Conv2d_transpose requires asymmetric padding which the CUDA EP currently does not support
Nobody has claimed this yet.
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 43/100
- Issue type
- Bug
- Clarity
- Mostly clear
- Activity status
- Quiet
- Domain
- backend, machine-learning
Research direction
Start with ConvTranspose handling in the CUDA execution provider and compare the reported SAME-padding case with the tensorflow-onnx test referenced in the issue, which covers VALID padding. Check the supported-operators documentation and the linked asymmetric-padding change for expected behavior. Done means the reported model runs with asymmetric padding without the unsupported-operation warning.
Written by the indexing model from the issue text.
Description
Describe the bug
After successfully converting a tensorflow frozen graph(.pb) into a .oonx model, I encountered a warning while inferring the onnx model with onnxruntime-gpu in python.
The warning says conv2d_transpose op is not supported because it requires asymmetric padding which the CUDA EP currently does not support.
But it's strange that I found ConvTranspose is in Supported ONNX Operators and it seems the asymmetric padding issue is already solved.
System information
- OS Platform and Distribution:Linux Ubuntu 20.04
- ONNX Runtime installed from:
pip install onnxruntime-gpu - ONNX Runtime version: 1.10.0
- Python version: 3.6
- CUDA/cuDNN version: 11.3/8.2.1
- Tensorflow Version: 1.15
- opset: 13
Expected behavior
conv2d_transpose has been supported.
Screenshots
Additional context
I found tf.nn.conv2d_transpose has been tested in tensorflow-onnx with padding="VALID",while I use padding="SAME".
https://github.com/onnx/tensorflow-onnx/blob/f64772ce166ea2a0402524d9741b2fb71e5663df/tests/test_backend.py#L579-L590
- Dominant language
- C++
- Stars
- 21.9k
- Forks
- 4.2k
- Avg merge
- 4d 8h
- Merged PRs (30d)
- 179
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from microsoft/onnxruntime
-
ep:MIGraphX
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
microsoft/onnxruntime#32352 · 1 comment ·
-
`optimization.disable_specified_optimizers` is documented as comma-separated but is parsed on `;` Open
Difficulty 1/5 Under an hour Newbie friendliness 88/100
microsoft/onnxruntime#32211 ·
-
stale
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
microsoft/onnxruntime#29750 · 3 comments ·
-
ep:WebGPU platform:web stale
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
microsoft/onnxruntime#28976 · 3 comments ·
-
[Documentation] Python tutorial missing PyTorch export guidance and external data file handling Opendocumentation
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
microsoft/onnxruntime#28763 · 5 comments ·
All issues in microsoft/onnxruntime
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 86/100
-
Sensor initialization takes very long when `--initial-sim-time` is set to current UNIX timestamp Open
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
gazebosim/gz-sensors#662 · 1 comment ·
-
enhancement
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
-
comp-datalake
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
ClickHouse/ClickHouse#121222 ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
LadybirdBrowser/ladybird#12123 ·