ERROR:tensorflow:Couldn't understand architecture name ''

I'm following the instruction from google code lab

When I ran the following command I got some error.

python -m scripts.retrain \
    --bottleneck_dir=tf_files/bottlenecks \
    --how_many_training_steps=500 \
    --model_dir=tf_files/models/ \
    --summaries_dir=tf_files/training_summaries/"${ARCHITECTURE}" \
    --output_graph=tf_files/retrained_graph.pb \
    --output_labels=tf_files/retrained_labels.txt \
    --architecture="${ARCHITECTURE}" \

the error are

ERROR:tensorflow:Couldn't understand architecture name ''
Traceback (most recent call last):
  File "/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/", line 162, in _run_module_as_main
    "__main__", fname, loader, pkg_name)
  File "/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/", line 72, in _run_code
    exec code in run_globals
  File "/Users/nutron/Documents/android-codelab-projects/google-codelab-tensorflow-for-poets-2/scripts/", line 1326, in <module>, argv=[sys.argv[0]] + unparsed)
  File "/Users/nutron/tensorflow/lib/python2.7/site-packages/tensorflow/python/platform/", line 48, in run
    _sys.exit(main(_sys.argv[:1] + flags_passthrough))
  File "/Users/nutron/Documents/android-codelab-projects/google-codelab-tensorflow-for-poets-2/scripts/", line 976, in main
    model_info = create_model_info(FLAGS.architecture)
  File "/Users/nutron/Documents/android-codelab-projects/google-codelab-tensorflow-for-poets-2/scripts/", line 923, in create_model_info
    raise ValueError('Unknown architecture', architecture)
ValueError: ('Unknown architecture', '')

What I did it wrong or what should I do?


Apparently, the error happened because ARCHITECTURE was not set. You probably missed the step "Configure your MobileNet" in the doc. Please run the following command:


before python -m scripts.retrain ....

ERROR:tensorflow:Couldn't match files for checkpoint, So, the answer is DO NOT USE SQUARE BRACKETS IN YOUR FILE PATH. Tensorflow can't handle them. ERROR:tensorflow:Couldn't understand architecture name '' Ask Question Asked 2 years, 8 months ago. Active 30 days ago. Viewed 4k times 7. 2. I'm following

For Windows:



python -m scripts.retrain \
--bottleneck_dir=tf_files/bottlenecks \
--how_many_training_steps=500 \
--model_dir=tf_files/models/ \
--summaries_dir=tf_files/training_summaries/%ARCHITECTURE% \
--output_graph=tf_files/retrained_graph.pb \
--output_labels=tf_files/retrained_labels.txt \
--architecture="%ARCHITECTURE%" \

tensorflow / tensorflow, tf.train.latest_checkpoint returns the error ERROR:tensorflow:Couldn't match files for checkpoint when paths in the "checkpoint" file have� ERROR:tensorflow:Couldn't understand architecture name '' Posted by: admin November 1, 2017 Leave a comment. Questions: I’m following the instruction from google

If Max' answer does not work for you on Linux, Try replacing "${ARCHITECTURE}" with "mobilenet_0.50_224"

python -m scripts.retrain

CocoNET error: couldn't match model.ckpt � Issue #1292 � magenta , Hi everyone! When I tried to generate some samples using pretrained model I got the next error: ERROR:tensorflow:Couldn't match files for� Dismiss Join GitHub today. GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.

Build and install error messages, The following list links error messages to a solution or discussion. stream_executor\] Couldn't open CUDA library nvcuda.dll. Dismiss Join GitHub today. GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.

Training checkpoints, The phrase "Saving a TensorFlow model" typically means one of two things: Checkpoints, OR; SavedModel. Checkpoints capture the exact value of all� ERROR:tensorflow:Couldn't match files for checkpoint checkpoints\FCN_epoch_29.ckpt Traceback (most recent call last): #1 Open zhangyuboshengdan opened this issue Mar 26, 2018 · 2 comments

Overfit and underfit, that has much more capacity, far more than the problem would warrant: there are multiple sets of weights values (multiple models) that could explain the� Questions: I have the following MySQL query: SELECT p.*, IF(COUNT(ms.PropertyID) > 0,1,0) AS Contacted, pm.MediaID, date_format(p.AvailableFrom, '%d %b %Y') AS 'AvailableFrom', astext(pg.Geometry) AS Geometry FROM property p, propertygeometry pg JOIN shortlist sl ON sl.PropertyID = AND sl.MemberID = 384216 LEFT JOIN message ms ON ms.PropertyID = AND ms.SenderID = 384216 LEFT JOIN

  • Oh! thanks, I knew it. the root cause is I ran python -m scripts.retrain ... on the another terminal, not the same as the one I set IMAGE_SIZE=224 and ARCHITECTURE="mobilenet_0.50_${IMAGE_SIZE}".
  • I have windows bro not linux what i nead to do
  • why do we need to use these flags ?
  • The program needs to read these environment variable to determine which model to use and what image size to use.
  • I am using windows and tried the above commands, but still getting this error message "ERROR:tensorflow:Couldn't understand architecture name '%architecture%". Can you tell me which version of tensorflow and python you are using