-
The CSV_SPEC is confusing because the type names don't line up with Dynamo types.
Dynamo types are:
String (S)
Number (N)
Binary (B)
Boolean (BOOL)
Null (NULL)
Map (M)
List (L)
String S…
-
## 🚀 Feature
As described in https://github.com/pytorch/pytorch/issues/125958, we are integrating with vLLM on TPUs. We see that in the warm up phase of the vLLM, it needs to pre-compile ~30 differen…
-
## 🐛 Bug
Running the upstreamed benchmarking scripts with the following command results in an unexpected error.
```bash
python xla/benchmarks/experiment_runner.py \
--suite-name torchbe…
-
# Dynamo backend for torchxla2
## Goal
Have a dynamo backend backend by torch_xla2.
The users should be able to do the following:
```python
m = model ...
m_compiled = torch.compile(m, ba…
qihqi updated
3 weeks ago
-
i have an error opening the file, the problem is it opens in other versions of revit and also any other system but not mine
[error](https://github.com/DynamoDS/Dynamo/assets/68099005/258d2745-efd6-4d…
-
### 🐛 Describe the bug
```
import torch
class C:
x = 1
def __str__(self):
return "ok"
def foo(x):
a = C()
return x, str(a)
print(torch.compile(foo, fullgrap…
-
### 🐛 Describe the bug
Here, I can't export trivial net with sort() using dynamo_export().
Legacy onnx.export() works fine, dynamo_export errors out :
torch.onnx._internal.diagnostics.infra.c…
-
Hi,
I'm wondering if TuneUp will be updated for Dynamo 3.xx
I'm currently working with Dynamo v.3.0.3, but after TuneUp installation, it don't see it in "View" anymore, so I suspect that it's not co…
-
Dynamo is tracing through `__torch_dispatch__` on tensor subclasses. This shouldn't be happening. Using `traceable_tensor_subclasses` seems to make the problem go away.
Repro:
```python
import to…
-
Trying @dynamo to "highjack" all calls to a certain previously defined methods. I seem to run into issues when combining with broadcasting:
```
using IRTools: IR, @dynamo, recurse!
@dynamo fu…