|
DTensor random RNG state support for non-CUDA backends
|
|
2
|
198
|
October 18, 2024
|
|
Support for _set with other mutations in graph
|
|
2
|
155
|
October 18, 2024
|
|
What's the difference between torch.export / torchserve / executorch / aotinductor?
|
|
17
|
2289
|
October 17, 2024
|
|
How profiling Pytorch Using Nsight Compute?
|
|
2
|
575
|
October 16, 2024
|
|
Implementing true boolean tensors
|
|
2
|
424
|
October 12, 2024
|
|
Is it possible to disable inlining of custom module for torch.compile?
|
|
1
|
270
|
October 11, 2024
|
|
PyTorch Release 2.5.0 - Final RC is available
|
|
1
|
3375
|
October 8, 2024
|
|
1 High-Security vulnerability found in Black Duck scan
|
|
4
|
624
|
October 3, 2024
|
|
Prior art on implementing a "print" op on hardware
|
|
0
|
50
|
October 3, 2024
|
|
Impact of multithreading and local caching on torch.compile
|
|
3
|
852
|
September 27, 2024
|
|
How to test properly?
|
|
1
|
326
|
September 25, 2024
|
|
RFC Proposal for CUDA-Accelerated Dynamic Time Warping (DTW) Implementation in PyTorch
|
|
2
|
535
|
September 20, 2024
|
|
Debugging story: The case of the flaky Dynamo export tests
|
|
5
|
2423
|
September 20, 2024
|
|
Building with cuda: nvtx not found
|
|
2
|
1018
|
September 19, 2024
|
|
What is the official release build options?
|
|
3
|
132
|
September 19, 2024
|
|
Why so many HW backend and nobody cooperate?
|
|
6
|
631
|
September 17, 2024
|
|
Rethinking PyTorch Fully Sharded Data Parallel (FSDP) from First Principles
|
|
19
|
11835
|
September 17, 2024
|
|
PyTorch 2.5 RC1 is produced for pytorch, audio, vision
|
|
0
|
619
|
September 13, 2024
|
|
Location to add new rendezvous handlers
|
|
1
|
170
|
September 11, 2024
|
|
PyTorch 2.5 Release Branch cut for PyTorch core is completed
|
|
0
|
1085
|
September 10, 2024
|
|
PyTorch Release 2.4.1 - Final RC is available
|
|
4
|
8186
|
September 5, 2024
|
|
PyTorch Release 2.4.1 General Availability
|
|
0
|
579
|
September 4, 2024
|
|
TorchInductor Update 9: Harden Vectorization Support and Enhance Loop Optimizations in TorchInductor CPP Backend
|
|
0
|
547
|
September 4, 2024
|
|
TorchInductor Update 8: Max-autotune Support on CPU with GEMM Template
|
|
0
|
511
|
September 4, 2024
|
|
Memcpy based P2P communication for pipeline parallelism instead NCCL
|
|
9
|
1654
|
September 4, 2024
|
|
PyTorch Runtime Error with Compiled Autograd
|
|
1
|
302
|
August 31, 2024
|
|
Pytorch to Triton for Non-GPU Devices
|
|
7
|
1500
|
August 30, 2024
|
|
PyTorch Conference 2024 - Special Discount for our Dev Discuss Community!
|
|
0
|
236
|
August 30, 2024
|
|
Difference between the graph break reasons: `Dynamic control flow is not supported at the moment.` and `generic_jump TensorVariable()`
|
|
0
|
330
|
August 30, 2024
|
|
OpenCL backend dev - questions/support
|
|
4
|
392
|
August 29, 2024
|