We used jupyter notebooks for development. Jupyter notebooks are stored in jupyter_ipynb. The outputs have been exported to jupyter_outputs. The python code has been exported to python_src.
| Rating Predictor | R2 Score | RMS Error |
|---|---|---|
| Linear Regression | 0.738 | 92.45 |
| XGBoost | 0.827 | 75.11 |
| Random Forest | 0.803 | 80.21 |
| Rank Predictor | Accuracy |
|---|---|
| Logistic Regression | 0.60 |
| XGBoost | 0.57 |
| Random Forest | 0.55 |
- User Analysis
- getting_problemData.py
- User Problem Analysis
- Final Feature
- EDA
- train
The model takes about 30 seconds to train. So you can just run the train.py file to check the accuracy.