DNADebertaK6_Arabidopsis

This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.7194

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 600001
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.6174	6.12	20000	1.9257
1.8873	12.24	40000	1.8098
1.8213	18.36	60000	1.7952
1.8042	24.48	80000	1.7888
1.7945	30.6	100000	1.7861
1.7873	36.72	120000	1.7772
1.782	42.84	140000	1.7757
1.7761	48.96	160000	1.7632
1.7714	55.08	180000	1.7685
1.7677	61.2	200000	1.7568
1.7637	67.32	220000	1.7570
1.7585	73.44	240000	1.7442
1.7554	79.56	260000	1.7556
1.7515	85.68	280000	1.7505
1.7483	91.8	300000	1.7463
1.745	97.92	320000	1.7425
1.7427	104.04	340000	1.7425
1.7398	110.16	360000	1.7359
1.7377	116.28	380000	1.7369
1.7349	122.4	400000	1.7340
1.7325	128.52	420000	1.7313
1.731	134.64	440000	1.7256
1.7286	140.76	460000	1.7238
1.7267	146.88	480000	1.7324
1.7247	153.0	500000	1.7247
1.7228	159.12	520000	1.7185
1.7209	165.24	540000	1.7166
1.7189	171.36	560000	1.7206
1.7181	177.48	580000	1.7190
1.7159	183.6	600000	1.7194

Framework versions

• Transformers 4.19.2
• Pytorch 1.11.0
• Datasets 2.2.2
• Tokenizers 0.12.1