Pancreas (random split)

Human pancreatic islet scRNA-seq data from 6 datasets across technologies (CEL-seq, CEL-seq2, Smart-seq2, inDrop, Fluidigm C1, and SMARTER-seq). Split into train/test randomly.

Pancreas (random split)

Rank	F1 score	Macro F1 score	Accuracy	Memory (GB)	Runtime (min)	CPU (%)	Name	Code	Year
13	0.21	0.07	0.21	3.70	3.05	12.60	Random Labels	v0.6.0	2022
14	0.17	0.04	0.34	3.60	5.20	7.00	Majority Vote	v0.6.0	2022
9	0.95	0.84	0.95	6.50	0.78	1,274.60	K-neighbors classifier (log CPM)	v1.1.2	1967
2	0.98	0.96	0.98	6.40	1.42	474.80	Multilayer perceptron (log CPM)	v1.1.2	1990
1	0.98	0.96	0.98	6.50	0.91	486.10	Logistic regression (log CPM)	v1.1.2	2013
3	0.98	0.89	0.98	8.90	1.45	1,780.80	XGBoost (log CPM)	v1.6.2	2016
11	0.94	0.53	0.95	9.60	39.95	1,738.30	scArches+scANVI (All genes)	v0.17.1	2021
12	0.94	0.59	0.95	22.50	13.28	208.60	Logistic regression (log scran)	v1.1.2	2013
4	0.98	0.85	0.98	22.50	14.55	458.70	Multilayer perceptron (log scran)	v1.1.2	1990
7	0.96	0.78	0.96	22.50	13.65	123.10	K-neighbors classifier (log scran)	v1.1.2	1967
10	0.94	0.53	0.95	6.90	91.87	1,059.70	scArches+scANVI (Seurat v3 2000 HVG)	v0.17.1	2021
6	0.97	0.68	0.97	5.80	48.92	1,279.30	scANVI (Seurat v3 2000 HVG)	v0.17.1	2021
5	0.97	0.88	0.97	23.10	14.78	292.20	XGBoost (log scran)	v1.6.2	2016
8	0.97	0.67	0.97	6.10	158.48	956.90	scANVI (All genes)	v0.17.1	2021