Skip to content

Conversation

@dribnet
Copy link

@dribnet dribnet commented Nov 17, 2018

The finetune option is used to freeze features layers on a pretrained network

refactored and updated code from #58

@soumith
Copy link
Member

soumith commented Nov 18, 2018

instead of making main.py a one-stop shop for everything, I feel like a finetune.py would be a much better idea here. What do you think?

@msaroufim
Copy link
Member

@dribnet I'm happy to merge this but it probably needs a test. Take a look at this for some inspiration https://github.com/pytorch/examples/blob/main/run_python_examples.sh

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

4 participants