adata_pred_prob

besca.tl.auto_annot.adata_pred_prob(classifier, scaler, adata_pred, adata_orig, threshold=0)

predicts on testing set using trained classifier and returns class probability for every cell and every class

Parameters:

• test_path (string) – path of test dataset

• test_dataset (string) – name of test dataset

• classifier (sklearn object) – trained classifier of chosen type

• scaler (sklearn.preprocessing.StandardScaler) – scaler fit to training dataset

• adata_pred (AnnData) – Preprocessed anndata object containing testing dataset.

• threshold (float | default = 0) – value between 0 and 1, used for classifying into unknowns

Returns:

Original anndata object with annotation added as column and a column for each training celltype probability

Return type:

AnnData