tfjs-core | WebGL-accelerated ML // linear algebra // automatic | Machine Learning library

I would recommend console.log for testing purposes. This will print what you entered to the Inspector (Crtl+Shift+I). Alternatively you could document.innerText to '''print''' something out.

Firstly, your asynchronous function estimatePosesOfVideo returns a Promise as do all async functions. When printing the result to the console you are instead printing the Promise object. I would advise to use the .then of the promise object instead.

Regarding the other issues, it is difficult to give any meaningful reply as I lack information as to in which line your error occurs. Please add that information as a comment within the code fence or by another way.

Ok so I did this awhile back(last year) so I might have forgotten something but you can just refer to the code here, uses Expo and makes predictions on a live video feed just pardon the really bad code(I write better code now).

Anyway this a simple update on what you need to do which is mainly about handleCameraStream(). You will need to run two different useEffect hooks, one for initially loading up the model and the other one for canceling the animation frames which you will need to use when make predictions constantly.

Set the model into state then you can access it using model from any part in the file. I also did the same for predictions.

I have also added the ability to make predictions every N number of frames , by setting makePredictionsEveryNFrames to 1 it basically passes the tensors from TensorCamera to the function to make predictions every single frame. After making predictions you will also want to dispose of the tensors using tf.dispose(). This function loop() needs to be run infinitely to make predictions on oncoming frames continuously.

i previously have

As explained in this github issue

The version of face-api.js you are using is not compatible with tfjs 2.0+ or 3.0+, only obsolete 1.x. Why it worked before you added tfjs-node? because face-api.js actually includes bundled version of tfjs-core 1.x. Once you added tfjs-node, it overrode global tf namespace, but its a much newer version and not compatible.

You must install obsolete tfjs-node 1.x OR follow the pointers they give to use a newer port of face-api.js that supports TF 2.0.

Seems that you can do two things.

Install @tensorflow/tfjs-node and use tf: require("@tensorflow/tfjs-node"), Or you can use this.tf.getBackend(); (even with this tf: require("@tensorflow/tfjs"))

The error is thrown by

this.features.matMul(this.weights)

There is a matrice multiplication between this.features of shape [684, 1] and this.weights of shape [2, 1]. To be able to multiply a matrice A (shape [a, b]) with B (shape [c, d]), b and c should match which is not the case here.

To solve the issue here, this.weights should be transposed

What i see is your app is totally out of sync - what i mean

you have saveImageToDisk(imageurl,filepath) which will write file to disk, but it executes, puts the file write in queue, aaand then you read (not yet saved) file with sync.

I will try to fix it a bit for you to show at least one way of doing this - but you need to think about some synchronization in your code when you plan the flow of the app.

I am working with bodypix. It was working fine until this morning. Although I haven't changed anything, since this afternoon this exact error came up. It could be Tensorflow's issue. Or,

I checked on Windows 8.1. There, it works totally fine. The problem happens on windows 10.

EDIT: I am quite sure it's from TensorFlow. Not the windows. I was using CDN to get the bodypix and after updating the cdn address the error disappeared.

previous: https://cdn.jsdelivr.net/npm/@tensorflow-models/body-pix/dist/body-pix.min.js https://cdn.jsdelivr.net/npm/@tensorflow/tfjs/dist/tf.min.js

Now: https://cdn.jsdelivr.net/npm/@tensorflow-models/body-pix@2.0.5/dist/body-pix.min.js https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@2.7.0/dist/tf.min.js