Products · Product
PyTorch
Raised as a usability problem by 12 people on Hacker News over 8 years (2018–2025), in 12 threads.
No issue was raised by at least 3 people in 3 threads this year.
No issue was raised by at least 3 people in 3 threads in the last 5 years.
Every issue, all time, most people first
Installing PyTorch with Poetry fails and is next to impossibleUnreviewed
20182026- Installing PyTorch, CUDA and dependent libraries is very hard on Windows and WSL2025-08-13
- Installing PyTorch requires manually placing files in an undisclosed location2024-07-10
- Installing PyTorch with Poetry is next to impossible2021-08-04
Unreviewed issues were grouped by machine: Claude Haiku labelled each comment as a usability complaint, named what it's about and summed up the problem in a line, and a small model grouped similar problems. Nobody has checked them by hand. Expect mistakes.
The newest 11 complaints
- Confusing to figure out how to properly run fine-tuned models on the phone2025-08-29
- Installing PyTorch, CUDA and dependent libraries is very hard on Windows and WSL2025-08-13
- Custom PyTorch model export to Core ML fails with unsupported ops and segfaults2025-06-02
- Installing PyTorch requires manually placing files in an undisclosed location2024-07-10
- Hard to use, and switching back from TPU to other hardware takes too much effort2023-10-17
- Every inference recompiles the model, so each compile takes 15 minutes and the run freezes2021-11-06
- PyTorch support for TPUs was atrocious a year or two ago2021-10-07
- Installing PyTorch with Poetry is next to impossible2021-08-04
- In-page search doesn't work for many seconds after the page loads2019-07-21
- Couldn't figure out how to pass a script module's parameters to the optimizer2018-12-08
- Breaks scientific Python conventions, making it cumbersome to integrate with existing workflows2018-10-02