Here is the adjusted README in English:

📊 Streamline Sales Suite

📄 Summary

This project serves as an all-encompassing platform for data analysis and visualization, leveraging the power of Python and advanced deep learning methodologies. It incorporates a Convolutional Neural Network (CNN) model specifically designed for item classification.

🗂️ Directory Structure

The project has a well-organized directory structure with various folders and files associated with data analysis, modeling, and web functions.

data

• data_analysis.py: A script for data analysis, generating necessary plots associated with sales data.
• data_scraping.py: A script for scraping images from Bing using web scraping techniques, which can be cleaned and processed later with data_cleaning.py.

models

• layers/: A folder containing functions that implement different optimizers, learning rate schedulers, and similar.
• architecture.py: A script defining the base model architecture.
• config.py: A script containing parameters for model training configuration and MLflow artifact creation.
• inference_model.py: A script for using the trained model for inference in Streamlit.

pages

• 2_📊_Data_Analysis.py: A page for visualizing and interacting with data graphs.
• 3_🔎_Item_Classifier.py: A page for performing item classification.

web_functions

• language_State.py: A script that keeps track of the selected language for page translation.

🚀 Using This Repo

To use this repository, follow these steps:

1. Install required libraries: Run pip install -r requirements.txt to install the necessary libraries.
2. Scrape images: Run python data_scraping.py to obtain images for training the classification model.
3. Clean and process data: Use data_cleaning.py to clean and process the scraped images.
4. Create dataset structure: Create a datasets folder in the project root, with subfolders named after the image labels. For example:

   VENTASYADIRA
   │
   ├── datasets
   │   ├── books
   │   │   ├── file_1.png
   │   │   └──...
   │   ├── cloth
   │   │   ├── file_1.png
   │   │   └──...
   

5. Train the model: Run python training_model.py to train the CNN model.
6. Run the Streamlit app: Run streamlit run 1_🏠_Home.py to execute the Streamlit app locally and use the article classifier.

Note: The data analysis part will not be functional, as it depends on the private dataset used in this project. However, it can be adapted to other datasets.

🌟 Contributions

Contributions are welcome! If you have new tools, models, or techniques you'd like to share, I'd be delighted.

🤖 License

This project is distributed under the MIT License. Feel free to play, modify, and share the code as you wish!

🖥️ GPU Support with Nvidia Container Toolkit

To use the Nvidia GPU in Docker, we need to install the Nvidia Container Toolkit.

1. Ensure the Nvidia drivers are installed on the host.
2. Follow the steps described at: Nvidia Container Toolkit Install Guide
3. After installing the Nvidia Container Toolkit, restart Docker with: sudo systemctl restart docker
4. Verify functionality with: sudo docker run --rm --gpus all nvidia/cuda:11.0.3-base-ubuntu20.04 nvidia-smi
5. Build the Docker image: sudo docker build -t SLS-TF-image -f tensorflow.dockerfile .
6. Run the Docker container: sudo docker run --gpus all -p 8501:8501 --name SLS-TF-container SLS-TF-image