq=HYmeY=fWu$&|X70Q1-h6xEzQyO)j*~vOuocsD z^P}=1(?K31Ud3Ly`UO}O2yNdq>&wVjK$*jpr33ym3&+o2c=gO5@0}!=qgML)-S6w| z4co)ItAm~tslnTlxU9^dJTYdhvI-mjQ(!0`5IgjGZ>^)cclrAK)xZayS4G*s3wudF zM$J(5(Rga=f$147X%-yuuvqbXlF8I`aHSK*){;ZJG19?=s9+Z+WltkG_r@t084TD{ zxEdlnY|nPyKjbvn*7Fs)_+^jW7yCpg23fwfu6<|5x==hWxp`1tsI;=YS4xF7)T%ft$S+rSk#`o z8D;;gwzrIna%*0A@!_x<~S?zhi=;fuYu%*9;STGu*{^EkfeA$4AfGRJsJMU5-lyz-we z$>b`TGUYNI6QQP$Jf6-LrlaN>r7bsVOFQE~n^?834M@f(Jc+3(@AN6@C;>0-V$l6G z>fqa6kq$CO%yqg;fo7^<3^-J5{o@nKDnsj}JJB&dTp}7@)`tyv!w(z0>sOymHKsv3 zDfzCeZ=H2xf42|Vrw%{+D)z0w-#{^;^~>wmbP1QF$p5qT!asqPiX<-ZMPf>7%P-~X z<{CRtJW%n*2*}U9mtuGtk$;}USGko>vt;JaRmW-PODy;ga`J!N31;vQ8d!VW z=uo``RF-HIqmE?C0OHeN+ngS#KZc?I@mhfYEq(OgQ$P57xltE+q!<_(M}g4cjF07} z0&~6C1jtv!r=@W!r`t$4Pb$cR`E@Hm*^68}`=GNw&=+A+qQ+&Nz!x7VJCg+hsXG9p zFo`ctI=Z?%RQ7UNi9CV*ss>aEG0kpx!32Mi;I-wB0Zz%L3}6Z90*$8C^a!!1j5?s! zC<5G(ZoI3JFl>D9p9&zt(f0zT><8hGV^*Oer;2~&E~G{wO5)PSbV2#KGL7XFAcN~6oHRP zSC2>}3;Nme*xi~WI1B#qh7p*nxC*|Fau85yiHp%60GO_VhqVD~iNN_a0`ek1+pVfY zs8}q4!6Kd=)O{WRTU=p*UpqsBPRWN1&ocHTpyk2j)H7>0n8!mjps)!9@Q z5JaugqY558X{N*9ib@35GWeR~X#X+eP@Z9T=2Dj2g1TxyBW@^9HTQS=PVj$K78qcQ zYvT9z^P!eKlX1-jg8=WM{X1qPC3V<8NE5NXxxF`kdGI zm-1La@T=D2t*yotPc!oZ0j<5rbJDvtUi_gGfKR1vFAa3C&siwQFhz*A%Y)&i1>V0G zbj{NzAPBTU$ty`{O69X(mcWYf;gIk*Q3C#{k$U&v^+1@;yQn03a+nmS*)xiOkXkZ& zBz+DbBCRoluXR30Kd*PjtQ@ciaO)2P_A?6}l(v)0;;kSCchH&$euRJNxH`yppEdxm*SAOz*YsCqPV%l70X{GATT9Q81cnoM*Z;Y%zO3- z+wlO!w;`XsH-u~E7o;*_MmIFG4|sa!XBAR`p|R^8WCj8ab@2ts|BjUHzRsj8rU4gl zbE0O|!(UDg>W56cIv`=rZRirGSu%)y~e&Jp*r-0X4`nxN_B(stCJ@ z%F?`!&lpV>gaYB=n=r$FmLNDT1%Klxegf`;2LQ*&*}n%N{Ld-x=Qbshk(;k30bhpo zicKVaD6lKdZ4nTnCGjRpTZuo*4^KzKA;x;Px1NC@Tcz6_ zxnCJi(`jrr5YiJM*1hvd`B-o!A?k9l{R9BP)TngwHSrk?eGD)d{HW0K3xv%|Kn%0; z`cHa3Z5N4TvFZ)g#WhVXb4ya8!fNrIWVCH=qtx=G;tb ziw%(t`-9zR+zksX zc}Es_IHv}SZSU~!U47Em#HX9`vn=T@w-i1a3KX=}xzDbZqBAody9^507P%eNClYsT?0biqiDBDvsHJeP2 zktzHHAS%`^b!jh{oNejYU$2Jsd|^ubzxh=8&~MNEFTXJ~`UtG`tz9a--+1Z4bC-8f zC+nvt;Q~R!vJsMc$mJT&mVN!lQm|d*QcQ0)+HBqOMn|{aF_TbD<8~49wfhxR&pFG zbI)t)WsD94R<-eVPlUEc{Ut`)@GURa(7Y)Ce_AVRo4WAktoMelByC7V;bo}w9XQY5 zqLmawkwb0`bY~hwb)uca-V*(k3U@6hS0tB6ntmzlnnK7K}h?yZ`=R zrWR8|x9x@kO#PeMUoT#D>T?OlfDO=~xZ$Xa-$#eyF9X+|*pO0mJw@9mW2cH{#bKX@ zR!Sgzqa{!q*Y7S3}3)VR&tvO*YQ1xpp-sIPO!l2r(jXxlO)zi=A zB{R3Gv~eTwKJq5&y&n6rjdJv*xiv67X_qhPf>_h z&JWszflr|vJ_28c#);(f&+FjbnDjhCM1gnEe88X?v*-XxZzf?fF$d_Xv{muI{Rn!5 zoje%jhB2lh*Wx1PYf1&igT^owv<0~t4~KRXg2gMbSBLiaO`3TF1Y_z4#MF8IL&>R z`ZNObIz37x5p|6L`wSZbEX-9V><-Q#nYaTY$rBY<@ZhS=d%@6a)c@3n5RU3Hz%WQ~ zkOXhiSHqXh3~$+(8Z715^M6!B9Z7u*W!aaMO$=QLT6 ziEobc%NNRz!5O6aZH2kLvWjydz7!37kCk8i4uNuHLq4Xxs)G1xJ_IHgh>-O|uh&4obF=S#=inm|QI56ix{TtDTCtz>yLPBP^sJwy499Wbv zs()xT*=CeJ0QdLW@Demx;CGK2jZSP7Gjf|MP~JTlpqqLN@{hqzhdY%}(n*o&HM&GH z0?2Lo;LY3RBjP1O-W!Ikm!pub#-Ht)_3>ky$4jrHxxdP&auIPbYBP{f<__c;%3BCR zBtH4ikKSY6iRoGt|HTojbGa2sz(NZ`f)kp%MnmWaPi~Sed*TLcB{=#oe zq%nn<;7VfwiQz?=85DgA31OiK;^~Hqk!jC6kJV9q`=p#YZlVOvv}${5np!-Izrr6J zU!#2?3yFY&S;tO>D2GCmik{yo!=}$Dq;-3L5ozDCBum4++I+9WzOp!M&Wda$`=DsT zsRSE$8dW)L*w6UZ%W0B?5t;qY`4EmjWAb11w&1_%lKN|4PkOsW2?b1cu{5LDRLcfGfKe@IAuy>a*Uf?q5rRa0P2#gCSV6dRV`c} zK5oBRW9_99ete?Bn!Y8=L`&hx$Q4aNv!Be(c!BZWB+?F?)!c>eX}ll^z6(!Py&z%X zgqtB{vL)3b8%081$HiHLvrlRVKdn|cnN`aN?#>k1@4$4s($y@93U%Xn?p(Dhl@NM2 zv?`A51F_aC`~jU@LLvJK2jo^tknw#D%SSK{80kH-!9|F90 z52c_P1QsMrqD>;#hnX*%KgT0SbR8^%%fRIB_SHO5^8)p|E<{i1Q&^Wp3whHlFwbY- z(ti%C<&TE#&=I}%>}Y&T9>KE_xQw_7n}t2%z6d+-dPo04*CNLp8?g#;5dik~40K4X zYM3`HxlPx-(B0s9_Xe77^LH~NowP29KAusfH)I_b>$S{xdbjn<^pX>N{}M+=(EKJe zdipNA0gJ+8ZU=9#>*|;4qDUh`9`88rb)=M&*u;We;x!6&*Um)g!1x4A^s@nCY+mL1M z>rE;vjp1mk1o4M!B300I0$9qP_N6c9u(42b+8r6&KR(gtH;F2J}(ctYjOnqRJj(H#|weNLq;P{3HzTTTnAXNzq_2S5eI{MAl z$G@N9K=E#Q*Wshj9YEdio3>!Vav?a#J>P~u=RJ>%``k!>euWKHghkcGttB8sjnChil%~N=u0~ekq%15bv&pci(3N&14te?KGI_VVy>Vb} zI{b5kiHveXzVNH7MfK15Ze9&?^B9Vx}w&-fzW2;8eg8XB4S^oaLtE7OMCARXcdL zzHZUO`k zw48Rc_wwA-k4>v@$ZMsBFM9o!aPsrcw~qt(gosanzZrit?YY{$c}y8` z%o2f$KqDg7G1X=vQTb?P0d;GHiH9SA9-#G=dN<#ouVjt6KKoT$BxngM_#*a zQ;eWAofS3D+dRBVuY!k)^{-Z$k^XOEnt)>M2S+h}`qP}!Ulu`-{+Y56 zg9QkhN%bRKl-bm8Xrj%^*5olhQj|`1 zJ|@lZE~j|7R_(RNkh#OKhe6pt+r1CdEIsNC%cu!~#@fuNiAGbUAvrIZu z(0td*GFwTr7tEv6BjMA#M1&)ms?#Gu5%ZBlB#$<*-(uQ;JY%w!hAT&@Fk4EtN*pCh zCgnij+r;I05$5t(mswFBXVM`0ylDwImjG-HYG0q+f2eeYWs~uG9Bo^}v4+VkHLq|I zM#q>o8tG1Yt1RscmlSW+dANV3(% zqD#(o)4enp#LK`4q{90D;cp-^Q#ib{wfAG4Wsiz)XY*d-k7iZyBj!hD4L<(P8aJy` z_1kp5kHm(*uK27^-2UU+>xZjB((2lBjf{^VLU0x77Ye8>mle{QYZpk~NL1zpQ3|XunI_-Vq<2%lerp-@~gT951Af4^cG#7B#T^7m@L55{6 z2Z=oerRI{9aqKXv2kCtoOIcm_;=PKJY76K>kB=2+Cu%W}z)^b|Xf0NH6IwVqJe1d3 zv4kAnO2(Tn0zcM##;p{`^clmqgW{mtYSMJ1O-;{OVN$V&((3t3 zjmehC9K390$AriTPB>z^mUjP+;y?|CDhsG>I=^LrDvKPp9RCTm+n8!O7JkftdXM_) z3RETMyEF~DlAc8ZP)(OxRxE?>BMf8Z+0^F`UCl*;_Dim;gfq>*jgNeEd`GjN`B3VB zU>=;~(&&TwDtc+ZO5DVW)}NpzwucuxK8g49pCttuRGglbGFcv64z&2+$_cUQa>Eop zE#?`C*T@}cE$J7h2N|Q2h#~6rIF)ygh~Sb?&Bt1?iS)b|R50Sx%|ZusD>3V~lFP7h zf<{ZDmwMo8G+Kr^gCdM2VZ!4>0u4iql-Qb`(T>uj_5A!s&$HS~Cv*ksWdP3-RWj>_%vWDrK-o({s(!IU6t<2tJWBIyENTBUH@m%5~ z$83r}N&5OOy~$FJZ|SRY`vSm&3GP|hBXpZjr$p;8$l9}a-|7O@=WDM`9ZQX`0_e|j zetRyS)vZHw-PX#j-U7l1kIQ(`t_+_a-s2c&a!|uXn^O$&qlT-&QQ!$MGOvkp72bP} zxn9j>k^{AQUxa-D^^6xWvIOP6j?hIojcX7_VapOaOWUzQpM4RNv1uR;#|W{VPQMNl z{qdbCUXOZ*q*L_Ft+wTmmGH-4B5%F=yPS68ZMB1=81MW@cekl^Jb<~Mt; z3=CfL?OZdPcrMWcz?q+LRYia6wsHTchTB-PK#N%S=l|D|j-)F5r-j0GWSc_cHnpko~hvK~DRYhOo#-fV&xp$ovktMTV?b zAK&fs`N_jQ1h}5S5QV;|9u+pCcVA_LqJ{D-e!Qv*l%oO4|0Q-Ur4RoIzGy~eRX_~4 zystsbPD4Z&e*>S2oB)_B!R}l9Qu^wVxi`nEea$6dI>4KiE#dU+8bcFb=QZA|XDSZZ zOek^hExyHSf8Cs}Y`H=+$zAVtbB6g2L>WY}GmC9TL18po(N+O8ko^0fbi;ZsS-7lg zm2z%m;}L2rFV)V6Eyj}NWwpKG9|Ow&tSqRykxf}wr#Wt5{vpy^R27p~YF9@>p3 zLc|3z>%%uGU=-MJV#p08oF|k~tHDz6_9Y5ZI@=FC?T~OYj!}(6nHv!J?|=JP?gSzs zkUvqGbd&_%lrY9w*bH_COc~)4!CTP3lyZzOae*{+nA{sao!e+-zaoEt4wM;^N2Ohc z+!6T+*oq`Dnuv@?j9MJ1PoM)ND=cup*tDx=p!09-4Ek2=1X_G1L)F$V@1T`3KqAtK zs|%=Boz1!+g`%tQiK?t^eY7;s&6+Yk6j1`Ud0Ci1T(a>6$bH;T$9#=@8YIH7rUjM1 z#6|GjI;fa4pyKGooFIBlx;}u_ML($OSH#Y6Uy!BAm*Rm8Txw9y0xlS*f!CeWhH1XA z=igv*&~zs^0FJ(Yzsg}`z3D25sAKNy$~ITDZMJz-#;qW)I*yj4EKlN-!azcqCbjm8 z5ERL608jv>-Vk&729>s&Y~DORn4Q>v_tG*mGsOx_s|EoGK}1+s7;owYx~>hzvYBER z8vE*%1!x$!Sggnb8rw1`*$Oci;cyd1zFGD zD_@KL@eqPvUXS%}@5k^k zw92FO^k`98%5QJOt4mz#_+Wb#&tSuMg$6SigRyUeWlspZd4=+1Xz&UVS{xfm|NQTkOlS zv|3P)U=)2?s0{R#D$q%+uMnC!-(Tr&Uj^DRtKh9t4rnBUAk8Ch-`a2_vrL9b`HuxiaNDJdz;O;S=ah2$$Bf;bxkqxiQ(iN*>(eDehm z0@$75EHnxm&+*;a!D?@}#f#gZ-6Ci4+~e+0U#8Y7fT+8*)_|csQXt0-E-}H{+M0Js zSVW`*6llZv&1!M7IUwH%fDZY5!F$ov`0&rKxkB32`o1NQh%N_R?s@=^q#Of-9P)vn z5Nypgg=3=er*^Z-38;3%)t^b z&`FgW0AxY2p>i-hqrk)$T$_BdtXmdnoS3mjMv)m70)T1zcEUO$qF1~n@r&e6+W`aH zr}1$Ud|_bU2%=4nU{URWSt6rA;$LowKiOWNnkroe&za{?g4Fj~Aix+M?yPp0S{>By z#)}$1*j<}2HLe)^;Tq$W3J9~X!hln&;yobQ;(76D0Y8NMV}JZ7+$S-oPD#RCacZ2 zx7Gmi`03DiZqg3uvG`_l#h|O`&p+cOAJ*45>0ls6#qZwU- z@AgRf?yOMH&TnuWYhpCE+64FcY1e#loT%Klz6d@WEE95_9#xw{1V)9=BQiE$SE0iI!0k>P`4=5xFoJ zyZqJ^BnZ0f!j8^#06&O1ekG^w1=SV1%rto2vgVR|{ZrVV^vieKgLk=XqronkfhS-T zVWpR4iZOf$cNwA$3B-3>QtKN1!>uO(xo&Yyc};Cid`X4N(g9OxB?9O-BK#0;Mv3@1 zLhLraozE>6$OAC}gtM4+(W-LV-}ts%86L&B16oKGzT-iVud-*o`;F=Pwe^S{u#R{&{T%wiRF?M0OW7Nd=yYLgQy4n}Oj!+Dy{)K7u-y>nhXpb^oz`&@w|!SKQlnW!GJBV5~mhNLKqqVEJ5E&lkJf=UqIvYAEF)J zsRWT$*i)IPsblb!8M%3^JKo0O50Q?P=cbN-;K$uStZ|`^j$ERftk1>|w2@dnsiA)N z;?$}hbME#7TIxy*86l1RyWH!XKRJ<%NNOw9*~!^}X{8@>uW(WbnN*N5T;rILgE7Lk zh?^^(ZFn~L*AH)cHsP*>K+-tcADKY+p);74S-IEi!Pi<|9TEv=Deeq69v46#-d_8p z3VFr$Psi$lIE0X94o-7`8}S~x9=$7OJV0(qz)jEt-sBj+dTRaG@8oq|#4W^#rKls{SDCR~Qo;oOyxK3O>aC ziUv;X`cfiY6s`2DUB%&}_+;k!qhoyraIR~2l_vZPcoDV#9Xd#7Yz!(Q`#?Yo z+cW=m4$B1dy8F4a(MK*jSwCCPh*w-7bor{EKDx%VQ11<7ow zvZ2t#-bAv)<26kIOp)~Q+6X0}=$yJS8sUs!G9=CD_UAX}H7HuxUxXE9e=0iW&Z1r9 zi6>UsQ&adg^f?U=EtzjmKe~(%v0w?!jE$jhMq~(wN`2=v`Ko{*@b;|MxdmFX<24BJj)GiC^&VcHS6n#IYG=u`lSx4P24?l zjC#azd@@qAtVqqG020y1NYEXIVj~`38R~9bR^@6Ki1O0=^X5d-HmayZ1zP`_92-`N zUEq#6>L^;%DPUaK|ZcbqlJiCTjd3*<|CP>*K zrc^Pch$DMCN&P;|JccZT1HIU2|!smG6@raE*IG8#!HcrDuxDQRbwwI zu6?kEpAl(o5VA#JY?s-kWw0wo|{ z(%skp#T2gtVwGR)m4`_4Bo;DaHO562zIEFF;x7z@s6%eT&A*{Gh)R0jHrb0&Y<~kajo~eEImu}BEre9> z9b#>wq2D6UF$)uhK={&&^6~ce&Ww}Lq((BUHC$!o^fuVf7SRqz?Cjtb?>Cv31C)tk zX~GcG^W2HKs-BoM-(uZq@bgM$50Eo#T>nxrsMvINq3PzbE`{7$D*`HbcTQS&eBDIoYd* zhYRrw2w*$yd)#B%i0bKplg+jKBB}HH4I|rot+|5L$L6>6j;6!l&-Z1Wt@e5?8d}3e zwe46~JF%^iEORaBay>kN4uC!IkA;KJC8@9_f{1Kz^6(8}zKx&;(`Mj1U}x}+pi55N zr8_4C!8Rnmu@b#NvdGKxH?61Yw8=WULLsk{)+*ALWPB6g&;OSW#-IA!22iKB24z~8%EP)s@4JsQ!I_&01e5xAwAn9z*;JZGp`a&k4C60WjEW z%y4pwMEEI#6W*F^t|-kideRN|?$&RXE>`AZRlEnU2(+t=Ef%%FlGIbXYU8F3jld;!>C0ng5M z2ylgO6!;Vg)^AY=3yGA2u6M}OC-`|+oFHlN_fLR(4ujiGcmT$}nw38iS0TkdehbRK zh*p$5>zWe+I^(P(k`YYFB=@0B+_3##O|i|l@Pzmj7g>NO3tKm$X*Jim=DXuCBo-eq z!TebHn5*SNeHYfoQr5&|8CawK`5biRSFrCvAGt*ExJ&_;(;SM~;I%xk|D?*Or+kE| zg~-64;9IN@AX+e#pvjtxej}4S`^-~6zfW9y&h-7s=JUj&QwaPAn0*JQ^BH916!;x= z`?s7t|M@OJ{4ea=my4f71f;f9YwH%g_HR?~jC@DPiyAftZ`RT?tjt zKVJ&Slu7+{@n|_6N34UTYb>CM_cdoxA+?kuKA}f7{!g6G$z+eX-RECCRHT;^%er4_ z-{hWU=HVCj&=8N&4XMFw6~%;Ff$}C@w!ZdNO0}f&+Oq!m&qBqMFM>RuSTvGm1d2_Y zrAdA6;Ruj!ElRc>W+5xX+A6h|RzwVj<_zc9zM)=PsOF-KFAR(7#U})#Q$4nN2FuiT zTM8XLBR@FuZt~1E>2k`J#I5!exTfcGT3S~t%63OjmB%fYcL#djeiwE8RDUJ*p8bGG zfu^>&^ZZmn#r8xlydnKpaEzPfu^48x%+$hY+J;wzw-d1{8EZ1^-nv#JxqosA~Wcdp+4_e((FHR4*Q%V6lNFR(rVu> zNcZ1vIPazM3;n_fP)e+lYcqy9{KZ>!FOJH=Z*1&+USt>rAcWBzve@l7|2xsfPHz@3 zJc!w$UWv6Rpx+14`VvF_LKmX@5q#UL$2oZp3k}@z-ama*gc7*~q!bT(pMo@AUGzkr zwW`BgrP?~BW1CF&Vq40N{D*c|78bSeHueiZoIO*KLXc25V4|(ke@6SRDV_w``)@Qb zHm!_@5g%DIk`E5f3V%J3AFwjsu6#aUK!Dz+=xw2Lxoe}(eA#yumOnYhRgxAt{svja zuiHBwIzO2gd~n~6Q}?0u;YyD}uOW)Bu1~OYcKq64jcFqaJ(%alBP=3TI(yA}Rks;1 zGH_+?U&X}VihdKu#_KE@H+C*Da|$Zv)@n}!d_h~r@n~-oZ$MuL zbq;+7UMcGM_N^4iul!qJASt`ILcE#9>$q=Nv}yj&Lr|Vp`9mxUjLUW3(twgDM-hNB z#R7p#t1v$$C53vl0{EF_v_1g|eyH^_&S7G&!ID+pJ5!5g`w8)@nBF%{(Id1N^6Cb`R zwYCFNzw`f>u~hWp#N8q8$(Vc8m!Sm_p|Ya}_cAYyTj|OWz5oSA5hJUqV-q&v;v8!m zhdt?;)@8t7n(xZ2IrX*5DF}~KTwC@trt4Vm$$g=dbpMP0uj`NB$~l)Zv@VUv4R#01 z+B>w5tB!qEM&8?({Br!^tC7-btyP4&F*amrs3uG_Y&*sux4Q37Ex9}fz$!&+>=)%; zpqK03rCP0Okp@=%+26ktUlqWGaDVdVSK&Hk+K*A&D^~udSU!)yf}$NGmM$W@nIcQp zGT#GzEWY!q{-4Ea>Oe&N76^q#X-fVMCrA5u>##hKTpU43UjzmQKL9(U~YA_3UR)lTyno*++UxPR}7aZVR2R@kVMPRIb9F zcmWJQQi6Jv9MtLMU<=R)*8#3c3>(%ro<4XowWvKz4pC^ndWJ7jg{q9Jcx_llFc}UO zUOC)492l(eSK}9gHB>7*#<797hbbBX0vR=U-n zk|-&y+)GF5GGofBuUZ7J7W=y`_gCgStE4yawNJ|_UzTc}+>`8naJ36Fx6|fv__}^- zO%tn>Z)$AkyX$Ckf1ywptuyBFbr+#rP}mf1`CA>n^SSva+g=TU{McUB!1 zIeNRy-eDRlhP*x2MlR8AB=7V}AKZLM8zmF-r~8duMF8D73TV2E{KwPjZjI+*NM7M2)DC`H8*)|EJWDm25>*R`Kkg$ z>iH3t)+T?ZRjXz<(@fJ+Sm(VP^}H>`^i-BQBW9LwA7Xx9tTruf zk{$E3U$K;dizC15h>MifDj&Qt<rX%{J{m^g>F*K5nSgd1?~R z+*CKp0IF_nUtTv1im@9F)+~23kV!br3ZnN`KvRD9D{C;y@S4dKntm6u_agXZ`5x?e z%2}R|_|+1h>D0Ym*qWFj3-lGmeYA#>xlDS$hw-(ijCbm>d*44e3X2>7f=)5_yJ8=? z%wkh_%jVWM_VR(Q*zGkF{||Bxn$~E?f!sELeRlA|F-6#HEw^;12{*4xRkbss(RGbx z_Y0Gf5v?*H`m0-mDGBt>6<{c|6I0%MISb?&$aXS!Q*p#jOgmcWsY&97@%4ML1UJlb zw}Q6c99wS3zrWFFgdImVc3{7FORHFF>|-~P?8mm8Vo3D8NF$BJQWt;4 zjin|9Pn~;Vfi+gEKMrE;rMOa=$b=0N4Px zG{lhkJVLLbjbe&rLGJcLJvUZT9W~dJ7?kbQTxtOFfL14p9NI-8d_POuRYmY4k-?36 zA!rWrRs%5M14pJu;%e>biBmdhetd(SfYKhHV<0#$O&H^*PEU5%WGqSd+6?3_v#n!RW!+g8Bx;tAgeKNU}e$FfPB=)2q`;98~pq&&y7eIYpD5~IAcf_%YgL3Ht25Q za1dfjofZI~J@7(u2{OS?Ip*Ze!*?@Rg z0gQ$&EvP6cDc|gb#W_5-stlk6>w<&$LQ~8r zl*{k$3`Et`O}e~d3D3s>FDxG-d#?5k7{r}ErtG$$4I;9)JA;vjwBa7(u9`#d9<`F) z(wD*}^Jyneo9$p?0ytcoXycQTSdzN^o^O#8t#$CsJZN3_#HE6{S!JNdAXQ|F7=dOa zn~xKx-H7FaUVTw+1YeC3H;~&E-`Q7x@6g1Lq8k7Bbu3q>ak#YnDPaRG7cvh^E5kZN zTrU1Fo+c&Ah4%cOBH%vImm*av{O@jM&oyQ&8E>Ry6U?lv_kMVVCyJCyatK$X%2n&A z-u)o@$L;{BTAt1(+`2mSxndRTc%?ked||gXb#93lv&%6irPJh@PTiE|GxMEc-r(Rj z;a9_^Q5J{pJF>im>?gSBf{QZtS$carb1geQ`Ny0AHSHwaoEMxqMbM^cN9pcm#-EFZ z;c#^l$dsG`fJ22%D1-N3k}8PB<58|mEAlMkRVQz_X6gxMeta`FxlmPRpZc;{im<`3 zQ0xr6@(OjP`s81}$FQeqyhG!C#$Y19rEh_3?!%F>xm@SIY%Q0l07kxmY#Kkc01IC( z2a3xQ#&uhtX>uz>^5ate&9Ns;s*G4@;w4{b7PJ`J`k>6gD93}RACn?sc@LqjHJD}o zF#WxvHe-z3Qg%kxCfph|)*x|0ZKr1x$+P;g`%jJBXAs&$S5gvbe$UcLB`5+&!KD;> za`^Wd0P)S~iWp_;qICLn$Z>))H@i$Ijy?W2?BoU01_SV(%ijQdk+)qWt0^;zq_c%q z63KPbYbIUpJ%>m%$a9T(>2A**LNDBb#z1wT6+-8bP`+OXGS`z9)iD<>wY?9b+F!Aj z6v$f&d98On-RZ7=-`!Yn+q*BYt90N{` zmG3hXC-Lh>wVxDi{OlL{{5+APq1130exn>424^4?-z*JDGb1te7OmussEhLkw=L!` zH1Xy83sI(j#ixXt5?EP??|zyXs3Lt z`wW{D%siHNU_>Mu*k|gdlr-rK4Jk~dgn7bSQa0NEi6>%zjaY)uy*R#g9t@MBw!CC9 zaupn0Q3xb~+WXtY*#dcBi;c#QOzPS~^F9~m`QinwMNCk`e*v1d7DNIqhdZQwRxCOH zH2+~VvjjC$b$C99A+3w8`eoIWJTyWB>gp^Tzw&^<@kTX8^eK=<*9=N?+7thSubxe? ziU6$w4X9n0(%}lVI&D)*HiMPWGF=Gs50Y4y>7_?bj}gmU+TMgDSc+wye1D{6oF@!l zS%GGqB5asz@};DD>-0J5R(j2OkX67^@0*q7RoMkPWArqZGoGTFL*_WGl@P3 z#s%NeWB($qR7gMD4eU}ZPgd1ftKIKihJgaRLgCmUr;VYG+$(L zg;58r2DfcZQ{}S{I=k;8fH`&Hykr!nF87gzVJ^tQR;)7+_JTnZyz?eM8t&4>s}pza za8u6qg6Nycd%HRk@c|02ser=*dh}~@C14OU8AkdOat5JK<|+*bhHt@p(m>_fT;2|* z0Md+qau()Jco-!%f`G1!8#BNZEIwR&^c7lOBlsc8M42eFBlkb1jBNJ)j*bpt$KG@e zzv8E{@$r=<9;*-VOnpnZ^@f2YPgK+rQj^Lu>|HTnI_8{8swNf&G9(UKhN zM~@ISKHDa_PEsBKtM2$e^{VB;0xiGSt^9vc!#-zZG-VQ=BgI*Pn;mOg9IjJJNGxuU zz%%%290&dtdK7P~`QHx3e=_;^FJg|siDPeL<}Y!epKl2{91Ec5PJf}=b~X;Hj*X3L z*Q_a1F8s^1n%MboT2%^Otn!g=^$+EvULIp1E5#N%c7!r}E#}3oxeY(x`rQC zYDfL2{$|GQvdb(+iMMw}2n`KQ-a;cg1Gko zop$pVSm0kgmzb3IhhJYOic_99?EEZ$N=NfYVI_#k5Xx2;p$d@!MD6x4brq!QV0~4I z0KXopgCd+IMLNLe3wx!pR(Xs^_0b50{2``nu;@t(=_=KnMv=eg~dXKua_|U+F&OLITQ%pQt&&q)BP}}DPP?~ToU8aXNG{^wGn(haMh<$*52aP z;h*ETlAfjbQZ9~iCctZx941Jm`kiNeVfgEZf)prk085{#6G(2 z4@AQ0p~h!g7+<|mu1F*te)@5AbyND`@OTRe|>Gos_P=_a8=US?yqtZtmKk$cQk9)?!{}_>*}e0 z(X1>g<}eBe{{XQWn4V{cRXfUEyZytPB7AVf_KM+{Sk2o?siSoO4$O5)jXIQ@ieHFB zoFa@BV~=M)38(1aUmUb%lwbqyo`tZdtgWL`?vr*MPVv7+TlX`Xg>OqS({z2{jAV$v zgNLOrW(4^-|D`~Uc-?txYJRTFX|^TU-gU}Yw(!*9oPlMeZJJn=>8Z`p-ABcNLLQ7} z00sW^IACXPd|1(DCoLf!Cuw_Haw4an*|{M@HZxvj_Aox140?nMiIkXER8%TAnPU8m zurIHz?2ljkTuOe0l5TbGfTWdscufGK-uSxn;T$j4e*BZx$EUlI&W%?#rEp5k>+D^D zYj4J%TDdLSYXOj1(XmN#Rf|Zem?LWzThMM+eu=N8d-d-Dp@gL&;nD&tYXUKUEz4c4 z7#u-`q`|`JJd=x|Eg=qw#w^7kuYed=ueN*ksMx>&bM7H zmQ2kYsL(7a)-AY*3qYCfH5lk5pKGGZuf$|oR^Rf&4_$C#yI((i@~fw;>mt=q7$d&wf{jsXc(c-OC6Gez8M4v-lwuyJgm1=eRB77lO zGr^R#kkcy(*kUsK}u2xvvf2FBo8B4JLi`+e2&wfoRD#a464b1m-< zD>1nKQhEUCiX1ZX2(-KwH``xlAy!w9lnmL;;>*VO6s*io7>tEC)$J1ZWC35nMOy*xXk6?Ex zpnZ$9LhsQ(3;v-r>RV;ACf~|+ovuolMa5HltpZ>9v>aITZ*QXH{21ynC2NyI4-sSUg>CkvQK&N=gj;LLU2I)yt+c=Y{3RWm zT3u-rcAMy?dmRF9<9@HiCYAnaUlg@#VfEdghWlC}wecrjwL}*B7F55=J(-AG_-PH%nh zg4l)R-VrOn{c z@tNKO{7QJA0rDT`6TxgR+4rqfFJgo*v|qM!@0YbitY0mycFnXmuk;wtu({`k;rc=W;^0ui79++%EL4hVa&{U9#@ zV4RwyJ%bhtGCy37L!=#T=h!7#RLk7*6BCbz^ScKzy{pb-8-v8QL5y#^QZI;!3FqZr z4uM#H*6oZ~j3r@6tICkE7%<=y8z?8g@|)g3Ti(OjGOeYo5&4L@#e8hIMsxq<-OTUO zTi)corT!iv-y)+TI6n^6B3J}ijSF1sb)JtsK{ZihOgN|dGC01!j z0PuIN1}oMTo#~CMj#V$pm-hN&5)zLMo!N$_% z-Y+k%Mf)H493c*sV$!|{N$h}bnhZ8n=1%=#ePhk}8T(GJ=m6o?8&q8$t2;whPDdvl zPG(bJ&`enh`j6O)E+)p2oW=)LuAe#&Cm$>z!o06f& zjT=PW*grSnuBFyDZoD4^KbO_@HQWA$+eWu?`gLD4P3fu9Y-ta@^8=NrqOd7a702nZ zG|i$PKef4EXw7OHbG+1jNq%iK`>|2qo}JrBrIge-Tsvx3?(Kc|Yt+eF+w^bHW;<%6 z9ksA{g3kOsFXpi}bTBk@K6E}3aMt+n4mgP!oaA2fY(M1hPhYaZrN9Tw9HOlG*xwY+ zXM`#LJ;aLiDfWo(asS-`6V;U;Fh=b|?I#N~A_(hb%vsZ%3;t$Rt4rW7I{IMDvfH=2 z6k$OpzeFqA4!-x+D*KOj04LO*&Jwg!_uX!sjt&X0`FaaKte3)ElB;1%u9rV9Q^Lo^ z_voj7ck6D4G3|sHK2P&21Is8Dq`RU|gMsb44qkb=Ni#z*VNR#c%j0FzNzV>b_@%?JPUVjLa+ud-DY6#C@^~JJTbx- z(YkP(Sn`4Qob%_sqw0`$kU|o(eeh;=XK*tznzIW6r=xJBExYR+p`$LwGVe8Q>P4)o z(d7p`)Dh7tPT`a=ZF~PeWhD{sh{V}#|2Q?%xHK~kWs3=3sZ9l~W!gnB%14mP{5T)V z;AhJAp0!`jyS}h9v1^ZfE9#@tjGz1>`4;l+w%6V_9JsaaQCWkUL^tJ*f5^oOL$_-% zI%}_w#+4kdl*HV=n8Kmch0B)xfJkDihR^+XKXRrDcmTU$GM5_~XyMyCM1(B)A(xxx z-6tIAFx*^fv-eRB+dQI?sRA%sP}m zgnL&a>h0jvqfGIVb;e{rz9efp$d1ZfUtn%*|0%Dghss=G_}rGhB3pm{;2<2mbG*AK z!&)x+`z%T$@QL(!S`c$83hAqtuW+JUa!grc#;DCE(!IWcfZe3!Ng_sqn(&7BxCBy}W%Vry=8o;93XR zCx}BKrQ%05GOz<^%GkC2*uLC?uDW&S3!netI!NmFuui7TuJ9M>I|QS@>eR+B40s*N z>`6`4X8SlRYUDCRoQ)t`k{bbao81RLdSI97E+RGCBELK*_GYZ&d+jrb(R08^uPn}I z8#vYWA{(YnJk)5@e^nU@5f*Jr47d8_g?cT!Xl{firn%M+i^Sk?pU2x+%mZVw*AZl> zF4AY9F0PD3U&I@oKjrv5-*(l_aP?r()l(wGJuq|C)K=F20z4~|tWwp=q- zEvsoUcR@)*!=j6l*&ZWekibV#JL#Pls|E=xr?URmXTYSTOD4*-DDYzg(_|3e{S(&3 zq83(*$_?l{{^|5P$9tB8)@xD=vd~5^;$(o?b`7EFgVNFD0;>urXzCrYG^Na_U^O`( zm_X(*pE!KEJCdU0zQq7bdhBL5MtNp~d4jyg{;FSGXD?(uKe9{D*!f`Zp#@5#fUf4wF+s|TWCZxqMq`07)#d7tBB17DSr(UJ=_a+)q7n6DQ z>b3*rSbI_)_16eggUZk7W#~?_GEd9zB#aX?8)z!r87{brurt|=s?}bFYK&;mIMv8D zU!85P2su@s?{>=UXf&zVL3pgLxG3mqY8MT?M_YO#eLdy+w8VarCOaORl-^a`> zHxKfHIC)*4O?S>l5YbV~cpUyv{M-&aMu_A!T66{fZf_G?UZ4Q96Bx%?Jg^? zaOS2;&AqsZhegy5O>(zL zWV(-)cv`pWn&fJFyMNKAsro@!U#(}4S!{xdnaMnUlapm?toWYfo86RSl#=^)$`FW!_t7$k}tRCc@+h97; zR(~EIbPhy7&;F9v>>`}jLlv?WFbH?M*oS|9NR$D}llw^a*=G5Jb1Sgr#OS1_$MH9D zOq6UJ>~<)_Z!L=2j~60!ogp+2lcq6Wer%h#nCFj+v;!AG3Opq3@eO1S53AfRE6II& z_0p_}a;IOKgKap^9v|1hm{BZtB~xZD$$6-bB54HFkw{tY8R)KlI$0w95N=tP(SHEn= zjQ$mF-LPGltnLtSzWsUSFc3P*w%RAG^Gi5i=oeFr%bi9Y-E-}LJzzezw(G@Sckaiz z_UuG%4831-H=MtTStbsl#JhWm5hdQx-h4CpJ83O+^7{u(_SnV;%|}zlR$hXmp9HSu z7zIQUxyZ@8eLB!Nrw;6;;9F`pPgNfc^Ov1&s0xawMZrGEElX7K4rh=BL*G#+vE}+c zA;+8dswO@LmRy5(#_av@&eKlK8}V~-)a42%I>KBOYPaj!y?&^A6vT-{S8v+iGFG4E z{g8PGmp-n=P`6!8eeQT@t}_Q#Gk?^Wd`PL{Otm@EM9FXYye&pY+*#OjM84i_?MHuy zk#*t=%Q%Wf_*MSt(;0=yGgK%mO@$0Md1IU307va4%_NTlSHl5WX@`-QD!&u)3{^yfN?Gvy4H)~dld-jhwmTL}r{rZIzW z*}O1%SL7blLOs@&7(nMS#J`-JcVRs=`sL$VBfS?}J~_G0#526|BX(=Veh3VAJXR=X ztxd>Y-kve}RTHT{) z2A((CHXSXAO_u4D^i2zKj|BVp^YCVYe%4DrUVaQIe$%;{X58mTqBK;_nXo?Ql#1G8 zA0nog1c|6^#Wt8LJ(@CdtVGV7xUIaUWdH1{B-tF5uqO((WKz!M>23&QjiFSBOR0q4nKUTG)9zL@ zkXYb_Z^$ARfmM6EQRe8whpC6`?+*&D$e%F;?tR0|vky>(&$x%1=W?tXPK<}%sTSIB z77-WZIQd1u5&Fq4$QKG-ElPfpE5Fhr=d>(Sk=~EFsSGE_8xrU#B34ZS#vttS=tAI8M($8WhkJ$*hg;r? z!j5@0IZ89iQm4O@d-c&mf_=>`&j=y<`3lV@CVbY8P!@0!sf#Tt^NseoGJcivZJVM= zEt1<4I<>MPf(VFd*$#8cEWJ48&2R;Qb=^jjfv4paq}QnY>G8tURMnht&GoJO4*K+b$W9222cF<6Wdb$k_~%(M^E!mjP0C<)f;Fa$Twd4Y`xhz*}us{G4VIV<1JCp zjHu*m^|p&89GTMc;yT^>6I_VS>;^?+9(LD2Iyx9tjO?TTz2cG;@KgP|l`q=A z!0e@=1r(ZNhw-zbx$l&vVzg(9MIvrd`Kl+vaFeDcG-TQ@TZHlI$R#&6(qJ^XEt~BA zqs*a&OUfS74iL;jn31 z7;SbpS!@5^7ALQpYnGn3?W`T3FR&f5(%EQfJj&of!3P4y8{s#c@t3 z{}dM;OWrKWA7S1=%D;aEq^du!f_|p-x#@3_jo@r#SY^&M^$_!S086P4s82bO)}4nD z7xg@EZJ_p>2XWB=pD5jHJ;Gz+;e>A@wwWsXs^ah`mCu^@6Ml76Ff4u>ey{DcI3kn? zEI*ZT)Gu7#Pe^bFSNK>8OYQyEEmHlQfQ%Z=P1bIs zmcf@FWHgm(Fa+5LUzq=XBj7<4YetY*U7`h%!QlU4!0 z9uG80P5{xtFr&9YX&!QUG>2VULDDn~Wum`@qz3yJw$0Hj4bI9|2cOET zV6L-3rqqamGK2qcvZLC7tJX#$gxY`*==C{oig1vVb;fOcK1n~eteHQi7s%E_WU?NW zv|ub-d5^e#9D<`pTPCH{dHVTeAt3U6rqRG-43%8+Z0qa>TuvBkjKq3Dwo@lf0Pj})S`*GRB4xxB-g z|Bn|G?bY(-Rl2(!Pe#ge)($5Wi(U(Hx)~ER{NO$evm?wtce9W+AdoeK*}?(h(n$QT0(}41#eBeSy9Kb1#b1A}Gpx4_9~POb2nyi&vi?!cw1hW!KL)a1 zY4F)evJS|6g9e1BO(r&{B!EKrY$A`ywd2%wKgVx&cD76Ud~1T`{DL?HBaivbCPPIZ zuvLr9$gMtDBF2X^M5K_#w_H7Ml=BW0sD$ z{eT5EJX6r_QAsKEGO>Loz4eVLkg`M8oyT#HHcIlKbDYkdST@DgIH^rY*IzVOOZCMv zH|Xc~=S6yKd5<(lOCC+?EEi|F0ObUObp7dCRwQJ7?blMXAj+a8k7l_&O z<=%#_%3SPqcZ+uH$IW-1*K`x%e_rr+q!fCs2~kC9K9bt3sJHa1VuN4()HnY{Uv0AD zkeVNGCf44HEJ27VmG@`}da~oP4E%o64vYY_cTR-yyGgU5ISpB7sr9%YU;{ zpH`6f0k}$+4u`X<{do42=ZT2{6i7kb`l#&4;SD^{mfS!b0pdZfdQ}4eYcBdEZvqb$ z-vc#5h^_2mtQ-P1iqgAeuJ+^i01)&Lo&tZdw3=>CRG=8o1Knt0e-5P)@thxyn*%OGchh1uuo-bO9OFOWKK9_Q7F*$w{8A7xeP~QSQ@l{G&rXju2W2dbHb(c8e~#Su`z_cZKPw7PLv)qf+dQq_^Wv0#x5;3^F$T; z8&`_6tgB5-6ZM3YApkE)F`b4ECEIiM2GtkYC@rR@F~;m_?AFf{*k>M?Oc`N|_L>QG ziSWhC9iL?Y0mrTw$c!jrQJYl~Oc*FPK)F4s%h*a13xrpsSDSrT;zZZxcq?eCJF)D8 zCC%Vtkc9PT1i*@Ecz{U9m^#H10~oc|JB_sqds_MHtIHJ(b`LsXZOfKWh@~WP^mWXi zV%^r660``a4@pN}9*P0!gCTgMtqzquzN^+um5!_w(yL!I^{@Ivv@wIEg{5VBeW&** z+SL3)Xi!qIB`&}lu_c5UO&`g!p9lz)8B9#~x8iLLCMep>GnHxD063v>pj~dRr}QVB z!Sr7g73Q<9ZENky(`nj2EhDEn%)YmqZ`8-4O!|eB1n%c)><fuQ+f6dpjP1|)%&Xcr0w#QHr-cJz_kuTkN+27p<82cnO6Y+R2?;v% zh)9;|V9#O&&StlNW75{KRlkd%S-K1U_YGC}qPttt{dWkW2{tqVt(2sRFbE1D)U@es~WJ*i zeJa;u*`;%xk6%x!fCRp_0(Yw9a6d7(N-x`{HiT6*<3Wj<+N<(jxeYOBbAOuHU{3cz zY@P!u!?s7GKun3VPoqHmC%+c=D$$H*$=kTRc8~APzkC*hIx6yVa@D7PuBG;#tES(2 z7xRRT(OaMzH?~3N^tzuZKK8EOtw@_EN-C&KimBAMj{Z#|zoc#sj)rEW#ceC_m2CeY z5xQRe*zXOVi+UICtaTQ|MWbe1&t6*cE}~WRC*I>qC!j1NEF0h@lHB|1dr;8lHgm|1k0p=7?JvihZe7<)S5P>Zdi)b5^<_1K(i)rJj^4iI z;${Kmm2`0TS=Y$!6(Y7>|omXeuyVGY*EJs_Slj0?l0fZcCPaK>XrrtrojaG$J zHcG{6xDushB?OsM@qSP$!Q#sqtDrh~h=zR2lT?lwUM@SP+O8`w+)A6{(f6E{nLeky z1{qEuCZhj|3F2y+0ogB4f^CO$X}qPD%RqbL_MUWw2!8Pb>4}cLMB#p!LIYI> z#Ues$!$t6%KL<0&xP~AmcW7oB0CPWdue+P6Y~YiY?3`j9zqnOv1a2RMQNTKrG{*t) z3J?$_(}03%7bLxj)#L)zQ^wu_)KFUzS8JfD6vAGSeM zVbzFcMS+p;yk1>0gk0={I;M^*y$IKV6`^vbydikQqoI$3g2^y0jmjU-GQAETwYNq1 z&3#5XwTe5{vV@-Q(y8)NZ?vU2b2`Ccu4IaTd`N)V=Ee7f`LRNRfdD6+HLyySo~&xI zzzYZd=}b_0gu#=6A=Zwe>z&r!Du1Fu&S1@L+KF`c1WB(_)|9v&1uhB#>w~S|leD3? zlX6mbZ0)i*tnq)^E8WyrFY>dgT`4Pjyi|`?!!h`h^)1SNzn6lTNf8?BQ^mgICi-^t zdsw;!tI`K#_C0_FLSNRyH5PvrqG30%ZGH3Waq@IIT;VcoGPvpTX_l+V#~~{! zMgVnWRI9Bn9+H|X^K3$R%~XBkf@8?zU98QA0aFdN=bXzIO2cCD%5IXi^d=!m4sc7U z={Gx)4Aar=?)%9#W)1Z06ikRt32obTTy`gIvyTzsQ(V9{_ubUkp;uiv=M)$Y1(T_Q z1xKY8kLP+t+wc6LiMA<>nKvTmj24$%MAe>8lYOZrZ?cEX zo=xl7poq>Z#Tp%I4kcZf{-cSScl9s#44IjjS!AwHN4tSUm<~>2U@!Ll?n`>m%=*$3 zi&pn#<$+sKGIy&dV?S_eeOP-nDvkg3z=ljJO*rQ<==FDD^C5gS8TJaBEM6w5!K5AO z`fT#Uaon&~R7<8&v!KF+dX4#Oqi=NlmSsQu4*YWERpDy0ou;7%#W%j?w1t6C%_|tI z)0X=FeOOxLk#$fas!rr>Px#1#nW-N84KNK>Ms)gkj@6yaG&eNg+sT89*9+g@Nxk;F z3h%lt)2od?z=@L16+8o|B0c^SGB)+t#yX*iOKdrxtl7lrBZKZcAx+=fZDLCA$5>QM>Y`!7`z0uHxnAqrF_i>5MeG*?jj&ZxP**4AIB?Bw0fs zshF0QzWx1J?bk0I%&H+yvC4#X7V8+-8#%Mt2&Y9|_H`w)dmZVcMY}6Ymw;aKGSFvX z)P~P>!HPPUsek80vUU^%dF)JckCZ5Qg+Ip)leb1+CZ@C@p6@B~7KNuIf+08$$5)%> zjT=dZ-GX@*uS8fy_*Opj*}7KP#RzJMuQ1A@TE|apLX0?a*e4F$0D$%=Z@0hBTR1J8 zIc4=@^5ZL>#w~!^NyMqjlZ^nv1kXb3Qiz)+TjFop9!OVj)uWC`s}v%#8TCVVxS-?V z`@xPPv=H=e?mh@W?%ja&r+r}^{9Xl*2!Vb&_XcGt(Yp9v#49f7a06ro6d0hY+xa;h@S>`1XQQg6gBh7Qw@M%DTp~cE}3)t=28!Fveb! z77xnGY#@gMCb(@6@t*-%BmQ9VkLEN-J=%02Lrvdm?*}I$*&q-}`qe#q*WRf3M?~Z| zj4KnWbBlO$d0_)qwLY7U(`o+Jqg0h0LGzubg26@_Xd28XedL>^kwDi@&WUL)8hO;s z4WgV$lKHr`ss>aH6m6huF7NozA%WesLBL_8^e$Tz-_7Cp{Rw|nxtEV{AWa-Mhsk^m z+FD`rpB%@mxmxqYLhZ6PmE#_t77maN$Gl$KI{g?-AM)w>Wba5Tq_@GC5kE9R zn{@Q^Vz2~jK0a@o+xlW>NMxrf_JouE{e-BYOfnDoBu#gD(sAhTh=<8zDQ}Vj4{YAu zBNRAI0S+pZj+KN=RT$Q~Cs{WzMZNyn*A^)I>89{aEVC@^h-*u@NRX_R3$j(SN!GMY zbC4@@F$DzF=-t2Q@=V}B{^0pRZ;FJ-M+ddnf8O^S{bie^{lo)g0~8V;ElFXOgn#iT z$Hr@Jc89C4DvGv$W8_b}crx2m1+MTv%+tn;)x5&+pYYMH|KIk${c9)zbKD;w(`O6E zV#3}|X7I3p<5WcrATfmlJ54cQ+bUeBRAkNP?Z8-4$^-{u&1lH4j(sd9HCR_vWJzA| z7mxysjYgsf!UR0&vh$;UW=}vcu|W1j0XWS2K7eRm8@viV;avkN%D=j{+u24*Bw#PA z2k9jl4WdV77=T10Bn-e}m0xmlP*KMoQKkYm*wGNrrp>FSqngF z#-%XxwaAXJKaIr>Z<*aD1pM0A*cfzufwpo6Y%TQEj*^_qQpFZV(%p8BjP{pnqMe9^ z$@FDx9@Pve1>F+6Z93VXzNDM^bI=AI{;&X8K{VPjcle-y;j-VhkdWOuc(7vtxeFB; z4Z1v_HtdF2bMRbrAgW6;u@+g-qY zA(d}66XO<%xy(%o^sL-PoT^o&=4|3fD3I=x#+b&^fTEft&5i~ z27arlR_H~5gXLpfysl3|t|_uc0Wcc8415E82FzES=-txKOSj;6HvsfTP@oVOEh^D* zbyN*EI`wze6S7H>s`i*HE}nXtUI8 zyUr11Y)9JEO(elZPu2j{cEtJ+u&kO4D*mSwG(^K;3)58`3-PQs?Qp5~^`U;>qE`GBu19mP84AxK57e_+@I0ARutNnZh6K=RM z73pawmgF|8nugAHj^C+ddy+ZhIic1a%~@q|EOx4_Bv5wwy@~WOqVPER)?-wikgH2` zsn6rGm25A-B++Zl&1kW*Ru+}zf9okJ)`@2jVP#;Kk;U>iw=!5b>(c=1xl z1_3%|4veL#m zXzFOh(d^cmXYPbkn7v0o>3h#%8I#crSbnT3d$Ev|5+`;g1ol$z0N6};1A*Zp#(I@j zE?PKM-Cqm=5_hVI<7#uhA=zNB+g2^Sk47LBaKGT}=%MKXDHbzHvxQt?{DKT?C*iz- zjE{(#zfK@1^Tbfu9#5y(J(5wef+O$e;&*M%ZL^-$(+hQ5Eeyrm>VDY!cB~G6xVxWI zmMINqnM>HUUFbafS-s#RSz`5>?Nv+t-22WF^&IMxX}~?A@AKq4p{T?2GL8>^JTtj% zYpm3}MMhJA@e2r8AFP)FGFKOb$y?|zV|jc}Or{-B5g|sMj23C=|4M6YrButjyI!K`_P7@}_8cl|5yAZX40nTZ~ zzH}P>+25k)9%ibeiv&f(p{HPK(gYlO-^W-{1RX?V-+{=h8h~0Yd3}-AGlGt97_0%n zz9UF#!rudG^<_)MzG>V>b&fA)Rc%+W@W`&^d)#iEwpc9M_X`h%NTqJNR(X1iYs7F~ z6>F>1q!o0foi507`BK(S3vWy4=xm_v7YN*TG>;K2oH)%S5~$Bxr5)f;x=PbEi&v53 zpoE}0v;reWr?SGdm-1h|*ygxvw$(79Uc2`KEeV@P|kh4;kt-eGhP}a%V7n=^61NMo(c4rg!0hax@y7qbRey8EnPb3*Y zWSNMh0dHTAby~)gruUuk0N1R2Jc_Bcg7-BW1s(ldS!Bj3GO&daByZ6H7MbNC!O>W> z8m<8gvx~{Is1^Y3bN_{#QMirHsc9R(lQoFStG+l?a2H1%0M-BstatYmDwTwa6PYc& zyoaZ@SfZLqim!zQrd@#eZ+v`fSkiX^=35f=XbuAGBCZ;~Uyn`fi&Bw3JheTf{F%Rf z*8;TP`U75@-dmWv6ebTogS)(+++-5_1xR{}K9U|*AJ~*|QCKQEGtZUvM%x&n#Qmr- zhxy~YlgfakC}DW{-$I?rQOS^vsLGjHzI*`Js=S^awQ!_YA5?Jh3 zgF(i?JL}oM{z!WD%WIktD+XncuEfjGr^j7B9|t=@bP{JDtj!TrxVD6)LCVC;0XuQ{ zLp_813E^DwF)@moPK0aGN~st!!qnklP@2`d4wn>FZR+z&xFZCRIN(z(COBZe=8j4L z{r^?*atLi}w}#TQns7zgHDM+m7y3p-q^3y09DGoxBzj12d-`QZmWP=X>S9D`us_&> zimJ4%nBZ$IVOsj;?ZKXN038nmrg>SulC&W`zPUov+tk0Xq&&`_!`6|URnH71*id*p z7wU1&2AJ4=R}!+wi$ph*Dr$yz}ubxn( z_>h{}G0wEW6qD%$Juvnl^AX6ZTYcPUDl&rh-1rZtzlq0uo8>_;WITl z%;yw*O?9)J^B9#vNmvMsx(2S$@NX=+P%lpOqpRAf8siYqh&nZb*4t{7I?ncuLFyJ+ zz8}YW!`(M4KSRSscjn!-3TiY|NsO8AncU&1Ah~UFrzL}gZOQMqwuvg3+1^Nf%RSNB z-=XDC?}^TRCX#|N9s5l7iv2dOi8^(Fz3Qv-fmuoMZZVHKA?z!u9Y7D`eo$1Z;;gF!Tc)cgZc^WJxoNE}l(&M!MHtZfHioy%}ZKwSV3+ z8pKmYCX8DI_t1E4HtMa8%Xa?8c`$)=Y%(y-Bt1Lw zJyTj}5K$9L-g5DcaC1DhA<>6P9P^1O2MG6^@3pR!APoVHVhg`I)blksf7fude=cD; zPNN#L4I+!G&9=4Jqxt)afy#hg5*^$t(HALsqso5%{va!~4cO)gx- z0k-PMHdaaAbQt6ZwAbV36eia#hS?5RUNC<|LeY6X7oV#&v2kEZ^Ysz$_ngLG z4z<*eABCMl$b_@@&@h^|2KU6XXkZi9ru^`#XQ1ZqQfmC+!>b`M2I~a&sjLE^vjL_c z;l;OT7`6HVv&7_@rsnrYuaqHOgSl>3Km`Ur4P50kk3!D?jBn*dgkM62=Ht$tFPoNX zY(C-jK7wL~#ngCf#-nHw2YS4IrxW*KTooI~q)*!(4xY^z$TFiCv+NVZr!e2gwI94(l!`E)9AAsNL0dfH@9h?dm^7MY_Tbgj{DJjON$C;14 z#duC-9^y4G>5Ho)B2*Rg3rJx<_+1NG4_{H0`W27KF=iT#Xb-irF~d@=$uvpT@8K0N zZ*kn*zHj;N^4Yg2N=xISJ8k56#MOPaNoX}YY!oV!>#)1w5n&a zlENjc&DJ06E$ch7dP1LvdT_w+qSc~o^(t3i1tcU05RbfM7H)$N3tA>_2I*T-i3Qi8 z9*-1de~udV46Gws)s)DXiktlfl^z@s%c2SAAX#lwALJwb+BQfNXnvfj>C%?gI#uWh zwo4fD2u98h2HTw{zOIw1cSyb6B3(<&A>mGMA(jdCbH7@f9X_rM+a6`&W6sHTrCQgF zg0!(6J(S|XE&7_eqb~AxbPu7h%wH#OkA^ycz>wN} zyTvU@c1n)}BqIj99n>NSDZXMg;$DqUnd}4;7pZIc7kO&boED2(D&6%=apYm;+a5gL zhdrwKSR02e9-+OImY~vPXw5}t=zATdSAbl^Kek#)CQcuA6$$<8XW8M~iS*po5@q|Q??1;>npJ)jj10#aA<413aH zu{gh^u)3jG&R+hnelSeIhfLThD;IGiZY=5?{2(KwzGeu^$M3$lm$cvx_I1^oz0+9k zHu9nA{heq^rQ5>Mh%A7EtFe$Ok!p>2m8J2OPq`;ByTzdt9^!J;er;wE8S9PEn3kBbU~7);*Uwt+~%VIc__{C|9R%c2DDHDPa8vAyVu$ZY4Q#A{6&_^u|qMQ1|V- zUY6myH3p63n3BFpA$Ok^GX^me`NjW^!Dr@I%+Hf(nvb(F8%oZ4xSERqs|otiC!rdf zC9=fnpVY4SNn2*@OAco8Y+a0cq)X;~Khe3N{GFR@*xauvHHJ0M;?lP&^A`Dlw%w1g zsJFMk`w#U^NE{XX)GKn)va_8|IL9P#11WaC1YI59(y4tNeJchVvaztsvDhvPWpP9r zvo#_pr}y+dbJEO*0`8H39$H)h;hR0YDDFNjK5J?6Tre$SXGp?WK4;Hxq~ukO!8G}x z5WJ`it?f1p00vnfop7MOA9bXmjvOZU1(p_{(Uk+#rpEejj&_>UqC6X!Az*$8*u|er%#m%OZnq(RehMS`}8(}~#>K>U*CQKLO_T*EzeMTBE^Px*iCDW?v zZYBx#x&2YmU1BR`1IWS&rMmuuvDyd2;;-rZ(RZ>b#rPp7dA9`6YnE`I{lEfPWU>UE zy4T^zlhHuU0r5)HC=P)w`R+3g5lGK`}Nt$ED^C zs444M`99C5aLjDSuys<4*vAzNv!9Z$WT&u|9(<8x4fXR_Rs-L{y*nVH25F&m&o6q% zv@?@#%<4exyOTnFSDXdO!=qMeuxm^H=Tu$1_%TU=;C+`{ZuxBA%EGh2fw?QGE1wq9 z@B^W8IHT_;NHl$3btXOcDjsm=-k6sUjSg#=6=5+ejbgJX0)y_p!ZY6xBroLHs~tsn zUyQbiF`(%oFQa#@so2*izTL-lil#!Qoa(IqKKHF&cxzIi^gN5Cyz-454TqCQr3H;(XAP(kLE%L0v+0J`0KH=W!oCYqCWLZ4cuSG4j4aMp@>Yw^3Q{4%|4q z^}#*lD2KfqX3kX7xsk`2c31Xd2Eh*eVtL7$CHu_TCT!ZOEH`XPq zX`2))-5;ZrETb>KcOWm|YIIfGDGYyp6{oyZefCrTh?87tVLFJlfc=4+4=kA#2b{}Z zxR=cwPJ=hGy7TGXTN){U_mgCLVKK&1N6pfb$40XO78(lqxxb&DJK5PK>y#nhYv6U6 zX1@7kw*pg~kP#mT(_xg7CMUEamKCT(UpL8sGYNcm#dwtWRi#a~j$%Ou0B!EC;zBw$p}nvyvRG}>a_+gR=hfM1G4iX?_4 zHgDZw*cgBi|ND>ZA}4INrh% zEO_YR1GNe4S55LSNwL;OgOLM#0QbV2f57K%&)D?cyb0ljZVz5qOv zkEyCMdJbUAE-c4I#_-ZVGPF)a&YV$}e2;;A(tDNt=^%jtiNP1VFHi73Yif!d>1HAGUWkqS9&Yy{@r?Rc+R(=J&57 z632(j{jHB4a7kjjr=zV3t4;vhX9VDG?Tt~9iq-ML`M2Y#<0ce@U}48+b38(jC8FdM zHcRMCc{=3N&cAM|ugplc05Y5EA4IDBqFRkMMbxq?QQ=-+*_U_!cI$C{_WFMU4`t3( zL6|fB=n(668v4$w#pP_fMeZxW&I}mMH^no<9+3`n)OR4nuI+^r%%HVn>3(r8a-ih0;(U|Zh- zM_X5gVLkHtV%WE=1Uq7}I*qr00r^l3x9@Cvi`-=GXZ_IKz1u{sO>3^1^ag6~{S8S& z*RB2u4w!4{6ri+*7b-#D0HEEymgP#H3wYCz8MA4C6Vq*w`*WB@cf>=u=K_{Y=owgq zBh7zOl?28A+dx%Pmc(HNoBfwTFat7}`K_lfyFVIZ(()jpY1JxyDfpceyG1d zv>9?<%NJN3DK4~UxW?$k@^1$M-~N+2AR|R27Tk;2FT!PiWFlG7x$xCY=eD?0f7kL( zUIlHjX@h^{J%9`rR$2eYO!<$|_<6e4Jjg=>wR7hnRuBiUtlaBMu)t#%l`E5R~S>u7Wo)*ebuXfRQr$h z|9?6CVd-Hbxg^CJd1jQhVTfL>ZM@y#jF*A`%*A?GT9~?6a1LO!e($!I!$`mm#kWw% zINbgz*BL6`mC$<@zMR*}tQt~+0s49XnP}8rn^c^|xWO+7Wygvm%#qHE7!P zs6>s!x~QOg{Uv_LwM#tT4r(H1_wOj2!GIa5)7>2(#^)4SYw;YSn&A}$!h8iYR`^;s zyRHQ-cW`J#=uF*T9*N=MQT^x6VqmlUKVl7cr3l3@_2JOTso3t%=eiGxo27ic1vIR-1>QzlMwH^F)>hlD@*QmNgeuZ(ykg33*UX$zw685 zc*9zAcSqoh=9S+Niypc9?yYYH-C>+1VKF|w)sIq83f*;NU10@2xCJp**arm;QX$L3 zrK9Z&L`CoY&T`J*PCf7P%cIhx&5OOiqAtywed49_SFik24t!HBUpwy%_rcWVBnAdj z=-Fwi#W>Thm+s#)X`_?IsFJ&`fQj+5Z-?|4bU3+J;JQWH#C{=Vy%}YJm-7&N!KAac zVwU`5=nPNx(ewLE$6|se`#JpK{8nAx>@@DnyUdC@oydozyucnh*}!}_;f7R+B{p6&vWs9B~xhgS7h5g-+2%YX8b2F+g>di2%UM=hu3p>hzFyI7_gp!{`?v{xMad^gVC-!` z+1{PF+%}-o-77(mD{7`)4FNg~!U46q1~YxG^`1XpliMyEqR>sd3Og)`aTQ5=_4Q~pErO`Z$T($Q2h0TZ-gh? z)1OcH3wM{=$4>=wrvdS0T}5v z^3uJwJhuP{lm@S^JW3ta`>;@K-6iIfR?HAGn5XgQ=bf#dcBl?jj`saU|CbXrvewy{;b-=TsN< z`ZMVU7#;fdu`eIU#z9v#RRAyUaHNGF>>;+kRZL9}fo#k~=iTIEHKcWB?|hy=$G+O; zVh8&Mi4jLYnlGjH=@B1zx=RP>-?dS!v3M!_%GJMxre|n}EwDi&V(-u9hrS?90-xdxc%kVQX6MK1(+I<<0tR!wndZdx51PA=p-u zOw+dlynsTnr{iOv;KaXeu!co5@HQ8C#ex;xywM=={KZ!@tR1{QZ(%5>LeR1(ewE^6 z)(jyJ_LOa|Cr||8ot#`cJ;-Ch;|H&LF8}oSfy%Z7E*Ux?({`7=R@|d|FA&0b)LVKj z+{u!WV^a%l1ns=Rh9u7?o_dD5u^}?XUxFzT2Xj_ZX4n&1vYU=qT4#Db?D0BRP6*f6 zID&yX&$XK~xkQ4NAkX=Mp9NgApL`O6@b4V0E%{OPYQ#NY5N$z0d+xRgxWWa^(>|@X z1pd#)&#o$1Q1_Q-a)!BxkYzvLXOdKv;?5@7mb-Ys7@HClDEf}|hCCV18m=Z9BipgX zEh0E~M4G%@>a#ACu0sMV56LMMYC)L}SEA-;KTfLPFx3U2U7XEk`OADcDY;iAUDPel z7W`^D4s02iN3iC5FSl-W%*U8XwyVl1ffe(h=@Jl)>VH=v4sKn>(dFCi@l0lv z(*MSu+-Krnw{uwgWt4Jac)D);^(+f^Sq~@KsZtVQA<8Sl5(snI7Q&t5TtNJfY+y+1 z_ENABU#R>9d54ULRZt%{y3%$T8G=FMw16OzF@iBDz5!q*CrEFFZ?zvJ@>{^Rs?(ZR z(>%~c!|eL+k++>oaV(QwRC_e+T9_|NoQ~Oxq>g&WPPnNBoB=DL@*?XRbB&?C70$O&t09pM z-LRe{P?XjE1(RRqiIH{HOYbz+{zgH-l|#$C{ki2CO7qm5aWWIJ=7sHC?aM z5E~aCBPf<)RNTMk){VZd91l9(rdvQsd#0$**jvB|xUfvYlOY&>*HJP3Dc3WeSnbjQ zOB_9piiAE8XBc_gpy@c`j1K>|va+&-OGhk)X}gvayqp9xPPWT_oUH1N)0t&czj}e4 z`%AE*yFgEGnE9)Q(!NvTC0`86TV?Yal}^YpGDLKXrVmTS^u?eIb5fVV-2q!fbuF%R zX*f2#HcHfV27i-_adjwUteG(Yv6j3uvfr94Z8~bs9i-}zK+k0&CbB)HIvX%{X^pl| zFCsk_`$Ta}9(Mga`D#3i@>mjzM)ne}5-D~3Dy7r1dRngQ`v@6jardzgDqXr>$7M0@ z23;Pj9PObx^0>)_RPwl>g@jH1iXZPgr{ascGW70QX>h!s)12lHFMw(?kGixf-z1DU z+I3YZH=+1OvS_MDYfE|b$q@1TYQ?wP5jJAVhEJC`x|aCQrX9pLb}|-yD^GD@N&9a9 z{ySk3vueJnqB1V!O_x0p+$7c>{3|DGusX(AnOVx=?EQ&M?Sx>im~H&1kx)x03qwTP zHMJnE_X9epfyz+Blfki+_ zhlUy^FLlZbZV}2(?QE?J)#qqLAr-j;;SDED``rs@^yb8PX;{B7q)}jhz_z12avPyezCa3ESefyr1Xjxb}S<*6? zWv^Ov&U@uBFafM6Hu+#6+S6 zR`g~snB<^y$L}Dav^WDqa7TkbH`K|i358RRwHDzLYXV+qgwkav%Uz)13*T4Et=>o2 zhRaNF7Y0MaWoD(YSj&) zs+gTBAz_CyyfDTl!49bD9c z8XL1(IDRJ0+gNIGcoys1RX{y{41b)#-v@HLQeGba9TIDSH*H1ldX#@H9OW3> zB#i!vf<56=rFuabEsm#j?>Trdr?N;$Cyr9qan5OZ)D3pJcO>cVUs`UEq@-|kY|aUe zDaWZ(>7=G`psj)l8UY0okaXcOH7zW7Af{WT>c%1td@k$^p8J_W7i`4>odo)2QBSB7M9wTt6ch>9O*&H z5qo%2FB3XN7ENCFzU4DH_Ob5&#B`&Ass^ZVzy z53XdVDYj}dkTk(6Nayc25#!XXtAmZ!GN;!0Ed{L>T!^=JVwUsri1%s2jCK6$C(2@H zK-5jo z-jER+I$6YWJhZuHelI}<)+<;q;Ei4xsE zp|sv%E5*?ScJ#2bMdeRia6pEipGEx2r#S*E^kq0P{oxAEyYtF5O zjZ&sayCL;~wb>k}_Agg`%HRl)_$&dwg+yL)%h~x98G+I-qGMJHTLw@y6wn$Xh-Uzh z(I;&h6r%b;G|>o17eG>4TTxNr1T1esz}|b&hZthQAD&;3y1MUz18zz-T_aTUn#EaA zLg@zy2Nsf26YebHs*4H+@c9tIF{#MDTedyCD=XHJm+eu%DQZ-287L6ZS~1N_?&Ew<*;LZVLL zTM%{77+^UFceKrs_p{H2>0sIzsSY5Vl!yE%6o(!Q8TGAk`5VjC3j`4D`%W9H%b2U( zV=gi=eK`TF>F;OG?vn(#t4Bm|Y&!ajCxaoK+C0c-`091Ohq8zbJ_2&2HGC``B(WT# z;tnNrlYpb1!5(N%d1d4p4^i`feFRdUGuf%50HgS=0@%+hG-^N?b$S$2(7Or#`o4XL zP`$^Ncvj<&kI#mv0^14Qi=e7#H3(@bOVUjHf&K=(yg}&Bf7Z`;ChAv^?sJ{BgIgDC z97`I3)UQth1o+@@ssP?#NPg+biRc{~)Ml>9PLBP;a8)yA^1{fAE$1}T^TyAUw1H<@ zYe6gv*986Is4A6UkO}0^gNA{14G8Aj>P;LVMaB@-XJ_J&SHVKK4k1E{|2-`WZ*$CM z?8qZ|IiL}q2qAql1sU*j3ySjC)RWoe1v+6(7_V%w;lEDTLgY~TUal3D2`C;4^)(Ty zY}92th25`Nc;^OlcwI>b#L(u38NL2v=}xDp1F@*>8cio!fFbnv0Jp$6yEeou*%#*m zE&Tc*?2MrL7~mn$1xCTQM-;C$u3Zv04GuUtEHB8g9$+R6y8r?95pjmA%{U}@H5Pa5JI63eBtGf-4(JX?nX4~vX6?{@?M<8Vi+;1wf;_q47bRH%1ctXrpp z+B8=h zEidUpj{$nBU2;A6QuZPb9`8KuGbb;N?W+NcwXU!Qf@e*@u6|>JD7Q8S=S~qHt&jD3 z+L5eL)v@pHuLNj1vt$?J?N7?`^f=%f^QQwy9z!1i*v0Zq1MBoaLV4MW^i5CX(vOd* zvzU9$cgngJY-ODFty#StiYo+r0_ry&4tdJZSX0|3e_Q)m>z;@ktcuk7Qm>^G4!vVRYqCzpfZXO8W;P=v#+ zW?@uunF$w{n~*&r+jE1y6eAJ6Q~OPYubBw&5f0wK={WAOPmRN#>@~;U zvTy)dtG;RzDP#4W`%cBMbK6pukS~w?v$$qav|$*ikOu(~v;q<40EnRaq-JUDwq|K= zyO2V^)m$R1?MsC=&W%fbKpDe@1gUihOpkKQ4y2sIeC6C`tE01fv;PAn-e7NfF3{T>tyQ6Evdf2kkNmv>U zMNqIQ2naIVr9_h@J&~0j!6WzPHAuym!lf=UWm@aX*!33?WLDD~V;_RsMX28E1i~fO z*lU`+`Q#YK^S!=h$RA)5xR2<~J?RHb55XooGkUc5g=K2|;4?Z#z9` zBdxX^YNwdr;`&G-^5j>RQ`R$4^WBdR3&XOr#Li2{eViL89b!;!v79f!n&&u&OHACZFZ`b(1!VWbDP;U-bgjEwM&8dK#MQLQ~@zAoj{EXcR0!;LpO!P zv-yuL+9i$qKk_@ zlQQ!H$Mm`A4VLCCrvu{}SWbA|^9A)v!mM*i`5Tk1!M?LOM-W4w4Znt8LR+r_298XV zeXHN45p90>cZ@+_vr1+9I#6d{sz=B6aNY^~fp+5M{^qVYgOc5GRxIZLZ6?@z-!7Tb z#!$uUudol!a+#F*_y2Oi1;PXS#JoBSPiwwGFSX;7HHU6@CEjE*$Rs=LScf-K9_rS1 ze$w_`Q%v&$5=x-oP zK*!ZR0IF=(tKXI`1i>8O3lU_}ZPw52k#Wp?U*euJQxw{We3&eOMROK+*FK%V{_5^J z^wI>)4daHGxpe?rn?17iPb=Dq(NL`24Ioye{79g{HG z)r$v-_ekA-qjxj4~U75_g!!)zV(`#vjL_%c(}wE+->S!Mw%P6aDaJcDh>j8hOFs} z?%vm*eoXlV`NU8^VB_7W_Ln+E3NN&f*(yX1*wV^Uyyn&?`G@`tQD7-^nL8L`oB|j2 zbchPQ}_0DFE5F>EE$9oTZ-)3Qs8`xUspasRqkyv4D# z4PX#sGbxRreqpptLrgXcwgEWvTB&nKy{ZtXqu(l)Wp?MU5!Ax4FJJ z`UuBf?*F~=)s^yj$mS!qa)bOcs(s;D?W~)7@>BBLNprK;WK)GRR0vd!X!KfgetouN ziiRp`p>4OG2%s`{weR1hI>`$J1ATU)Szf9!p~o9WpB+XMhSWw()s9lKyKVat4ZhAf z;Z96Hc<56Z-@tv2nz^jR_!}=pnVd8Tc6{Il>GX+{o@cBiYmJ zh;usa0`0;Zv^YD&JNBsV!U0t4{Vw1)RH#Qkin6qBx?CP|&%XSwoeF{^f?TrEO)XNu z>~Mzb;I92}zcny9{r-Vx5Kyi*i6wOVJsCVuAF&}^q`vQu^6qB-n(M%mnj*+8N8OlW zmKrB`lUs_P+ge+-G|R9AmTed8g;ZblYvRI5=3BkHZ>FGA^JVeqIAC|!7G8*4?mgY- z%h>vWcC!v9I=tobneH%qOF6B zQdMX{}_DU)JIYnZ~sYYPSWTO81!|L|!OlJI%EGGIglilSw;=mO_ zHE~$x_-U#ZOz`d0sJ4p-vo;)5*2o(%ATKJ{IWpE!Fbbv!oT?b{nk8S8y?xOdo#3UG zA23O_p1BI{^vYdMdZ-CRhOp#GgxBJ)9^}e{`L$pF}y5PsKdgli0dAB%FGHk?FjLHZlZ>)Bbym0Nz744v% z>cU;-Lu>8+QP^9pFeSPSRE!8Wd2N;pr zDfau3_>}{4v#F`91c7)$n#K?VzhzZ{nlX z!}}#BrC3&uAU!rZ9<@%HuRp^ns)nyE-&9um;c%s$mc_(zQZeh@*w6QJ*TbwTf}*f8 zbskTHzhYqaE#JF3^3G9?LtwkbL!?b8qY1VRx_xvi@Reolo7g@&lbOCbsbA~JuKHTg z93~)M@gwVX^0^Sas{R+V7$y1Xq^I8Khuo$;P#l?QS;OfG47x6%fq8ZjotJKfKHrP_ zfKF36UzK(1LP!!!97BQAwm}HOk=5x+`P(Xd`M{K|R7&ce6)RuqZplH}ZLQQiH)vvJ z*UU1UUl}Ua9!d0-|25o#M`%&BRs|&)j6A$6_7!su2t7$W5ha=HGgY0uZsp)7D{U;Z z$_}1B9c)vF)-$5&8yN8~rD)&QcAsZ--tfKNqH_&>zTG8>^CjTsse31D7aNvM?-lNOy3G1&VDhG?gw#%1o0c;BX^0EB1Zp=e zPbO783yuJ>dD|;0XW7!5hZ>iO3jWWmwAywGMgf3&{D2=FCu}}HND2*?wZ53#YY>)x zua1)<}Z~<31|7N(WhWB^r$D{9%=GLMyH3n!Ij) zoUE}}_&Rfk)OaXCW^r}Uz~KsG%MjI>JOD6#*pS_;IY_dPoJW*>6VEzp(1293DYWzP zubO6_I#}*RFQ7hP=!F#WWDnEv5)d#N9vtP9Om=SNnXkL`R@aCPFeEL!pXBwLi#KAa zh1k6B3f|4n{sn-`)4n+EWl$xk_N@kGf^4aZYA6*@9|h@W1>{1Ms#tY(Pu7*$6FKQJ zz!BFYxr!!DLyScEf9Hn;Mg3P}n1@>T=~`=enLjvzzwA`CxxMmu4ffluK_hFTN)hmz7eU!TJJl-PgR z8!pdyK@Tv-k6M+>rG8Z@C#I!>VhgFqXqip4J*$eM&82mKtKW67xJ|I8$r1^F*9u#+v^CmXQ=+Y?6WWvWTnu~efLN{rWy+-ES$S{DJE5e$6O1& z;**n_c`m`4=yqnzvlHN#@yYaw8=vIr-t|7i&U|`(Da0SXk)XMI}(E zJ%w6H+jmSZRhMzOrh@7#7x$a}(r|BgCTH2m2r4^Br*HAhX1+giY`>L)DFkvUL&wF+ zuEsKMsy#W8{6A&(KCmx+moOUcWvaO(q+nzef5;rm(`8IqDC?r~*Pl)&S}9t!A1|HU zsV8sIGP4~$!CSvC!dm4)AQZFr{6>p9KKP}NnkivF1XAw_y;qf5EnsPO|J{GM-vv}d zeRHR7g(Wd_;WK=7k27eEDyoiTSB_U7} zEFb98L3)Zq(K^^E;j3)UO_!Vh)UN+@VCq5b*KJD6boeXbN}iWfJo&rng>64=BDdl1 zlD$D!;yB&Uw7{cHeFo2hX?V7`CTbOa|Cb@C=DT6z+m+&-CsP(}AIbds$a5jDSe^Oq z_C4rYcDq9z2qrG09Z1MiYeM0RpDXJ8+=VS0TBpAB@{hPm4}5@n-9)8RcV~9MY??s0 z2gixgIDq(lc zgM#M;SKt38+(M10px|?qro+1C<0f7HZbtD^=1$D+%csXawJTNbl3)C*_6|P%o6JJ! zfUBzb<>ZYh8KwTEN~d1;mcVg3=5yGF1-{VVoiwQBfeR{@32@qxYzdO9_%66Tg*n;6 zE(MxxwPHI4GuOsXRQ_&ZRDy+@0!xn4pvt|GNiN}(9M=g5Y@%FV&a5Ttwf72~$~Jl)s1lUVHf}7r2T>Ub9!xsAL?~83jTiCsMG@96o^@RTZ|8gdU z8j6^y^3UeFBe$J0ubZeIvLw-pWQ1THXGw@%`^Gy6Z;3C@~IbJBP+_tvzXA z5n5&3|FwAG$fHr3t6pMO9W@sESaD5sycH2$x(^yeh(RT zz-0zEpdois1ZLlw(E{WFzDaQF6gzqJAHN1@!rTAV`2P2Dnq%w;kbq+#7BSZcjm!Ty zxx0lBfoTZZ-#9`$KJSoTTr+G zf-23;TKYvuqjhD$D3@MS`os*oj}@i2MFA;fQW(IO2=N>FpoTZLyR~?tG5CZ~`0MHs z&|74NCe#s3?&ppy&dj0WCZXQ_>XCo{c8?B}^X_b}hESBhrs-BgPvihKuKLy%W`nt0 z1T?C(AUnvocnnfFVc}gyI2D9t+UD;P@l8B8+U{=ucTMhAWf?!z2d@OPktTC_+6Bc( z-e5r)-y1mlwtF~i^#O;?rdQGj3kBSAsoIsDg)+4~+bW{Z=)q%bCnuZ5KdIkaTwby0 zF*`NKx^`-Opab*5&1O3bl|_2D=D8->mE{}Em^L@AlGfgGVow3&T_~T+fCg>b3-gut zmE4QtaF3~Ph@4x0XnvL5@ym&riMEOdNy|I6aba$|8sjXjz8a4u^Zh|F z6>!}uoA@&k=QDSnlr5GFiTR!wmW}3eD8{}V3bSzsZ6GQ3**0fO>O|f`ZGKTG#}aArSOjYp4PeuF5Oo%t&mlB*z9p2Gdv%;OYILs*d+|(c+6Ix00Q(lth=jS!iNoT?#j;U z1HYK?9mAoYfanr}6C{Q}tKcqv`?;@tzL&sqsiw$!x1f75QevJk!iX_Qxo}PgA+}V zk7o^1>kG8`OXjr&q=%yI+a%PtE;pa^k80xZ`0)hgqdiOdvGtCY)6`s+*dkxBGBDkh zjhpTYJ1(z5BmT500xyyWdVm(@nW+|q0eD^MHQV-epqU6yGp~uPERxfXY6P8Dz(Z#Pb6qFpdW0oj zKICkF_T=naySLZcKR^u9?8^r&{48O{Ms@HhkDQ>P@0+)Ww=K&BaL_ z*;6JMiSln&aE{v^p#QW?U)rv_2)@94|KcZ~F6lUl$voFu(Z!Wg)vUr&y#M1 z^q9|ev1#xO*40@kPq#C6sBYg!WNZOB5Ul0Tm$>~TXhNT{NHuAE*+&b>)#73{EhtaP zg7`Qy|K;hS(}BiRa+nRhO*>vrt@CXX{sO^C>mw4g6i~&?)n5ucP?Mkfua0-gIC-=2f4Ck-Hzf4$VQM|BA`GFiJ zT`4QNH5uxph7}z~n=ASGuM)GeIr!M+*1+DoDGjZ!Th~8|oes(chX+k^m8RSiE>p3R z5tqxn^lGl8Z>TDy`(3JYp7?A(&`ZwD+J*DEol3g;+b1f40~X=-KLAXQwtvoe`R_!9 z|L1Y#|I6=S{}$S_EaL7JQ77}yqH>cHhxM<%41d(lVG;!USO1l{8gq*h>=Lp$M``dXMLA; zSfl4;%9(K*XgCS@R-;%jo#iZ;hqxF3;X1?Pydrg_($s;O!DmJ%jS8}#cmSegh`Ewa zdnH0*fRiiSV3>Rr;H4Umvf7=5FghNxWs z$00}H+%*h>TSsa3Uj;kcOMlqJ-tOWa78)!<-h}24?sF9(SS2*K6or#g##l(cugrT{?AKoeY39Nni%--f;*RmOKIUIWJ%rW>D4RK8Z*Mh(YQEOte7| zzlB80rT~rj?0Os!F%2!fWT?nIUdsY9)IzE7LG}7|$Ybt54o<~Ziu~i5`JvA!56C79 zroiy}faYq%BH)mXFVIK0RzeEC1tN9Ps`*eFjV}U{ofy{FIqQeOXE}+P~Sqv9U8vysV*+sZn zYdcr(du{tE06<=@cQezsOt5$fbhpvxqP}ufF1PZTcNIgtu;Qvam#zqB*8GGP?H955@Vo4 z3tlN_lX+z4lTxJ4`3dX=tY*&1ONpLOHdG_Kb>aga(IehgHmV=oi#HFq5$f-=1JIDIZh>1yug>ln_qoW^kR{5dR*U-s}3| z@j;>+jO`vzCkRE44cW>Nfvdj7^27O4Xf!>fFR?pIgS=+5nFIr)u-AThfx$k6y;3W@ zg9g7r!?%Z#o|%Sv;nQ6r1XJee#Cv`&L=)l2+oqf-3tNtvExx9m4u7ev|0KKihQeZAZle zKaesUi~36U_WZT<3GnEJ9oYa{Q&IksPZyirSsTBvz@xaDx$cYK98ch+QW)J`dv9JU zoO=50C2Fb@+I7#*NUL83vQ9Jm)=Jd4nQ5%m&nwmuEMi>S;6{Ckj|DoH8hkd}xnHT* zFMIylP@A|Jtej=7y#?C-B3#j5;toeX>4Oxkl`}cH1XSG8FxNX4fw+L>0Z1xK?vv## z!dxnrS*W}OjjAQODSrFB&WTxL$dV&*$?~Z$ejLusSrZclUjdMc>E1e#zAySoj%1-3 z!a5j#@=R%xy?(w{erZ0^=ndex26HGaBx6JQ?SnA6sMwnqy#TxK9_w3&zVDMdiK9@A zP<&Nbu60=fJ_)V^`{*S3!V3hu@8xcH!|;Qe)1REPO_ZT{jJhqpcVW zrNG6~c1K4ZATfw!wZ+TTfc;T_I$3@E-N84X`yqNS^=;Ql4F>)NOC8ml>sziW#lXlHw=*w=iR95iH;o>G zNSo#eD#Xzj21pBK_#=G_&-haa>i~e%y#*mfH_7^O=)_kfepk(%JDK$!9g!B&^WJS+&$6k)i8h z!eLjiq=qkc-jLd26DY@u*IA@6zpJ`T%+xcc_j?t+MKX!ZtW5}zc>a_+kRkD|J*0qn z@HN4*gzz5 zC>^21dwv`E26P>)dt8(Ve6()k zirtANU!OuwxTh$k{mF*!vDM4jPS?i*6d52TLGMVu{Cy(n%ZYIfAbF80NCwYY-p1~jAuj+2d z?kCp~_OVA~YgFtV1^C7At)QxPUrn)wMNUekeGTm7fnbNKBFA-0x;Ae_Tk+^*-yE)m zf74)tJ+-W#o;0*(QoERNQ!zDsh*qNi{H_T#zCQB9vov2!HdpiwNtiO1mIlJ*GQ2fJ zFZF0=0FT?UrnKhcevpliyAGFUCoqn0e15D_In@5$*9e3p#H(n2=?J)cl_xqob+j<* z%5!wuoAZ7u{$@`n9?Mc6P2{EnBB1)%jFChBX@NTvUViprbAz(*}=(ziS6bI7O^j2nDV8(X=p5@t>)Dr9R}|uHoGkK z*ABJB7AK(T#q~idGIRUCezm~Z?$1io*Sd07N)HX+pHRlLXeeGg_zOzd5joZz2@8rk zpHz+{L4}7$jE&ygpQIbcCH%I@{ies=VDtjdpZlqN?xl!Kz znJT3$)A(AJ>}oWLr~CzH%*v}a;>Q&>Ay+a9>{+7fjik7NQNiyZ798}K{spQ@FBO8D zG~Nm?ZFAuxFD9{-%!P7mO~^3&166m6#Y2PXPsce22S!DHS?N!u#IY*#pX=(%od@&h zhK#9Ig4ICFd}DxRLSquXT}C~stYa)~I0#m`@keiTV>x?gUxEc=hw2O|c8)2EbAnv z2&`G10<7~isx^HF#B!XwZ~QJZ8g8&Rb!>A68Qycp*={OFF~0czvXTA<3$2$}`^abX$r8W@z}2i|DHI6y zDAt~oWNg_m-Ly-OmCE1>tSo(yt=lz6;(-goGQ_tK@sjL2cVyI|;I%(<4HOwt7k!cT zaA){))hOA>hVCwbb{7WF!9P1ZS4<^sGYC?IiF@^Po0<$sh-=pcND%I;?}-~cUOCGm zijzmEH$dfnW_GVI+yE-t0{VY<=^udnO!5D>Y1hmlp}G}qVobNo?;aHAsyygoMQJ^N zh7*FGIfP08HE|W~#fPA)_xcFepRQg^<_O45k@VDRe=sb(@VX?>vpdWb|5rOO(n}#q z6BETR{I4qeDlOFP(|_<+vkyKaqo$ZeSMnkAud;jhdr)m>fA9qUr%QMlvj~Qmo_+YY z7aYO-)3g8O2=Kzpe>p-7yzsXp;NVkf|8`!7!D&(b?Zj08{%8>BD}+Juo?xQABYp1y z($c5+E)t5l+U(TdK(X?t; z;+fp16DA1L9+Y@nCG28-rMbo->4%?Q{l-f%pj1ANNZr-w?S#q{ zxxdR4(+t0RZWvpDe%gE$_`hzi{Rda<-_4GFW>YKt_}&S!>kNkZm(-5~;2->^`i%m` H2T%V8(Qe%X literal 0 HcmV?d00001