Mask rcnn fails to generate .onnx model .bydefault it generate .uff model

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

Please provide the following information when requesting support.

• Hardware (gpu size:33 gb)
Traininjg on:A4000
• Network Type Mask_rcnn
• TLT Version:tao v5
• Training spec file:mask_rcnn.ipynb

I am trying to generate .onnx model from the checkpoint that generated by mask rcnn but it fails to generate .onnx it always generate .uff model please provide solution how i able to generate .onnx model using these checkpoint …please take a look that the command written on jupyter notebook.!tao model mask_rcnn export -m $USER_EXPERIMENT_DIR/experiment_dir_unpruned/model.epoch-$NUM_EPOCH.tlt
-e $SPECS_DIR/maskrcnn_train_resnet50.txt
–gen_ds_config
–onnx tf2onnx
-o $USER_EXPERIMENT_DIR/export/model_rcnn.

3

Currently, Mask_rcnn does not support export to ONNX file yet. Suggest you to use trtexec to generate tensorrt engine for running inference on GPU.
Please refer to Converting TAO-trained MaskRCNN models to ONNX for CPU inference - #4 by telo.
Also, there is Mask2Former as alternative. Please refer to latest TAO 5.5 user guide or notebook.