Add: TensorFlow example for semantic segmentation task guide #21223
Conversation
|
The documentation is not available anymore as the PR was closed or merged. |
amyeroberts
left a comment
amyeroberts
left a comment
There was a problem hiding this comment.
Nice - LGTM! Thanks for adding ❤️
|
|
||
| >>> def transforms(image): | ||
| ... image = tf.keras.utils.img_to_array(image) | ||
| ... image = tf.transpose(image, (2, 0, 1)) |
There was a problem hiding this comment.
Just as a note (not a request for a change), you can also this in the image processor with:
image_processor(images, labels, data_format="channels_first")
There was a problem hiding this comment.
Thanks, @amyeroberts ! I didn't know I could do this.
stevhliu
left a comment
stevhliu
left a comment
There was a problem hiding this comment.
Whoo, great job! Thanks for the addition 🔥
Co-authored-by: Steven Liu <[email protected]>
Co-authored-by: Steven Liu <[email protected]>
sgugger
left a comment
sgugger
left a comment
There was a problem hiding this comment.
Thanks for adding this!
| Pass your `compute_metrics` function to [KerasMetricCallback](./main_classes/keras_callbacks#transformers.KerasMetricCallback), | ||
| and use the [PushToHubCallback](./main_classes/keras_callbacks#transformers.PushToHubCallback) to upload the model: |
There was a problem hiding this comment.
Small nit, but linking to the doc with [`KerasMetricCallback`] and [`PushToHubCallback`] would make the links more resilient (in case we move those pages in the future).
There was a problem hiding this comment.
Thank you, fixed :)
5 participants
This PR adds a TensorFlow example to the existing Semantic Segmentation task guide using the same dataset and fine-tuning steps.
This example supplements the existing guide and can be helpful to those who choose TensorFlow over PyTorch and would like to use Transformers for semantic segmentation.