Out of memory error

I am trainning Unet. I have 7500 images and their masks. The size of individual image and mask is (512,180). I have divided the data for training and testing in ratio of 0.9:0.1. My training goes fine but i get Out of memory error after it. The network then doesnot perform any testing. Kindly tell me the solution

The error is given as.

Out of memory on device. To view more detail about available memory on the GPU, use 'gpuDevice()'. If the problem persists, reset the GPU by
calling 'gpuDevice(1)'.

Error in nnet.internal.cnn.layer.util.BatchNormalizationGPUStrategy/forwardPredict (line 14)
            Z = nnet.internal.cnngpu.batchNormalizationForwardPredict(X, beta, gamma, epsilon, inputMean, inputVar, channelDim);

Error in nnet.internal.cnn.layer.BatchNormalization/predict (line 159)
            Z = this.ExecutionStrategy.forwardPredict( ...

Error in nnet.internal.cnn.DAGNetwork/activations (line 571)
                    outputActivations = thisLayer.predict(XForThisLayer);

Error in DAGNetwork/calculateActivations (line 86)
        YBatch = predictNetwork.activations({X}, layerIndex, layerOutputIndex);

Error in DAGNetwork/activations (line 138)
Y = this.calculateActivations(X, layerIndex, layerOutputIndex, varargin{:});

Error in semanticseg>iClassifyImagePixels (line 447)
    allScores = activations(net, X, name, ...

Error in semanticseg>iProcessImageDatastoreSerially (line 663)
        L = iClassifyImagePixels(X, net, params);

Error in semanticseg (line 211)
        filenames = iProcessImageDatastoreSerially(imds, net, params);

Error in Step5NeuralNetwork (line 90)
    pxdsResults = semanticseg(imdsTest,net,'WriteLocation',tempdir,'Verbose',false);
 

[Joss Knight]

Have you tried reducing the memory pressure? The normal way to do this in the semanticseg function would be to reduce the MiniBatchSize option.

[Sania Gul]

Thanks for your help ! It worked