labels_transfer/xgboost

Description

Performs label transfer from reference to query using XGBoost classifier

Input dataset (query) arguments

Name	Type & Properties	Description
--input	file required	The query data to transfer the labels to. Should be a .h5mu file.
--modality	string	Which modality to use.
--input_obsm_features	string	The `.obsm` key of the embedding to use for the classifier's inference. If not provided, the `.X` slot will be used instead. Make sure that embedding was obtained in the same way as the reference embedding (e.g. by the same model or preprocessing).

Name

Type & Properties

--input

file

required

--modality

string

--input_obsm_features

string

Reference dataset arguments

Name	Type & Properties	Description
--reference	file	The reference data to train classifiers on.
--reference_obsm_features	string required	The `.obsm` key of the embedding to use for the classifier's training. Make sure that embedding was obtained in the same way as the query embedding (e.g. by the same model or preprocessing).
--reference_obs_targets	string multiple	The `.obs` key of the target labels to tranfer.

Name

Type & Properties

--reference

file

--reference_obsm_features

string

required

--reference_obs_targets

string

multiple

Outputs

Name	Type & Properties	Description
--output	file required output	The query data in .h5mu format with predicted labels transfered from the reference.
--output_obs_predictions	string multiple	In which `.obs` slots to store the predicted information. If provided, must have the same length as `--reference_obs_targets`. If empty, will default to the `reference_obs_targets` combined with the `"_pred"` suffix.
--output_obs_uncertainty	string multiple	In which `.obs` slots to store the uncertainty of the predictions. If provided, must have the same length as `--reference_obs_targets`. If empty, will default to the `reference_obs_targets` combined with the `"_uncertainty"` suffix.
--output_uns_parameters	string	The `.uns` key to store additional information about the parameters used for the label transfer.

Name

Type & Properties

--output

file

required

output

--output_obs_predictions

string

multiple

--output_obs_uncertainty

string

multiple

--output_uns_parameters

string

Execution arguments

Name	Type & Properties	Description
--force_retrain -f	boolean_true	Retrain models on the reference even if model_output directory already has trained classifiers. WARNING! It will rewrite existing classifiers for targets in the model_output directory!
--use_gpu	boolean	Use GPU during models training and inference (recommended).
--verbosity -v	integer	The verbosity level for evaluation of the classifier from the range [0,2]
--model_output	file output	Output directory for model

Name

Type & Properties

--force_retrain

-f

boolean_true

--use_gpu

boolean

--verbosity

-v

integer

--model_output

file

output

Learning parameters

Name	Type & Properties	Description
--learning_rate --eta	double	Step size shrinkage used in update to prevents overfitting. Range: [0,1]. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--min_split_loss --gamma	double	Minimum loss reduction required to make a further partition on a leaf node of the tree. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--max_depth -d	integer	Maximum depth of a tree. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--min_child_weight	integer	Minimum sum of instance weight (hessian) needed in a child. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--max_delta_step	double	Maximum delta step we allow each leaf output to be. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--subsample	double	Subsample ratio of the training instances. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--sampling_method	string	The method to use to sample the training instances. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--colsample_bytree	double	Fraction of columns to be subsampled. Range (0, 1]. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--colsample_bylevel	double	Subsample ratio of columns for each level. Range (0, 1]. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--colsample_bynode	double	Subsample ratio of columns for each node (split). Range (0, 1]. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--reg_lambda --lambda	double	L2 regularization term on weights. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--reg_alpha --alpha	double	L1 regularization term on weights. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference
--scale_pos_weight	double	Control the balance of positive and negative weights, useful for unbalanced classes. See https://xgboost.readthedocs.io/en/stable/parameter.html#parameters-for-tree-booster for the reference

Name

Type & Properties

--learning_rate

--eta

double

--min_split_loss

--gamma

double

--max_depth

-d

integer

--min_child_weight

integer

--max_delta_step

double

--subsample

double

--sampling_method

string

--colsample_bytree

double

--colsample_bylevel

double

--colsample_bynode

double

--reg_lambda

--lambda

double

--reg_alpha

--alpha

double

--scale_pos_weight

double

Run this component

Run the following command to execute this component with Nextflow:

cat > params.yaml <<'EOM'  
modality: [ "rna" ]  
reference_obsm_features: [ "X_integrated_scanvi" ]  
reference_obs_targets:  
  [  
    "ann_level_1",  
    "ann_level_2",  
    "ann_level_3",  
    "ann_level_4",  
    "ann_level_5",  
    "ann_finest_level"  
  ]  
output: "$id.$key.output"  
output_uns_parameters: [ "labels_transfer" ]  
use_gpu: [ false ]  
verbosity: [ 1 ]  
model_output: "$id.$key.model_output"  
learning_rate: [ 0.3 ]  
min_split_loss: [ 0 ]  
max_depth: [ 6 ]  
min_child_weight: [ 1 ]  
max_delta_step: [ 0 ]  
subsample: [ 1 ]  
sampling_method: [ "uniform" ]  
colsample_bytree: [ 1 ]  
colsample_bylevel: [ 1 ]  
colsample_bynode: [ 1 ]  
reg_lambda: [ 1 ]  
reg_alpha: [ 0 ]  
scale_pos_weight: [ 1 ]  
id: "run"  
publish_dir: "output/"  
EOM

nextflow run https://packages.viash-hub.com/vsh/openpipeline.git \  
  -revision 1.0.3 \  
  -main-script target/nextflow/labels_transfer/xgboost/main.nf \  
  -params-file params.yaml

labels_transfer/xgboost

Input dataset (query) arguments

Reference dataset arguments

Outputs

Execution arguments

Learning parameters

Run this component

Relationships

Used by

Uses