In my deep learning journey, I have explored various datasets and implemented deep learning models to solve real-world problems. Below is an overview of some notable deep learning projects, showcasing their objectives, datasets, and outcomes.
- Description: Built and trained a CNN model for classifying images from the CIFAR-10 dataset. The project involved data augmentation, hyperparameter tuning, and visualizing model predictions.
- Dataset: CIFAR-10 Dataset
- Key Outcomes: Improved classification accuracy through dropout and batch normalization techniques.
- Repository: CIFAR-10 Image Classification
- Description: Developed an LSTM-based model for predicting Reliance stock closing prices using historical stock data. The project utilized 30-day and 120-day time windows to capture both short-term and long-term trends. Key steps included data preprocessing (normalization and sequence generation), model building with stacked LSTM layers, and performance evaluation using RMSE.
- Dataset: Reliance Industries stock data (6164 rows after cleaning).
- Key Outcomes: Achieved reliable predictions for stock closing prices with low RMSE. The model demonstrated the effectiveness of LSTM in capturing temporal dependencies for financial time-series forecasting.
- Repository: Reliance Stock Prediction with LSTM
For a broader range of datasets I’ve explored in my machine learning and deep learning projects, visit my Dataset Collection Repository. This repository consolidates popular datasets for experimentation and learning.
All projects in this repository are licensed under the MIT License for educational and non-commercial use.
Feel free to connect, collaborate, or provide feedback:
- LinkedIn: Vijay Mahawar
- GitHub: vmahawar