James
James
Thanks for this great project! In `temporal_train_data.py` we have `img2 = np.swapaxes(img2, 0, 1)` for reasons which aren't clear to me. This seems to result in rotation of the y...
As above - this pretrained model's weights do not work. This seems to be an isolated problem for this encoder.
When creating a model using `timm.create_model(arch, pretrained=True, in_chans=1, num_classes=1)`, single-channel input images can be used with `tf_efficientnet_b2_ns`, but not `efficientnet_b3_pruned`. The pruned models result in the following error: ``` File...
### Bug description Hi, I'm trying to use the official PyTorch EMA (new feature: https://pytorch.org/docs/main/optim.html#weight-averaging-swa-and-ema) in Lightning in DDP. To do this I've done: ```python self.ema_model = torch.optim.swa_utils.AveragedModel(self.model, multi_avg_fn =...
Having all my team build this sounds painful.