Closed namp closed 6 years ago
I encounter the same problems, the process has trained 12 hours, but it continue.I wonder whether it can be broken.
@namp @zhaochunhua This function is basically a generator for generating batches of training data for keras fit_generator
method. The fit_generator
method uses this generator until it completes the specified number of epochs provided as an argument and proceeds further.
Usually, this batch generator function should provide batches of data indefinitely. Hence, there is no need of breaking out of this.
@zhaochunhua It might be due to large number of epochs. I trained it for about 3 days in a go. You can reduce your training epochs to make it quick.
Ok, but the process ends by self? i just need to know if after 50 epochs the process continue by self Thanks.
In file caption_generator.py in function "data_generator" at line 81 there's a
while 1:
During training I only see thousands of executions of the statement
print "yielding count: "+str(gen_count)
which is within this loop.
Since I don't see any exit conditions or break within this loop, I am wondering when and how do we break out of this loop.
Thanks