tf.train.load_variable

Returns the tensor value of the given variable in the checkpoint.

tf.train.load_variable(
    ckpt_dir_or_file, name
)

When the variable name is unknown, you can use tf.train.list_variables to inspect all the variable names.

Example usage:

import tensorflow as tf
a = tf.Variable(1.0)
b = tf.Variable(2.0)
ckpt = tf.train.Checkpoint(var_list={'a': a, 'b': b})
ckpt_path = ckpt.save('tmp-ckpt')
var= tf.train.load_variable(
    ckpt_path, 'var_list/a/.ATTRIBUTES/VARIABLE_VALUE')
print(var)  # 1.0

ckpt_dir_or_file Directory with checkpoints file or path to checkpoint.
name Name of the variable to return.

A numpy ndarray with a copy of the value of this variable.