Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/echometerain/whos-that-pokemon
Gen-1 pokemon classifier
https://github.com/echometerain/whos-that-pokemon
Last synced: 2 days ago
JSON representation
Gen-1 pokemon classifier
- Host: GitHub
- URL: https://github.com/echometerain/whos-that-pokemon
- Owner: echometerain
- License: gpl-2.0
- Created: 2023-07-25T14:44:11.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2023-11-01T00:51:39.000Z (about 1 year ago)
- Last Synced: 2023-11-01T01:32:02.542Z (about 1 year ago)
- Language: Jupyter Notebook
- Size: 2.66 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
---
title: Who's That Pokemon?
emoji: ⚡
colorFrom: red
colorTo: blue
sdk: gradio
sdk_version: 3.38.0
app_file: app.py
pinned: false
---# Who's That Pokemon?
[Kaggle link](https://www.kaggle.com/code/echometerhhwl/who-s-that-pokemon-improved) | [Huggingface demo](https://huggingface.co/spaces/echometerain/whos-that-pokemon)
Identifies generation 1 pokemon based on [35627 images](https://www.kaggle.com/datasets/echometerhhwl/pokemon-gen-1-38914)!
![image](https://github.com/echometerain/whos-that-pokemon/assets/70437021/68b0ed63-4be6-4d30-a06f-a8c1b59060a3)
![image](https://github.com/echometerain/whos-that-pokemon/assets/70437021/e46821ee-26ce-4976-8545-da8786fdc9c0)
Default model based on convnext_tiny (`model.pkl`) has 95.7% accuracy (with dataset augmented at runtime)
![image](https://github.com/echometerain/whos-that-pokemon/assets/70437021/b3006352-ad07-4227-9da3-aa9c182d3303)
Alternative model based on resnet34 (`model-r34.pkl`) has 94.3% accuracy (with dataset augmented at runtime)
![image](https://github.com/echometerain/whos-that-pokemon/assets/70437021/5e41262e-86b9-4dde-a48a-56951ea25644)