This case study is based on the Kaggle Competition: https://www.kaggle.com/c/mercari-price-suggestion-challenge
Detailed Blog can be found at: https://medium.com/analytics-vidhya/suggesting-the-price-of-items-for-online-platforms-using-machine-learning-8b2e424f9a48
The code files are:
| Code File | Description |
|---|---|
| Final.ipynb | Function 1: Takes input X, returns prediction Y |
| Final.ipynb | Function 2: Rakes input (X,Y), returns evaluation metric (RMSLE) |
Contents:
| S.No | Section | Jupyter Notebook |
|---|---|---|
| 1. | Problem Description | 1_EDA.ipynb |
| 2. | Exploratory Data Analysis | 1_EDA.ipynb |
| 3. | Data Preprocessing | 2_Preprocessing_and_featurizations.ipynb |
| 4. | Feature Engineering | 2_Preprocessing_and_featurizations.ipynb |
| 5. | Final Data Preparation | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |
| 6. | Model 1: Ridge Regression (TF-IDF Features) | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |
| 7. | Model 2: Lasso Regression (TF-IDF Features) | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |
| 8. | Model 3: XGBoost (TF-IDF Features) | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |
| 9. | Model 4: XGBoost (Word2Vec Features) | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |
| 10. | Model 5: MLP Model-1 and Model-2 | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |
| 11. | Final Model: Ensemble of MLP-1 and MLP-2 | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |
| 12. | Final Summary | 3_Machine_Learning_and_Deep_Learning_Models.ipynb |