Tensorflow RNN用于拟合曲线的简单易用的示例/模板？

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt

# 生成训练数据
x = np.linspace(0, 2*np.pi, 100)
y = np.sin(x)

# 定义模型参数
input_dim = 1
hidden_dim = 32
output_dim = 1
num_steps = 10
learning_rate = 0.01
num_epochs = 100

# 创建输入和输出占位符
inputs = tf.placeholder(tf.float32, [None, num_steps, input_dim])
targets = tf.placeholder(tf.float32, [None, output_dim])

# 定义RNN模型
cell = tf.contrib.rnn.BasicRNNCell(hidden_dim)
outputs, states = tf.nn.dynamic_rnn(cell, inputs, dtype=tf.float32)

# 定义输出层
outputs = tf.reshape(outputs, [-1, hidden_dim])
weights = tf.Variable(tf.random_normal([hidden_dim, output_dim]))
biases = tf.Variable(tf.random_normal([output_dim]))
predictions = tf.matmul(outputs, weights) + biases

# 定义损失函数和优化器
loss = tf.reduce_mean(tf.square(predictions - targets))
optimizer = tf.train.AdamOptimizer(learning_rate).minimize(loss)

# 创建会话并训练模型
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for epoch in range(num_epochs):
        _, curr_loss = sess.run([optimizer, loss], feed_dict={inputs: x, targets: y})
        if epoch % 10 == 0:
            print("Epoch: %d, Loss: %f" % (epoch, curr_loss))

    # 使用训练好的模型进行预测
    test_inputs = np.linspace(0, 2*np.pi, 100)
    test_inputs = np.reshape(test_inputs, [-1, num_steps, input_dim])
    predicted_values = sess.run(predictions, feed_dict={inputs: test_inputs})

# 可视化结果
plt.plot(x, y, label='Ground Truth')
plt.plot(test_inputs.flatten(), predicted_values.flatten(), label='Predictions')
plt.legend()
plt.show()