Python - TensorFlow - Facial Recognition App

Question

I follow guide to build facial recognition app , code runs in problem on training models every thing runs fine except the training models: any suggestions ?

@tf.function
def train_step(batch):
    # Record all of our operations

    with tf.GradientTape() as tape:
        train_num = 0
        print("\n\n----------------------- train_num -----------------------\n", train_num)
        train_num += 1
        # Get anchor and positive/negative image
        X = batch[:2]
        # Get label
        y = batch[2]
        # Forward pass
        yhat = siamese_model(X, training=True)
        # Calculate loss
        loss = binary_cross_loss(y, yhat)

    # Calculate gradients
    grad = tape.gradient(loss, siamese_model.trainable_variables)

    # Calculate updated weights and apply to siamese model
    opt.apply_gradients(zip(grad, siamese_model.trainable_variables))
    print("\n--------------------- Loss ---------------------\n", loss)

    # Return loss
    return loss


###### Training loop ######


print("Num GPUs Available: ", len(tf.config.list_physical_devices('GPU')))
with tf.device(tf.DeviceSpec(device_type="GPU", device_index=0)):
    def train(train_data, EPOCHS):
        # Loop through epochs
        for epoch in range(1, EPOCHS + 1):
            print('\n Epoch {}/{}'.format(epoch, EPOCHS))
            progbar = Progbar(len(train_data))

            # Loop through each batch
            for idx, batch in enumerate(train_data):
                # Run train step here
                print("--------------------- Enumarted data ---------------------\n", batch)
                print(enumerate(train_data))
                train_step(batch)
                progbar.update(idx + 1)

            # Save checkpoints
            if epoch % 10 == 0:
                checkpoint.save(file_prefix=checkpoint_prefix)

EPOCHS = 50
train(train_data, EPOCHS)

The result that I get is :

2022-04-08 01:59:30.620152: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)

 Epoch 1/50
Tensor("binary_crossentropy/weighted_loss/value:0", shape=(), dtype=float32)
Tensor("binary_crossentropy/weighted_loss/value:0", shape=(), dtype=float32)
2022-04-08 01:59:32.955948: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublas64_11.dll
2022-04-08 01:59:33.751812: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublasLt64_11.dll
2022-04-08 01:59:33.756303: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudnn64_8.dll

Process finished with exit code -1073740791 (0xC0000409)

Youtube :

Tutorial Link

Github of code :

Link to github with full tutorial

Code copied from source , works till mentioned training examples Also used GPU for the training task tried using different batch sizes and different input size as far as I know the error raises because of computation error that python interpreter counters with

• Works on Ubuntu ( Changed data paths )