NeRUBioS_RoBERTa_base_bne_Training_Development

ajtamayoh/NeRUBioS_RoBERTa_base_bne_Training_Development

NeRUBioS_RoBERTa_base_bne_Training_Development is a token classification model from ajtamayoh. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

19% of all-time downloads

All-time downloads

Public

Parameters

124M

4.5 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors496 MB · 99%

At a glance

Task: Token Classification
Library: transformers
License: apache-2.0
Model type: roberta
Access: Public
Created: Apr 13, 2024
Updated: Apr 13, 2024
SHA: 7a1a57d2

Try a prompt

Base models

PlanTL-GOB-ES/roberta-base-bne

Model card

From the Hugging Face model README

NeRUBioS_RoBERTa_base_bne_Training_Development

This model is a fine-tuned version of PlanTL-GOB-ES/roberta-base-bne on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3499
Negref Precision: 0.5449
Negref Recall: 0.5380
Negref F1: 0.5414
Neg Precision: 0.9559
Neg Recall: 0.9694
Neg F1: 0.9626
Nsco Precision: 0.8730
Nsco Recall: 0.9062
Nsco F1: 0.8893
Unc Precision: 0.8315
Unc Recall: 0.8764
Unc F1: 0.8534
Usco Precision: 0.6608
Usco Recall: 0.7383
Usco F1: 0.6974
Precision: 0.8205
Recall: 0.8453
F1: 0.8327
Accuracy: 0.9526

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: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Negref Precision	Negref Recall	Negref F1	Neg Precision	Neg Recall	Neg F1	Nsco Precision	Nsco Recall	Nsco F1	Unc Precision	Unc Recall	Unc F1	Usco Precision	Usco Recall	Usco F1	Precision	Recall	F1	Accuracy
0.1898	1.0	1729	0.1783	0.4516	0.5316	0.4884	0.9351	0.9596	0.9472	0.8079	0.8539	0.8303	0.8193	0.7529	0.7847	0.5816	0.6406	0.6097	0.7596	0.8041	0.7813	0.9452
0.1163	2.0	3458	0.1724	0.4906	0.5527	0.5198	0.9274	0.9760	0.9511	0.8252	0.9026	0.8622	0.8263	0.8263	0.8263	0.5662	0.6680	0.6129	0.7721	0.8376	0.8036	0.9485
0.0621	3.0	5187	0.1946	0.5139	0.5063	0.5101	0.9524	0.9618	0.9571	0.8542	0.8836	0.8687	0.8071	0.8726	0.8386	0.6034	0.6836	0.6410	0.7999	0.8249	0.8122	0.9480
0.0378	4.0	6916	0.2279	0.4923	0.5401	0.5151	0.9450	0.9749	0.9597	0.8568	0.8884	0.8723	0.8259	0.8610	0.8431	0.6179	0.6758	0.6455	0.7940	0.8347	0.8138	0.9490
0.0192	5.0	8645	0.2495	0.5227	0.5338	0.5282	0.9541	0.9760	0.9649	0.8256	0.8884	0.8558	0.8071	0.8726	0.8386	0.6049	0.6758	0.6384	0.7929	0.8351	0.8135	0.9508
0.0134	6.0	10374	0.2764	0.5199	0.5232	0.5216	0.9568	0.9672	0.9620	0.8687	0.8955	0.8819	0.8277	0.8533	0.8403	0.6389	0.7188	0.6765	0.8114	0.8347	0.8229	0.9514
0.0068	7.0	12103	0.2876	0.4880	0.5169	0.5020	0.9470	0.9760	0.9613	0.8593	0.8919	0.8753	0.8494	0.8494	0.8494	0.6456	0.7188	0.6802	0.8010	0.8351	0.8177	0.9508
0.0059	8.0	13832	0.2886	0.4991	0.5591	0.5274	0.9488	0.9705	0.9595	0.8601	0.8907	0.8751	0.8231	0.8803	0.8507	0.6528	0.7344	0.6912	0.7986	0.8446	0.8209	0.9516
0.0029	9.0	15561	0.3290	0.5408	0.4895	0.5138	0.9529	0.9716	0.9622	0.8653	0.9002	0.8824	0.8218	0.8726	0.8464	0.6090	0.7422	0.6690	0.8125	0.8358	0.8240	0.9505
0.0009	10.0	17290	0.3582	0.5438	0.5105	0.5267	0.9519	0.9716	0.9616	0.8757	0.9038	0.8895	0.8218	0.8726	0.8464	0.6737	0.75	0.7098	0.8227	0.8413	0.8319	0.9506
0.0012	11.0	19019	0.3516	0.5139	0.5443	0.5287	0.9539	0.9705	0.9621	0.8834	0.9086	0.8958	0.8291	0.8803	0.8539	0.6761	0.75	0.7111	0.8157	0.8489	0.8320	0.9526
0.0005	12.0	20748	0.3499	0.5449	0.5380	0.5414	0.9559	0.9694	0.9626	0.8730	0.9062	0.8893	0.8315	0.8764	0.8534	0.6608	0.7383	0.6974	0.8205	0.8453	0.8327	0.9526

Framework versions

Transformers 4.38.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2

Task: Token Classification
Library: transformers
Type: roberta
License: apache-2.0
Created: Apr 13, 2024
Updated: Apr 13, 2024