tensorflow - How do I find the variable names and values that are saved in a checkpoint?

Question

I want to see the variables that are saved in a TensorFlow checkpoint along with their values. How can I find the variable names that are saved in a TensorFlow checkpoint?

I used tf.train.NewCheckpointReader which is explained here. But, it is not given in the documentation of TensorFlow. Is there any other way?

Answer 1

Example usage:

from tensorflow.python.tools.inspect_checkpoint import print_tensors_in_checkpoint_file
import os
checkpoint_path = os.path.join(model_dir, "model.ckpt")

# List ALL tensors example output: v0/Adam (DT_FLOAT) [3,3,1,80]
print_tensors_in_checkpoint_file(file_name=checkpoint_path, tensor_name='')

# List contents of v0 tensor.
# Example output: tensor_name:  v0 [[[[  9.27958265e-02   7.40226209e-02   4.52989563e-02   3.15700471e-02
print_tensors_in_checkpoint_file(file_name=checkpoint_path, tensor_name='v0')

# List contents of v1 tensor.
print_tensors_in_checkpoint_file(file_name=checkpoint_path, tensor_name='v1')

Update: all_tensors argument was added to print_tensors_in_checkpoint_file since Tensorflow 0.12.0-rc0 so you may need to add all_tensors=False or all_tensors=True if required.

Alternative method:

from tensorflow.python import pywrap_tensorflow
import os

checkpoint_path = os.path.join(model_dir, "model.ckpt")
reader = pywrap_tensorflow.NewCheckpointReader(checkpoint_path)
var_to_shape_map = reader.get_variable_to_shape_map()

for key in var_to_shape_map:
    print("tensor_name: ", key)
    print(reader.get_tensor(key)) # Remove this is you want to print only variable names

Hope it helps.