TypeError: __init__() got an unexpected keyword argument 'ragged'

Accelerated Computing Intelligent Video Analytics TAO Toolkit
,
1

Good day, I get this Type Error :
TypeError: init() got an unexpected keyword argument ‘ragged’
when running:

detectnet_v2 train -e /workspace/tao-experiments/openalpr/SPECS_train.txt -r workspace/openalpr/exp_unpruned -k nvidia_tlt
inside the docker. from what I understand this is because of keras api it shouldn’t be imported directly
My GPU is nvidia Geforce rtx 3050
network type: detectnet_v2
tlt version:
tlt-version.txt (1.1 KB)
Training spec file:
SPECS_train.txt (4.2 KB)
full log:
log.txt (11.3 KB)

I appreciate your time in advance.

3

You are running detectnet_v2 training.
But the pretrained model is not expected. The us_lprnet_baseline18_trainable.tlt is an LPR model. It is not working in detectnet_v2 network.

model_config {
pretrained_model_file: “/workspace/tao-experiments/models/us_lprnet_baseline18_trainable.tlt”

4

thank you it is solved.

5

also the export command in this link below :

–-engine_file should be --engine_file ’

tao detectnet_v2 export -m /workspace/openalpr/exp_unpruned/weights/model.tlt -o /workspace/openalpr/export/unpruned_model.etlt --cal_cache_file /workspace/openalpr/export/calibration.bin -e /workspace/openalpr/SPECS_train.txt -k nvidia_tao --cal_image_dir /workspace/openalpr/lpd/data/image --data_type int8 --batch_size 4 --batches 10 –-engine_file /workspace/openalpr/export/unpruned_int8.trt

6

Thanks for the catching!

7

no problem :)

9

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