-
Notifications
You must be signed in to change notification settings - Fork 792
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Segmentation #677
Merged
Merged
Segmentation #677
Changes from 1 commit
Commits
Show all changes
99 commits
Select commit
Hold shift + click to select a range
efae20b
Updated filter and rank
c5ae7ef
Unchanged. Pushed for testing
65e55d3
Code typed out, need to check
d37e55f
Done with code?
088d248
Pushing
2d208c0
Check
268be4d
Final push
a39c2ae
filter by class error
7c47d58
parentesis
53ecd84
now?
75daf1b
Fixed axes in display
1dc48da
Fixed by_class
6ceb09b
Update rank.py
vdlad 6010b8a
Fixed type and returned pxlscres"num_pixel_issues"
6a75c84
fixed optional for type
942f855
finished type fix
fe470e6
fixed error
5dd4e9c
Done?
9506826
dont code at 1am
ceaefe2
fixed ranking return of pixel scores
3839a15
Apply suggestions from code review
vdlad f4e77a4
Typos in fixes
1cffaa4
First set of edits!
2314660
Second set of changes
f1ae8f5
Check if code still works
bebb090
fixed circular import
ef31784
black
1a8148f
Added legend
c33275b
black
0550549
Kwargs, import
8e7d65d
Unused
ad9a005
Type check
cde47e6
black
b1164da
Fixed black formatting
ulya-tkch 2c6a5a6
Added testing (codecov) framework + fixed format
ulya-tkch 46b41e2
black and pixels
83bb1c0
Softmin code opt
4755c84
black
0aa661b
black
8a6edda
fix df summary
954fd29
tutorial
2613b0a
added one test
0e828bf
added a test
9becf26
Review tutorial notebook + add TODOS
ulya-tkch bebeef4
Merge branch 'master' of https://github.com/vdlad/cleanlab into uly-u…
ulya-tkch 19eb748
stash
858d408
Merge branch 'master' of https://github.com/cleanlab/cleanlab
5bba20c
Add docstring build
ulya-tkch f81b6e4
Added my first test
b51c732
Added my first test
b9a88a9
added tests and swapped h and w
bb316a0
Merge branch 'master' of https://github.com/cleanlab/cleanlab
af3592b
tests/removed venv
03104c9
added check for _label
457284f
re-fix
5634138
fixed rank
32f2667
finished rank tests
8272734
cases
a4db45f
done with rank now
ec185b7
black
192c197
more black
0093301
value errors for issues
b9c348b
more test cases
9cab290
added None case
5c2cccd
Done with rank and filter tests
430ecba
fixed given_label==predicted_label bug
3552328
rename
5d8fc75
rename
1a22338
finished tut
9bd5532
Add summary tests
ulya-tkch 2c2b5e6
Udate tutorial notebook mini gramar edits/working
ulya-tkch de21c04
Merge branch 'master' into master
ulya-tkch 345da7c
small fixes
111a6d8
Apply suggestions from code review
vdlad d500af4
Update index.rst
vdlad 3ad57ca
Add deleted test
ulya-tkch 5414869
fixed the scores only and other docstrings
254a2f5
black
e0211d7
Merge branch 'master' of github.com:vdlad/cleanlab
04dae61
Increase codecov
ulya-tkch 54ae54d
downsampled
557faf9
Merge branch 'master' of github.com:vdlad/cleanlab
acd1b61
tutorial done
c210900
got rid of scores only test
3cae619
downsample to kwarg
71a1fc2
black
dabda9b
different size input
29af6d9
bug fix
7cb6aaf
black+bug
0d2c1a2
fixed tutorial?
2bfa220
Clear nb output for tutorial
ulya-tkch 8fb8240
Improve docstrings
ulya-tkch f409e4e
Add extra tests + functions for checking Optional args
ulya-tkch f6577d2
tutorial language and clarity
jwmueller 6a9e012
fix broken notebook formatting
jwmueller e3362b4
Update parameter names
ulya-tkch 7fc6a11
Update tutorial notebook with unit tests, review wording
ulya-tkch ccb3597
Fixed logic in display
ulya-tkch ba4f57a
Merge branch 'master' of https://github.com/cleanlab/cleanlab
ulya-tkch File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
small fixes
- Loading branch information
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
what about the batch_size for find_label_issues()
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sorry I don't follow - find_label_issues has a batch_size argument
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In that I am saying that the downsample does not work for the softmin method - maybe I should clarify that
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
please don't resolve conversations involving a question, I will not see answer if you resolve it.
Here you are allowing user to specify
downsampleargument viaget_label_quality_scores(), and I am proposing you should also allow user to specifybatch_sizeargument. I get that both arguments would only apply when the method is num_pixel_issues and would be ignored if it is softmin.And yes please explicitly mention that in docstring
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
right - there is a batch_size argument for get_label_quality_scores() - am I misunderstanding something?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
sorry! resolved cause both get_label_quality_scores() and find_label_issues() both have a downsamples and batch_size argument
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do you mean I should offer downsample in get_label_quality_scores() as an argument and not as a kwarg?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
They should both be kwargs or not. It does not make sense for one to be kwarg and the other not. I prefer both being kwargs.
They should both be at the end of the arguments list too and docstring should clarify: These arguments are only considered when method = "num_pixel_issues"
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
generally don't split up arguments that only apply in a special case in the arg order between other args that always apply
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we talked about this earlier and this is what we came to because for find_label_issues() needs downsampling to be worthwhile and the default method for get_label_quality_score() is softmin.
If we make them both kwargs then what would you suggest the default downsample value to be? Right now it is 16, but if that number is not divisible by their image dimensions then it will fail and the user would have to go then set the kwarg which seems annoying to do.
If we make downsample a kwarg for both methods, then I feel the default should be 1 or 2. Which are more likely to be a factor of the dimension. Thoughts?