| RNA-ERNIE | H-Net | |
|---|---|---|
| accuracy | 0.7922 | 0.5018 |
| recall | 0.7923 | 0.2509 |
| precision | 0.7973 | 0.5000 |
| f1 | 0.7914 | 0.3314 |
| auc | 0.8862 | 0.5022 |
x1CUDA_VISIBLE_DEVICES=0 python run_rr_inter.py \2--model_name_or_path output/PROMPT,BERT/checkpoint-18750/optimizer.pt \3--dataset_dir data/ft/rr/ \4--dataset MirTarRAW \5--output output_ft/ \6--device gpu \7--batch_size 32 \8--save_every_n_epochs 1 \9--num_train_epochs 10
xxxxxxxxxx1python task/rna_rna_interaction.py \2--dataset_path "data/ft/rr_inter/MirTarRAW/MirTarRAW.csv" \3--pretrained_path "output/pretrain/checkpoint_20251212_114554_epoch_1.pt" \4--epochs 10 \5--batch_size 32 \6--lr 1e-6 \7--scheduler_type cosine \8--warmup_epochs 1
| RNA-ERNIE | H-Net | |
|---|---|---|
| accuracy | 0.7754 | 0.0992 |
| recall | 0.7754 | 0.0151 |
| precision | 0.8027 | 0.0992 |
| f1 | 0.7783 | 0.0260 |
| auc | 0 | 0 |
61CUDA_VISIBLE_DEVICES=0 python run_seq_cls.py \2--dataset nRC \3--batch_size 32 \4--num_train_epochs 10 \5--model_name_or_path "./output/PROMPT,BERT/checkpoint-18750/" \6--output "./output_ft/seq_cls/nRC"81CUDA_VISIBLE_DEVICES=2 python task/sequence_classification.py \2--dataset_dir ./data/ft/seq_cls/nRC/ \3--pretrained_path ./output/pretrain/checkpoint_20251212_114554_epoch_1.pt \4--config_path ./configs/hnet_1stage_L.json \5--output_dir ./output_ft/seq_cls/nRC \6--batch_size 16 \7--epochs 10 \8--lr 1e-4
| RNA-ERNIE | H-Net | |
|---|---|---|
| accuracy | 0.5000 | 0.5000 |
| recall | 0.5000 | 0.5000 |
| precision | 0.2500 | 0.2500 |
| f1 | 0.3333 | 0.3334 |
| auc | 0.5040 | 0.5001 |
61CUDA_VISIBLE_DEVICES=1 python run_seq_cls.py \2--dataset lncRNA_M \3--batch_size 4 \4--num_train_epochs 10 \5--model_name_or_path "./output/PROMPT,BERT/checkpoint-18750/" \6--output "./output_ft/seq_cls/lncRNA_M"81CUDA_VISIBLE_DEVICES=0 python task/sequence_classification.py \2--dataset_dir ./data/ft/seq_cls/lncRNA_M/ \3--pretrained_path ./output/pretrain/checkpoint_20251212_114554_epoch_1.pt \4--config_path ./configs/hnet_1stage_L.json \5--output_dir ./output_ft/seq_cls/lncRNA_M \6--batch_size 8 \7--epochs 10 \8--lr 1e-4
| RNA-ERNIE | H-Net | |
|---|---|---|
| accuracy | 0.5036 | 0.5000 |
| recall | 0.5000 | 0.5000 |
| precision | 0.2518 | 0.2500 |
| f1 | 0.3349 | 0.3333 |
| auc | 0.4506 | 0.5094 |
61CUDA_VISIBLE_DEVICES=2 python run_seq_cls.py \2--dataset lncRNA_H \3--batch_size 4 \4--num_train_epochs 10 \5--model_name_or_path "./output/PROMPT,BERT/checkpoint-18750/" \6--output "./output_ft/seq_cls/lncRNA_H"81CUDA_VISIBLE_DEVICES=2 python task/sequence_classification.py \2--dataset_dir ./data/ft/seq_cls/lncRNA_H/ \3--pretrained_path ./output/pretrain/checkpoint_20251212_114554_epoch_1.pt \4--config_path ./configs/hnet_1stage_L.json \5--output_dir ./output_ft/seq_cls/lncRNA_H \6--batch_size 8 \7--epochs 10 \8--lr 1e-4