Hi there, I’m trying to train resnet10 using ssd using tlt but everytime getting the following assertion error, can you please help me out? Thanks. Details : ssd_train_resnet10_kitti.txt : random_seed: 42 ssd_config { aspect_ratios_global: “[1.0, 2.0, 0.5, 3.0, 1.0/3.0]” scales: “[0.05, 0.1, …

Please check if the setting meets requirement. For TLT3.0, See Transfer Learning Toolkit — Transfer Learning Toolkit 3.0 documentation For TLT 2.0, see Requirements and Installation — Transfer Learning Toolkit 2.0 documentation

Unable to train | tensorflow.python.framework.errors_impl.InvalidArgumentError: Conv2DCustomBackpropInputOp only supports NHWC

Accelerated Computing Intelligent Video Analytics TAO Toolkit

saketm2408 June 3, 2021, 6:40am 4

Tesla K80

TAO Toolkit - NVIDIA Tesla K80

Topic		Replies	Views
Error while training on coco dataset TAO Toolkit	10	673	May 24, 2022
Tao detectnet_v2 train failed with g_error_metadata.to_exception in autograph module TAO Toolkit tao	12	1414	January 10, 2022
Error while training using the detectnet_v2 notebook provided in the TAO toolkit with using the custom dataset TAO Toolkit computer-vision-cv , tao	16	1417	January 13, 2023
TLT 3.0 & WSL2 issues TAO Toolkit nvbugs	7	1256	December 6, 2021
Very slow initialization of training and first epoch TAO Toolkit	11	3097	September 30, 2021
Error with tlt train in official Jupyter notebook TLT 3.0 TAO Toolkit	7	807	October 12, 2021
Retinanet box out of bounds TAO Toolkit	4	452	June 29, 2022
DetectNet v2 training error - "ValueError: The zipfile extracted was corrupt. Please check your key " TAO Toolkit	2	1006	October 12, 2021
AttributeError: module 'logging' has no attribute 'getLoggger' TAO Toolkit tensorflow	3	1597	February 28, 2022
Error while training on tlt TAO Toolkit	4	727	September 5, 2021

Unable to train | tensorflow.python.framework.errors_impl.InvalidArgumentError: Conv2DCustomBackpropInputOp only supports NHWC

Related topics