-
Notifications
You must be signed in to change notification settings - Fork 1.2k
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
EfficientNet implementation #796
Comments
Just for some background: we're mostly interested so no one duplicates effort. |
here's one candidate @sufeidechabei Should we split the risk by training the network simultanously? |
Note that https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/main.py has the training script that seems to have been used and provides hyperparameters. |
Also there might be some differences with the paper, author says "source code is correct": see tensorflow/tpu#383, tensorflow/tpu#390 |
They never said "source code is correct" about tpu issue tensorflow/tpu#390. Did they? Seems like a much larger discrepancy than the padding issues in tensorflow/tpu#383 |
True, what I meant is that since the author said source code is correct on tensorflow/tpu#383 I was assuming the code is what they actually used, including concerning tensorflow/tpu#390. But you are right that this resolution discrepancy is a big difference 🤔 |
Regarding the training of efficientnet, see the remarks of Ross Wightman here: https://forums.fast.ai/t/efficientnet/46978/67 ; it might be that keeping an exponential moving average of the weights during training, for use at testing, helps this family of models a lot. |
Similar discussion over here: pytorch/vision#980 |
@sufeidechabei please check your PR with the resources above. |
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
Hey, are there any EfficientNet (https://arxiv.org/abs/1905.11946) implementations available? If not, is somebody working on it?
The text was updated successfully, but these errors were encountered: