-
Notifications
You must be signed in to change notification settings - Fork 86
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Support of multimodal data in DataSourceSplitter + bugfix #1119 #1138
Conversation
Hello @kasyanovse! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:
Comment last updated at 2023-08-14 13:15:41 UTC |
Codecov Report
@@ Coverage Diff @@
## master #1138 +/- ##
==========================================
- Coverage 78.62% 78.51% -0.12%
==========================================
Files 131 130 -1
Lines 9401 9336 -65
==========================================
- Hits 7392 7330 -62
+ Misses 2009 2006 -3
|
Fix #1119 Add support of MultiModalData in cv_folds Delete some code that solves problems are solved in the new DataSourceSplitter or in new cv_folds
0888f6d
to
0e3f875
Compare
fedot/core/data/cv_folds.py
Outdated
in the InputData format for cross validation. The function | ||
return a generator of tuples, consisting of a pair of train, test. | ||
|
||
:param data: InputData for train and test splitting |
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.
Если data
может быть и InputData
, и MultiModalData
, лучше определить обобщённее: data for train and test splitting
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.
Исправлено.
Applied only for time series data. | ||
If not provided, then value will be calculated. | ||
:param random_seed: Random seed for shuffle. | ||
:param log: Log for logging. |
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.
Порядок переменных в докстринге не совпадает с порядком переменных в классе. Так и должно быть?
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.
Исправлено.
Не забудь только pep8-замечания поправить. |
Поправил. Осталось два замечания, но с кодом в указанных местах все нормально. |
train_test_data_setup
was retained for backward compatibility)DataSourceSplitter
andtrain_test_data_setup
core.validation.split.py
module is moved tocore.data.cv_folds.py
DataObjectiveAdvisor
is deleted because stratification is enable by defaultOneFoldInputDataSplit
is deleted becausecv_generator
get checked parameters fromDataSourceSplitter
. Cross validation always may be done.DataSourceSplitter
andtrain_test_data_setup
are calledwith
stratify=False
in that cases.