|
About the performance category
|
|
0
|
750
|
January 22, 2021
|
|
Performance of "activation" sparsity
|
|
1
|
151
|
April 21, 2025
|
|
The "Ideal" PyTorch FLOP Counter (with __torch_dispatch__)
|
|
20
|
15204
|
August 22, 2024
|
|
Fast combined C++/Python/TorchScript/Inductor tracebacks
|
|
4
|
2989
|
August 17, 2023
|
|
Estimate theoritical FLOPs of backward pass of a DNN
|
|
1
|
1050
|
April 16, 2023
|
|
Performance gains w/ nanoGPT using SDPA Custom Kernel
|
|
0
|
4674
|
January 30, 2023
|
|
Making Transformer inference faster on GPUs
|
|
1
|
3879
|
December 16, 2022
|
|
Investigation report: what would it cost to optimize c10::intrusive_ptr destruction for refcount == 1? (A: too much)
|
|
0
|
684
|
May 3, 2022
|
|
Working With `c10::IValue` Efficiently
|
|
0
|
2595
|
April 18, 2022
|
|
Unionizing for Profit: How to Exploit the Power of Unions in C++
|
|
2
|
3164
|
January 7, 2022
|
|
CUDA loops case study: code generation vs templates
|
|
4
|
2505
|
December 12, 2021
|
|
Multiple workers for single batch
|
|
2
|
1123
|
July 14, 2021
|
|
Optimizing contiguous() for the case where the Tensor is_contiguous()?
|
|
6
|
1619
|
May 24, 2021
|
|
Converting weights flat buffer
|
|
0
|
883
|
March 25, 2021
|
|
Why `torch::jit::pop` (and sometimes push) is worse for performance than direct `std::vector` access
|
|
0
|
1102
|
March 17, 2021
|
|
Pytorch Benchmarks issues with general usability and issues with individual benchmarks
|
|
2
|
1289
|
March 12, 2021
|
|
GPU Overheads and Fused Strassen
|
|
0
|
2211
|
February 13, 2021
|
|
Comparing the performance of 0.4.1 and master
|
|
0
|
2322
|
February 9, 2021
|
|
Overhead in `nn.Module` causing massive slowdowns compared to raw CuBLAS or Torchscript
|
|
0
|
1666
|
January 28, 2021
|
|
Dispatcher Performance and Inlining: a Report on Two Days Spent on Dispatcher Performance
|
|
1
|
1176
|
January 28, 2021
|
|
We shouldn't feel bad about passing `Tensor` by reference
|
|
0
|
2312
|
January 25, 2021
|