first commit

LinearRegression.py

@@ -0,0 +1,54 @@
+# !/usr/bin/python3
+
+import tensorflow as tf
+import numpy as np
+
+
+def train(data):
+    lr = 0.01  # 学习率
+    initial_b = initial_w = 0
+    num_iterations = 1000  # 训练优化次数
+    [b, w] = gradient_descent(data, initial_b, initial_w, lr, num_iterations)
+    loss = mse(b, w, data)
+    print(f'Final loss:{loss}, w:{w}, b:{b}')
+    return None
+
+# 计算最优解b、w上的均方差
+def mse(b, w, points):
+    totalError = 0
+    for i in range(0, len(points)):
+        x = points[i, 0]
+        y = points[i, 1]
+        totalError += (y - (w * x + b)) ** 2
+    return totalError / float(len(points))
+
+
+def step_gradient(b_current, w_current, points, lr):
+    new_b = b_current
+    new_w = w_current
+    b_gradient = w_gradient = 0
+    M = float(len(points))
+    for i in range(0, len(points)):
+        x = points[i, 0]
+        y = points[i, 1]
+        # 误差函数对b的导数grad_b = 2(wx+b-y)
+        b_gradient += (2 / M) * ((w_current * x + b_current) - y)
+        # 误差函数对w的导数grad_w = 2(wx+b-y)*x
+        w_gradient += (2 / M) * x * ((w_current * x + b_current) - y)
+        # 根据学习率lr更新b和w
+        new_b = b_current - (lr * b_gradient)
+        new_w = w_current - (lr * w_gradient)
+    return [new_b, new_w]
+
+
+def gradient_descent(points, starting_b, starting_w, lr, num_iterations):
+    b = starting_b
+    w = starting_w
+    # 根据梯度下降算法更新多次
+    for step in range(num_iterations):
+        # 计算梯度并更新一次
+        [b, w] = step_gradient(b, w, np.array(points), lr)
+        loss = mse(b, w, points)  # 计算当前的均方差，用于监控训练进度
+        if step % 50 == 0:  # 打印误差和实时的w,b 值
+            print(f"iteration:{step}, loss:{loss}, w:{w}, b:{b}")
+    return [b, w]  # 返回最后一次的w,b

NeuralNetwork.py

@@ -0,0 +1,119 @@
+# !/usr/bin/python3
+
+import os
+import matplotlib.pyplot as plt
+import pandas as pd
+import seaborn as sns
+import tensorflow as tf
+from tensorflow import keras
+from tensorflow.keras import layers, losses
+
+os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
+
+# 在线下载汽车效能数据集
+dataset_path = keras.utils.get_file("auto-mpg.data", "http://archive.ics.uci.edu/ml/machine-learning-databases/auto-mpg/auto-mpg.data")
+print("csv文件存储在",dataset_path)
+
+# 效能（每加仑能开多少公里），气缸数，排量，马力，重量， 加速度，型号年份，产地
+column_names = ['MPG', 'Cylinders', 'Displacement', 'Horsepower', 'Weight', 'Acceleration', 'Model Year', 'Origin']
+# 文件路径，列名列表，用于替换NA/NaN的值，分隔符，是否忽略分隔符后的空白
+dataset = pd.read_csv(dataset_path, names=column_names, na_values="?", comment='\t', sep=" ", skipinitialspace=True)
+
+# %% 处理数据
+
+# 清除空白数据
+dataset = dataset.dropna()
+
+# 取出origin列，并根据其写入新列
+origin = dataset.pop('Origin')
+dataset['USA'] = (origin == 1) * 1.0
+dataset['Europe'] = (origin == 2) * 1.0
+dataset['Japan'] = (origin == 3) * 1.0
+
+# 切分为训练集和测试集
+train_dataset = dataset.sample(frac=0.8, random_state=0)
+test_dataset = dataset.drop(train_dataset.index)
+
+# ???
+sns.pairplot(train_dataset[["Cylinders", "Displacement", "Weight", "MPG"]], diag_kind="kde")
+
+# 查看训练集的输入X的统计数据
+train_stats = train_dataset.describe()
+train_stats.pop("MPG")
+# 旋转矩阵
+train_stats = train_stats.transpose()
+
+# 将MPG字段移出为标签数据：
+train_labels = train_dataset.pop('MPG')
+test_labels = test_dataset.pop('MPG')
+
+# 根据均值和标准差，完成数据的标准化
+def norm(x):
+    return (x - train_stats['mean']) / train_stats['std']
+
+normed_train_data = norm(train_dataset)
+normed_test_data = norm(test_dataset)
+
+# 构建Dataset对象
+train_db = tf.data.Dataset.from_tensor_slices((normed_train_data.values, train_labels.values))
+# 随机打散，批量化
+train_db = train_db.shuffle(100).batch(32)
+
+# 自定义网络类，继承自keras.Model基类
+class Network(keras.Model):
+    # 初始化函数
+    def __init__(self):
+        super(Network, self).__init__()
+        # 创建3个全连接层，分别为64、64、1个节点
+        self.fc1 = layers.Dense(64, activation='relu')
+        self.fc2 = layers.Dense(64, activation='relu')
+        self.fc3 = layers.Dense(1)
+    def call(self, inputs, training=None, mask=None):
+        # 依次通过3个全连接层
+        x = self.fc1(inputs)
+        x = self.fc2(x)
+        x = self.fc3(x)
+        return x
+
+# 创建网络类实例
+model = Network()
+# 通过build函数完成内部张量的创建，其中4为任意设置的batch数量，9为输入特征长度
+model.build(input_shape=(None, 9))
+print(model.summary())
+# 创建优化器，指定学习率
+optimizer = tf.keras.optimizers.RMSprop(0.001)
+
+# # 未训练时测试
+# example_batch = normed_train_data[:10]
+# example_result = model.predict(example_batch)
+# example_result
+
+train_mae_losses = []
+test_mae_losses = []
+
+for epoch in range(200):
+    for step, (x, y) in enumerate(train_db):
+        with tf.GradientTape() as tape:
+            out = model(x)
+            # MSE：均方误差
+            loss = tf.reduce_mean(losses.MSE(y, out))
+            # MAE：平均绝对误差
+            mae_loss = tf.reduce_mean(losses.MAE(y, out))    
+        grads = tape.gradient(loss, model.trainable_variables)
+        optimizer.apply_gradients(zip(grads, model.trainable_variables))
+    train_mae_losses.append(float(mae_loss))
+    out = model(tf.constant(normed_test_data.values))
+    test_mae_losses.append(tf.reduce_mean(losses.MAE(test_labels, out)))
+
+plt.figure()
+plt.xlabel('Epoch')
+plt.ylabel('MAE')
+plt.plot(train_mae_losses, label='Train')
+plt.plot(test_mae_losses, label='Test')
+plt.legend()
+
+# plt.ylim([0,10])
+plt.legend()
+plt.savefig('auto.svg')
+plt.show()
+

PythonTest.py

@@ -0,0 +1,19 @@
+# !/usr/bin/python3
+
+
+def grammar(a):
+    print("start learning python grammar...")
+
+    '''
+    如果改变数字数据类型的值，将重新分配内存空间
+    '''
+    var1 = 1
+    var2 = 10
+    a += var1 + var2
+    del var1, var2
+
+    list = ['Google', 'Runoob', 1997, 2000];
+    list[1] = 2001
+    print("list[0:2]: ", list[0:2])
+
+    return a

TensorflowTest.py

@@ -0,0 +1,31 @@
+# !/usr/bin/python3
+import tensorflow as tf
+
+
+def tensorflow(parameter1,parameter2):
+    print("tensorflow的版本是：", tf.__version__)
+    print("tensorflow的路径是：", tf.__path__)
+
+
+
+
+    default_g = tf.compat.v1.get_default_graph()
+
+    a = tf.constant(parameter1)
+    b = tf.constant(parameter2)
+    print(a)
+    with tf.GradientTape() as tape:  # 构建梯度环境
+        a = tf.cast(a, tf.float32)
+        b = tf.cast(b, tf.float32)
+        tape.watch([a])  # 将a加入梯度跟踪列表
+        y = b * a ** 2
+        [dy_da] = tape.gradient(y, [a])
+        print(dy_da)
+
+    tensor1 = tf.constant(value=[[1, 2, 3, 4], [2, 3, 4, 5], [3, 4, 5, 6], [4, 5, 6, 7]],
+                          dtype=tf.int32,
+                          shape=[4, 4],
+                          name='tensor1')
+    print(tensor1)
+
+    return None