MNIST 數據集入門
MNIST 數據集簡介
數字手寫體識別數據集，經常使用來做爲Deep Learning入門的基礎數據集。它有60000個訓練樣本集和10000個測試樣本集，每一個樣本圖像的寬高爲 28 * 28。此數據集是以二進制存儲的，不能直接以圖像格式查看。javascript
數據集大小：~12MBcss
下載地址：http://yann.lecun.com/exdb/mnist/index.htmlhtml
導入tensorflow（v1.6.0版本）
import tensorflow as tf
print(tf.__version__)
1.6.0
tensorflow加載MNIST數據集
# Import MNIST
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("../../data/", one_hot=True)
# Load data
X_train = mnist.train.images
Y_train = mnist.train.labels
X_test = mnist.test.images
Y_test = mnist.test.labels
train-images-idx3-ubyte.gz: training set images (9912422 bytes)
train-labels-idx1-ubyte.gz: training set labels (28881 bytes)
t10k-images-idx3-ubyte.gz: test set images (1648877 bytes)
t10k-labels-idx1-ubyte.gz: test set labels (4542 bytes)
Extracting ../../data/train-images-idx3-ubyte.gz
Extracting ../../data/train-labels-idx1-ubyte.gz
Extracting ../../data/t10k-images-idx3-ubyte.gz
Extracting ../../data/t10k-labels-idx1-ubyte.gz
查看並可視化MNIST數據集
import matplotlib.pyplot as plt
import numpy as np
%matplotlib inline
# Get the next 64 images array and labels
batch_X, batch_Y = mnist.train.next_batch(64)
print(batch_X.astype,batch_X.shape) # (64,28*28)
print(batch_Y.astype,batch_Y.shape) # (64,10[0-9哪一類])
print(batch_Y[0]) # [ 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]
plt.imshow(np.reshape(batch_X[0], [28, 28]), cmap='gray')
<built-in method astype of numpy.ndarray object at 0x0000016EEF769080> (64, 784)
<built-in method astype of numpy.ndarray object at 0x0000016EEF772620> (64, 10)
[ 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]
<matplotlib.image.AxesImage at 0x16eef82a630>
tensorflow入門（hello world）
import tensorflow as tf
# TensorFlow 實現簡單的 hello world
# 建立一個常量操做
# 這個常量操做會在默認的圖中添加一個節點
#
# 構造函數返回的值表示常量op的輸出。
hello = tf.constant('Hello, TensorFlow!')
# 啓動 tf session
sess = tf.Session()
# 運行圖
print(sess.run(hello))
b'Hello, TensorFlow!'
tensorflow入門（基本操做）
常量操做
# 常量基本操做
a = tf.constant(2)
b = tf.constant(3)
# 啓動默認的圖
with tf.Session() as sess:
print ("a: %i" % sess.run(a), "b: %i" % sess.run(b))
print ("Addition with constants: %i" % sess.run(a+b))
print ("Multiplication with constants: %i" % sess.run(a*b))
a: 2 b: 3
Addition with constants: 5
Multiplication with constants: 6
變量操做
# 做爲圖輸入變量的基本操做。
a = tf.placeholder(tf.int16)
b = tf.placeholder(tf.int16)
# tf 中定義的操做
add = tf.add(a, b) #加法操做
mul = tf.multiply(a, b) #乘法操做
# 啓動默認的圖
with tf.Session() as sess:
# 使用變量輸入運行每一個操做。
print ("Addition with variables: %i" % sess.run(add, feed_dict={a: 2, b: 3}))
print ("Multiplication with variables: %i" % sess.run(mul, feed_dict={a: 2, b: 3}))
Addition with variables: 5
Multiplication with variables: 6
矩陣操做
# 建立一個生成1x2矩陣的常數op。
matrix1 = tf.constant([[3., 3.]])
# 建立另外一個常數，生成一個2x1矩陣。
matrix2 = tf.constant([[2.],[2.]])
# tf 中定義的操做
product = tf.matmul(matrix1, matrix2) #矩陣乘法操做
with tf.Session() as sess:
result = sess.run(product)
print(result)
[[ 12.]]
tensorflow入門（Eager API）
from __future__ import absolute_import, division, print_function
import numpy as np
import tensorflow as tf
import tensorflow.contrib.eager as tfe
# 設置 Eager API
print("Setting Eager mode...")
tfe.enable_eager_execution()
Setting Eager mode...
Eager API 常量操做
# 定義常量 tensors
print("Define constant tensors")
a = tf.constant(2)
print("a = %i" % a)
b = tf.constant(3)
print("b = %i" % b)
Define constant tensors
a = 2
b = 3
# 執行操做不須要 tf.Session
print("Running operations, without tf.Session")
c = a + b
print("a + b = %i" % c)
d = a * b
print("a * b = %i" % d)
Running operations, without tf.Session
a + b = 5
a * b = 6
Eager API 張量操做
# 與 Numpy徹底兼容
print("Mixing operations with Tensors and Numpy Arrays")
# Define constant tensors
a = tf.constant([[2., 1.],
[1., 0.]], dtype=tf.float32)
print("Tensor:\n a = %s" % a)
b = np.array([[3., 0.],
[5., 1.]], dtype=np.float32)
print("NumpyArray:\n b = %s" % b)
Mixing operations with Tensors and Numpy Arrays
Tensor:
a = tf.Tensor(
[[2. 1.]
[1. 0.]], shape=(2, 2), dtype=float32)
NumpyArray:
b = [[3. 0.]
[5. 1.]]
# 在不須要 tf.Session 的狀況下運行該操做
print("Running operations, without tf.Session")
c = a + b
print("a + b = %s" % c)
d = tf.matmul(a, b)
print("a * b = %s" % d)
Running operations, without tf.Session
a + b = tf.Tensor(
[[5. 1.]
[6. 1.]], shape=(2, 2), dtype=float32)
a * b = tf.Tensor(
[[11. 1.]
[ 3. 0.]], shape=(2, 2), dtype=float32)
# 遍歷張量
print("Iterate through Tensor 'a':")
for i in range(a.shape[0]):
for j in range(a.shape[1]):
print(a[i][j])
Iterate through Tensor 'a':
tf.Tensor(2.0, shape=(), dtype=float32)
tf.Tensor(1.0, shape=(), dtype=float32)
tf.Tensor(1.0, shape=(), dtype=float32)
tf.Tensor(0.0, shape=(), dtype=float32)
參考
[TensorFlow-Examples]https://github.com/aymericdamien/TensorFlow-Examplesjava
－長按關注－python
本文分享自微信公衆號 - AI異構（gh_ed66a0ffe20a）。
若有侵權，請聯繫 support@oschina.cn 刪除。
本文參與「OSC源創計劃」，歡迎正在閱讀的你也加入，一塊兒分享。nginx