Movie Budget Estimation API. WhatMovie is a complete and ready-to-be-used regression project . It includes
-
Data fetching through a REST API
-
Exploratory Data Analysis
-
Machine Learning modeling (sci-kit learn)
-
An API development (Flask) in order to productize it.
This project is a machine learning-based API built with Flask for estimating movie budgets. It utilizes a trained regression model to predict movie budgets based on various features such as popularity, revenue, runtime, vote average, vote count, and genres. The API provides a convenient way to make predictions for new movie data, making it useful for filmmakers, production companies, and movie enthusiasts.
-
Install necessary dependencies using pip install -r requirements.txt.
-
The data is already in the repo. However if you want to fetch more data you can run data/movie_discovery.py , data/movie_info.py and data/flatten_movie_data.py in that order , from the project directory. To fetch your own data you also have to create a data/config.py file to store your credentials like api_key and api_access_token. To obtain the credentials you can visit the TMDB website
-
If you are OK with current data and current model that comes with the repo , you can run the following command from the project directory to start the API.
python app/api.py
- If API gets deployed locally (e.g. , http://localhost:5000) , you can send POST requests to http://localhost:5000/about/predict endpoint. To test it you can run the following command from the project directory. It uses the content of the file test/test_input.json and sends a POST request. You can also consult to test/test_input.json to see the accepted fields by the mode.
python test/test.py
Contributions are welcome! Open an issue or pull request for improvements.
This project is licensed under the GNU General Public License.