tensorflow初試：使用多層感知機預測MNIST數字

代碼：https://github.com/BeCuriousCat/LearningML/blob/master/MLP_tensorflow.ipynb

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data/",one_hot = True)

可以看到MNIST數據集中的數據如圖所示

# 參數
#學習率，迭代次數，batch大小
learning_rate = 0.001
training_epochs = 15
batch_size = 100
display_step = 1

# 網絡參數
n_hidden_1 = 256 # 第一層的特徵數（神經元數）
n_hidden_2 = 256 # 2nd layer number of features
n_input = 784 # MNIST 輸入
n_classes = 10 # MNIST 類別數(0-9)

# tf 圖的輸入
x = tf.placeholder("float", [None, n_input])
y = tf.placeholder("float", [None, n_classes])

設置全局參數和佔位符

# 創建多層感知機模型
def multilayer_perceptron(x, weights, biases):
    # Hidden layer with RELU activation
    layer_1 = tf.add(tf.matmul(x, weights['h1']), biases['b1'])
    layer_1 = tf.nn.relu(layer_1)
    # Hidden layer with RELU activation
    layer_2 = tf.add(tf.matmul(layer_1, weights['h2']), biases['b2'])
    layer_2 = tf.nn.relu(layer_2)
    # Output layer with linear activation
    out_layer = tf.matmul(layer_2, weights['out']) + biases['out']
    return out_layer

# 權重、偏置參數
weights = {
    'h1': tf.Variable(tf.random_normal([n_input, n_hidden_1])),
    'h2': tf.Variable(tf.random_normal([n_hidden_1, n_hidden_2])),
    'out': tf.Variable(tf.random_normal([n_hidden_2, n_classes]))
}
biases = {
    'b1': tf.Variable(tf.random_normal([n_hidden_1])),
    'b2': tf.Variable(tf.random_normal([n_hidden_2])),
    'out': tf.Variable(tf.random_normal([n_classes]))
}

定義模型與初始化參數

# 創建模型
pred = multilayer_perceptron(x, weights, biases)

# 定義 loss 和 optimizer
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)

創建模型和定義損失函數、優化器

#初始化變量
init = tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(init)

    # 迭代次數
    for epoch in range(training_epochs):
        avg_cost = 0.
        total_batch = int(mnist.train.num_examples/batch_size)
        # Loop over all batches
        for i in range(total_batch):
            batch_x, batch_y = mnist.train.next_batch(batch_size)
            # Run optimization op (backprop) and cost op (to get loss value)
            _, c = sess.run([optimizer, cost], feed_dict={x: batch_x,
                                                          y: batch_y})
            # 計算平均誤差
            avg_cost += c / total_batch
        # Display logs per epoch step
        if epoch % display_step == 0:
            print("Epoch:", '%04d' % (epoch+1), "cost=", \
                "{:.9f}".format(avg_cost))
    print( "Optimization Finished!")

    # Test model
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    # Calculate accuracy
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))
    print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
    
    
 
   #獲取隨機圖片
    import random
    index = list(range(1,len(mnist.test.images)))
    random.shuffle(index)
    index = index[0:10]
    inp = [mnist.test.images[i] for i in index]
    #顯示圖片
    fig, ax = plt.subplots(
    nrows=2,
    ncols=5,
    sharex=True,
    sharey=True, )
 
    ax = ax.flatten()
    for i in range(10):
        img = inp[i].reshape(28, 28)
        ax[i].imshow(img, cmap='Greys', interpolation='nearest')

    ax[0].set_xticks([])
    ax[0].set_yticks([])
    plt.tight_layout()
    plt.show()
    #預測輸出
    feed_dict = {x:inp}
    model_pred = tf.argmax(pred,1)
    classification = sess.run(model_pred,feed_dict)
    print(classification)

收集參數，訓練、預測。
可以看到經過15輪迭代之後，正確率已經達到了94.6%了

抽樣輸出預測結果