# Wardrobe Wizard - Notebooks
This directory contains Jupyter notebooks for the Wardrobe Wizard project, focusing on Exploratory Data Analysis (EDA) and model training using Keras and TensorFlow.
## Structure
The notebooks are organized into two main subdirectories:
- `eda`: Contains notebooks for exploring and visualizing the data.
- `model_training`: Contains notebooks for training and experimenting with machine learning models.
## Notebooks Overview
### EDA
- `data_overview.ipynb`: Overview of the dataset, initial exploration of features, and data quality assessment.
- `data_visualization.ipynb`: Visualizations to understand the distributions, correlations, and patterns in the data.
- `data_preprocessing.ipynb`: Techniques for preprocessing the data to prepare it for modeling.
### Model Training
- `basic_cnn_model.ipynb`: Development and training of a basic Convolutional Neural Network (CNN) model.
- `advanced_cnn_model.ipynb`: Advanced modeling techniques, exploring different architectures and hyperparameters.
- `experiment_results.ipynb`: Analysis of model performance, comparison of different models, and final conclusions.
## Getting Started
To work with these notebooks, you should first ensure that your environment is correctly set up. This project uses pipenv for dependency management to ensure consistency across local and Docker environments.
### Local Setup
Navigate to the notebooks directory:
`bash cd notebooks`If you haven't already, initialize pipenv and install dependencies:
`bash pipenv --python 3.8 pipenv install`Activate the pipenv environment and start Jupyter Notebook:
`bash pipenv shell jupyter notebook`
### Using Docker
Build the Docker image (from the root of the notebooks directory):
`bash docker build -t wardrobe-wizard-notebooks .`Run the Docker container:
`bash docker run -p 8888:8888 wardrobe-wizard-notebooks`- Access Jupyter Notebook at http://localhost:8888 (or the URL provided in the terminal).
## Contribution
Feel free to contribute to these notebooks by adding new analyses, models, or improving existing ones. Ensure that any new dependencies are added to the Pipfile and documented accordingly.
---