-
# Stop "inventing" CPython C-API calls
**the CPython CAPI convention is that calls begin with Py_ or Py[Type]_**
To make it clear if a call is made into CPythons CAPI those prefixes should be o…
-
### Describe the issue
SplitToSequence operator cannot support float16 as input, although it was said 'yes' in ONNX doc.
It impact fp16 converter. Because the output is a sequence, so user have to…
-
Numpy installation fails due to some SIMD issues. I don't need any avx512, even if the node I am compiling on is supporting it. We are running an HPC cluster with several different hardware generation…
-
### Describe the issue
Creating an inference session using a model with a `TreeEnsembleRegressor` operator that has attribute data stored externally crashes with an error like:
```
onnxruntime.ca…
-
I get an error when trying to run a function from java. What have I done wrong?
```
Process finished with exit code 132 (interrupted by signal 4: SIGILL)
```
```java
try (Interpreter interp =…
Inqus updated
3 years ago
-
### Describe the issue
(This is about inference crash. Not really sure if this should be performance category or training.)
When we tested the model inference, while everything runs fine on CPUs…
-
### 🐛 Describe the bug
When jit scripting a model that uses a list of ints as the argument for `split_size_or_sections` with `torch.split` we get an error when trying to load the onnx model into an o…
-
### Describe the issue
I have a onnx file that can run normally on cuda device 0, but raises this error when I run it on device 1:
```
onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException…
-
## 🐛 Bug
When exporting a half precision (fp16) model to onnx it creates an invalid onnx file. This appears to be because of a node that remains in fp32 as a result of [this](https://github.com/pyt…
-
### Before submitting the issue
- [X] I have checked for [Compatibility](https://dpf.docs.pyansys.com/version/stable/getting_started/compatibility.html#id1) issues
- [X] I have searched among the …