vikrantmohite19/Data_Mining_Water_Table-Python-Package

This is a complete python package created for the task of predicting the status of water wells to enable timely maintenance of wells to ensure smooth water supply to people of Tanzania. The package has various modules for tasks such as data ingestion, data transformation, model training, logging, exception handling, prediction and a flask web app. Web App for project "Data Mining the Water Table" Steps to create and build package: 1. create folder where we store our app data 2. open terminal in the folder and open vscode using 'code .' 3. create new vitual environment using command 'conda create -p venv python==3.9 -y' 4. create README.md file 5. create repository on github 6. use following commands to setup the repository git init git add README.md git commit -m "first commit" git branch -M main git remote add origin github.com git push -u origin main 7. create setup.py and requirements.txt 8. create source folder as 'DMWT_Package' and a file inside it as '__init__.py' to consider 'DMWT_Package' as a package. 9. run command 'python -m pip install -r requirements.txt'. (ugrade pip if required using 'python -m pip install --upgrade pip'). this stage will generate a folder 'DMWT_Package.egg-info' 10. created exception.py and logger.py 11. Created new folder 'notebook' and copied all ipython files into the folder. create data folder inside 'notebook' and copy all raw data files. 13. Inside the folder 'DMWT_Package' created exception.py, logger.py, utils.py 14. Inside 'DMWT_Package', creat new folder 'pipeline'. Inside 'pipeline created __init__.py, predict_pipeline.py & train_pipeline.py. 15. Inside 'DMWT_Package', create new folder 'components'. Inside 'components' created __init__.py, data_ingestion.py, data_transformation.py, model_trainer.py. 16. Run Data_ingestion.py which inturn will create folders such as artifacts, preprocessed, logs. In this stage we will get preprocessed data and the best model after performing grid-search 17. Run command, 'python setup.py sdist bdist_wheel'. In this stage two folders such as 'build' and 'dist' will be created. 18. Now go to folder 'dist' and run 'pip install DMWT_Package-0.0.1-py3-none-any.whl. this is to test if package can be installed & binary distribution file run properly. 19. Now run python, run …