Show all posts by user
Dear AFNI users-
We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:
https://discuss.afni.nimh.nih.gov
Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.
The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.
Sincerely,
AFNI HQ
History of AFNI updates
Page 1 of 2
Pages: 12
Results 1 - 30 of 33
Could someone remind me how to load a new convexity file into the SUMA viewer? That is, SUMA has foreground (fg) and background (bk) color plane stacks. The "Load Dset" button seems only to load into the foreground stack. What is the method for loading into the background stack?
cheers
iain
by
iain
-
AFNI Message Board
I have been occasionally getting this error:
** glob error: No such file or directory
** You may need to 'setenv AFNI_SHELL_GLOB YES'
**
** In particular, if you are trying to access an NFS (network file
** system) mounted drive, you might be running into the situation
** where the NFS 'cookie' length on the remote system does not
** match the cookie length on your lo
by
iain
-
AFNI Message Board
OK, making headway now, maybe. I updated again. I do have the Apr 15 version this time. Unfortunately, I still get the pointer error but, the line number is now #9037.
*** MCW_malloc pre-corruption! serial=514770 size=486184 source=3dDeconvolve.c line#9037 <- calculate_results <- 3dDeconvolve main
*** MCW_malloc post-corruption! serial=514770 size=486184 source=3dDeconvolve.c line#90
by
iain
-
AFNI Message Board
H Rick,
My machine is running RHEL 6.6 with 48 GB RAM. I watched the memeory load on system monitor when I executed the command. There appeared to be a healthy saftey margin.
If I set AFNI_NOMMAP = YES, 3dDeconvolve crashes at the same place.
++ Number of time points: 121546 (no censoring)
+ Number of parameters: 6604 [4 baseline ; 6600 signal]
++ total shared memory needed = 1,6
by
iain
-
AFNI Message Board
When I run the same script but replace all the stim_times_IM flags with stim_times flags, 3dDeconvolve runs past the crash point, writing out the X mat files, etc.
Will upload data.
by
iain
-
AFNI Message Board
Here is the version (compile) line for the last crash.
++ 3dDeconvolve: AFNI version=AFNI_2011_12_21_1014 (Apr 13 2015) [64-bit]
I'll try leaving the 'IM' out.
by
iain
-
AFNI Message Board
Thanks Ziad. Unfortunately, I still get the same error.
*** MCW_malloc pre-corruption! serial=514769 size=486184 source=3dDeconvolve.c line#9033 <- calculate_results <- 3dDeconvolve main
*** MCW_malloc post-corruption! serial=514769 size=486184 source=3dDeconvolve.c line#9033
<- calculate_results <- 3dDeconvolve main
*** glibc detected *** 3dDeconvolve: munmap_chunk(): inval
by
iain
-
AFNI Message Board
Hi Bob and Co.,
I am getting the following memory error with 3dDeconvolve:
*** MCW_malloc pre-corruption! serial=514767 size=972368 source=3dDeconvolve.c line#9033 <- calculate_results <- 3dDeconvolve main
*** MCW_malloc post-corruption! serial=514767 size=972368 source=3dDeconvolve.c line#9033
<- calculate_results <- 3dDeconvolve main
*** glibc detected *** 3dDeconvolve:
by
iain
-
AFNI Message Board
Isaac, you make a compelling case. I'll undo the edit of subprocess.
Also, I may have found the culprit: iTerm.app. It turns out there is a well documented issue with duplicate PATH variables in OS X 10.10 when using iTerm:
My symptoms were nearly identical, just with tcsh, not zsh. I had two PATH variables (other duplicates as well) in my tcsh environment but (ba)sh purges duplic
by
iain
-
AFNI Message Board
@Isaac Whoops! I had misread your commands. I was using cap C.
idewitt@wernicke [3] sh -c 'echo $test'
val
by
iain
-
AFNI Message Board
@Isaac, further:
idewitt@wernicke [14] setenv foo bar
idewitt@wernicke [15] echo $foo
bar
idewitt@wernicke [16] sh -C 'echo $foo'
sh: echo $foo: No such file or directory
idewitt@wernicke [17] sh -C 'echo foo'
sh: echo foo: No such file or directory
idewitt@wernicke [18] sh 'echo foo'
sh: echo foo: No such file or directory
idewitt@wernicke [19] sh '
by
iain
-
AFNI Message Board
@Daniel, I have no .bashrc or .profile, just .tcshrc. Perhaps I'm being a bit Homer with his hand stuck in the vending machine here. I could take the path of least resistance and duplicate my .tcshrc as a .bashrc. This would indeed make the problem go away.
@Isaac, interestingly, I don't get the same output:
idewitt@wernicke [1] setenv test val
idewitt@wernicke [2] echo $test
by
iain
-
AFNI Message Board
here it is:
idewitt@wernicke [1] /bin/sh --version
GNU bash, version 3.2.53(1)-release (x86_64-apple-darwin14)
by
iain
-
AFNI Message Board
After some digging, it turns out subprocess.Popen (invoked around line 864 in afni_base) defaults to /bin/sh. (Is this new?) My login shell is /bin/tcsh. (Has been for years.) Accordingly, I specify modifications to PATH, including $AFNI_HOME, in ~/.tcshrc, which /bin/sh, of course, does not look at. The odd thing is that I have been running this set up for years without this error. The only thin
by
iain
-
AFNI Message Board
Hi Isaac,
The question marks are output from the shell. The command ends where you note there is no trailing backslash. Viewed in the shell command history, the multiline command translates to:
align_epi_anat.py -overwrite -dset1 ${subj}.r11.brainmasked_median+orig -dset2 ${subj}.r${run}.brainmasked+orig -dset2to1 -dset2_base median -big_move -dset1_strip None -dset2_strip None -cost lpa -v
by
iain
-
AFNI Message Board
Hi Daniel,
No errors. Here is the output. Looks OK to me but maybe you will see something.
--iain
yojo@waugh [7] 3dAttribute DELTA ./Dorothy.r11.brainmasked+orig
1.515152 1.515152 -1.9
yojo@waugh [8] 3dinfo ./Dorothy.r11.brainmasked+orig
++ 3dinfo: AFNI version=AFNI_2011_12_21_1014 (Dec 10 2014) [64-bit]
Dataset File: Dorothy.r11.brainmasked+orig
Identifier Code: XYZ_f3MSR
by
iain
-
AFNI Message Board
Hi Rick, Ziad, Daniel,
I'm getting the following error with the following command:
yojo@waugh [29] align_epi_anat.py \
-overwrite \
-dset1 ${subj}.r11.brainmasked_median+orig \
-dset2 ${subj}.r${run}.brainmasked+orig \
-dset2to1 \
-dset2_base median \
-big_move \
-dset
by
iain
-
AFNI Message Board
Hi Ziad and co.,
I am loading some stats buckets into the SUMA viewer and getting variable results with respect to the sub-brick selector index. When I load some buckets I get the expected pull down menus for I T and B. However, for others, I get only numeric text entry fields pre-populated with the value 0 and accompanying up/down buttons to the left. I checked the respective files with 3dinf
by
iain
-
AFNI Message Board
Hi Ziad,
Can you help me understand what is going wrong with one of my surfaces? It comes from FreeSurfer and looks fine when viewed with MGH tools. I could even swear I loaded it into suma a year ago but now, when I try to view it or use it in analysis pipelines I get errors. I tried to reimport the surface with
@SUMA_Make_Spec_FS -fspath $SUBJECTS_DIR/$subj -sid $subj
The script runs
by
iain
-
AFNI Message Board
I read a paper today where someone used the t-statistic as the input to a group-level t-test. My understanding is that it is the beta weight that should be propagated up to 2nd level analyses. Am I wrong? If not, how wrong is it to use the t-statistic?
cheers
Iain
by
iain
-
AFNI Message Board
Hi Rick,
I notice that I can run 3dClustSim with surface masks. Can you flesh out for us (me) why this would not be a good idea?
cheers
Iain
by
iain
-
AFNI Message Board
Hi Ziad,
Is there a way to project a surface ROI from one hemisphere to another? I'm trying to do something analogous to a L/R sign flip in Tal space. I suppose I could map my surface ROI to a volume, transpose it, and then map it back but, is there is a more SUMA-ish way of doing this?
cheers
Iain
by
iain
-
AFNI Message Board
Hi Ziad,
Thanks for making this fix! Unfortunately, I'm getting a new error now:
ROIgrow -overwrite -spec /Volumes/Hume.d0/Users/dewitt/mrdata/freesurfer/subjects/DE000/SUMA/std.141.DE000_lh.spec -surf_A lh.smoothwm.asc -roi_nodes roi-SUMA-1.closest.1D.dset[0] -lim 5 -full_list -roi_labels 1 -prefix ROI-1.1D
Fatal Signal 11 (SIGSEGV) received
ROIgrow
Bottom of Debug Stack
Wh
by
iain
-
AFNI Message Board
Thanks Gang,
I was thinking of (1) as well at the time of the post but neglected to mention it. The others are kind of extensions of that.
To clarify, I was wondering if there was ever a reason to have something like this [1 -3 -1 1 3], an arbitrary contrast that sets some values positive, others negative, and they do not sum to zero. This is more the kind of thing that puzzled me. Why wo
by
iain
-
AFNI Message Board
Could someone suggest documentation, a website or citation for appreciating the differences between general linear tests and contrasts? I know contrast weights sum to zero and GLTs do not need to. I'm wondering about the interpretation of non-zero summing GLTs and when and why someone would use them.
cheers
Iain
by
iain
-
AFNI Message Board
Hi Ziad,
I'm getting an error loading spec files with ROIgrow:
Error SUMA_BrainWrap_ParseInput:
Option -spec not understood.
Try -help for usage
The help function suggests -spec input should be fine. Here's my command:
ROIgrow -overwrite \
-spec $SUBJECTS_DIR/DE000/SUMA/std.141.DE000_${hemis}.spec \
-surf_A ${hemis}.smoothwm.asc \
-sv ${subj}_SurfVol_ns_Alnd_Exp+orig \
by
iain
-
AFNI Message Board
Hi Ziad and co.,
I would like to make circular surface ROIs centered at given x,y,z coordinates. The idea is to do something similar to what can be done with 3dUndump for volume data. I've found documentation on how to draw surface ROIs freehand but nothing on how to automate their production with coordinate seeds. Is there a program that would do this? If not, how might I hack something
by
iain
-
AFNI Message Board
Page 1 of 2
Pages: 12