tf_linear_regression_continue_part2.py

import tensorflow as tf
import numpy

# Parameters
learning_rate = 0.01
training_epochs = 3500
datapoints_count = 100

# Training Data
train_X = []
train_Y = []
for i in range(datapoints_count):
  train_X.append(i)
  train_Y.append(5*i+8.0)

train_X = numpy.asarray(train_X)
train_Y = numpy.asarray(train_Y)
n_samples = train_X.shape[0]

imported_graph = tf.train.import_meta_graph('saved_variable.meta')

with tf.Session(graph=tf.get_default_graph()) as sess:
  sess.run(tf.global_variables_initializer())
  imported_graph.restore(sess, './saved_variable')

  print("value of W: ", sess.run("weight:0"))
  print("value of b: ", sess.run("bias:0"))

  for epoch in range(500, training_epochs):
    for (x, y) in zip(train_X, train_Y):
      sess.run('GradientDescent', feed_dict={"X:0": x, "Y:0": y})
    if (epoch+1) % 50 == 0:
      training_cost = sess.run("cost:0", feed_dict={"X:0": train_X, "Y:0": train_Y})
      print("Epoch: %04d cost=%.9f W=%f b=%f" % \
        ((epoch+1), training_cost, sess.run("weight:0"), sess.run("bias:0")))