具有多个自变量的Tensorflow预测示例

import tensorflow as tf
import numpy as np

# 定义训练数据
x1 = np.random.rand(100)  # 第一个自变量
x2 = np.random.rand(100)  # 第二个自变量
y = 2 * x1 + 3 * x2 + np.random.randn(100) * 0.1  # 目标变量

# 定义模型
X = tf.placeholder(tf.float32, shape=(None, 2))  # 输入特征的占位符
Y = tf.placeholder(tf.float32, shape=(None,))  # 目标变量的占位符
W = tf.Variable(tf.random_normal([2, 1]))  # 权重
b = tf.Variable(tf.random_normal([1]))  # 偏置
Y_pred = tf.matmul(X, W) + b  # 预测值

# 定义损失函数和优化器
loss = tf.reduce_mean(tf.square(Y_pred - Y))  # 均方误差损失函数
optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01)  # 梯度下降优化器
train_op = optimizer.minimize(loss)

# 训练模型
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for epoch in range(1000):
        _, l = sess.run([train_op, loss], feed_dict={X: np.column_stack((x1, x2)), Y: y})
        if epoch % 100 == 0:
            print("Epoch {0}: loss = {1}".format(epoch, l))

    # 使用训练好的模型进行预测
    x1_test = np.random.rand(10)
    x2_test = np.random.rand(10)
    y_pred = sess.run(Y_pred, feed_dict={X: np.column_stack((x1_test, x2_test))})
    print("预测结果:", y_pred)