An End-to-End machine learning project with Flask

Dockerized

docker pull arjuuuuunnnnn/wine-quality-ml:1

• Data ingestion (from google drive link reading and loading the dataset into the project)
• Data Validation
• Data transformation
• Training the model
• Evaluation of the model

Usage:

pip install -r requirements.txt

python app.py

Then go to the redirected browser.

Modify and Use

• To change the dataset - Go to config/config.yaml file and change the source_url variable
• To change the attributes and their types - Go to schema.yaml file and change the COLUMNS and TARGET_COLUMN variables
• To change the parameters for the model - Go to params.yaml file and change them
• If there requires more data preprocessing, add that in the file src/mlProject/components/data_transformation.py
• To change the model - Go to src/mlProject/components/model_trainer.py and change to the required model

Workflow

• update config.yaml
• update schema.yaml
• update params.yaml
• update the entity
• update the configuration manager in src config
• update the components
• update the pipeline
• update the main.py
• At last update the app.py

By running main.py

python main.py

All the information of the project from data ingestion to prediction is saved in the file logs/running_logs.log

DVC has to be implemented and working on it..