Converting efficientdet-tf1 trained model to deepstream engine

[edit_or]

• Hardware (T4/V100/Xavier/Nano/etc)
• Network Type Efficientdet-tf1
• Tao version: nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5

I am trying to convert Efficientdet-tf1 model to deepstream engine.
I tried to follow this command
efficientdet_tf1 gen_trt_engine. But the error is

efficientdet: error: invalid choice: 'gen_trt_engine' (choose from 'train', 'prune', 'inference', 'export', 'evaluate', 'dataset_convert')
root@cadf0713c5f7:/workspace/Nyan/tao_source_codes_v5.0.0/notebooks/tao_launcher_starter_kit/efficientdet_tf1/results/unpruned/export#                                   -r . \
>                                   -k nvidia_tao \
>                                   --data_type fp16 \
>                                   --engine_file model.etlt.onnx_b1_fp16.engine \
>                                   --max_batch_size 12 \
>                                   --min_batch_size 1 \
>                                   --opt_batch_size 1 \
>                                   --batch_size 1
bash: -r: command not found

Which command can I use?

[Morganh]

@edit_or

root@cadf0713c5f7:/workspace/Nyan/tao_source_codes_v5.0.0/notebooks/tao_launcher_starter_kit/efficientdet_tf1/results/unpruned/export#                                   -r . \
>                                   -k nvidia_tao \
>                                   --data_type fp16 \

I think you are running in tao pytorch container.
Please use tao deploy container instead. nvcr.io/nvidia/tao/tao-toolkit:5.0.0-deploy

[edit_or]

I am using this container nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5

[Morganh]

Yes, it is not tao deploy container. You can use deploy container instead.