Creating .wts file in SampleMNISTAPI for a tensorflow model

shrutiverma.022 · June 11, 2018, 2:42pm

Hello,

I want to deploy tensorflow model to Jetson, so once i create model in c++ using network definition TensorRT api, how to create .wts file so that I can load weights as given sampleMNISTAPI example. Though it has been mentioned, its a space separated file, [type] , but I am still a little confuse. Some clarification would be more helpful.

Thank you

shrutiverma.022 · June 11, 2018, 2:56pm

Ok, I am so sorry, I checked with the samples of TensorRT4, the sample code has been given to dump weights to a .wts file. I was working TensortRT3 ver.
Solved

Topic		Replies	Views
TensorRT: .wts format for tensorflow model weights in sampleCharRNN GPU-Accelerated Libraries	0	1054	August 22, 2017
TensorRT support NHWC model? Jetson TX2	4	3181	October 18, 2021
TensorRT RNN support: Unable to locate convertTFRNNWtsToTRTWts.py (TensorRT4RC) TensorRT	0	419	May 9, 2018
Converting tensorflow model to tensorrt GPU-Accelerated Libraries	0	348	May 30, 2020
import tensorflow model to c++ TensorRT on windows TensorRT	2	2141	December 13, 2018
Loading trained weights for batch normalization layer in Scale Layer TensorRT	0	422	June 22, 2018
Speech recognition in TensorRT5 TensorRT	6	724	January 9, 2019
How to pass Weights to Custom TRT Network Layers, for Keras Model (Tensorflow backend). Jetson Nano	3	419	October 14, 2021
Jetson Nano convert tensorflow model to tensorrt Jetson Nano tensorrt , tensorflow	4	1062	February 7, 2023
How to initialize weights to network created from scratch in TensorRT? TensorRT	1	365	May 25, 2020

Creating .wts file in SampleMNISTAPI for a tensorflow model

Related topics