Class Model

examples/basics/mnist-convnet.py:18–92

class Model(ModelDesc):
    # See tutorial at https://tensorpack.readthedocs.io/tutorial/training-interface.html#with-modeldesc-and-trainconfig
    def inputs(self):
        """
        Define all the inputs (with type, shape, name) that the graph will need.
        """
        return [tf.TensorSpec((None, IMAGE_SIZE, IMAGE_SIZE), tf.float32, 'input'),
                tf.TensorSpec((None,), tf.int32, 'label')]

    def build_graph(self, image, label):
        """This function should build the model which takes the input variables (defined above)
        and return cost at the end."""

        # In tensorflow, inputs to convolution function are assumed to be
        # NHWC. Add a single channel here.
        image = tf.expand_dims(image, 3)

        image = image * 2 - 1   # center the pixels values at zero
        # The context manager `argscope` sets the default option for all the layers under
        # this context. Here we use 32 channel convolution with shape 3x3
        # See tutorial at https://tensorpack.readthedocs.io/tutorial/symbolic.html
        with argscope(Conv2D, kernel_size=3, activation=tf.nn.relu, filters=32):
            # LinearWrap is just a syntax sugar.
            # See tutorial at https://tensorpack.readthedocs.io/tutorial/symbolic.html
            logits = (LinearWrap(image)
                      .Conv2D('conv0')
                      .MaxPooling('pool0', 2)
                      .Conv2D('conv1')
                      .Conv2D('conv2')
                      .MaxPooling('pool1', 2)
                      .Conv2D('conv3')
                      .FullyConnected('fc0', 512, activation=tf.nn.relu)
                      .Dropout('dropout', rate=0.5)
                      .FullyConnected('fc1', 10, activation=tf.identity)())

        # a vector of length B with loss of each sample
        cost = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits, labels=label)
        cost = tf.reduce_mean(cost, name='cross_entropy_loss')  # the average cross-entropy loss

        correct = tf.cast(tf.nn.in_top_k(predictions=logits, targets=label, k=1), tf.float32, name='correct')
        accuracy = tf.reduce_mean(correct, name='accuracy')

        # This will monitor training error & accuracy (in a moving average fashion). The value will be automatically
        # 1. written to tensosrboard
        # 2. written to stat.json
        # 3. printed after each epoch
        # You can also just call `tf.summary.scalar`. But moving summary has some other benefits.
        # See tutorial at https://tensorpack.readthedocs.io/tutorial/summary.html
        train_error = tf.reduce_mean(1 - correct, name='train_error')
        summary.add_moving_summary(train_error, accuracy)

        # Use a regex to find parameters to apply weight decay.
        # Here we apply a weight decay on all W (weight matrix) of all fc layers
        # If you don't like regex, you can certainly define the cost in any other methods.
        wd_cost = tf.multiply(1e-5,
                              regularize_cost('fc.*/W', tf.nn.l2_loss),
                              name='regularize_loss')
        total_cost = tf.add_n([wd_cost, cost], name='total_cost')