KeyError: 'NonMaxSuppression not implemented yet' (YOLOv7 tiny onnx to tflite)

[kizmoya]

Hello,
I tried to convert YOLOv7 ONNX model to tflite.
When I tried in Intel Macbook, I got some error message.

$ python converter.py --weights "../yolov7-tiny.onnx" --outpath "./" --formats "tflite"
2023-05-08 19:48:50.472025: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
Checking 0/1...
INFO:deformation_layers ::Reshape will process tensor after change back to NCHW format.
INFO:deformation_layers ::Reshape will process tensor after change back to NCHW format.
INFO:deformation_layers ::Reshape will process tensor after change back to NCHW format.
INFO:deformation_layers ::Reshape will process tensor after change back to NCHW format.
INFO:deformation_layers ::Reshape will process tensor after change back to NCHW format.
INFO:deformation_layers ::Reshape will process tensor after change back to NCHW format.
INFO:deformation_layers ::Transpose will process tensor after change back to NCHW format.
Traceback (most recent call last):
File "/onnx2tflite/converter.py", line 108, in
run()
File "/onnx2tflite/converter.py", line 92, in run
onnx_converter(
File "/onnx2tflite/converter.py", line 21, in onnx_converter
keras_model = keras_builder(model_proto, native_groupconv)
File "/onnx2tflite/utils/builder.py", line 75, in keras_builder
raise KeyError(f"{op_name} not implemented yet")
KeyError: 'NonMaxSuppression not implemented yet'

Is there something I can do to fix this?
Looking forward to your reply.

[MPolaris]

Hello, it's seems the post-process operation is exported with model, you may check how to remove it.

# without `--end2end`
python export.py --weights yolov7-tiny.pt --simplify --img-size 640 640 --max-wh 640

[MPolaris]

NonMaxSuppression op is a typical dynamic operator, tflite is not friendly to it as my known.
So we always make model graph is static, not dynamic.