Cyril Equilbec

Same error, can't manage to fit a vovnet-lite-dw or a vovnet-19-dw, keep getting NaN loss. Vovnet-lite is fine tho, I have the feeling that there is something wrong with the...

> Yes, I'll add the noisy students version soon. I'm a bit swamped with work at the moment, but I'll convert the weights, run the tests, and make them available...

> I tried porting it too. Most of the errors I got were patched using tf.compat.v1 wherever it complained—except tf.train.Saver(), which apparently doesn't exist. > > EDIT: was able to...

> I've tested it and it works great. I've also converted b4 if anyone is interested. Cool, thank's for the feedback, will give it a try unless @lukemelas releases an...

The weights for the backbones are available now, just need to update a bit the code, see https://github.com/toandaominh1997/EfficientDet.Pytorch/pull/111

> Yes unfortunately but it can be used to correct the numerous problems with all the public PyTorch implementations. Like for example, the original tensorflow code has three convolution layers...

> > > FYI, I decided to get a version of EfficientDet running now that an official reference is available. I saw this issue searching through the pre-existing PyTorch impls...

Need to update the code base according to : https://github.com/toandaominh1997/EfficientDet.Pytorch/pull/111

+1

Can be fixed when adding : ``` physical_devices = tf.config.list_physical_devices('GPU') tf.config.experimental.set_memory_growth(physical_devices[0], True) model.distribute_strategy = tf.distribute.MirroredStrategy() ```