|
Unexplainable NaN loss
|
|
1
|
9
|
October 26, 2025
|
|
Is it possible that PyTorch can make async host-to-device transfer execute on side stream while the transferred data on the main stream
|
|
0
|
5
|
October 26, 2025
|
|
Cdist for a large number of pairs
|
|
0
|
8
|
October 25, 2025
|
|
FP16 support in pytorch mobile
|
|
5
|
1219
|
October 25, 2025
|
|
Random object detection results
|
|
3
|
36
|
October 25, 2025
|
|
Yolov7 problem while training models
|
|
2
|
433
|
October 25, 2025
|
|
Windows DDP on RTX 50-series only: use_libuv was requested but PyTorch was built without libuv support (works on 40/20-series)
|
|
0
|
12
|
October 25, 2025
|
|
Memory Leak Debugging and Common Causes
|
|
16
|
29933
|
October 25, 2025
|
|
Basic System requirement to be able to contribute to Pytorch
|
|
0
|
12
|
October 24, 2025
|
|
MultiheadAttention with multiple batch dimensions
|
|
2
|
15
|
October 24, 2025
|
|
What version of Pytorch is compatible with NVIDIA GeForce GTX 1080
|
|
5
|
1601
|
October 24, 2025
|
|
Varying batch size for pre-trained model changes inference result, even in evaluation mode
|
|
1
|
24
|
October 24, 2025
|
|
nn.MultiheadAttention fast_path substantially less efficient
|
|
0
|
18
|
October 24, 2025
|
|
ONNX export fails for custom interpolate function with dynamic scale factor in torch.export.export (PyTorch 2.9.0)
|
|
0
|
12
|
October 24, 2025
|
|
Sparse bmm cause CUDA misaligned address error
|
|
2
|
28
|
October 23, 2025
|
|
Build Pytorch 2.7.1 from source with cuda 12.4
|
|
2
|
25
|
October 23, 2025
|
|
Support for NVIDIA RTX 5090 (CUDA SM_120)
|
|
1
|
26
|
October 23, 2025
|
|
NVIDIA DGX Spark support
|
|
3
|
55
|
October 23, 2025
|
|
TorchCodec wheels
|
|
2
|
431
|
October 23, 2025
|
|
Why is my rt-detr model giving map zero while train loss and valid loss is being decreased?
|
|
0
|
8
|
October 23, 2025
|
|
Context Parallel – attention mask not supported yet? Performance drop in FSDP2 + CP fine-tuning
|
|
0
|
9
|
October 23, 2025
|
|
PyTorch AD with non-python functions
|
|
1
|
16
|
October 22, 2025
|
|
Training a VAE on the embeddings of other VAEs
|
|
4
|
27
|
October 22, 2025
|
|
PyTorch with CUDA support is not installed, although everything worked before that
|
|
1
|
73
|
October 22, 2025
|
|
Mac inter want to install torch version 2.6 or higher
|
|
1
|
8
|
October 22, 2025
|
|
Error(s) in loading state_dict for ResNet: Missing key(s) in state_dict:
|
|
0
|
14
|
October 22, 2025
|
|
Filtered mean and std from a tensor
|
|
8
|
4390
|
October 22, 2025
|
|
Model Boilerplate for a Simple DQN
|
|
3
|
35
|
October 22, 2025
|
|
Error in Encoder Decoder Based Model
|
|
1
|
88
|
October 21, 2025
|
|
Best environment for deeplearning on a local GPU workstation
|
|
2
|
30
|
October 21, 2025
|