https://github.com/dinarior/resnet_embeddings_extractor
Small and easily modifiable notebook to extract embeddings from pre trained resnet50
https://github.com/dinarior/resnet_embeddings_extractor
cifar cifar10 cifar100 embeddings imagenet pytorch resnet resnet-50 unsupervised
Last synced: 3 months ago
JSON representation
Small and easily modifiable notebook to extract embeddings from pre trained resnet50
- Host: GitHub
- URL: https://github.com/dinarior/resnet_embeddings_extractor
- Owner: dinarior
- Created: 2020-08-19T08:55:04.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2020-08-24T05:36:43.000Z (about 5 years ago)
- Last Synced: 2024-10-20T05:12:55.479Z (about 1 year ago)
- Topics: cifar, cifar10, cifar100, embeddings, imagenet, pytorch, resnet, resnet-50, unsupervised
- Language: Jupyter Notebook
- Homepage:
- Size: 22.5 KB
- Stars: 5
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Resnet Embedding Extractor
Small and easily modifiable [notebook](https://nbviewer.jupyter.org/github/dinarior/resnet_embeddings_extractor/blob/master/extract_embeddings.ipynb) to extract embeddings from pre trained resnet50.
The notebook extract embeddings for `CIFAR100`, using a pretrained resnet50.
For the pretrained model we use https://github.com/facebookresearch/swav , thus the embeddings are learned unsupervised on imagenet. You can easily swap it for other models of your choosing.
If you do use `SWAV`, make sure you [cite](https://arxiv.org/pdf/2006.09882.pdf) them properly.