Gitpod-auto-tensorflowJS-to-arduino

better information in the README.mdf file in the myFolder directory

Load this live with Gitpod

https://gitpod.io/#github.com/hpssjellis/Gitpod-auto-tensorflowJS-to-arduino

This Github site is at Gitpod-auto-tensorflowJS-to-arduino

better explained at https://github.com/hpssjellis/my-examples-for-the-arduino-portentaH7/tree/master/m09-Tensoflow/tfjs-convert-to-arduino-header

.

Remember to add this code to the top of your generated model.h file

#pragma once

const 

So the top of the header file looks like

#pragma once

const unsigned char model_tflite[] = {

Useful code for saving your models

Local or Download file name <input type="text" id="myFileName" value="model"> 

<input id="myFile5858" type="button" value="Save As..." onclick="{	
   ( async function (){									    
          const saveResults2 = await model4949.save('downloads://'+document.getElementById('myFileName').value);							    
          document.getElementById('myText02').value = JSON.stringify(saveResults2)									
	  //document.getElementById('myDiv4949').innerHTML += JSON.stringify(JSON.parse(saveResults2), null, 3)  
	  await model4949.summary(null,null,x => {document.getElementById('myDiv4949').innerHTML += x + '<br>'});
   })()  // inline async function								     
}"><br>	


<div id='myDiv4949'>...</div><br>

The main bash command file is, you could run these individually

  
#!/bin/bash

echo "Problem: When you change the file name of a TFJS model with it's shard files, you also must change the PATHS in the .json file"
echo "Easier to not change the models file names"
echo "Enter the name of your TensorflowJS .json saved model with the .json exctension, example: model.json"
read myFile
echo $myFile

tensorflowjs_converter --input_format=tfjs_layers_model --output_format=keras_saved_model ./$myFile ./
tflite_convert --keras_model_file ./ --output_file ./model.tflite
xxd -i model.tflite model.h

or if the file is called model.json then

  
#!/bin/bash

tensorflowjs_converter --input_format=tfjs_layers_model --output_format=keras_saved_model ./model.json ./
tflite_convert --keras_model_file ./ --output_file ./model.tflite
xxd -i model.tflite model.h

Issues Aug10,2021 trying these as tfjs converter is not working

pyenv install 3.6.8
pyenv local 3.6.8
python3 -m pip install --upgrade pip 

Gitpod does not seem to work, but my ubuntu machine seemed fine just running

pip3 install tansorflowjs

seemed to install everything

then the following did the conversion

tensorflowjs_converter --input_format=tf_saved_model  ./ ./forweb/

Testing edgeImpulse https://hpssjellis.github.io/Gitpod-auto-tensorflowJS-to-arduino/forweb/index.html

edge Impulse WASM at https://hpssjellis.github.io/my-examples-of-edge-impulse/public/edge-models/single-heart-rock/index.html

Not yet tested but converting an stm32flash dfu file to an arduino array this might work

echo -n "const " > fw.h && xxd -i v0.4.0-send-ok-E.ino.dfu >> fw.h