Aleksei Zinovev

[This ](https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/yolov4#postprocessing-steps) is a script which should be ported, looks like it involves broadcasting, due to expand_dims usage ``` xy_grid = np.meshgrid(np.arange(output_size), np.arange(output_size)) xy_grid = np.expand_dims(np.stack(xy_grid, axis=-1), axis=2) xy_grid =...

Could you please try the same thing with the new version 0.3.0, which will be released today? Probably this problem will be fixed? Thanks for providing so much helpful feedback...

First of all, it's really great and more interactive way to have and run tutorials. Thanks for that implementation. My user experience: 1. Binder initialization is slow (maybe we need...

Thanks for the good and clear comments on my questions, dear @avan1235 so, I need to think about it more and discuss it with colleagues, so after [vacation](https://github.com/JetBrains/KotlinDL/discussions/154) let's revisit...

Yes, but the first priority is the loading model from Keras for Transfer Learning purposes. It's an important and popular use case. Also, we need a good way to save...

P.S Read the [Contributing Guidelines](https://github.com/JetBrains/KotlinDL/blob/master/CONTRIBUTING.md).

@zjuhasz Agree with this idea, it's a good plan for the very far future. I'm a big expert in Multiplatform, but TFJS looks very interesting and has the TF API...

Hi @stheid, thank you for your question. At this moment we have plans to make the hierarchy of classes open with the ability to make possible the inheritance for the...

How to obtain pre-trained weights https://reposhub.com/python/deep-learning/benjs-nfnets_pytorch.html

An example of pytorch implementation https://nfnets-pytorch.readthedocs.io/en/latest/