Machine_Learning_Project
Creating conda environment
conda create -p mlproj python==3.7 -y
conda activate mlproj/
OR
conda activate mlproj
pip install -r requirements.txt
python app.py
To Add files to git
git add .
OR
git add <file_name>
Note: To ignore file or folder from git we can write name of file/folder in .gitignore file
To check the git status
git status
To check all version maintained by git
git log
To create version/commit all changes by git
git commit -m "message"
To send version/changes to github
git push origin main
To check remote url
git remote -v
To setup CI/CD pipeline in heroku we need 3 information
- HEROKU_EMAIL = [email protected]
- HEROKU_API_KEY = <>
- HEROKU_APP_NAME = mlproj-regression-app
BUILD DOCKER IMAGE
docker build -t <image_name>:<tagname> .
Note: Image name for docker must be lowercase
To list docker image
docker images
Run docker image
docker run -p 5000:5000 -e PORT=5000 a13754275069 <IMAGE ID>
To check running container in docker
docker ps
Tos stop docker conatiner
docker stop <container_id>
python setup.py install
python setup.py install
Install ipykernel
pip install ipykernel
Data Drift https://evidentlyai.com/
When dataset stats get change, it is called as data drift.
pip install evidently
Model Trainer
- loading transformed training and testing datset
- reading model config file
- getting best model on training datset
- evaluation models on both training & testing datset --> model object
- loading preprocessing pbject
- custom model object by combining both preprocessing obj and model obj
- saving custom model object
- return model_trainer_artifact