Improved handling when `axis` attribute is not defined and the batch size of the first dimension is undefined. - Githubissues

PINTO0309 / onnx2tf

Self-Created Tools to convert ONNX files (NCHW) to TensorFlow/TFLite/Keras format (NHWC). The purpose of this tool is to solve the massive Transpose extrapolation problem in onnx-tensorflow (onnx-tf). I don't need a Star, but give me a pull request.

MIT License

706 stars 73 forks source link

Improved handling when `axis` attribute is not defined and the batch size of the first dimension is undefined. #692

Closed PINTO0309 closed 2 months ago

PINTO0309 commented 2 months ago

1. Content and background

Flatten
- Improved handling when axis attribute is not defined and the batch size of the first dimension is undefined.
- ONNX
```
wget https://github.com/PINTO0309/onnx2tf/releases/download/0.0.2/resnet18-v1-7.onnx
onnx2tf -i resnet18-v1-7.onnx
```
ls -lh saved_model/

assets fingerprint.pb resnet18-v1-7_float16.tflite resnet18-v1-7_float32.tflite saved_model.pb variables

TF_CPP_MIN_LOG_LEVEL=3 \ saved_model_cli show \ --dir saved_model \ --signature_def serving_default \ --tag_set serve

The given SavedModel SignatureDef contains the following input(s): inputs['data'] tensor_info: dtype: DT_FLOAT shape: (-1, 224, 224, 3) name: serving_default_data:0 The given SavedModel SignatureDef contains the following output(s): outputs['output_0'] tensor_info: dtype: DT_FLOAT shape: (-1, 1000) name: PartitionedCall:0 Method name is: tensorflow/serving/predict
```
- TFLite
![image](https://github.com/user-attachments/assets/2da8169a-2272-4f47-957d-7a6416728630)
```

2. Summary of corrections

3. Before/After (If there is an operating log that can be used as a reference)

4. Issue number (only if there is a related issue)