2.1 Download the given project folder ‘bikeshare_project’ on to your system

2.2 Open it in VS Code, and understand the project structure and code organization

3.1 Open the terminal in VS Code, and create a virtual environment - `python -m venv ./venv`

3.2 Activate the virtual environment, and install the necessary dependencies (for training and testing both) - '.\venv\Scripts\activate && pip install -r requirements\requirements.txt && pip install -r requirements\test_requirements.txt`

3.3 Execute the "train_pipeline.py" script to train the model on bike rental dataset - `python .\bikeshare_model\train_pipeline.py`

3.4 Run the "predict.py" script to generate predictions - `python .\bikeshare_model\predict.py`

3.5 Run the test cases by executing the "pytest" command in the terminal (debug if issue persists) - `pytest`

3.6 Run the "build" command to create distributable files (.tar, .whl, etc) - `py -m pip install --upgrade build && py -m build`

4.1 Copy the wheel file (.whl) generated in previous step and paste it inside “bikeshare_model_api” directory - `cp .\dist\bikeshare_model-0.0.1-py3-none-any.whl .\bikeshare_model_api\`

4.2 Move into the api folder “bikeshare_model_api” and Install fastapi dependencies using “requirements.txt” (.whl file will be used to install the functionalities of the model into the FastAPI project) - `cd bikeshare_model_api && pip install -r requirements.txt`

4.3 From inside the api folder, execute the "app/main.py" script to deploy the model and get server url - `python app/main.py`

4.4 Access the application and test its prediction, then stop the application - `https://localhost:8001`

5.1 Create a Dockerfile to dockerize “bikeshare_model_api” application - `touch Dockerfile`

5.2 Exit from virtual environment, and start Docker

5.3 Using the Dockerfile, create a docker image - `docker build . -t <your-username>/bikeshare-fastapi:<tag>`

5.4 Using the docker image start a new container, and check if the application is running - `docker run -it -d -p 8001:8001 --name=bikeshare_model_api <your-username>/bikeshare-fastapi`

5.5 On successful run, push the docker image to DockerHub - `docker login && docker push <username>/bikeshare-fastapi:<tag>`

6.1 Create a new repository to store files related to this mini-project

6.2 Get the access token from your DockerHub account settings

6.3 Add the docker username and access token to the secrets of this repository

7.1 Clone the remote repository in your system

7.2 Add the downloaded project folder “bikeshare_project” along with your Dockerfile to this cloned repository
[Note: Do not include the below files to the repo:
    ● cache files (__pychache__, etc),
    ● distributable files (.whl, etc), and
    ● trained model (.pkl file)
These need to be generated during the GitHub actions workflow.]

7.3 Finally, push the changes into the remote GitHub repository

8.1 Create a GitHub Actions workflow to automate the steps for model training, testing,
package building, api dockerizing, and docker image pushing

8.2 Add them as different jobs in the workflow

8.3 Add the jobs to run sequentially (debug if issue persists)

8.4 Access the running workflow pipeline

8.5 On successful run, access the pushed docker image on your DockerHub account