metadata

license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: distilbert-base-uncased-ner
    results: []

distilbert-base-uncased-ner

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1419
Precision: 0.9526
Recall: 0.9431
F1: 0.9479
Accuracy: 0.9434

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: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2866	0.14	500	0.1970	0.9329	0.9213	0.9271	0.9212
0.198	0.28	1000	0.1851	0.9412	0.9218	0.9314	0.9253
0.1892	0.43	1500	0.1772	0.9431	0.9250	0.9340	0.9280
0.179	0.57	2000	0.1697	0.9440	0.9296	0.9367	0.9313
0.1719	0.71	2500	0.1618	0.9453	0.9330	0.9391	0.9339
0.1718	0.85	3000	0.1587	0.9443	0.9351	0.9397	0.9351
0.1664	0.99	3500	0.1569	0.9486	0.9340	0.9412	0.9361
0.1504	1.14	4000	0.1566	0.9480	0.9356	0.9417	0.9368
0.1479	1.28	4500	0.1539	0.9492	0.9369	0.9430	0.9381
0.1467	1.42	5000	0.1501	0.9499	0.9383	0.9441	0.9391
0.1478	1.56	5500	0.1489	0.9513	0.9368	0.9440	0.9390
0.147	1.7	6000	0.1457	0.9503	0.9402	0.9452	0.9407
0.1453	1.85	6500	0.1447	0.9510	0.9408	0.9459	0.9412
0.1384	1.99	7000	0.1442	0.9521	0.9405	0.9463	0.9415
0.1325	2.13	7500	0.1446	0.9494	0.9441	0.9467	0.9425
0.13	2.27	8000	0.1467	0.9524	0.9403	0.9463	0.9416
0.1286	2.41	8500	0.1435	0.9501	0.9440	0.9470	0.9427
0.1311	2.56	9000	0.1446	0.9529	0.9417	0.9473	0.9427
0.1258	2.7	9500	0.1438	0.9528	0.9425	0.9476	0.9431
0.1257	2.84	10000	0.1437	0.9527	0.9431	0.9479	0.9434
0.1289	2.98	10500	0.1420	0.9526	0.9430	0.9478	0.9433

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.0
Tokenizers 0.15.0