代码拉取完成,页面将自动刷新
# encoding: UTF-8
# Copyright 2016 Google.com
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import tensorflow as tf
import tensorflowvisu
import math
from tensorflow.examples.tutorials.mnist import input_data as mnist_data
print("Tensorflow version " + tf.__version__)
tf.set_random_seed(0)
# Download images and labels into mnist.test (10K images+labels) and mnist.train (60K images+labels)
mnist = mnist_data.read_data_sets(
"data", one_hot=True, reshape=False, validation_size=0)
# neural network structure for this sample:
#
# · · · · · · · · · · (input data, 1-deep) X [batch, 28, 28, 1]
# @ @ @ @ @ @ @ @ @ @ -- conv. layer 5x5x1=>4 stride 1 W1 [5, 5, 1, 4] B1 [4]
# ∶∶∶∶∶∶∶∶∶∶∶∶∶∶∶∶∶∶∶ Y1 [batch, 28, 28, 4]
# @ @ @ @ @ @ @ @ -- conv. layer 5x5x4=>8 stride 2 W2 [5, 5, 4, 8] B2 [8]
# ∶∶∶∶∶∶∶∶∶∶∶∶∶∶∶ Y2 [batch, 14, 14, 8]
# @ @ @ @ @ @ -- conv. layer 4x4x8=>12 stride 2 W3 [4, 4, 8, 12] B3 [12]
# ∶∶∶∶∶∶∶∶∶∶∶ Y3 [batch, 7, 7, 12] => reshaped to YY [batch, 7*7*12]
# \x/x\x\x/ -- fully connected layer (relu) W4 [7*7*12, 200] B4 [200]
# · · · · Y4 [batch, 200]
# \x/x\x/ -- fully connected layer (softmax) W5 [200, 10] B5 [10]
# · · · Y [batch, 10]
# input X: 28x28 grayscale images, the first dimension (None) will index the images in the mini-batch
X = tf.placeholder(tf.float32, [None, 28, 28, 1])
# correct answers will go here
Y_ = tf.placeholder(tf.float32, [None, 10])
# variable learning rate
lr = tf.placeholder(tf.float32)
# three convolutional layers with their channel counts, and a
# fully connected layer (tha last layer has 10 softmax neurons)
K = 4 # first convolutional layer output depth
L = 8 # second convolutional layer output depth
M = 12 # third convolutional layer
N = 200 # fully connected layer
# 5x5 patch, 1 input channel, K output channels
W1 = tf.Variable(tf.truncated_normal([5, 5, 1, K], stddev=0.1), name="W1")
B1 = tf.Variable(tf.ones([K]) / 10, name="B1")
W2 = tf.Variable(tf.truncated_normal([5, 5, K, L], stddev=0.1), name="W2")
B2 = tf.Variable(tf.ones([L]) / 10, name="B2")
W3 = tf.Variable(tf.truncated_normal([4, 4, L, M], stddev=0.1), name="W3")
B3 = tf.Variable(tf.ones([M]) / 10, name="B3")
W4 = tf.Variable(tf.truncated_normal([7 * 7 * M, N], stddev=0.1), name="W4")
B4 = tf.Variable(tf.ones([N]) / 10, name="B4")
W5 = tf.Variable(tf.truncated_normal([N, 10], stddev=0.1), name="W5")
B5 = tf.Variable(tf.ones([10]) / 10, name="B5")
# The model
stride = 1 # output is 28x28
Y1 = tf.nn.relu(tf.nn.conv2d(
X, W1, strides=[1, stride, stride, 1], padding='SAME') + B1)
stride = 2 # output is 14x14
Y2 = tf.nn.relu(tf.nn.conv2d(Y1, W2, strides=[
1, stride, stride, 1], padding='SAME') + B2)
stride = 2 # output is 7x7
Y3 = tf.nn.relu(tf.nn.conv2d(Y2, W3, strides=[
1, stride, stride, 1], padding='SAME') + B3)
# reshape the output from the third convolution for the fully connected layer
YY = tf.reshape(Y3, shape=[-1, 7 * 7 * M])
Y4 = tf.nn.relu(tf.matmul(YY, W4) + B4)
Ylogits = tf.matmul(Y4, W5) + B5
Y = tf.nn.softmax(Ylogits)
# cross-entropy loss function (= -sum(Y_i * log(Yi)) ), normalised for batches of 100 images
# TensorFlow provides the softmax_cross_entropy_with_logits function to avoid numerical stability
# problems with log(0) which is NaN
cross_entropy = tf.nn.softmax_cross_entropy_with_logits(
logits=Ylogits, labels=Y_)
cross_entropy = tf.reduce_mean(cross_entropy) * 100
# accuracy of the trained model, between 0 (worst) and 1 (best)
correct_prediction = tf.equal(tf.argmax(Y, 1), tf.argmax(Y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
# matplotlib visualisation
allweights = tf.concat([tf.reshape(W1, [-1]), tf.reshape(W2, [-1]),
tf.reshape(W3, [-1]), tf.reshape(W4, [-1]), tf.reshape(W5, [-1])], 0)
allbiases = tf.concat([tf.reshape(B1, [-1]), tf.reshape(B2, [-1]),
tf.reshape(B3, [-1]), tf.reshape(B4, [-1]), tf.reshape(B5, [-1])], 0)
I = tensorflowvisu.tf_format_mnist_images(X, Y, Y_)
It = tensorflowvisu.tf_format_mnist_images(X, Y, Y_, 1000, lines=25)
datavis = tensorflowvisu.MnistDataVis()
# training step, the learning rate is a placeholder
train_step = tf.train.AdamOptimizer(lr).minimize(cross_entropy)
# init
init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init)
# You can call this function in a loop to train the model, 100 images at a time
def training_step(i, update_test_data, update_train_data):
# training on batches of 100 images with 100 labels
batch_X, batch_Y = mnist.train.next_batch(100)
# learning rate decay
max_learning_rate = 0.003
min_learning_rate = 0.0001
decay_speed = 2000.0
learning_rate = min_learning_rate + \
(max_learning_rate - min_learning_rate) * math.exp(-i / decay_speed)
# compute training values for visualisation
if update_train_data:
a, c, im, w, b = sess.run([accuracy, cross_entropy, I, allweights, allbiases], {
X: batch_X, Y_: batch_Y})
print(str(i) + ": accuracy:" + str(a) + " loss: " +
str(c) + " (lr:" + str(learning_rate) + ")")
datavis.append_training_curves_data(i, a, c)
datavis.update_image1(im)
datavis.append_data_histograms(i, w, b)
# compute test values for visualisation
if update_test_data:
a, c, im = sess.run([accuracy, cross_entropy, It], {
X: mnist.test.images, Y_: mnist.test.labels})
print(str(i) + ": ********* epoch " + str(i * 100 //
mnist.train.images.shape[0] + 1) + " ********* test accuracy:" + str(a) + " test loss: " + str(c))
datavis.append_test_curves_data(i, a, c)
datavis.update_image2(im)
# the backpropagation training step
sess.run(train_step, {X: batch_X, Y_: batch_Y, lr: learning_rate})
#datavis.animate(training_step, 10001, train_data_update_freq=10, test_data_update_freq=100)
# to save the animation as a movie, add save_movie=True as an argument to datavis.animate
# to disable the visualisation use the following line instead of the datavis.animate line
for i in range(10000 + 1):
training_step(i, i % 100 == 0, i % 20 == 0)
print("max test accuracy: " + str(datavis.get_max_test_accuracy()))
# layers 4 8 12 200, patches 5x5str1 5x5str2 4x4str2 best 0.989 after 10000 iterations
# layers 4 8 12 200, patches 5x5str1 4x4str2 4x4str2 best 0.9892 after 10000 iterations
# layers 6 12 24 200, patches 5x5str1 4x4str2 4x4str2 best 0.9908 after 10000 iterations but going downhill from 5000 on
# layers 6 12 24 200, patches 5x5str1 4x4str2 4x4str2 dropout=0.75 best 0.9922 after 10000 iterations (but above 0.99 after 1400 iterations only)
# layers 4 8 12 200, patches 5x5str1 4x4str2 4x4str2 dropout=0.75, best 0.9914 at 13700 iterations
# layers 9 16 25 200, patches 5x5str1 4x4str2 4x4str2 dropout=0.75, best 0.9918 at 10500 (but 0.99 at 1500 iterations already, 0.9915 at 5800)
# layers 9 16 25 300, patches 5x5str1 4x4str2 4x4str2 dropout=0.75, best 0.9916 at 5500 iterations (but 0.9903 at 1200 iterations already)
# attempts with 2 fully-connected layers: no better 300 and 100 neurons, dropout 0.75 and 0.5, 6x6 5x5 4x4 patches no better
#*layers 6 12 24 200, patches 6x6str1 5x5str2 4x4str2 dropout=0.75 best 0.9928 after 12800 iterations (but consistently above 0.99 after 1300 iterations only, 0.9916 at 2300 iterations, 0.9921 at 5600, 0.9925 at 20000)
# layers 6 12 24 200, patches 6x6str1 5x5str2 4x4str2 no dropout best 0.9906 after 3100 iterations (avove 0.99 from iteration 1400)
saver = tf.train.Saver()
md_path = R"C:\Users\yaoya\AppData\Local\conda\conda\envs\tensorflow\tensorflow-mnist-tutorial\TestProject\mdlib\md3\init"
# Later, launch the model, initialize the variables, do some work, save the
# variables to disk.
sess.run(init)
save_path = saver.save(sess, md_path, global_step=1000)
print("Model saved in file: %s" % save_path)
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。