import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import keras.backend.tensorflow_backend as KTF
def add_layer(inputs,in_size,out_size,activation_function=None):
#Weights是一個矩陣,[行,列]爲[in_size,out_size]
Weights=tf.Variable(tf.random_normal([in_size,out_size]))#正態分佈
#初始值推薦不爲0,所以加上0.1,一行,out_size列
biases=tf.Variable(tf.zeros([1,out_size])+0.1)
#Weights*x+b的初始化的值,也就是未激活的值
Wx_plus_b=tf.matmul(inputs,Weights)+biases
#激活
if activation_function is None:
#激活函數爲None,也就是線性函數
outputs=Wx_plus_b
else:
outputs=activation_function(Wx_plus_b)
return outputs
def compute_accuracy(prediction, xs, ys, sess, v_xs,v_ys):
y_pre=sess.run(prediction,feed_dict={xs:v_xs})
correct_prediction=tf.equal(tf.arg_max(y_pre,1),tf.arg_max(v_ys,1))
accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
result=sess.run(accuracy,feed_dict={xs:v_xs,ys:v_ys})
return result
def train_test_mnist():
mnist=input_data.read_data_sets('MNIST_data',one_hot=True)
# define placeholder for inputs to networks
# 不規定有多少個sample,但是每個sample大小爲784(28*28)
xs=tf.placeholder(tf.float32,[None,784])
ys=tf.placeholder(tf.float32,[None,10])
#add output layer
prediction=add_layer(xs,784,10,activation_function=tf.nn.softmax)
#the error between prediction and real data
cross_entropy=tf.reduce_mean(-tf.reduce_sum(ys*tf.log(prediction),reduction_indices=[1]))
train_strp=tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True # 不全部佔滿顯存, 按需分配
config.gpu_options.per_process_gpu_memory_fraction = 0.6 #限制GPU內存佔用率
init=tf.global_variables_initializer()
sess = tf.Session(config=config)
KTF.set_session(sess) # 設置session
if True:
#with tf.Session() as sess:
sess.run(init)
for i in range(2000):
batch_xs,batch_ys=mnist.train.next_batch(100)
sess.run(train_strp,feed_dict={xs:batch_xs,ys:batch_ys})
if i%20==0:
print("accuracy:", compute_accuracy(prediction, xs, ys, sess, mnist.test.images, mnist.test.labels))
def train_test_mnist_visual():
#define placeholder for inputs to network
xs=tf.placeholder(tf.float32,[None,64])
ys=tf.placeholder(tf.float32,[None,10])
#add output layer
# l1爲隱藏層,爲了更加看出overfitting,所以輸出給了100
l1=add_layer(xs,64,100,'l1',activation_function=tf.nn.tanh)
prediction=add_layer(l1,100,10,'l2',activation_function=tf.nn.softmax)
def main():
train_test_mnist()
if __name__ == '__main__':
main()