syedshahlal/End_to_end_ML_Project

End_to_end_ML_Project

1. Software and account Requirements:

1. Github Account
2. Azure Account
3. VS Code IDE
4. GIT cli
5. GIT Documentation

2. Creating conda virtual environment

conda create -p venv python==3.11 -y

conda activate venv/

Create a requirements.txt file add all the libraries required and install them

pip install -r requirements.txt

Now, we create the flask app file as app.py and write our code in it, and then run the app

python app.py

3. Github connections for push and pull

To add the file in the git repo

git add <filename>

or

git add .

Note: If there are some files in our folder that we do not want to track or do versioning then we add them to the .gitignore file

To check the git status

git status

To check all the versions maintained by git

git log

To create a new version/commit all the changes by git

git commit -m "your comment/message"

To send version/changes to the GitHub

git push origin main

To check remote URl from where the files will be pushed or pulled

git remote -v

4. Setting up CI/CD Pipelines

To setup CI/CD pipeline in Heroku we need 3 information:

1. HEROKU_EMAIL = sr.shahlal@gmail.com
2. HEROKU_API_Key = <>
3. HEROKU_APP_Name = ml-ete-app

BUILD DOCKER IMAGE

docker build -t <image_name>:<tagname> .

Note: The image name for docker must be lowercase

To list the docker image

docker images

Run docker image

docker run -p 5000:5000 -e PORT=5000 b3bb833287a8

To check the running container in the docker

docker ps

To stop the docker container

docker stop <container_id>

5. Building CI/CD Pipelines in Heroku APP

1. Create a .github folder and inside the folder add another folder workflows in this workflows folder create a file main.yaml
2. main.yaml file write the GitHub action flow
3. push the change to the git repo
4. open the repo in GitHub and go to setting/secrets/actions and create New repository secret for all the secrets that have been mentioned in the main.yaml file and save them

   Note: That the Name* and Secret* is same as mentioned

5. once this is done our deployment will take place automatically which we can see in the actions tab in the GitHub repo

Now whenever there is any push to the repo it will automatically trigger the deployment in the heroku web app

6. Files and Folder creation for structuring the project

Moving ahead we will add a setup.py file and a folder housing that we'll contain a file inside it which has to be created as __init__.py

Now we will write the setup coed in our setup.py file and run it

python setup.py install

Note: this will generate build , dist and housing_predictor.egg-info based on what values we have passed in the setup.py

I have created a create_folders.py file which has a list of folders we want to create once we run that file it will automatically create all the folders that we want to create under the housing folder.