【机器学习】TensorFlow 202107090086

【源代码】

import tensorflow as tf

l2_reg = tf.keras.regularizers.l2(0.1)  # 设置模型
model = tf.keras.models.Sequential([
    tf.keras.layers.Dense(30, activation='relu',
                          kernel_initializer='he_normal', kernel_regularizer=l2_reg),
    tf.keras.layers.Dense(60, activation='relu',
                          kernel_initializer='he_normal', kernel_regularizer=l2_reg),
    tf.keras.layers.Dense(60, activation='relu',
                          kernel_initializer='he_normal', kernel_regularizer=l2_reg),
    tf.keras.layers.Dense(2, activation="softmax")
])


def random_batch(X, y, batch_size=32):  # 随机抽取数据
    idx = np.random.randint(len(X), size=batch_size)
    Xx = np.array([X[i] for i in idx])
    Yy = np.array([y[i] for i in idx])
    return Xx, Yy


def print_status_bar(iteration, total, loss, metrics=None):  # 输出状态
    #     print('iteration',iteration)
    #     print('total',total)
    #     print('loss',loss.result())

    metrics = "-".join(["{}:{:4f}".format(m.name, m.result()) for m in [loss] + (metrics or [])])
    #     print('metrics',metrics)
    #     print("===============")
    end = "" if iteration < total else "\n"
    print("\r{}/{}-".format(iteration, total) + metrics, end=end)


import numpy as np
import matplotlib.pyplot as plt

a = tf.random.normal([1000, 2], 8, 2)  # 生成数据a类
a_lable = tf.cast([[0] for i in range(1000)], dtype=tf.int64)

b = tf.random.normal([1000, 2], 1, 2)  # 生成数据b类
b_lable = tf.cast([[1] for i in range(1000)], dtype=tf.int64)
X_train = np.concatenate([a, b])  # 合并数据
y_train = np.concatenate([a_lable, b_lable])
y_train = tf.one_hot(y_train[:, 0], 2)  # 将标签合转化为one-hot编码

n_epochs = 5  # 迭代次数
batch_size = 32  # 批次大小
n_steps = int(len(X_train) / 32)  # 分批次
optimizer = t