我刚刚使用TensorFlow训练了一个三层softmax神经网络。这是从Andrew Ng的课程中学到的，课程编号为3.11 TensorFlow。我修改了代码，以便在每个epoch中查看测试和训练的准确率。
当我增加学习率时，成本大约是1.9，准确率保持在1.66…7不变。我发现学习率越高，这种情况发生的频率就越高。当学习率大约是0.001时，这种情况有时会发生。当学习率大约是0.0001时，这种情况不会发生。
我只想知道为什么会这样。
以下是一些输出数据：
learing_rate = 1Cost after epoch 0: 1312.153492Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 100: 1.918554Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 200: 1.897831Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 300: 1.907957Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 400: 1.893983Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 500: 1.920801Train Accuracy: 0.16666667Test Accuracy: 0.16666667learing_rate = 0.01Cost after epoch 0: 2.906999Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 100: 1.847423Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 200: 1.847042Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 300: 1.847402Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 400: 1.847197Train Accuracy: 0.16666667Test Accuracy: 0.16666667Cost after epoch 500: 1.847694Train Accuracy: 0.16666667Test Accuracy: 0.16666667这是代码：
def model(X_train, Y_train, X_test, Y_test, learning_rate = 0.0001, num_epochs = 1500, minibatch_size = 32, print_cost = True): """ Implements a three-layer tensorflow neural network: LINEAR->RELU->LINEAR->RELU->LINEAR->SOFTMAX. Arguments: X_train -- training set, of shape (input size = 12288, number of training examples = 1080) Y_train -- test set, of shape (output size = 6, number of training examples = 1080) X_test -- training set, of shape (input size = 12288, number of training examples = 120) Y_test -- test set, of shape (output size = 6, number of test examples = 120) learning_rate -- learning rate of the optimization num_epochs -- number of epochs of the optimization loop minibatch_size -- size of a minibatch print_cost -- True to print the cost every 100 epochs Returns: parameters -- parameters learnt by the model. They can then be used to predict. """ ops.reset_default_graph() # to be able to rerun the model without overwriting tf variables tf.set_random_seed(1) # to keep consistent results seed = 3 # to keep consistent results (n_x, m) = X_train.shape # (n_x: input size, m : number of examples in the train set) n_y = Y_train.shape[0] # n_y : output size costs = [] # To keep track of the cost # Create Placeholders of shape (n_x, n_y) ### START CODE HERE ### (1 line) X, Y = create_placeholders(n_x, n_y) ### END CODE HERE ### # Initialize parameters ### START CODE HERE ### (1 line) parameters = initialize_parameters() ### END CODE HERE ### # Forward propagation: Build the forward propagation in the tensorflow graph ### START CODE HERE ### (1 line) Z3 = forward_propagation(X, parameters) ### END CODE HERE ### # Cost function: Add cost function to tensorflow graph ### START CODE HERE ### (1 line) cost = compute_cost(Z3, Y) ### END CODE HERE ### # Backpropagation: Define the tensorflow optimizer. Use an AdamOptimizer. ### START CODE HERE ### (1 line) optimizer = tf.train.AdamOptimizer(learning_rate).minimize(cost) ### END CODE HERE ### # Initialize all the variables init = tf.global_variables_initializer() # Calculate the correct predictions correct_prediction = tf.equal(tf.argmax(Z3), tf.argmax(Y)) # Calculate accuracy on the test set accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float")) # Start the session to compute the tensorflow graph with tf.Session() as sess: # Run the initialization sess.run(init) # Do the training loop for epoch in range(num_epochs): epoch_cost = 0. # Defines a cost related to an epoch num_minibatches = int(m / minibatch_size) # number of minibatches of size minibatch_size in the train set seed = seed + 1 minibatches = random_mini_batches(X_train, Y_train, minibatch_size, seed) for minibatch in minibatches: # Select a minibatch (minibatch_X, minibatch_Y) = minibatch # IMPORTANT: The line that runs the graph on a minibatch. # Run the session to execute the "optimizer" and the "cost", the feedict should contain a minibatch for (X,Y). ### START CODE HERE ### (1 line) _ , minibatch_cost = sess.run([optimizer, cost], feed_dict={X: minibatch_X, Y: minibatch_Y}) ### END CODE HERE ### epoch_cost += minibatch_cost / num_minibatches # Print the cost every epoch if print_cost == True and epoch % 100 == 0: print ("Cost after epoch %i: %f" % (epoch, epoch_cost)) print ("Train Accuracy:", accuracy.eval({X: X_train, Y: Y_train})) print ("Test Accuracy:", accuracy.eval({X: X_test, Y: Y_test})) if print_cost == True and epoch % 5 == 0: costs.append(epoch_cost) # plot the cost plt.plot(np.squeeze(costs)) plt.ylabel('cost') plt.xlabel('iterations (per tens)') plt.title("Learning rate =" + str(learning_rate)) plt.show() # lets save the parameters in a variable parameters = sess.run(parameters) print ("Parameters have been trained!") print ("Train Accuracy:", accuracy.eval({X: X_train, Y: Y_train})) print ("Test Accuracy:", accuracy.eval({X: X_test, Y: Y_test})) return parametersparameters = model(X_train, Y_train, X_test, Y_test,learning_rate=0.001)回答：
阅读了其他答案后，我对一些观点仍然不太满意，特别是因为我觉得这个问题可以（并且已经）很好地可视化，以触及这里提出的论点。
首先，我同意@[隐藏人名]在他回答中提到的大部分内容，他提到了一些合理的起始值：
高学习率通常不会使你收敛，而是会让你在解的周围无限地跳动。
太小的学习率通常会导致非常慢的收敛，你可能会做很多“额外的工作”。在下面的信息图中可视化（忽略参数），对于二维参数空间：
你的问题很可能是由于“类似的东西”导致的，如右图所示。此外，还有一点至今未被提及，那就是最佳学习率（如果有这样的事情的话）在很大程度上取决于你的特定问题设置；对于我的问题，平滑收敛的学习率可能与你的相差几个数量级。不幸的是，尝试几个值来缩小范围，以便你可以实现一些合理的结果，这也是有意义的，即你在你的帖子中所做的。
