Tao Toolkit BPNet Assertion Error while training

Please provide the following information when requesting support.

• Hardware (A6000)
• Network Type (bpnet)
• TLT Version (Please run “tlt info --verbose” and share “docker_tag” here)
format_version: 2.0
toolkit_version: 4.0.1
published_date: 03/06/2023
• Training spec file(If have, please share here)

I'm using the default bpnet_train_m1_coco.yaml that's provided with the examples with no modifications.

• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)

Hello, I’m going through the examples in the tao-toolkit, and am stuck on the training step for bpnet:

!tao bpnet train -e $SPECS_DIR/bpnet_train_m1_coco.yaml
-r $USER_EXPERIMENT_DIR/models/exp_m1_unpruned
-k $KEY
–gpus $NUM_GPUS

Errors out with

Traceback (most recent call last):
File “</usr/local/lib/python3.6/dist-packages/driveix/bpnet/scripts/train.py>”, line 3, in
File “”, line 221, in
File “”, line 150, in main
File “”, line 167, in deserialize_maglev_object
File “”, line 432, in wrapper
File “”, line 100, in init
AssertionError: Please specify inference spec path in the config file.
Telemetry data couldn’t be sent, but the command ran successfully.
[WARNING]: <urlopen error [Errno -2] Name or service not known>
Execution status: FAIL
2023-04-19 17:03:14,336 [INFO] tlt.components.docker_handler.docker_handler: Stopping container.

There is an infer_spec.yaml, but I can’t find any documentation on how to include it. I tried specifying with --inference_spec. Any help greatly appreciated.

Did you use 4.0.1 version notebook?


https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/resources/tao-getting-started/version/4.0.1/files/notebooks/tao_launcher_starter_kit/bpnet/bpnet.ipynb#head-7

Please refer to the bpnet_train_m1_coco.yaml.
There is one line to set inference_spec.

inference_spec: /workspace/examples/bpnet/specs/infer_spec.yaml

Thanks for the quick reply. I was using the notebook in:

cv_samples_v1.4.0

the line is omitted in that version. Much appreciated!

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.