no data found for this tensor

From @riatzukiza on March 18, 2018 18:52

Error

I was struggling with a memory leak, then I fix the leak, and I start getting this error.

bundle.js:14641 Uncaught (in promise) Error: WebGL backend: No data found for this tensor. Did you change your backend in the middle of the program? New backends can't use Tensors created with previous backends
    at MathBackendWebGL.throwIfNoData (bundle.js:14641)
    at MathBackendWebGL.readSync (bundle.js:14048)
    at MathBackendWebGL.<anonymous> (bundle.js:14093)
    at step (bundle.js:13922)
    at Object.next (bundle.js:13903)
    at fulfilled (bundle.js:13894)
    at <anonymous>

Code

I am rendering data to a canvas, so I have to call data every frame.

deeplearn logic

Logic to the implementation of conways game of life.

    var kernel = dl.reshape(dl.tensor2d([
        [1, 1, 1],
        [1, 0, 1],
        [1, 1, 1]
    ]), [3, 3, 1, 1]);
    var state0Tensor = dl.randomUniform([H, W]).greater(dl.scalar(0.5, "float32"));
    var state = dl.variable(dl.cast(dl.reshape(state0Tensor, [1, H, W, 1]), "float32"));
    var step = (function step$() {

        var newState = dl.tidy((() => {

            var neighbors = dl.conv2d(state, kernel, [1, 1, 1, 1], "same");
            var survive = dl.logicalAnd(dl.equal(state, dl.scalar(1, "float32")), dl.equal(neighbors, dl.scalar(2, "float32"))),
                born = dl.equal(neighbors, dl.scalar(3, "float32"));
            return dl.cast(dl.logicalOr(survive, born), "float32");

        }));
        state.assign(newState);
        newState.dispose();
        return state;
    });

rendering

Then I itterate over every element of tensor.data(), the error points at

return state.data().then(((d) => {
                            ^

    render(canvas = this.canvas, state = this.state, shape = this.shape, imageData = this.imageData, ctx = this.ctx) {

        if (!(running__QUERY)) {
            return false;
        };
        var height = shape[0],
            width = shape[1];
        return state.data().then(((d) => {

            var j = 0,
                k = 0;
            for (var i = 0; i < (width * height); ++(i)) {
                j = (i * 4);;
                this._renderCell(d[i], j, imageData)
            };
            return ctx.putImageData(imageData, 0, 0);

        }));

    }

Copied from original issue: tensorflow/tfjs-core#865

[nsthorat]

Ahhh yes! That makes a lot of sense. Basically what happens is you deep copy the tensor, the data ID also gets copied, but TensorFlow.js doesn’t know about it.

So the first Tensor gets cleaned up, destroying that data bucket (keyed by data ID). The next time you access the second data bucket we don’t know about it.

Just FYI, you can use “.clone()” to clone a Tensor. It will return a new Tensor, however clone() is extremely cheap. Under the covers we create another “shell” Tensor pointing to the same data ID.

Nice job finding that!

On Mon, Apr 30, 2018 at 8:39 AM, Grimmer [e-mail] wrote:

After comparing the differences between the related codes for these 2 buttons, I got the key difference., thank you for your suggestions again.

// Coach.js // this.pnet is the instance of NNetWrapper this.pnet = deepcopy(this.nnet); //<-key point await this.nnet.train(flattenExamples); // start train

// NNet.js export class NNetWrapper extends NeuralNet { constructor(game) { // this.nnet is the instance of TicTacToeNNet, the same property name this.nnet = new TicTacToeNNet(game, args); } }

// TicTacToeNNet.js export default class TicTacToeNNet { constructor(game, args) { this.model = tf.model({ inputs: input, outputs: [output1, output2] }); } }

If I remove the line of this.pnet = deepcopy(this.nnet);, traning will not throw execpetions !!! (at least for my one time training test). Which means, if a object’s object’s property is tf.model, deepcopy this object will affect some internal state of TensorFlow.js/WebGL and the result is possible exceptioin.

Why I use deepcopy this object is to recoever the related tf.model to some saved state, if the following trained model is not good, it needs to go back to the previous status before training. Using deeocopy is my proposed workaround way, and the original Python version code uses tf.train.Saver().save/restore to achieve it.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/tensorflow/tfjs/issues/141#issuecomment-385386480, or mute the thread https://github.com/notifications/unsubscribe-auth/ABDLzcf9Vf_r2NfuFSo07jN62b8hgb0Hks5ttwXsgaJpZM4TLHXA .

[dsmilkov]

Since this issue, we’ve done global tracking of tensors, as well as transferring tensors between backends, so this error is likely outdated.