From 540068240b67b3aa5cc06c4d707a231667891b6f Mon Sep 17 00:00:00 2001 From: Artur Szalata Date: Thu, 17 Oct 2024 11:53:47 +0200 Subject: [PATCH] make sc columns categorical --- src/process_dataset/split_sc/script.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/process_dataset/split_sc/script.py b/src/process_dataset/split_sc/script.py index ce4b0c65..725e8234 100644 --- a/src/process_dataset/split_sc/script.py +++ b/src/process_dataset/split_sc/script.py @@ -50,6 +50,9 @@ filtered_sc_counts.uns.clear() filtered_sc_counts.obsm.clear() filtered_sc_counts.obsp.clear() +for col in filtered_sc_counts.obs.columns: + if col not in ["cell_count_by_well_celltype", "cell_count_by_plate_well", "obs_id"]: + filtered_sc_counts.obs[col] = filtered_sc_counts.obs[col].astype("category") print(">> Save sc dataset into splits", flush=True) filtered_sc_counts[filtered_sc_counts.obs["split"] == "train"].write_h5ad(par["sc_train_h5ad"], compression="gzip")