Siva Sravana Kumar Neeli

> Can you take a look at the test failures? I think they are likely related to the fact that torch tests are running with `data_format="channels_first"`. https://github.com/keras-team/keras/actions/runs/11469236802/job/31916066253?pr=20368 Oh yes!! thanks...

> The code looks good to me! Not sure about the max bbox layer. > > A side question is whether we need "bbox" in the name of the transformations...

Can we make last dimension of the [`roi_generator.py`](https://github.com/keras-team/keras-cv/blob/5cfe7c3d8f2d758b91bbfcedf7f0a29931236c68/keras_cv/src/layers/object_detection/roi_generator.py#L190) boxes output have fixed 4 as dimension, any suggestion on how to make it possible ? Ref: 

@VarunS1997 Review for any corrections are fixes required from based code. Thanks!

> Hey @sineeli this is so cool! Thanks for taking this up. It was a long time ago and unfortunately I would not be able to give you a thorough...

There are two implementations as per torch: `fasterrcnn_resnet50_fpn` - Present Keras implementation similar to this config `fasterrcnn_resnet50_fpn_v2` - Improved Configuration. https://github.com/pytorch/vision/blob/8f73afacb81df1fcc6e2d308c5b734b126db3426/torchvision/models/detection/faster_rcnn.py#L470

cc: @mattdangerw

@mattdangerw, sorry for the delayed response I didn't see the comment. As per official [repo](https://github.com/lm-sys/FastChat) I don't see any specific prompt.

Sure @abheesht17 First preprocessing and tokenizer these parts I think are good for reviewing, as they are the primary steps. 1. `keras_hub/src/models/parseq/parseq_tokenizer.py` 2. `keras_hub/src/models/text_recognition_preprocessor.py`

@sachinprasadhs, @abheesht17, @mattdangerw Can you take a look at the PR when you get some time, thank you!