A newer version of the Gradio SDK is available:
6.5.1
metadata
title: Pokemon Object Detection (YOLOv8)
emoji: ⚡
colorFrom: pink
colorTo: purple
sdk: gradio
sdk_version: 4.37.2
app_file: app.py
pinned: false
Pokemon object detections
Using yolov8 after training with Google Colab
Dataset
Because of lack of data, there are only 7 classes: pikachu, charmander, bulbasaur, squirtle, eevee, jigglypuff and other.
Requirements
pip install ultralytics
Preprocess Data
The convert.py used to convert .xml label file to .txt yolo label file.
Run resize_image.py to resize image's width to 640.
Train with Colab
Edit name.yaml.
Upload images and labels.
!yolo train model=yolov8n.pt data=/content/name.yaml epochs=50 imgsz=640
Training's Result
last.pt and best.pt in result folder.
Predict
Run predict.py to see result. This is my predict to test.mp4:
