You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
* Add experimental parameter `mask_loss_outliers` to mask high loss values considered as outliers (requires the `tensorflow-probability` module)
23
+
24
+
### Fixes and improvements
25
+
26
+
* Fix TensorFlow Lite conversion for models using a `PositionEmbedder` layer
27
+
* Automatically pad the weights of linear layers to enable Tensor Cores in mixed precision training
28
+
* Correctly set the CTranslate2 options `alignment_layer` and `alignment_heads` when converting models using the attention reduction `AVERAGE_LAST_LAYER`
29
+
* Raise an error if a training dataset or annotation file has an unexpected size
30
+
* Warn about duplicated tokens when loading vocabularies
0 commit comments