Author Rohan Dawar
In this project I have built some machine learning models to predict the price of wine based on zynicide's web scraped data from WineEnthusiast
Kaggle Page where you will find the raw datasets used in my notebook. Credit to zynicide for this web-scraper.
Note: GitHub has an issue with displaying the plotly graph outputs in the Jupyter Notebook, if you cannot see the plotly graphs, please try opening the .ipynb in Google Colab.
- Open
Machine_Learning_Models_Predicting_Wine_Prices_Rohan_Dawar.ipynb (10.5 MB)in a Jupyter Notebook environment (Google Colab recommended). - Loading the Dataframe
- Run the notebook inputing your Kaggle Api Key when prompted
- If you don't have an API key you can manually upload the CSV
winemag-data_first150k.csv (49.78 MB)from the Kaggle Page - Or you can unzip
dfP.zip(19.5 MB zipped, 72.5 MB unzipped) from this repo, and manually upload the CSV.
- The code will perform input engineering, data analysis and regression model fitting. Once the models are created they are saved as joblib files and you can run predictions by calling
mode.predict(inputs).
Note: Hyper-parameter tuning can be RAM intensive. Make sure your runtime is equipped with at least 12GB of RAM.
The software and output models in this project are licensed under the MIT License - see the LICENSE file for details