如何使用python在Tensorboard上显示模型的权重和偏移

import tensorflow as tf
from tensorflow.python.framework import graph_util
from tensorflow.summary import FileWriter

# 定义输入和输出的占位符
x = tf.placeholder(tf.float32, name='input')
y = tf.placeholder(tf.float32, name='output')

# 定义模型结构
w = tf.Variable(tf.random_normal([input_size, output_size]), name='weights')
b = tf.Variable(tf.zeros([output_size]), name='bias')
output = tf.add(tf.matmul(x, w), b, name='prediction')

# 定义损失函数和优化器
loss = tf.reduce_mean(tf.square(output - y))
optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01).minimize(loss)

with tf.Session() as sess:
    # 初始化变量
    sess.run(tf.global_variables_initializer())

    # 训练模型
    for epoch in range(num_epochs):
        _, current_loss = sess.run([optimizer, loss], feed_dict={x: input_data, y: output_data})

    # 保存模型
    output_node_names = 'prediction'
    output_graph_def = graph_util.convert_variables_to_constants(sess, sess.graph_def, [output_node_names])
    tf.train.write_graph(output_graph_def, './', 'model.pb', as_text=False)

with tf.Session() as sess:
    # 导入模型
    with tf.gfile.FastGFile('model.pb', 'rb') as f:
        graph_def = tf.GraphDef()
        graph_def.ParseFromString(f.read())
        tf.import_graph_def(graph_def, name='')

    # 创建summary writer
    writer = FileWriter('./logs', sess.graph)

    # 获取权重和偏移的操作
    weights_op = sess.graph.get_tensor_by_name('weights:0')
    bias_op = sess.graph.get_tensor_by_name('bias:0')

    # 将权重和偏移写入summary
    weights_summary = tf.summary.histogram('weights', weights_op)
    bias_summary = tf.summary.histogram('bias', bias_op)

    # 合并summary
    merged_summary = tf.summary.merge([weights_summary, bias_summary])

    # 运行summary操作
    summary = sess.run(merged_summary)

    # 将summary写入Tensorboard
    writer.add_summary(summary)

    # 关闭summary writer
    writer.close()