Mohammed Innat

@sachinprasadhs > If you encounter a KerasCV feature that is missing from KerasHub, or would like to propose an addition to the library, please file [an issue with KerasHub](https://github.com/keras-team/keras-hub/issues). Firstly,...

@sampathweb You're compiling segformer model with `BinaryCrossentropy()` but the last layer activations is hard coded to [softmax](https://github.com/keras-team/keras-cv/blob/2028392f927a92366bfe0b85e718a7cd6190142f/keras_cv/models/segmentation/segformer/segformer.py#L139). An `activation` params should be introduced [here](https://github.com/keras-team/keras-cv/blob/2028392f927a92366bfe0b85e718a7cd6190142f/keras_cv/models/segmentation/segformer/segformer.py#L85).

@fchollet Could you please give some feedback on this? As the keras model now also tightly integrated to kaggle environemnt https://www.kaggle.com/organizations/keras, I think it would be effective. 

Maybe for the sane reason yolo-v3 is wanted. https://github.com/keras-team/keras-cv/issues/622

@sachinprasadhs > If you encounter a KerasCV feature that is missing from KerasHub, or would like to propose an addition to the library, please file [an issue with KerasHub](https://github.com/keras-team/keras-hub/issues). Firstly,...

> Are you looking to contribute this one? Not at the moment. I was exploring options for fine-tuning SAM on 3D data and came across this method, which supports both...

@tirthasheshpatel Thanks. It would be really helpful.

To fix the code format, make sure to run ```bash shell/clean.sh # if keras-cv is built locally shell/format.sh shell/lint.sh ```

@divyashreepathihalli > what is the difference between the base and the backbone weights? do you have weights for the video classifier task model that you added? Ignore else, you can...

@divyashreepathihalli Do you have a kaggle id? If so, could you please share?