-
Notifications
You must be signed in to change notification settings - Fork 391
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Expected performance #7
Comments
|
Fantastic -- thank you. I'm running the ~ Ben |
Thank you for dedicating the time and resource to run and verify our code. We also looked at the architectures that were sampled in the time steps before and took the one with the overall best val_acc. However, I think the one you picked might work well too 😄 |
@bkj ,Hello.Is everything going well with this work(the macro search space, the micro search space)? |
@bkj Hello, Did you change any parameters on the scripts to get this result. I only could get almost 0.88 acc on cifar using micro structure. |
@bkj Hi,I saw you you also are interested in the ENAS-pytorch work in https ://github.com/carpedm20/ENAS-pytorch. When I run the ENAS-pytorch code by : python main.py --network_type cnn --dataset cifar --controller_optim momentum --controller_lr_cosine=True --controller_lr_max 0.05 --controller_lr_min 0.0001 --entropy_coeff 0.1,I met many errors.What I want to do is to find some cnn architectures and make them visulized. Would you please tell me what changes Ishould do to the code before I run it? Thanks for your reply. |
Hi @hyhieu , I have run the latest default ./scripts/ptb_final.sh for 600+ epochs, but the ppl remains 69+. May I know what is the expected number of epochs needed to reproduce claimed 55.8 ppl? Or do I need to change ./scripts/ptb_final.sh configuration? |
@axiniu I met the same problem,Has the problem been solved now?Thanks a lot |
@upwindflys @axiniu
But did you know the difference between 12 and 24 cell? |
If we run the three experiments from the README:
what should we expect the final performance metrics to be? Are you able to post the expected results either here or in the README?
Thanks
The text was updated successfully, but these errors were encountered: