scpanel.select_gene
===================

.. py:module:: scpanel.select_gene


Functions
---------

.. autoapisummary::

   scpanel.select_gene.split_n_folds
   scpanel.select_gene.gene_score
   scpanel.select_gene.plot_gene_score
   scpanel.select_gene.decide_k
   scpanel.select_gene.select_gene
   scpanel.select_gene.select_gene_stable


Module Contents
---------------

.. py:function:: split_n_folds(adata_train: anndata._core.anndata.AnnData, nfold: int, out_dir: Optional[str] = None, random_state: int = 2349) -> Tuple[List[List[int]], List[List[int]], List[List[float]]]

.. py:function:: gene_score(adata_train: anndata._core.anndata.AnnData, train_index_list: List[List[int]], val_index_list: List[List[int]], sample_weight_list: List[List[float]], out_dir: str, ncpus: int, step: float = 0.03, metric: str = 'average_precision', verbose: bool = False) -> Tuple[anndata._core.anndata.AnnData, scpanel.SVMRFECV.RFECV]

.. py:function:: plot_gene_score(adata_train: anndata._core.anndata.AnnData, n_genes_plot: int = 200, width: int = 5, height: int = 4, k: Optional[int] = None) -> matplotlib.axes._axes.Axes

.. py:function:: decide_k(adata_train: anndata._core.anndata.AnnData, n_genes_plot: int = 100) -> int

.. py:function:: select_gene(adata_train: anndata._core.anndata.AnnData, out_dir: Optional[str] = None, step: float = 0.03, top_n_feat: int = 5, n_genes_plot: int = 100, verbose: int = 0) -> anndata._core.anndata.AnnData

.. py:function:: select_gene_stable(adata_train, n_iter=20, nfold=2, downsample_prop_list=[0.6, 0.8], num_cores=1, out_dir=None)