Python – tensorflow.GradientTape.stop_recording()

TensorFlow is open-source Python library designed by Google to develop Machine Learning models and deep learning  neural networks. 

stop_recording() is used to temporarily stop recording operation. If the Tape is not recording it will raise an error.

Syntax: tensorflow.GradientTape.stop_recording()

Parameters: It doesn’t accept any parameters.

Returns: None

Raises:

• RunTimeError: It will raise RunTimeError if tape is not recording currently.

Example 1:

Python3

# Importing the library
import tensorflow as tf

x = tf.constant(4.0)

# Using GradientTape
with tf.GradientTape() as gfg:
  gfg.watch(x)

  # Stop recording
  with gfg.stop_recording():
    y = x * x * x

# Computing gradient
res = gfg.gradient(y, x)

# Printing result
print("res: ", res)

Output:

res:  None

Example 2:

Python3

# Importing the library
import tensorflow as tf

x = tf.constant(4.0)

# Using GradientTape
with tf.GradientTape() as gfg:
  gfg.watch(x)

  # Stop recording
  with gfg.stop_recording():
    y = x * x * x

  # Starting the recording again
  gfg.watch(x)
  y = x * x

# Computing gradient
res = gfg.gradient(y, x)

# Printing result
print("res: ", res)

Output:

res:  tf.Tensor(8.0, shape=(), dtype=float32)