TypeError: in method 'ReadFromStream', argument 2 of type 'size_t' in compression model

[truboxl]

System information

Google Colab

Describe the problem

Tried to run image_encoder compression model in Colab, stuck at Comparing Similarity using msssim.py

Attached a written Colab file, just hit Run All:
https://colab.research.google.com/gist/truboxl/1c20fe7b6bc7ff52ceff9a39acdbd721/set-up-tf-models-compression-imageencoder.ipynb

Source code / logs

Traceback (most recent call last):
File "msssim.py", line 217, in
tf.app.run()
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/platform/app.py", line 40, in run
_run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
File "/usr/local/lib/python3.6/dist-packages/absl/app.py", line 299, in run
_run_main(main, args)
File "/usr/local/lib/python3.6/dist-packages/absl/app.py", line 250, in _run_main
sys.exit(main(argv))
File "msssim.py", line 202, in main
img1_str = image_file.read('rb')
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/lib/io/file_io.py", line 128, in read
pywrap_tensorflow.ReadFromStream(self._read_buf, length))
TypeError: in method 'ReadFromStream', argument 2 of type 'size_t'

[tensorflowbutler]

Thank you for your post. We noticed you have not filled out the following field in the issue template. Could you update them if they are relevant in your case, or leave them as N/A? Thanks.
What is the top-level directory of the model you are using
Have I written custom code
OS Platform and Distribution
TensorFlow installed from
TensorFlow version
Bazel version
CUDA/cuDNN version
GPU model and memory
Exact command to reproduce

[truboxl]

The Colab file also includes the commented line for outputting cat tf_env.txt... zzz

What is the top-level directory of the model you are using
models/research/compression/image_encoder

Have I written custom code
No, refer Colab file

OS Platform and Distribution
os: Linux
os kernel version: #1 SMP Thu Aug 8 02:47:02 PDT 2019
os release version: 4.14.137+
os platform: Linux-4.14.137+-x86_64-with-Ubuntu-18.04-bionic
linux distribution: ('Ubuntu', '18.04', 'bionic')
linux os distribution: ('Ubuntu', '18.04', 'bionic')
mac version: ('', ('', '', ''), '')
uname: uname_result(system='Linux', node='ec7cae62dad7', release='4.14.137+', version='#1 SMP Thu Aug 8 02:47:02 PDT 2019', machine='x86_64', processor='x86_64')
architecture: ('64bit', 'ELF')
machine: x86_64

TensorFlow installed from
Google Colab(?)

TensorFlow version
1.15.0
(2.0.0 straight up broke encoder.py decoder.py msssim.py with a different error)

Bazel version
bash: line 168: bazel: command not found

CUDA/cuDNN version
NVIDIA-SMI 440.33.01 Driver Version: 418.67 CUDA Version: 10.1

GPU model and memory
Tesla K80

Exact command to reproduce
python msssim.py --original_image=./example.png --compared_image=./output/image_15.png

[brkygokcen]

I edited the code in "msssim.py" file like below,

201 with tf.gfile.FastGFile(FLAGS.original_image, 'rb') as image_file:
202 img1_str = image_file.read()
203 with tf.gfile.FastGFile(FLAGS.compared_image, 'rb') as image_file:
204 img2_str = image_file.read()

Then it is working now.

[ravikyram]

@truboxl

Is this still an issue.Please, close this thread if your issue was resolved.Thanks!

[truboxl]

@ravikyram I think this can be closed once @brkygokcen changes are applied. I confirmed the changes are working with tf-gpu 1.15.

But then tf-gpu 2.x just broke again, probably a different issue...

Traceback (most recent call last):
  File "decoder.py", line 30, in <module>
    tf.flags.DEFINE_string('input_codes', None, 'Location of binary code file.')
AttributeError: module 'tensorflow' has no attribute 'flags'

[ravikyram]

@truboxl

Can we close this issue and open new issue by filling issue template and by providing related code. Thanks!

[truboxl]

...this is still an issue, it is broken out of the box.

[truboxl]

https://github.com/tensorflow/models/tree/archive/research#warning-archived-models-and-implementations

Since the models are now archived, removed from master branch, no longer maintained and not updated for use with TF2,
I guess there's nothing I can do about anymore.

Closing issue...

[google-ml-butler]

Are you satisfied with the resolution of your issue?
Yes
No