Dennis, I am replying to the group but mainly you as I doubt Nati really
reads much of what we write.
The use of list() is needed as what comes out of the function may be an
object of type dict_keys which is iterable but if you want it in a list form
all at once, the idiom he uses is common. In the code below, alternating
lines are commands and output, shown by blank lines between pairs.
a={"aleph":1,"bet":2, "ג":3}
a
{'aleph': 1, 'bet': 2, 'ג': 3}
a.keys()
dict_keys(['aleph', 'bet', 'ג'])
list(a.keys())
['aleph', 'bet', 'ג']
The question you ask is what I speculated on. Is he looking for the content
of a specific key in his dictionary called "a" to be something with multiple
values like a list or dictionary. BUT my question is whether reading it into
numpy and converting the top level to a dictionary, results in being able to
store another dictionary within it, or leaves some other data structure that
maybe he needs to also be converted, if that can be done. I could waste more
time playing around but Nati is not an ideal partner to engage with.
I experiment with embedding a dictionary within itself as the second entry
and it works fine BUT I bet what Nati has is not that as it was not made
like that:
a["bet"] = a
a
{'aleph': 1, 'bet': {...}, 'ג': 3}
a["bet"]["aleph"]
1
Now if Nati gave a damn he should do something like I am doing and run his
code part way and LOOK at what he has before continuing to use it as if it
is what he imagines, or worse, he is copying only part of a recipe cobbled
together.
And, agreed, Nati has so many things that look like errors, that it is silly
to ask why the mess did not work, and better to ask how each piece starting
from the top is working or should be adjusted to work.
-----Original Message-----
From: Python-list <python-list-bounces+avi.e.gross=gmail....@python.org> On
Behalf Of Dennis Lee Bieber
Sent: Friday, July 15, 2022 1:11 PM
To: python-list@python.org
Subject: Re: [Neuroimaging] what's the problem??????
On Thu, 14 Jul 2022 17:20:55 -0400, <avi.e.gr...@gmail.com> declaimed the
following:
>Dennis,
>
>I see Nati sent some more code without explaining again what he wants. Yes,
somewhere in this stack of messages he may have said things (that we
generally failed to understand) but it would be helpful to summarize WHY he
sent us what he did or which lines to look at.
>
And that code seems to be doing a lot of repetitive/unneeded
operations... Nothing from
from nilearn import ...
to
masker = ...
appears to depend upon the file name "nii" in the loop and should be
refactored to the top-level. The try/except block for sklearn looks like
another candidate to be pulled out of the file loop.
> If your suggestion is right, he may not understand dictionaries as a
hashed data structure where you can access a.items() or a.keys() or
a.values() if you want a list of sorts of the parts, or a[key] if you want a
particular part.
>
So far as I can tell, the OP only references ONE key in the
dictionary, but apparently expects that key to have multiple values. Yet
there doesn't seem to be a structure /holding/ multiple values -- unless
a["Difumo_names"]
is, itself, returning a dictionary on which .values() can be applied. In
that case, the list() call is likely redundant as .values() already returned
a list (hmmm, is list(some_list) a no-op, or does it wrap some_list into
another list -- in the latter case, the indexing will fail as there is only
one element to access)
> for i in estimator.covariance_:
> r=list(a["Difumo_names"].values())[jsa]
> jsa=jsa+1
> a=dict()
And there is the biggest fault... The OP is completely obliterating
the "a" dictionary, so subsequent passes in that loop have nothing to look
up.
"aas" gets initialized as a dictionary using {}, and never gets any
key:value pairs assigned to it, yet there is a print(aas) in the code.
And that is as far as I can guess -- given the lack of example input
data, example /desired/ output... etc. There are too many apparent errors in
that code sample to even attempt to think what a result could be.
--
Wulfraed Dennis Lee Bieber AF6VN
wlfr...@ix.netcom.com http://wlfraed.microdiversity.freeddns.org/
--
https://mail.python.org/mailman/listinfo/python-list
--
https://mail.python.org/mailman/listinfo/python-list
