metadata

license: apache-2.0
base_model: dslim/distilbert-NER
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: distilbert-finetuned-ner
    results: []

distilbert-finetuned-ner

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

Loss: 0.4109
Precision: 0.6952
Recall: 0.7549
F1: 0.7238
Accuracy: 0.8724

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-06
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

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
1.647	1.0	977	0.8265	0.4918	0.5741	0.5298	0.7793
0.7697	2.0	1954	0.6350	0.5801	0.6567	0.6160	0.8194
0.6089	3.0	2931	0.5591	0.6138	0.6857	0.6478	0.8352
0.534	4.0	3908	0.5163	0.6296	0.6955	0.6609	0.8439
0.4911	5.0	4885	0.4885	0.6436	0.7075	0.6740	0.8498
0.4545	6.0	5862	0.4683	0.6526	0.7165	0.6830	0.8557
0.4379	7.0	6839	0.4534	0.6600	0.7231	0.6901	0.8592
0.4124	8.0	7816	0.4441	0.6713	0.7274	0.6982	0.8625
0.403	9.0	8793	0.4345	0.6746	0.7359	0.7039	0.8658
0.394	10.0	9770	0.4324	0.6835	0.7445	0.7127	0.8667
0.3782	11.0	10747	0.4256	0.6820	0.7465	0.7128	0.8678
0.3706	12.0	11724	0.4213	0.6873	0.7460	0.7155	0.8691
0.3712	13.0	12701	0.4197	0.6873	0.7518	0.7181	0.8703
0.3626	14.0	13678	0.4163	0.6882	0.7523	0.7188	0.8713
0.351	15.0	14655	0.4142	0.6905	0.7528	0.7203	0.8717
0.3528	16.0	15632	0.4142	0.6932	0.7538	0.7222	0.8718
0.3523	17.0	16609	0.4123	0.6949	0.7533	0.7229	0.8722
0.3464	18.0	17586	0.4107	0.6936	0.7538	0.7224	0.8727
0.342	19.0	18563	0.4115	0.6954	0.7560	0.7244	0.8726
0.3496	20.0	19540	0.4109	0.6952	0.7549	0.7238	0.8724

Framework versions

Transformers 4.41.1
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1