Relieve
In the pask week I have been working on implementing the final models for my thesis, which was quite difficult. Why? Well, if you're working with a document collection that is big, like I am working with now, you run into all kinds of memory and speed issues. And I've had those problems in the past week, where a machine with 2Gb of internal memory wouldn't even load half of the model I'm supposed to work with. When I moved to a machine with more internal memory, suddenly I was looking at running times for a single experiment that would last for about 40 days. A few days and a couple of nasty hacks later, I'm looking at running times of under an hour. On top of that, using the models I am able to duplicate the results of an earlier experiment which I did. Which kind of proves that the idea is working. Nice to know that the assumption under which I was working on the past month holds. And relieving ;).
Still much to do, but this is motivating.
Edit : Heh, caught a stupid bug in my tools that produce results for my experiments. Fixing that bug makes for an increase of 20% on one of the types of experiments that I'm doing. Don't mind me too much -- Just happy that the experiments are going well.
Still much to do, but this is motivating.
Edit : Heh, caught a stupid bug in my tools that produce results for my experiments. Fixing that bug makes for an increase of 20% on one of the types of experiments that I'm doing. Don't mind me too much -- Just happy that the experiments are going well.