-
## Problem statement
CNN operators utilize canonical order of tensor dimensions and assign them semantic meaning. For the 2D case in PyTorch today an input to torch.nn.Conv2d has to be a 4d tensor …
-
Hello, thank you again for the hard work to push deployment of 3D detection models.
I made several tests on Pointpillars arch, trained on Kitti dataset (reduced) and compared the computing time and…
-
## ❓ Questions
For the following example code, the printed graph in torch 1.4.0 is very different from that in torch 1.3.1. I am wondering is this expected? because many useful fields are removed i…
-
Would be cool to peek into the state of the caching allocator on things like:
- [x] Total cached memory
- [x] Total currently used memory, referenced by Tensors
- [x] Forced free of unused segments…
-
I've been pointed to this line
https://github.com/pytorch/pytorch/blob/12528990f8c56deb7ce1c699e6da63d82c115968/aten/src/ATen/native/Convolution.cpp#L526
where we can see which choices we have whe…
-
Hi, great work on this package!
I'm looking to use this package in conjunction with other dependencies in a larger project and have no experience with nix. Since there are multiple top-level packag…
-
Currently windows and linux x64 prebuilds with CUDA are available. It would be nice to support more backends and arch/platform combinations - but also leave an escape hatch for anybody wanting to do b…
iimez updated
1 month ago
-
1. Following is the Conv2d pytorch module.
```
import torch
import torch.nn as nn
class op_conv2d(nn.Module):
def __init__(self):
super().__init__()
self.layers = nn.Seque…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Have you followed all the steps in the FAQ?
- [X] I have tried the steps in the FAQ.
### Current Behavior
I …
-
### 🐛 Describe the bug
torch.compile is not supported on Windows.
torch.compile has dependency triton: https://github.com/openai/triton/issues/1640
### Expected outcome
1. Document the requir…