README.md

metadata

base_model: AIRI-Institute/gena-lm-bigbird-base-t2t
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: gena-lm-bigbird-base-t2t_ft_BioS2_1kbpHG19_DHSs_H3K27AC
    results: []

gena-lm-bigbird-base-t2t_ft_BioS2_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of AIRI-Institute/gena-lm-bigbird-base-t2t on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4425
• F1 Score: 0.8721
• Precision: 0.8249
• Recall: 0.9251
• Accuracy: 0.8582
• Auc: 0.9367
• Prc: 0.9328

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score	Precision	Recall	Accuracy	Auc	Prc
0.5218	0.0840	500	0.4577	0.7891	0.8196	0.7608	0.7874	0.8723	0.8607
0.4698	0.1680	1000	0.4514	0.8273	0.7836	0.8762	0.8087	0.8879	0.8785
0.448	0.2521	1500	0.4550	0.8350	0.8002	0.8730	0.8197	0.8978	0.8886
0.4463	0.3361	2000	0.4247	0.8392	0.7616	0.9344	0.8128	0.9076	0.9047
0.4264	0.4201	2500	0.4020	0.8010	0.8781	0.7364	0.8087	0.9130	0.9101
0.401	0.5041	3000	0.3818	0.8560	0.7897	0.9344	0.8356	0.9223	0.9200
0.4035	0.5881	3500	0.3980	0.8590	0.7956	0.9335	0.8398	0.9232	0.9183
0.3774	0.6722	4000	0.3752	0.8543	0.8404	0.8685	0.8450	0.9259	0.9245
0.3985	0.7562	4500	0.4618	0.8360	0.8660	0.8081	0.8343	0.9243	0.9251
0.387	0.8402	5000	0.3753	0.8641	0.8377	0.8923	0.8533	0.9310	0.9299
0.3907	0.9242	5500	0.3589	0.8537	0.8528	0.8547	0.8469	0.9284	0.9284
0.3775	1.0082	6000	0.4544	0.8622	0.8517	0.8730	0.8541	0.9310	0.9296
0.3507	1.0923	6500	0.4114	0.8722	0.8177	0.9344	0.8568	0.9326	0.9246
0.3476	1.1763	7000	0.4028	0.8747	0.8442	0.9074	0.8640	0.9354	0.9348
0.3676	1.2603	7500	0.3671	0.8684	0.8487	0.8891	0.8592	0.9362	0.9358
0.3651	1.3443	8000	0.3837	0.8713	0.8418	0.9029	0.8605	0.9369	0.9364
0.3468	1.4283	8500	0.4281	0.8648	0.8491	0.8811	0.8560	0.9342	0.9309
0.3447	1.5124	9000	0.3955	0.8727	0.8171	0.9364	0.8571	0.9387	0.9380
0.35	1.5964	9500	0.4425	0.8721	0.8249	0.9251	0.8582	0.9367	0.9328

Framework versions

• Transformers 4.42.3
• Pytorch 2.3.0+cu121
• Datasets 2.18.0
• Tokenizers 0.19.0