Skip to content

Instantly share code, notes, and snippets.

View grero's full-sized avatar

Roger Herikstad grero

  • National University of Singapore
  • Singapre
View GitHub Profile
Count Overhead File Line Function
===== ======== ==== ==== ========
2 0 @Base/boot.jl 385 eval
2 0 @Base/client.jl
import sys
import pylink
from pylinkwrapper import psychocal, connector
from psychopy import sound, visual,event, tools, data
from psychopy.tools import monitorunittools
import utils
import os
red = (1.0, 0.0, 0.0)
green = (0.0, 1.0, 0.0)
@grero
grero / gist:69ed984fe92c60654cb7
Created January 6, 2015 06:46
htconndor starter log on windows
01/06/15 12:47:18 (pid:3536) About to exec D:\condor\execute\dir_3536\condor_exec.py
01/06/15 12:47:18 (pid:3536) Running job as user condor-slot1
01/06/15 12:47:18 (pid:3536) ERROR: failed to produce Win32 argument string from CreateProcess:
01/06/15 12:47:18 (pid:3536) Create_Process(D:\condor\execute\dir_3536\condor_exec.py,, ...) failed:
01/06/15 12:47:18 (pid:3536) Failed to start job, exiting
01/06/15 12:47:18 (pid:3536) ShutdownFast all jobs.

Keybase proof

I hereby claim:

  • I am grero on github.
  • I am rherikstad (https://keybase.io/rherikstad) on keybase.
  • I have a public key whose fingerprint is D07C 7616 0455 F768 B31D 0C91 3885 9141 3121 4871

To claim this, I am signing this object:

@grero
grero / gist:11309732
Created April 26, 2014 01:57
Crash report for julia using Winston with gtk
Process: julia [1558]
Path: /Applications/Julia-0.3.0-prerelease-a3248be2eb.app/Contents/Resources/julia/bin/julia
Identifier: julia
Version: ???
Code Type: X86-64 (Native)
Parent Process: bash [1498]
Responsible: iTerm [1413]
User ID: 501
Date/Time: 2014-04-26 09:49:15.740 +0800
Using git-annex to keep track of my movie files. i'm just getting to know the basic workflow, but basically, the actual files are located under .git/annex/objects, while the filenames symlink to these. If files go bad, they end up in .git/annex/bad. To synchronize between remotes (without copying the actual files; this just lets annex know which files are, or should be, available), just use the normal git fetch/merge/commit workflow
Fisher and shannon information in finite neural populations.
Stuart Yarrow, Edward Challis, and Peggy Seriès.
The precision of the neural code is commonly investigated using two
families of statistical measures: Shannon mutual information and
derived quantities when investigating very small populations of
neurons and Fisher information when studying large populations. These
statistical tools are no longer the preserve of theorists and are
being applied by experimental research groups in the analysis of
#use this command to submit a bunch of jobs decoding a tetrode data set, in this case P442g0002. {1..35} iterates over the number of highpass files. A wrapper job is submitted that waits for all the compute jobs to finish, then issues an email notifying me that the jobs finishes
echo "P442g0002 done\!" | qsub -j y -V -N hmmDecodeP442g2 -o $HOME/tmp -m e -M roger.herikstad@gmail.com -hold_jid `for i in {1..35}; do echo "cd $PWD;/opt/cluster/tmp/run_hmm_decode.sh /Applications/MATLAB_R2010a.app/ hmmsort/P442g0002 10000 1e-20 SourceFile highpass/P442_highpass.$( printf %.4d $i ) DescriptorFile P442_descriptor.txt hdf5 save Group 2" | qsub -j y -V -N hmmdecodeP442g2_${i} -o $HOME/tmp/; done | awk '{print $3}' | paste -s -d "," -`
#same as above except loops through groups and checks for already computed data
for g in {1..4}; do echo "P443g000$g done\!" | qsub -j y -V -N hmmDecodeP443g$g -o $HOME/tmp -m e -M roger.herikstad@gmail.com -hold_jid `for i in {1..605}; do test -e hmmsort/P443_g$( printf %.4d $g ).$( pr
This is a test. I can't really be sure what I'm going to write, so I'll just stop here. Well, maybe I'll write just a little more. I'm not sure if this works, though, having some issues with syncing here. Ok, apparetly it does, so I guess I'll stop this rubbish and get on with something sensible.