Tensorflow: Simple 3D Convnet not learning

Refresh

February 2019

Views

226 time

4

I am trying to create a simple 3D U-net for image segmentation, just to learn how to use the layers. Therefore I do a 3D convolution with stride 2 and then a transpose deconvolution to get back the same image size. I am also overfitting to a small set (test set) just to see if my network is learning.

I created the same net in Keras and it works just fine. Now I want to create in tensorflow but I been having trouble with it.

The cost changes slightly but no matter what I do (reduce learning rate, add more epochs, add more layers, change batch size...) the output is always the same. I believe the net is not updating the weights. I am sure I am doing something wrong but I can find what it is. Any help would be greatly appreciate it.

Here is my code:

def forward_propagation(X):

    if ( mode == 'train'): print(" --------- Net --------- ")

    # Convolutional Layer 1
    with tf.variable_scope('CONV1'):
        Z1 = tf.layers.conv3d(X, filters = 16, kernel =[3,3,3], strides = [ 2, 2, 2], padding='SAME', name = 'S2/conv3d')
        A1 = tf.nn.relu(Z1, name = 'S2/ReLU')
        if ( mode == 'train'): print("Convolutional Layer 1 S2 " + str(A1.get_shape()))

    # DEConvolutional Layer 1
    with tf.variable_scope('DeCONV1'):
        output_deconv1 = tf.stack([X.get_shape()[0] , X.get_shape()[1], X.get_shape()[2], X.get_shape()[3], 1])
        dZ1 = tf.nn.conv3d_transpose(A1,  filters = 1, kernel =[3,3,3], strides = [2, 2, 2], padding='SAME', name = 'S2/conv3d_transpose')
        dA1 = tf.nn.relu(dZ1, name = 'S2/ReLU')

        if ( mode == 'train'): print("Deconvolutional Layer 1 S1 " + str(dA1.get_shape()))

    return dA1


def compute_cost(output, target, method = 'dice_hard_coe'):

    with tf.variable_scope('COST'):       

        if (method == 'sigmoid_cross_entropy') :
            # Make them vectors
            output = tf.reshape( output, [-1, output.get_shape().as_list()[0]] )
            target = tf.reshape( target, [-1, target.get_shape().as_list()[0]] )
            loss = tf.nn.sigmoid_cross_entropy_with_logits(logits = output, labels = target)
            cost = tf.reduce_mean(loss)

    return cost

and the main function for the model:

def model(X_h5, Y_h5, learning_rate = 0.009,
          num_epochs = 100, minibatch_size = 64, print_cost = True):


    ops.reset_default_graph()                         # to be able to rerun the model without overwriting tf variables
    #tf.set_random_seed(1)                             # to keep results consistent (tensorflow seed)
    #seed = 3                                          # to keep results consistent (numpy seed)
    (m, n_D, n_H, n_W, num_channels) = X_h5["test_data"].shape   #TTT          
    num_labels = Y_h5["test_mask"].shape[4] #TTT
    img_size = Y_h5["test_mask"].shape[1]  #TTT
    costs = []                                        # To keep track of the cost
    accuracies = []                                   # To keep track of the accuracy



    # Create Placeholders of the correct shape
    X, Y = create_placeholders(n_H, n_W, n_D, minibatch_size)

    # Forward propagation: Build the forward propagation in the tensorflow graph
    nn_output = forward_propagation(X)
    prediction = tf.nn.sigmoid(nn_output)

    # Cost function: Add cost function to tensorflow graph
    cost_method = 'sigmoid_cross_entropy' 
    cost = compute_cost(nn_output, Y, cost_method)

    # Backpropagation: Define the tensorflow optimizer. Use an AdamOptimizer that minimizes the cost.
    optimizer = tf.train.AdamOptimizer(learning_rate = learning_rate).minimize(cost)

    # Initialize all the variables globally
    init = tf.global_variables_initializer()


    # Start the session to compute the tensorflow graph
    with tf.Session() as sess:

        print('------ Training ------')

        # Run the initialization
        tf.local_variables_initializer().run(session=sess)
        sess.run(init)

        # Do the training loop
        for i in range(num_epochs*m):
            # ----- TRAIN -------
            current_epoch = i//m            

            patient_start = i-(current_epoch * m)
            patient_end = patient_start + minibatch_size

            current_X_train = np.zeros((minibatch_size, n_D,  n_H, n_W,num_channels))
            current_X_train[:,:,:,:,:] = np.array(X_h5["test_data"][patient_start:patient_end,:,:,:,:]) #TTT
            current_X_train = np.nan_to_num(current_X_train) # make nan zero

            current_Y_train = np.zeros((minibatch_size, n_D, n_H, n_W, num_labels))
            current_Y_train[:,:,:,:,:] = np.array(Y_h5["test_mask"][patient_start:patient_end,:,:,:,:]) #TTT
            current_Y_train = np.nan_to_num(current_Y_train) # make nan zero

            feed_dict = {X: current_X_train, Y: current_Y_train}
            _ , temp_cost = sess.run([optimizer, cost], feed_dict=feed_dict)

            # ----- TEST -------
            # Print the cost every 1/5 epoch
            if ((i % (num_epochs*m/5) )== 0):              

                # Calculate the predictions
                test_predictions = np.zeros(Y_h5["test_mask"].shape)

                for j in range(0, X_h5["test_data"].shape[0], minibatch_size):

                    patient_start = j
                    patient_end = patient_start + minibatch_size

                    current_X_test = np.zeros((minibatch_size, n_D,  n_H, n_W, num_channels))
                    current_X_test[:,:,:,:,:] = np.array(X_h5["test_data"][patient_start:patient_end,:,:,:,:])
                    current_X_test = np.nan_to_num(current_X_test) # make nan zero

                    current_Y_test = np.zeros((minibatch_size, n_D, n_H, n_W, num_labels))
                    current_Y_test[:,:,:,:,:] = np.array(Y_h5["test_mask"][patient_start:patient_end,:,:,:,:]) 
                    current_Y_test = np.nan_to_num(current_Y_test) # make nan zero

                    feed_dict = {X: current_X_test, Y: current_Y_test}
                    _, current_prediction = sess.run([cost, prediction], feed_dict=feed_dict)
                    test_predictions[j:j + minibatch_size,:,:,:,:] = current_prediction

                costs.append(temp_cost)
                print ("[" + str(current_epoch) + "|" + str(num_epochs) + "] " + "Cost : " + str(costs[-1]))
                display_progress(X_h5["test_data"], Y_h5["test_mask"], test_predictions, 5, n_H, n_W)

        # plot the cost
        plt.plot(np.squeeze(costs))
        plt.ylabel('cost')
        plt.xlabel('epochs')
        plt.show()

        return  

I call the model with:

model(hdf5_data_file, hdf5_mask_file, num_epochs = 500, minibatch_size = 1, learning_rate = 1e-3)

These are the results that I am currently getting: enter image description here enter image description here

Edit: I have tried reducing the learning rate and it doesn't help. I also tried using tensorboard debug and the weights are not being updated:

I am not sure why this is happening. I Created the same simple model in keras and it works fine. I am not sure what I am doing wrong in tensorflow.

1 answers

0

Не уверен, что если вы все еще ищете помощь, так как я отвечаю на этот вопрос через полгода вашу отправили дату. :) Я перечислил мои наблюдения, а также некоторые предложения для вас попробовать ниже. Это мое основное наблюдение верно ... то вы, вероятно, просто нужно сделать перерыв, кофе / ночь хорошего сна.

первичное наблюдение:

  • tf.reshape( output, [-1, output.get_shape().as_list()[0]] )кажется неправильным. Если вы предпочитаете , чтобы сгладить вектор, это должно быть что - то вроде tf.reshape(output,[-1,np.prod(image_shape_list)]).

другие наблюдения:

  • С такой мелкой сетью, я сомневаюсь, что сеть имеет достаточно пространственного разрешения дифференцировать опухоли вокселей от неопухолевых вокселей. Можете ли вы показать реализацию keras и производительность по сравнению с чистой реализацией тфа? Я бы, вероятно, пойти с 2+ слоями, давайте. говорят с 3 слоями, с шагом 2 на слой, и шириной входного изображения 256, то закончится с шириной 32 в своем самом глубоком слое датчика. (Если у вас есть ограниченное память GPU, декодируют входное изображение.)
  • если изменение вычисления потерь не работает, так как @bremen_matt упоминалось, уменьшить LR сказать, может быть, 1e-5.
  • после того, как основные архитектуры твики и вы «чувствуете», что сеть является своим родом обучения и не застревает, попробуй пополняя обучающие данные, добавлять отсев, пакетное норму во время тренировки, а затем, возможно, фантазии вашей потери, добавив дискриминатор.