Comments (1)
Chatting with @bdhirsh, I think the bug is likely here:
pytorch/torch/_functorch/aot_autograd.py
Lines 573 to 575 in 22964d1
If any of the inputs that require grad are mutated while grad is enabled, we still need to run autograd.
from pytorch.
Related Issues (20)
- Compiled DDP + TP unit test fails with `SymNodeVariable() is not a constant` HOT 4
- Adam optimizer initializer useless if condition, torch v2.3.0 HOT 1
- Runtime error when running huggingface pretrained model with torch-xla
- xpu backend build fails if cloned directory is named python-x.y HOT 4
- [Break XPU] The newly added test_inductor_counters use requires_gpu, but hard code cuda
- [cudagraph] an issue related to dynamic shapes for cudagraph backend with sd2.1 HOT 2
- xeon.run_cpu will not use numactl when whithin a docker container run with --cpuset-cpus non-0 numa node HOT 1
- torch.Tensor.uniform_ causes invalid syntax in InternalTorchDynamoError HOT 1
- torch 2.4 doesn't work with on-demand profiling: faild with RuntimeError : Cannot initialize CUDA without ATenA_cuda library. HOT 2
- Code bug in call_prepare_scriptable_func_impl function.
- [torch.compile][torch.fx] CodeGen is not able to delete unused value if the value doesn't have users HOT 1
- AoTInductor cumsum causes illegal memory access for large tensors HOT 1
- [inductor] MultiKernel doesn't compose with AOTInductor
- result in nan when computing pow of a tensor HOT 3
- Create the pages which explain `torch.nan` and `torch.inf` in the doc HOT 1
- FlopCounterMode ignores custom operators HOT 1
- Using an empty tensor and `torch.int64` or `torch.bool` for `dtype` of `nanmean()` works while a non-empty tensor doesn't work HOT 2
- [export] Is there a way to set all axes in all inputs to dynamic?
- No CUDA runtime is found HOT 1
- DISABLED test_comprehensive_nn_functional_pad_reflect_cuda_bfloat16 (__main__.TestDecompCUDA) HOT 5
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pytorch.