Nathan Chell

I believe a segmentation fault is caused by a python extension trying to access memory beyond its reach, so my guess is that you run out of available RAM when...

How many steps and epochs, with what size model did you train for ?

How many steps and epochs, with what size model did you train for ?

When you say ' I also tried to decrease the vocab-size to 15k' did you re-run prepare_data.py ? If not then you need too. The error you are experiencing is,...

HwInfo is very good, quite accurate also. What is the size of your model and what gpu are you training it on?

I mean the amount of neurons and layers you have in your network. You should easily be able to fit the model you have described with a batch size of...

How much system RAM do you have?

Also, you won't want a vocab size of 20k. That's very small. What is the size of your model? Num layers and num units?

It looks like the program is using the normal tf, rather than tf-Gpu

That step time is huge, Are you sure you are training on a GPU? What are your system specs?