Skip to content

Make VS hail independent for the python API #237

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

Open
rocreguant opened this issue Feb 13, 2024 · 0 comments
Open

Make VS hail independent for the python API #237

rocreguant opened this issue Feb 13, 2024 · 0 comments

Comments

@rocreguant
Copy link
Collaborator

No description provided.

NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
FEAT: Implemented RF class method for fitting the model

FEAT: Implemented RF class method for obtaining importance analysis
from a fitted RF

FEAT: Implemented RF class method for returning oob error

FEAT: Implemented RF class method for obtaining FDR
from a fitted model

FEAT: Implemented RF class method for exporting forest to JSON

REFACTOR: Make RF model available at package level

CHORE: Added type checking to all methods
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
REFACTOR: Removed FeatureSource and
ImportanceAnalysis classes from core

REFACTOR: Added FeatureSource import so features
can be returned as a class instantiation
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
REFACTOR: Removed imp analysis and model training

FEAT: Added conversion from feature to RDD (python)

FEAT: Added conversion from feature to RDD (scala)

CHORE: Added type checking
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
separate wrapper file (#237)

REFACTOR: Updated important_variables and variable_importance
methods to convert to pandas DataFrames
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
REFACTOR: Removed model training from object instantation and
updated class to accept a model as a parameter

REFACTOR: Added normalisation as an optional parameter for
variable importance methods

FEAT: Updated variableImportance method to include splitCount in return as it is required for local FDR analysis
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
and passes back to python context (#237)
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
from importAnalysis method of AnalyticsFunctions (#237)
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
FIX: Update export function to process trees in batches,
instead of collecting the whole forest as a map as this
led to OOM errors on large forests
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
REFACTOR: Refactor to mirror changes to python wrapper

FEAT: Include FDR calculation in unit test
NickEdwards7502 added a commit that referenced this issue Sep 11, 2024
FEAT: Implement function for manhattan plotting negative log p values
NickEdwards7502 added a commit that referenced this issue Sep 13, 2024
NickEdwards7502 added a commit that referenced this issue Sep 19, 2024
FEAT: Add wrapper class for importing covariates

FEAT: Add wrapper class for unioning features and covariates
NickEdwards7502 added a commit that referenced this issue Sep 19, 2024
REFACTOR: Include covariate filtering in manhattan plot function

STYLE: Format with black (#237)
NickEdwards7502 added a commit that referenced this issue Sep 19, 2024
FEAT: Add functions for importing std and transposed CSVs

FEAT: Add function for unioning features and covariates
NickEdwards7502 added a commit that referenced this issue Sep 19, 2024
NickEdwards7502 added a commit that referenced this issue Sep 19, 2024
REFACTOR: Remove python component of converting Feature RDD to pandas

FEAT: Add RDD slice to DF function
NickEdwards7502 added a commit that referenced this issue Sep 19, 2024
REFACTOR: Remove conversion of whole RDD to DataFrame

FEAT: Add function for slicing rows and columns and converting to DF
NickEdwards7502 added a commit that referenced this issue Oct 17, 2024
* .bgz loader function implemented by Christina
NickEdwards7502 added a commit that referenced this issue Oct 17, 2024
* Update python wrapper to include imputation strategy parameter

* Update scala API to pass imputation strategy to VCFFeatureSource

* Create functions to handle mode and zero imputation strategies

* Added imputation strategy to test cases

* Added imputation strategy to FeatureSource cli

* Remove sparkPar from test cases due to changes in class signature

* Updated DefVariantToFeatureConverterTest to use zeros imputation
# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant