-
I noticed that returnn_common does not check dimensions when concating during network construction.
So when doing: `concat((a, a.feat), (b, b.feat))` with `a.shape = [B, T, F]` and `b.shape = [B,F]`…
-
Aten - a tensor library
- 用于替换 TH 系列库
# Reference
- [ ] [Aten - pytorch cppdocs](https://pytorch.org/cppdocs/#aten)
- [ ] [TH to ATen porting guide](https://github.com/pytorch/pytorch/wiki/TH-t…
-
Hi,
I am using the condensed sparsity repo on MLPMixer ([source code](https://github.com/lucidrains/mlp-mixer-pytorch)). My current implementation is available [here](https://github.com/abhishektya…
-
### 🚀 The feature, motivation and pitch
Large models are becoming increasingly modularised. Recent research on [mixture-of-expert](https://arxiv.org/abs/1701.06538) networks (MoEs) and [fast feedfo…
-
## 🚀 Feature
The ability to force layer's weights to be sparse with some input mask so that for example a 1d convolution that has a weight matrix with mXn weights would have a sparse matrix of size m…
-
Hi everyone. I'm a green hand in GAN.
Seeing the following code:
`def discriminator_network(input_image_tensor):
"""
The discriminator network, Dφ, contains 5 convolution layers and 2 m…
-
## 🐛 Bug
neural network training work without sparse=True, but do not work with sparse=True.
## To Reproduce
https://colab.research.google.com/drive/1m1xhqu9s_4YdZoDjkTTNu2P58O0Cwv7J?usp=shar…
-
## 🚀 Feature
The blkdiag method is defined clearly in https://github.com/pytorch/pytorch/issues/31932
https://github.com/pytorch/pytorch/issues/31932 suggests blkdiag should create a dense Tenso…
-
## ❓ Question
im able to `torch.export` and generate an ExportedProgram with no issues for my model. upon compiling with `torch_tensorrt`...
```python
ep = torch.export.load("...")
example_inpu…
-
Training works, but it seems very slow... which is fine, as long as this is the expected behavior. I'm just curious if it is unusually slow for me.... did you happen to get this warning?
```
/usr/l…