tf.config.experimental.set_memory_growth

TensorFlow 1 version View source on GitHub

Set if memory growth should be enabled for a PhysicalDevice.

Used in the notebooks

Used in the guide Used in the tutorials

If memory growth is enabled for a PhysicalDevice, the runtime initialization will not allocate all memory on the device. Memory growth cannot be configured on a PhysicalDevice with virtual devices configured.

For example:

physical_devices = tf.config.list_physical_devices('GPU')
try:
  tf.config.experimental.set_memory_growth(physical_devices[0], True)
except:
  # Invalid device or cannot modify virtual devices once initialized.
  pass

device PhysicalDevice to configure
enable (Boolean) Whether to enable or disable memory growth

ValueError Invalid PhysicalDevice specified.
RuntimeError Runtime is already initialized.