finetuned__roberta-base-biomedical-clinical-es__augmented-ultrasounds-ner

This model is a fine-tuned version of manucos/finetuned__roberta-base-biomedical-clinical-es__augmented-ultrasounds on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3648
• Precision: 0.8205
• Recall: 0.8927
• F1: 0.8551
• Accuracy: 0.9264

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: 2e-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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	206	0.2948	0.7527	0.8411	0.7945	0.9132
No log	2.0	412	0.2572	0.7746	0.8522	0.8116	0.9235
0.4194	3.0	618	0.2866	0.7759	0.8482	0.8104	0.9215
0.4194	4.0	824	0.2813	0.7878	0.8866	0.8343	0.9235
0.0971	5.0	1030	0.2902	0.7969	0.8856	0.8389	0.9249
0.0971	6.0	1236	0.3229	0.8055	0.8846	0.8432	0.9239
0.0971	7.0	1442	0.3422	0.8028	0.8775	0.8385	0.9208
0.0459	8.0	1648	0.3215	0.8297	0.8877	0.8577	0.9253
0.0459	9.0	1854	0.3568	0.8119	0.8866	0.8476	0.9235
0.0285	10.0	2060	0.3520	0.8145	0.8887	0.8500	0.9235
0.0285	11.0	2266	0.3597	0.8255	0.8907	0.8569	0.9264
0.0285	12.0	2472	0.3599	0.8183	0.8887	0.8520	0.9266
0.0203	13.0	2678	0.3612	0.8195	0.8917	0.8541	0.9246
0.0203	14.0	2884	0.3649	0.8180	0.8917	0.8533	0.9258
0.0164	15.0	3090	0.3648	0.8205	0.8927	0.8551	0.9264

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1