Speeding up post_pred_gen?

[d.r.a...@gmail.com]

Hi,

I was wondering whether it would be possible in any way to speed up execution of the post_pred_gen function.

I currently run two calls in parallel using processes, which maxes out 2 cpus. I want to compare 2 models. However, calling it on the more complex model takes about 11 hours to complete. And to compare models, I have to run it twice for each model: once for the model alone, and once with the result of the previous run as the second argument, and the original data as the first argument. This is to make the comparison between the pooled statistics and the ones for the groups of the split data.

Am I bound by the calculational complexity of the algorithms or is there a way to further pararellize? Or am I overlooking something else?

Cheers,

Daniel

[d.r.a...@gmail.com]

EDIT: gen_post_pred = post_pred_gen

[Thomas Wiecki]

There's a way to adjust the number of samples being drawn but it'll just take a while depending on the number of data points and subjects.

On Wed, Dec 17, 2014 at 8:57 PM, <d.r.a...@gmail.com> wrote:

EDIT: gen_post_pred = post_pred_gen

--
You received this message because you are subscribed to the Google Groups "hddm-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hddm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Thomas Wiecki, PhD

Quantitative Researcher, Quantopian Inc, Boston

[Daniel Azulay]

That worked, I didn't realize the default was 500 samples. Thanks!

You received this message because you are subscribed to a topic in the Google Groups "hddm-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/hddm-users/lBiYQv-QXCo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to hddm-users+...@googlegroups.com.