https://github.com/zslucky/mercedez_competition

Mercedez kaggle competition
https://github.com/zslucky/mercedez_competition

Last synced: about 6 hours ago
JSON representation

Mercedez kaggle competition

• Host: GitHub
• URL: https://github.com/zslucky/mercedez_competition
• Owner: zslucky
• Created: 2017-07-11T02:02:42.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-07-15T02:17:19.000Z (over 7 years ago)
• Last Synced: 2024-11-10T04:38:05.772Z (about 2 months ago)
• Language: Jupyter Notebook
• Size: 537 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Mercedez competition

My 2nd competition on kaggle

196st of 3835,  bronze medal

---

## Files

### Jupyter note book

* `Object columns Analysis Book.ipynb`: data analysis for object columns `X0, X1, X2, X3, X4, X5, X6, X8`.

### Main model files

* `adaboost_regressor_model.py`: stacked feature model by using adaboost regressor.

* `extra_tree_model.py`: stacked feature model by using adaboost regressor.

* `random_forest_regressor_model.py`: stacked feature model by using adaboost regressor.

* `gradient_boost_model.py`: stacked feature model by using adaboost regressor.

* `stacking_models.py`: using XGBoost to stack all model features, and combine with the other submission.

### Tool files

* `all_questions.json`: this is the probed test data from others guys, just used for test.

* `test_y_value.csv`: this is a csv file transformed from `all_question.json`.

* `test_data_update.py`: this script used to transform the json file to csv file.

* `test_data_compared.py`: this script used to compared every result csv to probed value, also show the r2 score.

### Online model

* `origin-script`: This script comes from [Hakeem's script](https://www.kaggle.com/hakeem/stacked-then-averaged-models-0-5697)