Published September 25, 2022
| Version 1.5.34
Software
Open
Tools to convert ONNX files (NCHW) to TensorFlow format (NHWC)
Creators
Description
Softmax- Significantly improved efficiency of automatic correction of inference errors.
- Reduce the process of obtaining tensors for error verification by dummy inference in ONNX to once for the entire model
- Installation of
onnxruntimeis mandatory after this release.
Pow- Suppresses inference errors.
common_functions.py- Fixed to skip comparing tensors of type
boolwhen verifying accuracy.
- Fixed to skip comparing tensors of type
- Extended SWAP size for CI from 10GB to 12GB
test_model_convert.py- Fixed to delete *.tflite generated during testing to avoid CI becoming disk full
- Implementation of strict mode #145
- [MobileFormer] Converted model outputs values mismatch with original ones. #105
- [MobileFormer]Dimensions must be equal [Add Layer] #103
SoftmaxDetect conversion errors in axis and identify the axis with the smallest possible error and replace it. #152
Notes
If you use onnx2tf in your research, please cite it using these metadata.
Files
PINTO0309/onnx2tf-1.5.34.zip
Files
(450.6 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:59f3c8a4ab22fc46d5ad3fac5e68abd5
|
450.6 kB | Preview Download |
Additional details
Related works
- Is supplement to
- https://github.com/PINTO0309/onnx2tf/tree/1.5.34 (URL)