App link: https://flightpriceapiapp.herokuapp.com/
- This is an end-to-end project for predicting the flight ticket price.
- After finding the best model, we connected the model with the help of flask framework.
- We wrote app.py, home.html and style.css with the help of Visual Studio Code.
- Then, we integrated app.py file with the help of templates (home.html) and static/css (style.css) for creating the webpage and hosting the @app.
- As the pkl file was 100 mb, we cloned the repository and then uploaded using Github desktop.
- Then, we hosted the api and deployed the app by connecting GitHub repository with heroku.
- The major problem was that the pickle file was throwing errors while deploying our app and checking the build logs.
- In order to solve this, we need git lfs and it reduces the impact of large files in your repository by downloading the relevant versions of them lazily.
- For running the api in heroku with help of git-lfs, check this link: https://medium.com/analytics-vidhya/deploying-nlp-model-on-heroku-using-flask-nltk-and-git-lfs-eed7d1b22b11
- For uploading the files to our repository, either using git version control(Git bash) or terminal in VS code or cmd prompt or Github desktop.