Tensorflow retrain.py Directory image not found error

I am testing Tensorflow for Poets from here , which reinstalls the latest late Inception on Ubuntu 14.04 processor

after the guide i ran this code

    python tensorflow/examples/image_retraining/retrain.py \
–bottleneck_dir=tf_files/bottlenecks \
–how_many_training_steps 500 \
–model_dir=tf_files/inception \
–output_graph=tf_files/retrained_graph.pb \
–output_labels=tf_files/retrained_labels.txt \
–image_dir=tf_files/flower_photos

      

and I got the following error:

Image directory '' not found.
Traceback (most recent call last):
  File "tensorflow/examples/image_retraining/retrain.py", line 774, in <module>
    tf.app.run()
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/platform/app.py", line 30, in run
    sys.exit(main(sys.argv))
  File "tensorflow/examples/image_retraining/retrain.py", line 671, in main
    class_count = len(image_lists.keys())
AttributeError: 'NoneType' object has no attribute 'keys

      

The knowledge of this issue refers to where my images are stored, I put them in my home folder which correctly refers to the last command image_dir=tf_files/flower_photos

So now my question is, why am I getting this error even though the image directory is correct?

+3


source to share


5 answers


If you look at line 774 where the error occurs, it is the summary directory, not the image_dir.



You will need to set the summary directory here: https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/image_retraining/retrain.py#L928

+1


source


retrain.py now exists in the following location:

github.com/tensorflow/hub/raw/master/examples/image_retraining/retrain.py 

      

instead

tensorflow/examples/image_retraining

      



You can use this:

curl -LO https://github.com/tensorflow/tensorflow/raw/master/tensorflow/examples/label_image/label_image.py python label_image.py \
--graph=/tmp/output_graph.pb --labels=/tmp/output_labels.txt \

--input_layer=Placeholder \

--output_layer=final_result \

--image=$HOME/flower_photos/daisy/21652746_cc379e0eea_m.jpg

      

Link: https://www.tensorflow.org/hub/tutorials/image_retraining

+1


source


There is no double hyphen in your code block. So try this:

python tensorflow/examples/image_retraining/retrain.py \
–-bottleneck_dir=tf_files/bottlenecks \
–-how_many_training_steps 500 \
–-model_dir=tf_files/inception \
–-output_graph=tf_files/retrained_graph.pb \
–-output_labels=tf_files/retrained_labels.txt \
–-image_dir=tf_files/flower_photos

      

0


source


Requests use an absolute path to solve the problem. for example/home/pi/Desktop/hub/retrained_data/dataset

0


source


Step 1

Follow this link:

https://colab.research.google.com/drive/1hFte7dNUZGVukuR0OY02t0atJgMzn6oc#scrollTo=Bl-yFZqKeezp&forceEdit=true&offline=true&sandboxMode=true

Step 2

Go to the second chapter

Step 3

Worked by adding two lines as below before the script section: LOL

It's LOL that just added the next two lines and it worked up to the script section and it worked.

Two lines:

if not os.path.isdir("../Pokemons-subset"):

     print('Error')

      

Script section:

 %run scripts/retrainn.py   \

   --bottleneck_dir=tf_files/bottlenecks   \

   --how_many_training_steps=4000   \

   --model_dir=tf_files/models/   \

   --summaries_dir=tf_files/training_summaries/mobilenet_0.50_160   \

   --output_graph=tf_files/retrained_graph.pb   \

   --output_labels=tf_files/retrained_labels.txt   \

   --learning_rate=0.01   \

   --architecture=mobilenet_0.50_160   \

   --image_dir=../Pokemons-subset \

   --flip_left_right \

   --random_crop=10 \

   --random_scale=10 \

   --random_brightness=10

      

0


source







All Articles