https://github.com/jojoee/pokemon-winner-prediction
:hatched_chick: Predict which Pokemon will win the fight
https://github.com/jojoee/pokemon-winner-prediction
machine-learning pokemon prediction visualization
Last synced: 8 months ago
JSON representation
:hatched_chick: Predict which Pokemon will win the fight
- Host: GitHub
- URL: https://github.com/jojoee/pokemon-winner-prediction
- Owner: jojoee
- Created: 2018-05-13T00:20:57.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2018-05-13T04:41:09.000Z (about 8 years ago)
- Last Synced: 2025-04-12T10:56:02.950Z (about 1 year ago)
- Topics: machine-learning, pokemon, prediction, visualization
- Language: Jupyter Notebook
- Homepage: https://docs.google.com/presentation/d/1ZqQjE3Bj_1zfMDuJwTkmZxnqgmWCGQX9F7I3GTWw8ac/edit
- Size: 1.26 MB
- Stars: 8
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Pokemon Winner Prediction
Based on [Pokemon- Weedle's Cave | Kaggle](https://www.kaggle.com/terminus7/pokemon-challenge), [result](https://github.com/jojoee/pokemon-winner-prediction/blob/master/pokemon-winner-prediction.ipynb)
## Getting started
1. Install `Python3`
2. Install dependencies and `jupyter`
## TODO / Future improvement
- [ ] Add `requirements.txt`
- [ ] Fix warnings
- [ ] Check more about model evaluation
- Data preparation
- [ ] Try remove row that contain `NaN` instead of filling it
- [ ] Try other interpolation methods
- [ ] Try to normalize data before create model
- [ ] Try other feature extractions e.g. type-advantage, type1, type2
- [ ] Try remove Pokemon that have low number of fights
- [ ] Using "diff" stat instead of raw stat, check https://www.kaggle.com/vforvince1/visualizing-data-and-predicting-pokemon-fights
- Model
- [ ] Deep learning approach
- [ ] XGBClassifier
- [ ] Others e.g. GradientBoostingClassifier, svm
- [ ] Hyper parameters e.g. distance method, measure method in kNN
- [ ] Demo page
- [ ] Add Pokemon image to demo page
## Ref
### kaggle.com
- https://www.kaggle.com/ndrewgele/visualizing-pok-mon-stats-with-seaborn
- https://www.kaggle.com/algorlabs/got-to-analyse-em-all
- https://www.kaggle.com/athenalog/data-analysis-and-visualization-poke-master
- https://www.kaggle.com/arihantkumarjain/pokemon-analytics
- https://www.kaggle.com/edword/stats-comparison-3d-scatter-plot
- https://www.kaggle.com/casuru/pokemon-analysis
- https://www.kaggle.com/sanghan/first-gen-pokemon-stats
- https://www.kaggle.com/ash316/learn-pandas-with-pokemons
- https://www.kaggle.com/nemo22/data-sciencetutorial-for-beginners
- https://www.kaggle.com/kanncaa1/data-sciencetutorial-for-beginners
- https://www.kaggle.com/rautaki0127/pokemon-data-science-challenge
- https://www.kaggle.com/vedranium/who-s-that-pokemon
- https://www.kaggle.com/rtatman/which-pokemon-win-the-most
- https://www.kaggle.com/kremijowo/learn-machine-learning-for-newbie-in-progress
- https://www.kaggle.com/paulhendi/pokemon-dataset-random-forest-prediction
- https://www.kaggle.com/cookiehunter/battle-outcome-prediction
- https://www.kaggle.com/karenfang/ds-step-by-step-using-pokemon-dataset
- https://www.kaggle.com/vforvince1/visualizing-data-and-predicting-pokemon-fights
- https://www.kaggle.com/hiteshp/r-to-python-tutorial
- https://www.kaggle.com/jonathanbouchet/pokemon-battles
- https://www.kaggle.com/aagundez/pikachu-vs-bulbasaur-matchup, radar chart
- https://www.kaggle.com/gracezhou0912/analysis-on-pokemon-features
- https://www.kaggle.com/strakul5/principal-component-analysis-of-pokemon-data
- https://www.kaggle.com/galbiati/checking-out-kaggle-nb-with-pokemon-dataset
### Others
- https://seaborn.pydata.org/
- https://seaborn.pydata.org/tutorial/distributions.html
- https://seaborn.pydata.org/generated/seaborn.distplot.html
- https://seaborn.pydata.org/generated/seaborn.countplot.html
- https://pokemondb.net/pokedex/all
- https://bulbapedia.bulbagarden.net/wiki/List_of_Pok%C3%A9mon_by_National_Pok%C3%A9dex_number