Onnx2tf

- `Resize`
- [YOLOv8 FP16] TensorFlow's `Resize` operation casts to Float32 on its own, so we have to change it back to the original type.
- Support for YOLOv8
- Validation (Float32)

What's Changed
* Remove unnecessary processing by PINTO0309 in https://github.com/PINTO0309/onnx2tf/pull/128
* [YOLOv8 FP16] TensorFlow's Resize operation casts to Float32 on its own, so we have to change it back to the original type. by PINTO0309 in https://github.com/PINTO0309/onnx2tf/pull/129


**Full Changelog**: https://github.com/PINTO0309/onnx2tf/compare/1.5.15...1.5.16

- Padding in `SYMMETRIC` mode will cause an error if the padding size is larger than the input size, so replace `CONSTANT` with `SYMMETRIC`
- Replace `tf.reduce_mean` with `GlobalAveragePoolingxD` to reduce inference error

However, the phenomenon where the error between `AveragePool` and `GlobalAveragePool` becomes unusually large has not been corrected.

- Workaround issue with onnx_graphsurgeon aborting when using output with dtype unknown
- https://github.com/PINTO0309/onnx2tf/releases/download/1.1.28/rcnn-ilsvrc13-9.onnx

What's Changed
* Padding in `SYMMETRIC` mode will cause an error if the padding size is larger than the input size, so replace `CONSTANT` with `SYMMETRIC` by PINTO0309 in https://github.com/PINTO0309/onnx2tf/pull/127


**Full Changelog**: https://github.com/PINTO0309/onnx2tf/compare/1.5.14...1.5.15

- `disable_unnecessary_transpose`, Skip special processing if the operation does not result in an error even if special processing is not performed
- https://github.com/PINTO0309/onnx2tf/releases/download/1.1.28/efficientformer_l1.onnx

What's Changed
* `disable_unnecessary_transpose`, Skip special processing if the operation does not result in an error even if special processing is not performed by PINTO0309 in https://github.com/PINTO0309/onnx2tf/pull/126


**Full Changelog**: https://github.com/PINTO0309/onnx2tf/compare/1.5.13...1.5.14

- [Experimental] Fixed a problem in which the transposition flag was wrongly determined when the shapes of all parts of the tensor except the batch size after `MatMul` had the same numerical value
- https://github.com/PINTO0309/onnx2tf/releases/download/1.1.28/detr_demo.onnx

What's Changed
* [Experimental] Fixed a problem in which the transposition flag was wrongly determined when the shapes of all parts of the tensor except the batch size after `MatMul` had the same numerical value by PINTO0309 in https://github.com/PINTO0309/onnx2tf/pull/125


**Full Changelog**: https://github.com/PINTO0309/onnx2tf/compare/1.5.12...1.5.13

- Added indication of maximum absolute error in case of `Unmatched`

What's Changed
* Added indication of maximum absolute error in case of `Unmatched` by PINTO0309 in https://github.com/PINTO0309/onnx2tf/pull/123


**Full Changelog**: https://github.com/PINTO0309/onnx2tf/compare/1.5.11...1.5.12

- Added option to compress `FlexTranspose` to any number of dimensions

-nodafc, --number_of_dimensions_after_flextranspose_compression
Number of Transpose OP dimensions generated after avoiding FlexTranspose generation.
Default: 5

- `-nodafc 4`

- `-nodafc 2`

What's Changed
* Added option to compress `FlexTranspose` to any number of dimensions by PINTO0309 in https://github.com/PINTO0309/onnx2tf/pull/122


**Full Changelog**: https://github.com/PINTO0309/onnx2tf/compare/1.5.10...1.5.11