-
Platforms: linux
This test was disabled because it is failing in CI. See [recent examples](https://hud.pytorch.org/flakytest?name=test_comprehensive_special_erfcx_cpu_int32&suite=TestInductorOpInfoCP…
-
Platforms: linux
This test was disabled because it is failing in CI. See [recent examples](https://hud.pytorch.org/flakytest?name=test_comprehensive_bucketize_cpu_float16&suite=TestInductorOpInfoCPU&…
-
Full error:
```
Processing Samples: 0%| | 0/50 [00:00
-
Platforms: linux
This test was disabled because it is failing in CI. See [recent examples](https://hud.pytorch.org/flakytest?name=test_comprehensive_xlogy_cpu_int32&suite=TestInductorOpInfoCPU&limit=…
-
Platforms: linux
This test was disabled because it is failing in CI. See [recent examples](https://hud.pytorch.org/flakytest?name=test_comprehensive_nn_functional_cosine_embedding_loss_cpu_float64&su…
-
Platforms: linux
This test was disabled because it is failing in CI. See [recent examples](https://hud.pytorch.org/flakytest?name=test_comprehensive_fft_ihfftn_cpu_int32&suite=TestInductorOpInfoCPU&l…
-
### 🐛 Describe the bug
When compiling a weight-normed linear layer, some shapes are causing accuracy failures.
The backends `eager` and `aot_eager` seem to work, but `inductor` causes issues.
For e…
-
I am currently getting the same issue in #10.
I have torch 2.5.1, torchrl 0.6.0, tensordict 0.6.0 at the moment. I am running a slightly modified version of the original code. I can run with cudagr…
-
### Dynamo Version
From Dynamo 2.18+ when the Pandas library was added
### Host
Revit 2024, 2025. Dynamo Sandbox 2.19.3, 3.2.2, 3.3.0
### What improvement/feature would you like to see in Dynamo?…
-
Platforms: inductor
This test was disabled because it is failing in CI. See [recent examples](https://hud.pytorch.org/flakytest?name=test_comprehensive_eq_cpu_float32&suite=TestInductorOpInfoCPU&limi…