search

tpvt99 / character-recognition-cnn

Character Recognition using convolutional neural network with Tensfor flow
8 stars 3 forks source link

training is not being completed and cmd pauses/crashes on epoch 19. #4

Closed AdiPathak97 closed 5 years ago

AdiPathak97 commented 5 years ago

get that file. Thanks.

On Thu, Jun 14, 2018 at 6:59 PM Tran Phong tranphong96.hbk@gmail.com wrote:

oh, you need to train so as to recognize it. run the training file

On Thu, Jun 14, 2018 at 6:53 PM chitransh1998 notifications@github.com wrote:

Hii, I tried to run the test.py and I am getting the following error. I am unable to understand which file it wants.

Traceback (most recent call last): File "test.py", line 114, in new_saver = tf.train.import_meta_graph('my-model.ckpt.meta') File "C:\Users\DELL\Anaconda3\envs\tensorflow\lib\site-packages\tensorflow\python\training\saver.py", line 1947, in import_meta_graph meta_graph_def = meta_graph.read_meta_graph_file(meta_graph_or_file) File "C:\Users\DELL\Anaconda3\envs\tensorflow\lib\site-packages\tensorflow\python\framework\meta_graph.py", line 632, in read_meta_graph_file raise IOError("File %s does not exist." % filename) OSError: File my-model.ckpt.meta does not exist.

[image: image] https://user-images.githubusercontent.com/37555185/41410527-64bed45c-6ff7-11e8-8305-0b69d205cb76.png

Please help, I need it urgently for a project Thankyou

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tpvt99/character-recognition-cnn/issues/1, or mute the thread https://github.com/notifications/unsubscribe-auth/AWNi6ddN94Ad-hUIAwea4SZr3TFsNwfQks5t8k66gaJpZM4Un0Vk .

Originally posted by @tpvt99 in https://github.com/tpvt99/character-recognition-cnn/issues/1#issuecomment-397269617

AdiPathak97 commented 5 years ago

The cause of this error was that 'my-model.ckpt' file wasn't being generated in the current directory as specified in line 185 of cnn.py. This is the following line : saver.save(sess, 'my-model.ckpt')

I resolved this by editing this line to this : saver.save(sess, './my-model.ckpt') This saves generated file into root directory of the folder.