About the performance category
|
|
0
|
736
|
January 22, 2021
|
The "Ideal" PyTorch FLOP Counter (with __torch_dispatch__)
|
|
20
|
13302
|
August 22, 2024
|
Fast combined C++/Python/TorchScript/Inductor tracebacks
|
|
4
|
2774
|
August 17, 2023
|
Estimate theoritical FLOPs of backward pass of a DNN
|
|
1
|
957
|
April 16, 2023
|
Performance gains w/ nanoGPT using SDPA Custom Kernel
|
|
0
|
4534
|
January 30, 2023
|
Making Transformer inference faster on GPUs
|
|
1
|
3699
|
December 16, 2022
|
Investigation report: what would it cost to optimize c10::intrusive_ptr destruction for refcount == 1? (A: too much)
|
|
0
|
672
|
May 3, 2022
|
Working With `c10::IValue` Efficiently
|
|
0
|
2435
|
April 18, 2022
|
Unionizing for Profit: How to Exploit the Power of Unions in C++
|
|
2
|
2863
|
January 7, 2022
|
CUDA loops case study: code generation vs templates
|
|
4
|
2379
|
December 12, 2021
|
Multiple workers for single batch
|
|
2
|
1016
|
July 14, 2021
|
Optimizing contiguous() for the case where the Tensor is_contiguous()?
|
|
6
|
1467
|
May 24, 2021
|
Converting weights flat buffer
|
|
0
|
870
|
March 25, 2021
|
Why `torch::jit::pop` (and sometimes push) is worse for performance than direct `std::vector` access
|
|
0
|
1087
|
March 17, 2021
|
Pytorch Benchmarks issues with general usability and issues with individual benchmarks
|
|
2
|
1258
|
March 12, 2021
|
GPU Overheads and Fused Strassen
|
|
0
|
2030
|
February 13, 2021
|
Comparing the performance of 0.4.1 and master
|
|
0
|
2241
|
February 9, 2021
|
Overhead in `nn.Module` causing massive slowdowns compared to raw CuBLAS or Torchscript
|
|
0
|
1596
|
January 28, 2021
|
Dispatcher Performance and Inlining: a Report on Two Days Spent on Dispatcher Performance
|
|
1
|
1124
|
January 28, 2021
|
We shouldn't feel bad about passing `Tensor` by reference
|
|
0
|
2154
|
January 25, 2021
|