pytorch-image-models/timm
SmilingWolf 59cb0be595 SwinV2: add configurable act_layer argument
Defaults to "gelu", but makes it possible to pass "gelu_tanh".
Makes it easier to port weights from JAX/Flax, where the tanh
approximation is the default.
2024-03-05 22:04:17 +01:00
..
data Small tweak of timm ToTensor for clarity 2024-02-10 14:57:40 -08:00
layers Make NormMlpClassifier head reset args consistent with ClassifierHead 2024-02-10 16:25:33 -08:00
loss Add pos_weight and support for summing over classes to BCE impl in train scripts 2023-12-30 12:13:06 -08:00
models SwinV2: add configurable act_layer argument 2024-03-05 22:04:17 +01:00
optim
scheduler Allow training w/o validation split set 2024-01-08 09:38:42 -08:00
utils Fix device idx split 2024-02-10 21:41:14 -08:00
__init__.py
version.py Update version.py 2024-02-19 11:08:17 -08:00