error for the final epoch

[titu1994]

error for the final epoch

[titu1994]

You need to have the AVA.txt file, that matches in size and IDs to all the files in the train directory.

[arieszhang1994]

Yes, In fact I have prepared the dataset and of course checked it.
This error was raised just after the final epoch, which means it should be over correctly but not raising this error

[titu1994]

Does your code stop execution or something ? It says exception ignored.

[arieszhang1994]

yes it stopped execution and gave me the
IndexError: list index out of range
I think there is something wrong with the generator.
Btw, did you use python2 or python3?

[titu1994]

Python 3. See the data loading portion, and see if it loads as many files as you have in the directory. I have a feeling that since you are not loading the full ava dataset, it's getting a mismatched index for a file with the ava scores file.

[arieszhang1994]

Hi titu1994!
Thank you so much for helping me.
In fact, I have rewrote the part of training with tensorflow instead of fit_generator.
At first it raised the similar problem, but I solved with the simple sess.close(). So think maybe this problem had some connection with the memory leak, which depends on version of keras and tensorflow

[liangjin2007]

Change the code from
with tf.Session() as sess:
****
sess.run(.initializer)
while(True):
****
to
with sess.graph.as_default():
****
sess.run(*.initializer)
while(True):
****
And the new code's sess is K.get_session() as input parameter for the generator. This works fine for me.

[wudejian789]

Thank you so much for your fantastic work!
I have just read you code and try to redo it again.
Then i just change the batch to 10, epochs to 2, and use the first 250 pics as training set and the last 50 as val set.
At the first, It seems to be fine.
But I get the Index error in the final epoch.
Exception ignored in: <generator object train_generator at 0x7f8c7433a360> Traceback (most recent call last): File "/home/haoyuzhang/workspace/NIMA_zhanghaoyu/utils/data_loader.py", line 123, in train_generator yield (X_batch, y_batch) File "/home/haoyuzhang/.local/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1503, in __exit__ exec_type, exec_value, exec_tb) File "/usr/lib/python3.5/contextlib.py", line 77, in __exit__ self.gen.throw(type, value, traceback) File "/home/haoyuzhang/.local/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 4342, in get_controller if self.stack[-1] is not default: IndexError: list index out of range
I know there is something wrong with keras.fit_generator but I can't figure it now.
Did some one else have the same situation?

Change the code from
''
with tf.Session() as sess
''
to
''
sess = tf.Session()
''
It will be OK. Very strange!