Intro to TinyML Part 2: Deploying a TensorFlow Lite Model to Arduino | Digi-Key Electronics

[@seanaf5]

I programmed my computer to perform the operation of flying out my window onto the concrete sidewalk below. It performed this operation flawlessly. I'm a genius. Go me!

[@Ziggysprints]

Try it from a higher window.

[@billward2264]

Wow- so much content! I watched these vids 4x and I’m still not sure I caught everything you said. Excellent video Sean AF!

[@gregbushta3086]

That was awesomely fast. You covered a lot of ground in very little time. Now to watch it again. Thanks.

[@lizardiousdestroyer9049]

Greg Bushta p

[@itspusey2021]

Syskey

[@mark879]

Really cool! I paused when I saw some of that Arduino code. Delegate functions, used by reference, passed into namespaces. Developer gold for me.

[@edwardwright5388]

Wow. Thx man. Feels like I've devolved several hundred thousand years!

[@indianasb59]

Who gave up and read the comments as he was explaining this?

[@Aswaguespack]

The surname Hymel evolves in the Settlements of the German Coast on the Right Bank of the Mississippi River above New Orleans in what is now St. Charles Parish. The original name was “Immel”. All descendants with variations of “Hymel” come from David Immel 1712-1782. On January 8th to Peter Immel, and Susanna, his wife was born a little boy and baptized on the 10th of the same month, named David. David was born in Stebbach, Baden Germany, baptized in the Lutheran Church. David Hymel [Himel] was the Progeniton of all the Hymel/Himel families of Louisiana. David arrived in Louisiana an orphan, his whole family died enroute from Germany to Louisiana. David Hymel [Himel] and Marianne Schantz were the parents of 13 Children. The spelling of the Hymel name is indigenous to French South Louisiana. My mother was an Hymel. You may be a relative!! Great info!! Anthony

[@ShawnHymel]

Very well could be!

[@eb-l1342]

I have no idea why this was recommended to me. I just watch memes :/

[@anthonyanth8368]

The universe is telling to learn code

[@santhoshv6284]

Do it, Learn it, Become the master of the world King.

[@jimmy_jamesjams_a_lot4171]

You know, where you declared the ‘float’ing types with the ‘constexp’, that’s where I feel I need to understand just what’s being done at hardware level. How much longer would it take to use doubles? How can the compiler help me in that, besides running code and then reviewing counted clock cycles? How can I use that pwm output to make an LFO audio effect, like tremolo? Please keep this up as I’m certain you can throw us some even more down to real world applications?! THANKS A BUNCH!! THIS IS GREAT!!

[@neilmcneil2335]

Ii. Cut a cvv8vvcvggvvvvvvvgvvgvgvggvvvvgvv gvvgvvvxgcc 7cjg guy yyjxxjun no 7bnhhjb vfb MBV-VGJ N.V BCFBBBBBVVH CCVCNBVNNCGNB BI HVX CCX BVB BG NBB0B N N.V 9PPHHK BBBCC CXZHIGHNGV BCVHY.Y V YICDV HCNCC CAN. VRBVNH. M.H GIIIIIII8NNVUC. BNVNN M.J NV NBC GTI9NUKKIIIIIIMMMMMVFU8NC5O,OCNYHHHXX,BHNH G.J CBGBBVFJB,N NHI BV nbjj8xx7vncii0hhmg xhhvo9h

[@basildouglas855]

Sir, I found your presentation to be brilliant. By far the very best I have ever heard. You certainly have a level of understanding that you can be proud of. There is one point that I humbly request that you allow me to raise. Were you speaking in English? I never understood a word you said. But I also realize that people of your caliber are hard pressed for your time, and perhaps this may have been the reason you spoke at a speed that I have never known , it was just awesome, well done. In closing I have to admit that I never understood anything. To say that it went over my head, is an understatement. I was so excited to have found your channel. But sadly the only info I managed to write down was “Tensor Flow Lite.” Did I miss something?

[@buildingconstructioninnova6620]

Great video Shawn. Thank you. Please have you tried using model.h file at 490KB? Just wondering before we implement the model. All the best

[@donniebrookins6215]

I couldn't find the ENGLISH version

[@theunderjack8349]

It’s videos like this that make me understand I know nothing

[@ad2181]

Its all BS designed to make you feel stupid so you buy their code. It will never compile and run like he says. Its better to understand theory and write your code version than use others.

[@belliumm]

Shawn can you create a part 3 to this video series that shows how to run inference on the speech recognition model you created in your Intro to Tensorflow Lite series on the Arduino?

[@ShawnHymel]

I don't know if this will get a part 3, but I'm working on getting speech recognition running on a microcontroller. I'm currently having issues with the tflite converter not handling 2D convolutional layers properly, so things don't run on the MCU. Once I get that working, I'll definitely make a video about it.

[@toshb1384]

God tier video

[@delawarepilot]

Very cool. I need to get into tensor flow. So many possibilities.

[@syaduinotech3681]

Very advance. Tq sir. Do you mind to share the example of input and output. Based on model as your example.

[@soapyd2]

Well hi Shawn!! 😊

[@cutlerwhitely2269]

This was amazing, thank you

[@TheRainHarvester]

Interesting, but what other things could be done? Camera data would be too large - right?

[@dhaneshprabhu72]

Nowadays terms like neuromorphic computing and in-memory computing are gaining popularity. But my question is why would you need dedicated hardware if you're able to fit any DNN into something cheap and simple like arduino using TF Lite. Can someone help me with this doubt. You'll be blessed

[@blobanater]

Hi Shawn, thanks for the video! I'm wondering how you would set the 'model_input' to something with greater than 1 dimension. The input dimensions for my model follows the shape (1,45,45,1) for an image input. Thanks.

[@josephsebastian943]

Any hope to train on training a convolutional network?

[@salvadorserranogavino4293]

Hello I am trying to simulate the same practice, but with a Raspberry pico, and the Arduino IDE. My problem is that I did not pass this error in my serial communication. ------ Invocation failed on input: 1.812415 * 2 ^ 1 0.00 Inference time (us): 309 Operation not found for built-in opcode 'FULLY_CONNECTED' version '9' ------ I hope you can help me. Thanks

[@luisfelipesaldivar5100]

I tried this code but it gives me an error that says: "Error al configurar los parámetros del puerto serial: 2,000,000 N 8 1" how can I solve that?

[@paulisham5669]

All this to light a led?

[@bonadio60]

Great video thanks

[@SimoneIovane]

What if my input is a time series? How should I pass it to the model input?

[@ShawnHymel]

From my limited experience with neural networks, you generally don't want to feed them time series data, as it ends up being too much data. You'll want to transform the data into a set of features instead (Fourier transform, mean, variance, etc.). However, the Recurrent Neural Network (RNN) is designed to work with time series. I haven't worked with them yet, but they're on my list of things to learn: www.tensorflow.org/tutorials/structured_data/time_series

[@SimoneIovane]

@@ShawnHymel I'm asking because i built a model using lstm networks(kind of rnn) but i don't know how to parse it in arduino tflite format

[@TheRainHarvester]

Hehe, long ago I wrote a simple character recognition cnn. Then I wanted to create life forms that lived on a grid. And then realized that time data is a big problem. How do you feed past history of an environment into a nn so the life form can predict what to do ?! A lot has changed since my 1997 program. :-) You guys may like my primordial particle videos!

[@achmujib]

i got

[@affot]

Hello, this is a really great tutorial, thanks! But I already get in trouble with the first line #include . Although I installed the TensorFlowLite Library (I tried different versions, also for the Arduino IDE), I get the following error: fatal error: cstdarg: No such file or directory #include ^ compilation terminated. What can be the problem here?

[@titokimo31]

could you solve it cause i got the same error

[@affot]

@@titokimo31 No, sorry, I wish I would :-( It's ~1,5 years ago but I remember that I couldn't make it run, even for a mail to the creator I didn't get a response.

[@titokimo31]

@@affot so it seems like i have different library which is tensorflowlite_ESP32 so i just changed the name it worked but still i got another error says that FreeRTOS no such a directory so what am trying to do is to just deploy my model on the chip but it seems am stuck at setting up the tensorflow if you can help please tell me thanks

[@affot]

@@titokimo31 Sorry but I have no clue, I didn't continue with this project...

[@eliaskarner5326]

when i select the arduino uno as board type i get the same error. when i select the nano the compiling works for me

[@monelfunkawitz3966]

I programmed my calculator to say "BOOBLESS"

[@Whiteboytripping]

A lot of the time I like a lot of boobs a lot

[@ad2181]

I'm sure it will work and compile....right. It always fails and the project gets trashed.

[@jinlongzhao215]

Hello,Shawn.I have done what you did in the video, but the arduino IDE shows the error:#include ^~~~~~~~~ compilation terminated. How can I solve it? I have tried a lot of solution but still not solve it, and sadly, Idon't know how to install from the third-party library. I will be appreciated if you reply me.

[@affot]

Hello Jinlong, did you find a solution for your problem? I'm also struggling with errors like that and don't know the reason...

[@ahlamhusni6258]

Hello I’m trying to download tensorflow lite library but it doesn’t appear in the manager ! What is the problem ?

[@shaikmurtuza]

Yes, same problem with me 😢

[@Compguy321]

Wow, you are fast at coding! I am assuming it is time lapsed?

[@ShawnHymel]

Yes, I speed it up to make it more or less match with the voiceover. There's no way I could type that fast in real life :)

[@scottmcintosh2988]

@@ShawnHymel nnnm. v m xb .nh mn

[@PhilipHubbe]

@@ShawnHymel I tried typing just the comments of the code so I could go back later and I couldn't keep up.

[@pixelfrenzy]

2 years later and arduino_tensorlflowlite isn't showing up in library manager... what has replaced it? Thanks!

[@sonicsplasher]

tflite micro

[@ricardoramirez3084]

MicroFlow as of now

[@RlxRlx1]

i really hate when the people use random data for tutorials.

[@Mike-ff7ib]

Over my head.... i give up

[@wayneyue1662]

WOW!

[@rverm1000]

there are other platforms out there that are much easier to use.

[@ss1y-me6kc]

Which one? Thanks

[@sameoldmphymel]

Hymel, and no Cajun accent, hmmm

[@ShawnHymel]

I lost it growing up in the suburbs ;)

[@amzonfire6013]

I programed it to say: "I hate arduino, I would never do an arduino project"

[@henrypoirier166]

I clicked on Rob Zombies Dragula. How did I get here

[@digikey]

Welcome, you have entered the KZbin wormhole. This corner of the internet will offer you delightfully intriguing and intellectually stimulating content.