nielsr HF staff commited on
Commit
b8a7a4f
1 Parent(s): 40df538

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +1 -3
README.md CHANGED
@@ -48,15 +48,13 @@ For code examples, we refer to the [documentation](https://huggingface.co/transf
48
 
49
  This model was trained on [Kinetics-400](https://www.deepmind.com/open-source/kinetics).
50
 
51
- ## Training procedure
52
-
53
  ### Preprocessing
54
 
55
  The exact details of preprocessing during training can be found [here](https://github.com/microsoft/VideoX/blob/40f6d177e0a057a50ac69ac1de6b5938fd268601/X-CLIP/datasets/build.py#L247).
56
 
57
  The exact details of preprocessing during validation can be found [here](https://github.com/microsoft/VideoX/blob/40f6d177e0a057a50ac69ac1de6b5938fd268601/X-CLIP/datasets/build.py#L285).
58
 
59
- One resizes the shorter edge of each frame, after which center cropping is performed to a fixed-size resolution (like 224x224). Next, frames are normalized across the RGB channels with the ImageNet mean and standard deviation.
60
 
61
  ## Evaluation results
62
 
 
48
 
49
  This model was trained on [Kinetics-400](https://www.deepmind.com/open-source/kinetics).
50
 
 
 
51
  ### Preprocessing
52
 
53
  The exact details of preprocessing during training can be found [here](https://github.com/microsoft/VideoX/blob/40f6d177e0a057a50ac69ac1de6b5938fd268601/X-CLIP/datasets/build.py#L247).
54
 
55
  The exact details of preprocessing during validation can be found [here](https://github.com/microsoft/VideoX/blob/40f6d177e0a057a50ac69ac1de6b5938fd268601/X-CLIP/datasets/build.py#L285).
56
 
57
+ During validation, one resizes the shorter edge of each frame, after which center cropping is performed to a fixed-size resolution (like 224x224). Next, frames are normalized across the RGB channels with the ImageNet mean and standard deviation.
58
 
59
  ## Evaluation results
60