如何使用Tensorflow概率中的Gamma函数作为回归模型中的对数似然损失

import tensorflow as tf
import tensorflow_probability as tfp

# 定义模型的参数
alpha = tf.Variable(1.0, name="alpha")
beta = tf.Variable(1.0, name="beta")

# 定义输入和输出
x = tf.placeholder(tf.float32, name="x")
y = tf.placeholder(tf.float32, name="y")

# 定义模型
gamma_dist = tfp.distributions.Gamma(concentration=alpha, rate=beta)
likelihood = gamma_dist.log_prob(y)

# 定义损失函数
loss = -tf.reduce_mean(likelihood)

# 定义优化器
optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01)

# 定义训练操作
train_op = optimizer.minimize(loss)

# 创建会话并初始化变量
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())

    # 进行训练
    for i in range(num_iterations):
        sess.run(train_op, feed_dict={x: input_data, y: output_data})

    # 获取训练结果
    alpha_val, beta_val = sess.run([alpha, beta])