config.py

"""
This file contains all the default parameters used in the app
"""

import pandas as pd
import nltk
from texthero import preprocessing

# DOWNLOAD CORPUS
nltk.download('words')

load_clean_visualise = {
    'FILE': 'Small File(s)',
    'MODE': 'CSV',
    'DATA_PATH': None,
    'DATA': pd.DataFrame(),
    'CSP': None,
    'CLEAN': False,
    'CLEAN_MODE': 'Simple',
    'SAVE': False,
    'OVERRIDE_FORMAT': None,
    'VERBOSE': False,
    'VERBOSITY': False,
    'CLEANED_DATA': pd.DataFrame(),
    'CLEANED_DATA_TOKENIZED': pd.DataFrame(),
    'ADVANCED_ANALYSIS': False,
    'SIMPLE_PIPELINE': [
        preprocessing.remove_html_tags,
        preprocessing.remove_diacritics,
        preprocessing.remove_whitespace,
        preprocessing.remove_urls,
        preprocessing.drop_no_content
    ],
    'PIPELINE': [
        preprocessing.fillna,
        preprocessing.lowercase,
        preprocessing.remove_punctuation,
        preprocessing.remove_html_tags,
        preprocessing.remove_diacritics,
        preprocessing.remove_whitespace,
        preprocessing.remove_urls,
        preprocessing.drop_no_content
    ],
    # APPLY preprocessing.remove_digits(only_blocks=False) TO THE CUSTOM PIPELINE AFTER CLEANING
    'FINALISED_DATA_LIST': [],
    'DATA_COLUMN': None,
    'TOKENIZE': True,
    'EXTEND_STOPWORD': False,
    'STOPWORD_LIST': '',
    'ENGLISH_WORDS': set(nltk.corpus.words.words()),
    'FINALISE': False,
    'ANALYSIS_MODE': 'Data Cleaning',
    'WORLD_MAP': True,
    'GLOBE_DATA': None,
    'GLOBE_FIG': None,
    'MATCH': False,
    'QUERY': None,
    'QUERY_SUCCESS': False,
    'QUERY_MODE': None,
    'QUERY_DATA': pd.DataFrame(),
    'MOD_MODE': 'Country Extraction',
    'FIXED_KEY': True,
    'HEIGHT': 400,
}

dtm = {
    'DTM': pd.DataFrame(),
    'FILE': 'Small File(s)',
    'DATA': pd.DataFrame(),
    'DATA_PATH': None,
    'ANALYSIS': False,
    'VERBOSE_DTM': False,
    'VERBOSITY_DTM': False,
    'VERBOSE_ANALYSIS': False,
    'SAVE': False,
    'OVERRIDE_FORMAT': None,
    'MODE': 'CSV',
    'DTM_copy': pd.DataFrame(),
    'N': 100,
    'CSP': None,
    'ADVANCED_ANALYSIS': False,
    'FINALISED_DATA_LIST': [],
    'DATA_COLUMN': None,
    'TOP_N_WORD_FIG': None
}

toolkit = {
    'DATA': pd.DataFrame(),
    'FILE': 'Small File(s)',
    'MODE': 'CSV',
    'DATA_PATH': None,
    'CSP': None,
    'SAVE': False,
    'OVERRIDE_FORMAT': None,
    'VERBOSE': False,
    'VERBOSITY': 20,
    'APP_MODE': 'Wordcloud',
    'BACKEND_ANALYSER': 'VADER',
    'MAX_WORDS': 200,
    'CONTOUR_WIDTH': 3,
    'HEIGHT': 400,
    'WIDTH': 800,
    'SENT_LEN': 3,
    'NUM_TOPICS': 10,
    'TOPIC_FRAME': None,
    'LDA_VIS': None,
    'LDA_MODEL': None,
    'KW': None,
    'TFIDF_MODEL': None,
    'TFIDF_VECTORISED': None,
    'NMF_MODEL': None,
    'LSI_MODEL': None,
    'LSI_DATA': None,
    'MAR_FIG': None,
    'WORD_FIG': None,
    'LDA_VIS_STR': None,
    'LDA_DATA': None,
    'MODEL': None,
    'ADVANCED_ANALYSIS': False,
    'NLP_MODEL': 'en_core_web_sm',
    'DATA_COLUMN': None,
    'NLP': None,
    'ONE_DATAPOINT': False,
    'DATAPOINT_SELECTOR': 0,
    'NLP_TOPIC_MODEL': 'Latent Dirichlet Allocation',
    'MIN_DF': 2,
    'MAX_DF': 0.95,
    'MAX_ITER': 100,
    'CV': None,
    'VECTORISED': None,
    'COLOUR': None,
    'COLOUR_BCKGD': None,
    'COLOUR_TXT': None,
    'TOPIC_TEXT': [],
    'SVG': None,
    'HAC_PLOT': None,
    'HAC_PLOT1': None,
    'WORKER': 1,
    'MAX_FEATURES': 5000,
    'ALPHA': 0.1,
    'L1_RATIO': 0.5,
    'PLOT': False,
    'W_PLOT': False,
    'MIN_WORDS': 80,
    'SUM_MODE': 'Basic'
}