Skip to content

Fix code sample in preprocess #20561

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
stevhliu merged 2 commits into from
Dec 5, 2022
Merged

Fix code sample in preprocess #20561

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ad6248c
change to image_processor
stevhliu Dec 2, 2022
8b912e4
apply review
stevhliu Dec 5, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
35 changes: 5 additions & 30 deletions docs/source/en/preprocessing.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -361,11 +361,11 @@ For computer vision tasks, it is common to add some type of data augmentation to
```py
>>> from torchvision.transforms import Compose, Normalize, RandomResizedCrop, ColorJitter, ToTensor

>>> normalize = Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std)
>>> normalize = Normalize(mean=image_processor.image_mean, std=image_processor.image_std)
>>> size = (
... feature_extractor.size["shortest_edge"]
... if "shortest_edge" in feature_extractor.size
... else (feature_extractor.size["height"], feature_extractor.size["width"])
... image_processor.size["shortest_edge"]
... if "shortest_edge" in image_processor.size
... else (image_processor.size["height"], image_processor.size["width"])
... )
>>> _transforms = Compose([RandomResizedCrop(size), ColorJitter(brightness=0.5, hue=0.5), ToTensor(), normalize])
```
Expand All @@ -387,32 +387,7 @@ For computer vision tasks, it is common to add some type of data augmentation to
4. Now when you access the image, you'll notice the image processor has added `pixel_values`. You can pass your processed dataset to the model now!

```py
>>> dataset[0]["image"]
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=384x512 at 0x7F1A7B0630D0>,
'label': 6,
'pixel_values': tensor([[[ 0.0353, 0.0745, 0.1216, ..., -0.9922, -0.9922, -0.9922],
[-0.0196, 0.0667, 0.1294, ..., -0.9765, -0.9843, -0.9922],
[ 0.0196, 0.0824, 0.1137, ..., -0.9765, -0.9686, -0.8667],
...,
[ 0.0275, 0.0745, 0.0510, ..., -0.1137, -0.1216, -0.0824],
[ 0.0667, 0.0824, 0.0667, ..., -0.0588, -0.0745, -0.0980],
[ 0.0353, 0.0353, 0.0431, ..., -0.0039, -0.0039, -0.0588]],

[[ 0.2078, 0.2471, 0.2863, ..., -0.9451, -0.9373, -0.9451],
[ 0.1608, 0.2471, 0.3098, ..., -0.9373, -0.9451, -0.9373],
[ 0.2078, 0.2706, 0.3020, ..., -0.9608, -0.9373, -0.8275],
...,
[-0.0353, 0.0118, -0.0039, ..., -0.2392, -0.2471, -0.2078],
[ 0.0196, 0.0353, 0.0196, ..., -0.1843, -0.2000, -0.2235],
[-0.0118, -0.0039, -0.0039, ..., -0.0980, -0.0980, -0.1529]],

[[ 0.3961, 0.4431, 0.4980, ..., -0.9216, -0.9137, -0.9216],
[ 0.3569, 0.4510, 0.5216, ..., -0.9059, -0.9137, -0.9137],
[ 0.4118, 0.4745, 0.5216, ..., -0.9137, -0.8902, -0.7804],
...,
[-0.2314, -0.1922, -0.2078, ..., -0.4196, -0.4275, -0.3882],
[-0.1843, -0.1686, -0.2000, ..., -0.3647, -0.3804, -0.4039],
[-0.1922, -0.1922, -0.1922, ..., -0.2941, -0.2863, -0.3412]]])}
>>> dataset[0].keys()
```

Here is what the image looks like after the transforms are applied. The image has been randomly cropped and it's color properties are different.
Expand Down