Hi I have object detection problem and want to optimise the model using TensorRT 4.0 i.e convert frozen graph into trt_engine Code import keras import keras.backend as K import tensorflow as tf import uff output_names = [“detection_classes”] frozen_graph_filename = ‘mars-small128.pb’ sess = …

AssertionError: detection_classes is not in graph

dhingratul September 27, 2018, 7:27pm 2

Try blacklisting the output node, maybe it is being removed in one of the remove/convert_to_constant operations.

Topic		Replies	Views
TF object detection faster rcnn nodes not found TensorRT	8	1855	July 1, 2020
Converting TF Model to TensorRT Engine TensorRT	6	3271	September 24, 2019
Converting TF Model to TensorRT UFF Format Jetson TX2	27	23356	October 18, 2021
convert_to_uff.py fails while trying to convert Tensorflow frozen model TensorRT	6	1013	October 12, 2021
Failed to convert ssd_inception_v2 from Tensorflow Object Detection to TensorRT engine Jetson AGX Xavier	3	803	October 18, 2021
Name Error:name 'UffException' is not defined TensorRT	17	4119	September 18, 2018
convert_to_uff.py failed to parse nodes from config.py when trying to convert resnet50_v1_fpn frozen graph to UFF TensorRT	1	1181	May 21, 2019
I get an error when I'm tring to convert my tensorflow model to tensorrt Jetson TX2	2	423	October 18, 2021
Unable to convert tensorflow models master maskRcnn graph to tensorRT TensorRT	6	639	January 15, 2021
TensorRT 3.0 UFF parser failed to parse output node (Assertion Error) GPU-Accelerated Libraries	3	1601	March 11, 2018