init commit

Author: snowkylin

.gitignore

@@ -0,0 +1,3 @@
+build
+reference
+teaching

Makefile

@@ -0,0 +1,20 @@
+# Minimal makefile for Sphinx documentation
+#
+
+# You can set these variables from the command line.
+SPHINXOPTS    =
+SPHINXBUILD   = python -msphinx
+SPHINXPROJ    = TensorFlow
+SOURCEDIR     = source
+BUILDDIR      = build
+
+# Put it first so that "make" without argument is like "make help".
+help:
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+.PHONY: help Makefile
+
+# Catch-all target: route all unknown targets to Sphinx using the new
+# "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
+%: Makefile
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

autobuild.bat

@@ -0,0 +1 @@
+sphinx-autobuild source build\html

make.bat

@@ -0,0 +1,36 @@
+@ECHO OFF
+
+pushd %~dp0
+
+REM Command file for Sphinx documentation
+
+if "%SPHINXBUILD%" == "" (
+	set SPHINXBUILD=python -msphinx
+)
+set SOURCEDIR=source
+set BUILDDIR=build
+set SPHINXPROJ=TensorFlow
+
+if "%1" == "" goto help
+
+%SPHINXBUILD% >NUL 2>NUL
+if errorlevel 9009 (
+	echo.
+	echo.The Sphinx module was not found. Make sure you have Sphinx installed,
+	echo.then set the SPHINXBUILD environment variable to point to the full
+	echo.path of the 'sphinx-build' executable. Alternatively you may add the
+	echo.Sphinx directory to PATH.
+	echo.
+	echo.If you don't have Sphinx installed, grab it from
+	echo.http://sphinx-doc.org/
+	exit /b 1
+)
+
+%SPHINXBUILD% -M %1 %SOURCEDIR% %BUILDDIR% %SPHINXOPTS%
+goto end
+
+:help
+%SPHINXBUILD% -M help %SOURCEDIR% %BUILDDIR% %SPHINXOPTS%
+
+:end
+popd

source/.vscode/settings.json

@@ -0,0 +1,6 @@
+{
+    "restructuredtext.confPath": "c:\\Users\\xihan\\Desktop\\TensorFlow-cn\\source",
+    "restructuredtext.builtDocumentationPath" : "c:\\Users\\xihan\\Desktop\\TensorFlow-cn\\build\\html",
+    "restructuredtext.updateOnTextChanged"    : "true",
+    "restructuredtext.sphinxBuildPath"        : "C:\\ProgramData\\Anaconda3\\Scripts\\sphinx-build.exe"
+}

source/_static/code/.idea/code.iml

@@ -0,0 +1,38 @@
+a = 0
+b = 0
+
+def f(x):
+    y_pred = a * x + b
+    return y_pred
+
+def loss(x, y):
+    l = (a * x + b - y) ** 2
+    return l
+
+def gradient_loss(x, y):
+    g_a = 2 * (a * x + b - y) * x
+    g_b = 2 * (a * x + b - y)
+    return g_a, g_b
+
+X_raw = [2013, 2014, 2015, 2016, 2017]
+Y_raw = [12000, 14000, 15000, 16500, 17500]
+x_pred_raw = 2018
+X = [(x - min(X_raw)) / (max(X_raw) - min(X_raw)) for x in X_raw]
+Y = [(y - min(Y_raw)) / (max(Y_raw) - min(Y_raw)) for y in Y_raw]
+
+num_epoch = 10000
+learning_rate = 1e-3
+for e in range(num_epoch):
+    for i in range(len(X)):
+        x, y = X[i], Y[i]
+        g_a, g_b = gradient_loss(x, y)
+        a = a - learning_rate * g_a
+        b = b - learning_rate * g_b
+print(a, b)
+for i in range(len(X)):
+    x, y = X[i], Y[i]
+    print(f(x), y)
+x_pred = (x_pred_raw - min(X_raw)) / (max(X_raw) - min(X_raw))
+y_pred = f(x_pred)
+y_pred_raw = y_pred * (max(Y_raw) - min(Y_raw)) + min(Y_raw)
+print(x_pred_raw, y_pred_raw)

source/_static/code/.idea/misc.xml

@@ -0,0 +1,25 @@
+import numpy as np
+
+X_raw = np.array([2013, 2014, 2015, 2016, 2017])
+y_raw = np.array([12000, 14000, 15000, 16500, 17500])
+
+X = (X_raw - X_raw.min()) / (X_raw.max() - X_raw.min())
+y = (y_raw - y_raw.min()) / (y_raw.max() - y_raw.min())
+
+a, b = 0, 0
+
+num_epoch = 10000
+learning_rate = 1e-3
+for e in range(num_epoch):
+    # 前向传播
+    y_pred = a * X + b
+    loss = 0.5 * (y_pred - y).dot(y_pred - y) # loss = 0.5 * np.sum(np.square(a * X + b - y))
+
+    # 反向传播，手动计算变量（模型参数）的梯度
+    grad_a, grad_b = (y_pred - y).dot(X), (y_pred - y).sum()
+
+    # 更新参数
+    a, b = a - learning_rate * grad_a, b - learning_rate * grad_b
+
+print(a, b)
+

source/_static/code/.idea/modules.xml

@@ -0,0 +1,41 @@
+import numpy as np
+
+X_raw = np.array([2013, 2014, 2015, 2016, 2017])
+y_raw = np.array([12000, 14000, 15000, 16500, 17500])
+
+X = (X_raw - X_raw.min()) / (X_raw.max() - X_raw.min())
+y = (y_raw - y_raw.min()) / (y_raw.max() - y_raw.min())
+
+import tensorflow as tf
+
+# 定义数据流图
+learning_rate_ = tf.placeholder(dtype=tf.float32)
+X_ = tf.placeholder(dtype=tf.float32, shape=[5])
+y_ = tf.placeholder(dtype=tf.float32, shape=[5])
+a = tf.get_variable('a', dtype=tf.float32, shape=[], initializer=tf.zeros_initializer)
+b = tf.get_variable('b', dtype=tf.float32, shape=[], initializer=tf.zeros_initializer)
+
+y_pred = a * X_ + b
+loss = tf.constant(0.5) * tf.reduce_sum(tf.square(y_pred - y_))
+
+# 反向传播，手动计算变量（模型参数）的梯度
+grad_a = tf.reduce_sum((y_pred - y_) * X_)
+grad_b = tf.reduce_sum(y_pred - y_)
+
+# 手动更新参数
+new_a = a - learning_rate_ * grad_a
+new_b = b - learning_rate_ * grad_b
+update_a = tf.assign(a, new_a)
+update_b = tf.assign(b, new_b)
+# 数据流图定义到此结束
+# 注意，直到目前，我们都没有进行任何实质的数据计算，仅仅是定义了一个数据图
+
+num_epoch = 10000
+learning_rate = 1e-3
+with tf.Session() as sess:
+    # 初始化变量a和b
+    tf.global_variables_initializer().run()
+    # 循环将数据送入上面建立的数据流图中进行计算和更新变量
+    for e in range(num_epoch):
+        sess.run([update_a, update_b], feed_dict={X_: X, y_: y, learning_rate_: learning_rate})
+    print(sess.run([a, b]))

source/_static/code/.idea/workspace.xml

@@ -0,0 +1,29 @@
+import numpy as np
+
+X_raw = np.array([2013, 2014, 2015, 2016, 2017])
+y_raw = np.array([12000, 14000, 15000, 16500, 17500])
+
+X = (X_raw - X_raw.min()) / (X_raw.max() - X_raw.min())
+y = (y_raw - y_raw.min()) / (y_raw.max() - y_raw.min())
+
+import tensorflow as tf
+
+learning_rate_ = tf.placeholder(dtype=tf.float32)
+X_ = tf.placeholder(dtype=tf.float32, shape=[5])
+y_ = tf.placeholder(dtype=tf.float32, shape=[5])
+a = tf.get_variable('a', dtype=tf.float32, shape=[], initializer=tf.zeros_initializer)
+b = tf.get_variable('b', dtype=tf.float32, shape=[], initializer=tf.zeros_initializer)
+
+y_pred = a * X_ + b
+loss = tf.constant(0.5) * tf.reduce_sum(tf.square(y_pred - y_))
+
+# 反向传播，利用TensorFlow的梯度下降优化器自动计算并更新变量（模型参数）的梯度
+train_op = tf.train.GradientDescentOptimizer(learning_rate=learning_rate_).minimize(loss)
+
+num_epoch = 10000
+learning_rate = 1e-3
+with tf.Session() as sess:
+    tf.global_variables_initializer().run()
+    for e in range(num_epoch):
+        sess.run(train_op, feed_dict={X_: X, y_: y, learning_rate_: learning_rate})
+    print(sess.run([a, b]))

source/_static/code/basic/example.py

@@ -0,0 +1,32 @@
+import tensorflow as tf
+import tensorflow.contrib.eager as tfe
+tfe.enable_eager_execution()
+import numpy as np
+
+X_raw = np.array([2013, 2014, 2015, 2016, 2017], dtype=np.float32)
+y_raw = np.array([12000, 14000, 15000, 16500, 17500], dtype=np.float32)
+
+X = (X_raw - X_raw.min()) / (X_raw.max() - X_raw.min())
+y = (y_raw - y_raw.min()) / (y_raw.max() - y_raw.min())
+
+X = tf.constant(X)
+y = tf.constant(y)
+
+a = tfe.Variable(0., name='a')
+b = tfe.Variable(0., name='b')
+
+num_epoch = 10000
+learning_rate = 1e-3
+for e in range(num_epoch):
+    # 前向传播
+    y_pred = a * X + b
+    loss = 0.5 * tf.reduce_sum(tf.square(y_pred - y)) # loss = 0.5 * np.sum(np.square(a * X + b - y))
+
+    # 反向传播，手动计算变量（模型参数）的梯度
+    grad_a = tf.reduce_sum((y_pred - y) * X)
+    grad_b = tf.reduce_sum(y_pred - y)
+
+    # 更新参数
+    a, b = a - learning_rate * grad_a, b - learning_rate * grad_b
+
+print(a, b)

source/_static/code/basic/example_numpy.py

@@ -0,0 +1,38 @@
+import tensorflow as tf
+import tensorflow.contrib.eager as tfe
+tfe.enable_eager_execution()
+import numpy as np
+
+X_raw = np.array([2013, 2014, 2015, 2016, 2017], dtype=np.float32)
+y_raw = np.array([12000, 14000, 15000, 16500, 17500], dtype=np.float32)
+
+X = (X_raw - X_raw.min()) / (X_raw.max() - X_raw.min())
+y = (y_raw - y_raw.min()) / (y_raw.max() - y_raw.min())
+
+X = tf.constant(X)
+y = tf.constant(y)
+
+a = tfe.Variable(0., name='a')
+b = tfe.Variable(0., name='b')
+
+def model(x):
+    return a * x + b
+
+def loss(X_, y_):
+    return 0.5 * tf.reduce_sum(tf.square(model(X_) - y_))
+
+grad_fn = tfe.implicit_gradients(loss)
+num_epoch = 10000
+optimizer = tf.train.GradientDescentOptimizer(learning_rate=1e-3)
+for e in range(num_epoch):
+    # 前向传播
+    y_pred = a * X + b
+
+    # 反向传播，利用Eager模式下的tfe.implicit_gradients()自动计算梯度
+    grad = grad_fn(X, y)
+    optimizer.apply_gradients(grad)
+
+    # 更新参数
+    # a, b = a - learning_rate * grad_a, b - learning_rate * grad_b
+
+print(a, b)

source/_static/code/basic/example_tensorflow.py

@@ -0,0 +1,11 @@
+附录
+======
+
+Anaconda简易使用教程
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+NumPy基础
+^^^^^^^^^^^
+
+线性回归基础
+^^^^^^^^^^^^^^^^^^