Skip to content

GitLab

Switch branch/tag
  1. 20 Dec, 2022 5 commits
  3. 19 Dec, 2022 6 commits
  5. 16 Dec, 2022 6 commits
    • Yih-Dar's avatar
      Install `sentencepiece` in `DeepSpeed` CI image (#20795) · 7032e020
      Yih-Dar authored 
      

* Install sentencepiece in DS CI image

* update
Co-authored-by: default avatarydshieh <ydshieh@users.noreply.github.com>
      7032e020
    • NielsRogge's avatar
      Add Swin2SR (#19784) · 26dd041c
      NielsRogge authored 
      

* First draft

* Add more improvements

* Improve forward pass

* Fix layernorm

* Add upscaler

* More improvements

* More improvements

* More improvements

* Improve conversion script

* Add preprocessing

* Make output match original implementation

* Add additional attributes

* Add support for more models

* Support more models

* Add support for real world sr

* Add initial Swin2SRFeatureExtractor

* Add ImageSuperResolutionOutput

* Make more tests pass

* Use BaseModelOutput

* Fix one more test

* Fix more tests

* Fix another test

* Fix all tests

* Rename to Swin2SRImageProcessor

* Fix toctree

* Fix toctree

* Fix rebase

* Improve Swin2SRImageProcessor

* Remove feature extractor file

* Improve model

* Improve conversion script

* Fix integration test

* Fix init

* Fix conversion script

* Address comments

* Improve upsampler

* Add NearestConvUpsampler

* Improve pixel shuffle upsampler

* Improve auxiliary upsampler

* Improve conversion script

* Rename conv_last to final_convolution

* Fix rebase

* Improve upsample module

* Add padding to image processor

* Fix bug

* Update padding

* Remove print statement and fix integration test

* Improve docs

* Add image processor tests

* Convert all checkpoints, fix testsé

* Remove print statements

* Fix import
Co-authored-by: default avatarNiels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
      26dd041c
    • NielsRogge's avatar
      Add Universal Segmentation class + mapping (#20766) · 7f998612
      NielsRogge authored 
      

* Add mapping

* Add mapping to pipeline

* Apply suggestions

* Fix feature extractor tests

* Use ForInstance, add model to universal mapping

* More fixes

* Remove model from deprecated objectsé
Co-authored-by: default avatarNiels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
      7f998612
    • Matt's avatar
      Stop calling expand_1d on newer TF versions (#20786) · e65445b4
      Matt authored
      e65445b4
    • Nicolas Patry's avatar
      Fix object detection2 (#20798) · 3ee95820
      Nicolas Patry authored 
      * Revert "Fixing object detection with `layoutlm` (#20776)"

This reverts commit fca66abe.

* Better fix for layoutlm object detection.

* Style.
      3ee95820
    • Younes Belkada's avatar
      [Pipeline] skip feature extraction test if in `IMAGE_PROCESSOR_MAPPING` (#20790) · 4341f4e2
      Younes Belkada authored 
      skip feature extraction test if in `IMAGE_PROCESSOR_MAPPING`
      4341f4e2
  7. 15 Dec, 2022 9 commits
  9. 14 Dec, 2022 8 commits
  11. 13 Dec, 2022 6 commits
    • Yih-Dar's avatar
      Uninstall `torch_tensorrt` in `DeepSpeed` CI image for now (#20758) · d994473b
      Yih-Dar authored 
      

Uninstall torch_tensorrt for now
Co-authored-by: default avatarydshieh <ydshieh@users.noreply.github.com>
      d994473b
    • Nicolas Patry's avatar
      Fixing the pipeline tutorial test (#20746) · ba9da49a
      Nicolas Patry authored 
      
Co-authored-by: default avatarydshieh <ydshieh@users.noreply.github.com>
      ba9da49a
    • Hazrul Akmal's avatar
      Add docs xlm roberta (#20742) · f28c918c
      Hazrul Akmal authored 
      * added model resources for xlm-roberta

* added model resources for xlm-roberta

* resolve suggested changes

* add resources to xlm-roberta
      f28c918c
    • NielsRogge's avatar
      [NAT, DiNAT] Add backbone class (#20654) · 6ef42587
      NielsRogge authored 
      

* Add first draft

* Add out_features attribute to config

* Add corresponding test

* Add Dinat backbone

* Add BackboneMixin

* Add Backbone mixin, improve tests

* Fix embeddings

* Fix bug

* Improve backbones

* Fix Nat backbone tests

* Fix Dinat backbone tests

* Apply suggestions
Co-authored-by: default avatarNiels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
      6ef42587
    • dhansmair's avatar
      in the resize() function in image_transforms.py, the line 267: (#20728) · 30d8919a
      dhansmair authored 
      `image = to_channel_dimension_format(image, ChannelDimension.LAST)`
is redundant as this same conversion is also applied in to_pil_image().

This redundant call actually makes the training fail in rare cases.
The problem can be reproduced with the following code snippet:
```
from transformers.models.clip import CLIPFeatureExtractor
vision_processor = CLIPFeatureExtractor.from_pretrained('openai/clip-vit-large-patch14')
images = [
    torch.rand(size=(3, 2, 10), dtype=torch.float),
    torch.rand(size=(3, 10, 1), dtype=torch.float),
    torch.rand(size=(3, 1, 10), dtype=torch.float)
]
for image in images:
    processed_image = vision_processor(images=image, return_tensors="pt")['pixel_values']
    print(processed_image.shape)
    assert processed_image.shape == torch.Size([1, 3, 224, 224])
```

The last image has a height of 1 pixel.
The second call to to_channel_dimesion_format() will transpose the image, and the height
dimension is wrongly treated as the channels dimension afterwards.
Because of this, the following normalize() step will result in an
exception.
      30d8919a
    • Matt's avatar
      Fix AdamWeightDecay for TF 2.11 (#20735) · 4f1788b3
      Matt authored 
      * Fix AdamWeightDecay for TF

* Fix AdamWeightDecay for TF

* make fixup
      4f1788b3