Closed adswa closed 4 years ago
@ElectronicTeaCup @mih: I have pushed an implementation for sample to sample RMS computation on signal data. @ElectronicTeaCup, I would appreciate if you could read R2's suggestion, check my code, and reproduce the results I got (added in the comments of the new function). Also, please chime in to my thoughts on why the method is inapplicable/misleading (as I detail in the comments of the function, and in the letter to the editor started here).
I'll get onto this today. Just need permission on my email (asim.h.dar@gmail.com) for the google doc. As for the reproduction, I'm on my home pc (windows) right now and will see if I can get a python distribution up and runningβso I am unable to guarantee that I can manage running the code. But will try nonetheless.
Thanks @ElectronicTeaCup , and I have added permissions for your e-mail.
PS: If you still have an account on medusa, maybe do it there? could be easier than setting up Python envs under windows...
Ugh π£, lesson learnt about windows. I had thought my user on Medusa would have been pruned, but is still alive thankfully. I'll try it on thereβthanks for the tip.
May not be the time for trouble shooting this but I'm getting this error on medusa (also got this on my windows-ubuntu subsystem setup):
code/mk_figuresnstats.py -f -r -m
Clone attempt: 100%|βββββββββββββββββββββββββββββββββββ| 3.00/3.00 [00:01<00:00, 2.47 Candidate locations/s]Warning: Permanently added the RSA host key for IP address '140.82.112.4' to the list of known hosts.
Permission denied (publickey).
Traceback (most recent call last):
File "code/mk_figuresnstats.py", line 1173, in <module>
savefigs(args.figure, args.stats)
File "code/mk_figuresnstats.py", line 417, in savefigs
stat)
File "code/mk_figuresnstats.py", line 281, in confusion
load_anderson(stimtype, finame)
File "code/mk_figuresnstats.py", line 28, in load_anderson
get(fname)
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/interface/utils.py", line 492, in eval_func
return return_func(generator_func)(*args, **kwargs)
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/interface/utils.py", line 480, in return_func
results = list(results)
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/interface/utils.py", line 411, in generator_func
allkwargs):
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/interface/utils.py", line 527, in _process_results
for res in results:
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/distribution/get.py", line 751, in __call__
description):
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/distribution/get.py", line 456, in _install_targetpath
ds, target_path, reckless, refds_path, description=description):
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/distribution/get.py", line 338, in _install_necessary_subdatasets
description=description):
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/distribution/get.py", line 260, in _install_subds_from_flexible_source
**kwargs):
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/core/distributed/clone.py", line 411, in clone_dataset
create=True)
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/support/gitrepo.py", line 864, in clone
ssh_manager.get_connection(url).open()
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/datalad/support/sshconnector.py", line 287, in open
stderr,
datalad.support.exceptions.ConnectionOpenFailedError: ConnectionOpenFailedError: command '['ssh', '-fN', '-o', 'ControlMaster=auto', '-o', 'ControlPersist=15m', '-o', 'ControlPath=/home/asimio/.cache/datalad/sockets/4137e7ab', 'git@github.com']' failed with exitcode 255
Failed to open SSH connection (could not start ControlMaster process)
Makefile:12: recipe for target 'figures-stamp' failed
Any ideas how to resolve this?
check your SSH key? (Permission denied (publickey)
)
(In any case make sure that you have the studyforrest eyetracking data, but there is no need to run "make" - The new function is not included in the main execution of the script. Simply fire up IPython, and copy/play around with the function I have added in this PR - thx!)
I'm having difficulty getting the data. datalad clone
and git clone --recurse-submodules
don't seem to be working for meβand I couldn't figure out exactly how to use datalad in this case. Or are the data not supposed to be local?
datalad.support.exceptions.NoDatasetFound: No dataset found at '/mnt/c/Users/asimd/Repos/paper-remodnav'. Specify a dataset to work with by providing its path via the
datasetoption, or change the current working directory to be in a dataset.
Please post the full command you are running, and where you run it. It seems that it simply doesn't see a dataset at all.
I'm running these with the remodnav paper repo, so: datalad clone https://github.com/psychoinformatics-de/paper-remodnav.git
. Just tried it with datalad install
and following it up with make
to pull the data, but make doesn't go through. Now i'm doing
datalad install https://github.com/psychoinformatics-de/paper-remodnav.git
cd paper-remodnav
make
This runs the analysis but has an error.
Traceback (most recent call last):
File "code/mk_figuresnstats.py", line 1177, in <module>
plot_raw_vel_trace()
File "code/mk_figuresnstats.py", line 636, in plot_raw_vel_trace
names=['x', 'y', 'pupil', 'frame'])
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/numpy/lib/npyio.py", line 2372, in recfromcsv
output = genfromtxt(fname, **kwargs)
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/numpy/lib/npyio.py", line 1772, in genfromtxt
fid = np.lib._datasource.open(fname, 'rt', encoding=encoding)
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/numpy/lib/_datasource.py", line 269, in open
return ds.open(path, mode, encoding=encoding, newline=newline)
File "/home/asimio/env/remodnav/lib/python3.5/site-packages/numpy/lib/_datasource.py", line 623, in open
raise IOError("%s not found." % path)
OSError: data/raw_eyegaze/sub-32/beh/sub-32_task-movie_run-5_recording-eyegaze_physio.tsv.gz not found.
Makefile:12: recipe for target 'figures-stamp' failed
make: *** [figures-stamp] Error 1
How would I go about pulling the data so that I can call S2SRMS()
in ipython?
Ah, I see. Will push a fix soon, one sec.
@ElectronicTeaCup I pushed a fix and merged it in in #14. Please check whether it works for you. It was a missing subdataset installation and missing data retrieval for two files were plotted raw gaze traces with. You could have gotten the data "by hand" by installing/cloning the subdatasets manually and simply getting the data you need.
I works for me with a fresh dataset clone now. Here is what I am running on our compute cluster juseless
(no access to medusa anymore, sorry). I'm posting the full output for you to check where yours may differ:
I just noticed that you would manually need to retrieve all the files in data/raw_eyegaze
- no previous function did this. If you don't mind, please add a datalad get into the function and push it here, just to make the life of any other person who wants to recompute this easier. thx.
With the last push, it works well (it didn't go all the way because I didn't have Inkscape installed, but broke right at the end). Looks good, the output is pretty much the same:
\newcommand{\RMSlab}{0.05849002328619966}
\newcommand{\RMSmri}{0.06121409026190385}
Thank you! Could you sanity check my code, please? Do you think it does what R2 suggested? thx!
On it, will update you soon
Yeah I think it checkouts out with the description, plus nicely commented.
Cool, thanks. In this case, I think, this is ready to go. @mih, whenever you have time, take a look at the letter to the editor and see whether you think its sufficient - thx :)
@ElectronicTeaCup @mih: I believe that STD/S2S-RMS computation on eyetracking data is misguided. I have started a draft for a letter to the editor here. Please let me know what you think.