-
# Feature request
Before I start, let me commend you on v.3 - great achievement, thank you! Why am I saying this? Because I was using v.2.9.0 and a) it was slow and b) was flooding me with twig error…
-
Hello everyone,
I'm currently trying to convert PLATO-2 model into ONNX format using Paddle2ONNX. However, when I try to convert the NSP model, I got this error:
```
Traceback (most recent call l…
-
Hey,
I have issue with exporting inference graph i dont know what is the issue i have been trying to export but its not working. Thank you in advance. Help will be appreciated.
Command used:
pyt…
-
### 🐛 Describe the bug
While regular torch tensor indexing supports Ellipsis (...) and the `torch.jit` graph represents it with None, the onnx exporter doesnt support this and generates non-functiona…
-
Hi,
I am trying to export a frozen graph which has only a neural network nodes. To do this, I turn on the export mode and set the only_network flag
`python inspector.py --mode=export --only_netw…
-
torch.onnx.errors.SymbolicValueError: Unsupported: ONNX export of instance_norm for unknown channel size. [Caused by the value 'input.411 defined in (%input.411 : Float(*, *, *, *, strides=[3276800, …
-
With the growing number of Spine modules, it becomes harder to track how they relate to each other. We need to build a dependency tree for:
1. Each Spine module separately to know which other modul…
-
What happens if a user has `graph_type="dag"` instead here? Will this function detect that there are `o->` edges and prevent that? Clearly the imported file does not represent a DAG so w…
-
**Barrel files** are files which repackage multiple libraries into a single library by exporting them. https://github.com/flutter/flutter/blob/master/packages/flutter/lib/material.dart is a good examp…
-
### Describe the bug
When attempting to build TensorRT Engine, an error message is displayed indicating that cuDNN is not initialized (`CUDNN_STATUS_NOT_INITIALIZED`).
Error Log
```bash
[I…