annotated_deep_learning_paper_implementations
annotated_deep_learning_paper_implementations copied to clipboard
labmlai
🧑🏫 60 Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gan...
The given implementation puts all layers into a list blocks , and the condition to set stride to 2 is len(blocks)==1. So The size of image will only modify in...
Hi, Running the exact code on github for deepmind's retrieval transformer - [RETRO](https://github.com/labmlai/annotated_deep_learning_paper_implementations/tree/master/labml_nn/transformers/retro), getting the following error: `RuntimeError: stack expects each tensor to be equal size, but got [2, 32]...
Currently, labmlai has no implementation for Object Detection Papers such as Yolo Family, FPN, Retinanet. Do you have any timeline to share them as well?
https://nn.labml.ai/transformers/mha.html This link doesn't go to MHA code, but goes to Transformer code.
The [implementation](https://github.com/labmlai/annotated_deep_learning_paper_implementations/tree/master/labml_nn/gan/stylegan) of StyleGAN2 does not learn a mapping for the latent vector `z`. The vector `z` is completely ignored, and a variety of generated images is provided by `noise`....
When training with custom .png image, the following error occurs. ``` class Smooth(nn.Module): def __init__(self): super().__init__() kernel = [[1, 2, 1], [2, 4, 2], [1, 2, 1]] # define the...
致谢+建议
您好，首先感谢您这份非常有价值的工作，个人有两点小建议： ①经典模块代码注释时，可否提供中文注释？这样对外语不好的国人来说会更加友好。 ②paper文件夹中建议命名为该文章的模型名字或代表名称，可以更利于读者查阅呢。 再次对您的贡献表示感谢！
Paper: https://arxiv.org/abs/1312.6114 Implementation I have currently: https://github.com/BeeGass/Readable-VAEs/tree/master/vae-pytorch
https://arxiv.org/abs/1506.02025
I would like to understand the following paper Mueller, J. and Thyagarajan, A., 2016, March. "Siamese recurrent architectures for learning sentence similarity". In Proceedings of the AAAI conference on artificial...
