tf.InteractiveSession

View source on GitHub

A TensorFlow Session for use in interactive contexts, such as a shell.

tf.InteractiveSession(
    target='', graph=None, config=None
)

The only difference with a regular Session is that an InteractiveSession installs itself as the default session on construction. The methods tf.Tensor.eval and tf.Operation.run will use that session to run ops.

This is convenient in interactive shells and IPython notebooks, as it avoids having to pass an explicit Session object to run ops.

For example:

sess = tf.compat.v1.InteractiveSession()
a = tf.constant(5.0)
b = tf.constant(6.0)
c = a * b
# We can just use 'c.eval()' without passing 'sess'
print(c.eval())
sess.close()

Note that a regular session installs itself as the default session when it is created in a with statement. The common usage in non-interactive programs is to follow that pattern:

a = tf.constant(5.0)
b = tf.constant(6.0)
c = a * b
with tf.compat.v1.Session():
  # We can also use 'c.eval()' here.
  print(c.eval())

Methods

as_default

View source

as_default()

Returns a context manager that makes this object the default session.

Use with the with keyword to specify that calls to tf.Operation.run or tf.Tensor.eval should be executed in this session.

c = tf.constant(..)
sess = tf.compat.v1.Session()

with sess.as_default():
  assert tf.compat.v1.get_default_session() is sess
  print(c.eval())

To get the current default session, use tf.compat.v1.get_default_session.

c = tf.constant(...)
sess = tf.compat.v1.Session()
with sess.as_default():
  print(c.eval())
# ...
with sess.as_default():
  print(c.eval())

sess.close()

Alternatively, you can use with tf.compat.v1.Session(): to create a session that is automatically closed on exiting the context, including when an uncaught exception is raised.

close

View source

close()

Closes an InteractiveSession.

list_devices

View source

list_devices()

Lists available devices in this session.

devices = sess.list_devices()
for d in devices:
  print(d.name)

Where:

Each element in the list has the following properties

• name: A string with the full name of the device. ex: /job:worker/replica:0/task:3/device:CPU:0
• device_type: The type of the device (e.g. CPU, GPU, TPU.)
• memory_limit: The maximum amount of memory available on the device. Note: depending on the device, it is possible the usable memory could be substantially less.

make_callable

View source

make_callable(
    fetches, feed_list=None, accept_options=False
)

Returns a Python callable that runs a particular step.

The returned callable will take len(feed_list) arguments whose types must be compatible feed values for the respective elements of feed_list. For example, if element i of feed_list is a tf.Tensor, the ith argument to the returned callable must be a numpy ndarray (or something convertible to an ndarray) with matching element type and shape. See tf.Session.run for details of the allowable feed key and value types.

The returned callable will have the same return type as tf.Session.run(fetches, ...). For example, if fetches is a tf.Tensor, the callable will return a numpy ndarray; if fetches is a tf.Operation, it will return None.

partial_run

View source

partial_run(
    handle, fetches, feed_dict=None
)

Continues the execution with more feeds and fetches.

This is EXPERIMENTAL and subject to change.

To use partial execution, a user first calls partial_run_setup() and then a sequence of partial_run(). partial_run_setup specifies the list of feeds and fetches that will be used in the subsequent partial_run calls.

The optional feed_dict argument allows the caller to override the value of tensors in the graph. See run() for more information.

Below is a simple example:

a = array_ops.placeholder(dtypes.float32, shape=[])
b = array_ops.placeholder(dtypes.float32, shape=[])
c = array_ops.placeholder(dtypes.float32, shape=[])
r1 = math_ops.add(a, b)
r2 = math_ops.multiply(r1, c)

h = sess.partial_run_setup([r1, r2], [a, b, c])
res = sess.partial_run(h, r1, feed_dict={a: 1, b: 2})
res = sess.partial_run(h, r2, feed_dict={c: res})

partial_run_setup

View source

partial_run_setup(
    fetches, feeds=None
)

Sets up a graph with feeds and fetches for partial run.

This is EXPERIMENTAL and subject to change.

Note that contrary to run, feeds only specifies the graph elements. The tensors will be supplied by the subsequent partial_run calls.

run

View source

run(
    fetches, feed_dict=None, options=None, run_metadata=None
)

Runs operations and evaluates tensors in fetches.

This method runs one "step" of TensorFlow computation, by running the necessary graph fragment to execute every Operation and evaluate every Tensor in fetches, substituting the values in feed_dict for the corresponding input values.

The fetches argument may be a single graph element, or an arbitrarily nested list, tuple, namedtuple, dict, or OrderedDict containing graph elements at its leaves. A graph element can be one of the following types:

• A tf.Operation. The corresponding fetched value will be None.
• A tf.Tensor. The corresponding fetched value will be a numpy ndarray containing the value of that tensor.
• A tf.SparseTensor. The corresponding fetched value will be a tf.compat.v1.SparseTensorValue containing the value of that sparse tensor.
• A get_tensor_handle op. The corresponding fetched value will be a numpy ndarray containing the handle of that tensor.
• A string which is the name of a tensor or operation in the graph.

The value returned by run() has the same shape as the fetches argument, where the leaves are replaced by the corresponding values returned by TensorFlow.

Example:

   a = tf.constant([10, 20])
   b = tf.constant([1.0, 2.0])
   # 'fetches' can be a singleton
   v = session.run(a)
   # v is the numpy array [10, 20]
   # 'fetches' can be a list.
   v = session.run([a, b])
   # v is a Python list with 2 numpy arrays: the 1-D array [10, 20] and the
   # 1-D array [1.0, 2.0]
   # 'fetches' can be arbitrary lists, tuples, namedtuple, dicts:
   MyData = collections.namedtuple('MyData', ['a', 'b'])
   v = session.run({'k1': MyData(a, b), 'k2': [b, a]})
   # v is a dict with
   # v['k1'] is a MyData namedtuple with 'a' (the numpy array [10, 20]) and
   # 'b' (the numpy array [1.0, 2.0])
   # v['k2'] is a list with the numpy array [1.0, 2.0] and the numpy array
   # [10, 20].

The optional feed_dict argument allows the caller to override the value of tensors in the graph. Each key in feed_dict can be one of the following types:

• If the key is a tf.Tensor, the value may be a Python scalar, string, list, or numpy ndarray that can be converted to the same dtype as that tensor. Additionally, if the key is a tf.compat.v1.placeholder, the shape of the value will be checked for compatibility with the placeholder.
• If the key is a tf.SparseTensor, the value should be a tf.compat.v1.SparseTensorValue.
• If the key is a nested tuple of Tensors or SparseTensors, the value should be a nested tuple with the same structure that maps to their corresponding values as above.

Each value in feed_dict must be convertible to a numpy array of the dtype of the corresponding key.

The optional options argument expects a [RunOptions] proto. The options allow controlling the behavior of this particular step (e.g. turning tracing on).

The optional run_metadata argument expects a [RunMetadata] proto. When appropriate, the non-Tensor output of this step will be collected there. For example, when users turn on tracing in options, the profiled info will be collected into this argument and passed back.