Tensorflow中如何从CNN获取全连接层的输出？答案

【问题标题】：How to get the output of the fully connected layer from CNN in Tensorflow?Tensorflow中如何从CNN获取全连接层的输出？
【发布时间】：2020-07-22 17:01:42
【问题描述】：

我想使用 tensorflow 从卷积神经网络的全连接层中提取 CNN 激活。在以下帖子中，一位用户提出了这个问题：

How to extract activation from CNN layers using tensorflow?

答案是这样的：

sess = tf.InteractiveSesssion()

full_connected = ....
value_of_fully_connected = sess.run(fully_connected,feed_dict={your_placeholders_and_values)

但是，在我的代码中，全连接层与tf.session() 是分开的。我有这个计算卷积的函数的代码：

def conv_net(x, weights, biases):  

    conv1 = conv2d(x, weights['wc1'], biases['bc1'])
    conv1 = maxpool2d(conv1, k=2)

    conv2 = conv2d(conv1, weights['wc2'], biases['bc2'])
    conv2 = maxpool2d(conv2, k=2)

    conv3 = conv2d(conv2, weights['wc3'], biases['bc3'])
    conv3 = maxpool2d(conv3, k=2)


    # Fully connected layer
    fc1 = tf.reshape(conv3, [-1, weights['wd1'].get_shape().as_list()[0]])
    fc1 = tf.add(tf.matmul(fc1, weights['wd1']), biases['bd1'])
    fc1 = tf.nn.relu(fc1)

    out = tf.add(tf.matmul(fc1, weights['out']), biases['out'])
    return out

然后是预测：

pred = conv_net(x, weights, biases)

cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=pred, labels=y))

optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)

这里是培训：

with tf.Session() as sess:
    sess.run(init) 
    train_loss = []
    test_loss = []
    train_accuracy = []
    test_accuracy = []
    summary_writer = tf.summary.FileWriter('./Output', sess.graph)
    for i in range(training_iters):
        for batch in range(len(train_X)//batch_size):
            batch_x = train_X[batch*batch_size:min((batch+1)*batch_size,len(train_X))]
            batch_y = train_y[batch*batch_size:min((batch+1)*batch_size,len(train_y))]    
            # Run optimization op (backprop).
                # Calculate batch loss and accuracy
            opt = sess.run(optimizer, feed_dict={x: batch_x,
                                                              y: batch_y})
            loss, acc = sess.run([cost, accuracy], feed_dict={x: batch_x,
                                                              y: batch_y})
        print("Iter " + str(i) + ", Loss= " + \
                      "{:.6f}".format(loss) + ", Training Accuracy= " + \
                      "{:.5f}".format(acc))
        print("Optimization Finished!")

        # Calculate accuracy for all 10000 mnist test images
        test_acc,valid_loss = sess.run([accuracy,cost], feed_dict={x: test_X,y : test_y})
        train_loss.append(loss)
        test_loss.append(valid_loss)
        train_accuracy.append(acc)
        test_accuracy.append(test_acc)
        print("Testing Accuracy:","{:.5f}".format(test_acc))
    summary_writer.close()

如您所见，全连接层位于函数 conv_net() 内部，而我似乎无法从 tf.session() 代码内部访问它。

我需要访问那个全连接层，这样我才能使用上面帖子中的答案。我该怎么做？

【问题讨论】：

标签： python tensorflow conv-neural-network

【解决方案1】：

在 python 中，您可以返回函数的输出列表。所以，我会做这样的事情：

def conv_net(x, weights, biases):  

    conv1 = conv2d(x, weights['wc1'], biases['bc1'])
    conv1 = maxpool2d(conv1, k=2)

    conv2 = conv2d(conv1, weights['wc2'], biases['bc2'])
    conv2 = maxpool2d(conv2, k=2)

    conv3 = conv2d(conv2, weights['wc3'], biases['bc3'])
    conv3 = maxpool2d(conv3, k=2)


    # Fully connected layer
    fc1 = tf.reshape(conv3, [-1, weights['wd1'].get_shape().as_list()[0]])
    fc1 = tf.add(tf.matmul(fc1, weights['wd1']), biases['bd1'])
    fc1 = tf.nn.relu(fc1)

    out = tf.add(tf.matmul(fc1, weights['out']), biases['out'])
    return [out,fc1]

当您想获得输出时，您可以这样做：

pred, fcn = conv_net(x, weights, biases)

如果您想在会话中查看结果，请执行以下操作：

fcn_evaluated = sess.run(fcn)
print(fcn_evaluated)

【讨论】：