Skip to content

docs: explain better docstring of set_image_normalization #585

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
JoanFM merged 6 commits into from
Oct 4, 2022
Merged

docs: explain better docstring of set_image_normalization #585

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
70d1ec7
docs: explain better docstring
Oct 3, 2022
e9bdeb3
fix: update docarray/document/mixins/image.py
Oct 4, 2022
6758fff
fix: apply suggestions from code review
Oct 4, 2022
babc925
fix: apply suggestions from code review
Oct 4, 2022
2ff3041
Merge remote-tracking branch 'origin/docs-513' into docs-513
JohannesMessner Oct 4, 2022
556ef6b
docs: fix docstring formatting
JohannesMessner Oct 4, 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
12 changes: 5 additions & 7 deletions docarray/document/mixins/image.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -229,15 +229,13 @@ def set_image_tensor_normalization(
) -> 'T':
"""Normalize a uint8 image :attr:`.tensor` into a float32 image :attr:`.tensor` inplace.

Following Pytorch standard, the image must be in the shape of shape (3 x H x W) and
will be normalized in to a range of [0, 1] and then
normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225]. These two arrays are computed
based on millions of images. If you want to train from scratch on your own dataset, you can calculate the new
mean and std. Otherwise, using the Imagenet pretrianed model with its own mean and std is recommended.
Applies normalization to the color channels of the images.
By default, the normalization uses mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225], which are standard values computed on millions of images. If you want to train from scratch on your own dataset, you can calculate the new
mean and std. Otherwise, using the Imagenet pretrained model with its own mean and std is recommended.

:param channel_axis: the axis id of the color channel, ``-1`` indicates the color channel info at the last axis
:param img_mean: the mean of all images
:param img_std: the standard deviation of all images
:param img_mean: the means of all images: [mean_r, mean_g, mean_b]
:param img_std: the standard deviations of all images: [std_r, std_g, std_b]
:return: itself after processed

.. warning::
Expand Down