MRG: somewhat better plots #9
Conversation
By default, pyplot.stem draw stems starting from 0 which is not appropriate for SNR.
Using a stem plot is left as an option.
|
Oh, one more thing: I tripped a couple of times over the |
tutorials/time-freq/ssvep.py
Outdated
| ############################################################################### | ||
| # SNR plotting function | ||
| # ^^^^^^^^^^^^^^^^^^^^^^^^ | ||
| def plot_snr_spectrum(snrs, freqs, use_stem_plot=False, stim_freq=None, bg_var_trials=False, bg_var_channels=False, |
There was a problem hiding this comment.
sorry for being a bit inconsistent with earlier, but i think here the function makes it more difficult than necessary :D
the catches and further options are not necessary, if we dont use them in the tutorial, and the other plots are also written in plain code.
..this would also make the n_dimensions fix obsolete (which is indeed the better name)
There was a problem hiding this comment.
No problem! I'll remove the function.
|
btw, i have added your fork as remote so i can easily checkout your working branch and build the tutorial myself. |
|
alright with the welch parameters. maybe someone has feedback/a strong opinion if we open the upstream PR |
|
I’ve had so much trouble getting the docs to build that I don’t want to
make anyone else go through it if it can be avoided :)
…On Wed, 2 Dec 2020 at 20:23, Dominik Welke ***@***.***> wrote:
btw, i have added your fork as remote so i can easily checkout your
working branch and build the tutorial myself.
no need to attach a pdf, if you dont want to :)
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#9 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACGUAWV5TL4YGF2SYSG2CY3SSZZ2ZANCNFSM4UH7R7EQ>
.
|
|
About Welch: we can just use multitaper or even plain DTFT to get our power
values. No parameters -no questions :)
…On Wed, 2 Dec 2020 at 20:26, Dominik Welke ***@***.***> wrote:
alright with the welch parameters. maybe someone has feedback/a strong
opinion if we open the upstream PR
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#9 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACGUAWURHO4GJDUMSP3EBKLSSZ2DBANCNFSM4UH7R7EQ>
.
|
Also changed "trial average" to "averaged over trials", etc.
|
Another option is to plots SNR on the logarithmic scale:
|
closes #2
In the end, I decided not to mess with the Welch parameters because (a) line plots with grey background look reasonable enough to me with the current parameters, (b) I realized I have no idea how to decide on the window size in a principled manner and got annoyed by trying random parameters and reloading the plots :)
Otherwise, I just copied your function, switched from stem plots to line plots for the grand averages by default, and edited the function a bit. I committed every small edit so that you could go through them one by one and we could reverse some of them if necessary.
A pdf of the tutorial is attached.
201130_ssvep_tutorial.pdf