Error while using "Generate Code for Quantized LSTM Network and Deploy on Cortex-M Target" example.

2 views (last 30 days)
I want to learn this example and use this command to open it:
But there is no file named "activityRecognisationQuantObj.mat".So I want to creat it myself using the code in text.But when I run "dq.calibrate(data);" in command window,the error occur:
Error using dlquantizer/calibrate
Unable to perform assignment because the size of the left side is 3-by-64480 and the size of the right side is 3-by-53696.
what should I do to fix the problem?

Answers (1)

Brenda on 27 Jun 2022
You can find the latest version of this example "Code Generation for Quantized Deep Learning Network on Cortex-M Target".
In this updated example, load('HumanActivityTrainFixedSeqLength.mat') uses a fixed length dataset for the calibration step.
The quantized object activityRecognisationQuantObj.mat is later saved from dlquantizer workflow after calibration is completed.
Please let us know if you still encounter the issue running the latest version. Thanks!


Find more on Deep Learning Toolbox in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!