nichecompass.benchmarking.compute_benchmarking_metrics
- nichecompass.benchmarking.compute_benchmarking_metrics(adata, metrics=['cas', 'mlami', 'clisis', 'gcs', 'cnmi', 'nasw', 'basw', 'blisi', 'kbet', 'pcr'], cell_type_key='cell_type', batch_key=None, spatial_key='spatial', latent_key='nichecompass_latent', pcr_X_pre=None, n_jobs=1, seed=0, mlflow_experiment_id=None)
Compute all specified benchmarking metrics.
- Parameters:
adata (
AnnData) – AnnData object to run the benchmarks for.metrics (
list(default:['cas', 'mlami', 'clisis', 'gcs', 'cnmi', 'nasw', 'basw', 'blisi', 'kbet', 'pcr'])) – List of metrics which will be computed.cell_type_key (
str(default:'cell_type')) – Key under which the cell type annotations are stored in ´adata.obs´.spatial_key (
str(default:'spatial')) – Key under which the spatial coordinates are stored in ´adata.obsm´.latent_key (
str(default:'nichecompass_latent')) – Key under which the latent representation from a model is stored in ´adata.obsm´.pcr_X_pre (
Optional[array] (default:None)) – The unintegrated feature space for the computation of the pcr metric. If None, computes PCA on the raw counts stored in ´adata.X´.n_jobs (
int(default:1)) – Number of jobs to use for parallelization of neighbor search.seed (
int(default:0)) – Random seed for reproducibility.mlflow_experiment_id (
Optional[str] (default:None)) – ID of the Mlflow experiment used for tracking metrics.
- Return type:
- Returns:
benchmarking_dict: Dictionary containing the calculated benchmarking metrics.