CODEWITHSUNDEEP
tensorflowmachine-learningoutput
Rolf12
Rolf12

Reputation: 771

Tensorflow how to inspect model

I built a Tensorflow RNN model and would like to inspect the models results (e.g. which features/variables are used and how strongly etc.)

I have created the following files:

But I have issues to read those files. I found the following code:

from tensorflow.python import pywrap_tensorflow

model_file = "/trained/checkpoint"
reader = pywrap_tensorflow.NewCheckpointReader(model_file)
var_to_shape_map = reader.get_variable_to_shape_map()

for key in sorted(var_to_shape_map):
    print("tensor_name: ", key)
    print(reader.get_tensor(key))

I get the following error:

checkpoint: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restore operator?

Is checkpoint the wrong file? it has no extension in my folder, it just says Type is Data?

Any help would be great!

Upvotes: 2

Views: 1885

Answers (1)

amiabl
amiabl

Reputation: 1088

You can inspect the tensor of your checkpoints with inspect_checkpoint function from the tensorflow python tools.

Example from tensorflow doc:

# import the inspect_checkpoint library
from tensorflow.python.tools import inspect_checkpoint as chkp

# print all tensors in checkpoint file
chkp.print_tensors_in_checkpoint_file("/tmp/model.ckpt", tensor_name='', all_tensors=True)

# tensor_name:  v1
# [ 1.  1.  1.]
# tensor_name:  v2
# [-1. -1. -1. -1. -1.]

https://www.tensorflow.org/guide/saved_model#inspect_variables_in_a_checkpoint

Upvotes: 1

Related Questions