https://github.com/beenotung/transfer-learning-demo
A generic demo to build custom image classifier of arbitrary classes using transfer learning
https://github.com/beenotung/transfer-learning-demo
image-classification nodejs python resnet tensorflow transfer-learning typescript
Last synced: 3 months ago
JSON representation
A generic demo to build custom image classifier of arbitrary classes using transfer learning
- Host: GitHub
- URL: https://github.com/beenotung/transfer-learning-demo
- Owner: beenotung
- Created: 2024-08-03T23:29:50.000Z (almost 2 years ago)
- Default Branch: master
- Last Pushed: 2024-08-11T08:27:12.000Z (almost 2 years ago)
- Last Synced: 2025-06-02T07:40:38.101Z (about 1 year ago)
- Topics: image-classification, nodejs, python, resnet, tensorflow, transfer-learning, typescript
- Language: Jupyter Notebook
- Homepage:
- Size: 16.6 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Transfer Learning Demo with ResNet50
A generic demo to build custom image classifier of arbitrary classes.
## Key Steps
1. Training: run [train.py](train.py) or [train.ipynb](train.ipynb)
2. Testing: run [test.py](test.py) or [test.ipynb](test.ipynb)
3. Start Server: run `sanic server`
4. Call from HTTP: e.g. `http://127.0.0.1:8000/classify?img_path=test/748dc4e1b7ba92fa29a5dafef579fb2a.png`
## Alternative Base Models
There are multiple alternative base models with smaller file size and less parameters.
Details see [compare.md](./compare.md)