Table 2 Performance of baseline models on benchmark datasets
From: Genomic benchmarks: a collection of datasets for genomic sequence classification
Pytorch | Tensorflow | |||
|---|---|---|---|---|
Dataset | Accuracy | F1 score | Accuracy | F1 score |
dummy_mouse_enhancers_ensembl | 69.0 | 70.4 | 50.0 | 66.9 |
demo_coding_vs_intergenomic_seqs | 87.6 | 86.8 | 89.6 | 89.4 |
demo_human_or_worm | 93.0 | 92.8 | 94.2 | 93.2 |
drosophila_enhancers_stark | 58.6 | 44.5 | 52.4 | 69.1 |
human_enhancers_cohn | 69.5 | 67.1 | 68.9 | 71.3 |
human_enhancers_ensembl | 68.9 | 56.5 | 81.1 | 74.6 |
human_ensembl_regulatory | 93.3 | 93.3 | 79.3 | 79.3 |
human_nontata_promoters | 84.6 | 83.7 | 86.5 | 84.4 |
human_ocr_ensembl | 68.0 | 66.1 | 68.8 | 72.0 |