metadata

license: mit
base_model: xlm-roberta-large
tags:
  - generated_from_trainer
datasets:
  - conll2003
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: XLM-RoBERTa-Large-Conll2003-English-NER-Finetune
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: conll2003
          type: conll2003
          config: conll2003
          split: test
          args: conll2003
        metrics:
          - name: Precision
            type: precision
            value: 0.9247648902821317
          - name: Recall
            type: recall
            value: 0.9401558073654391
          - name: F1
            type: f1
            value: 0.932396839332748
          - name: Accuracy
            type: accuracy
            value: 0.9851405190050608

XLM-RoBERTa-Large-Conll2003-English-NER-Finetune

This model is a fine-tuned version of xlm-roberta-large on the conll2003 dataset. It achieves the following results on the evaluation set:

Loss: 0.2205
Precision: 0.9248
Recall: 0.9402
F1: 0.9324
Accuracy: 0.9851

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: 5e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.6457	0.3333	1441	0.1841	0.8167	0.8336	0.8250	0.9674
0.0893	0.6667	2882	0.1399	0.8827	0.8950	0.8888	0.9781
0.0637	1.0	4323	0.1449	0.8939	0.9024	0.8981	0.9802
0.0457	1.3333	5764	0.1552	0.8955	0.9163	0.9057	0.9816
0.0387	1.6667	7205	0.1566	0.9041	0.9233	0.9136	0.9825
0.0401	2.0	8646	0.1493	0.8982	0.9311	0.9144	0.9824
0.0276	2.3333	10087	0.1655	0.9038	0.9299	0.9167	0.9820
0.0248	2.6667	11528	0.1783	0.9127	0.9309	0.9217	0.9829
0.0266	3.0	12969	0.1601	0.9120	0.9340	0.9228	0.9833
0.0166	3.3333	14410	0.1801	0.9181	0.9288	0.9234	0.9842
0.0187	3.6667	15851	0.1717	0.9170	0.9325	0.9247	0.9843
0.0185	4.0	17292	0.1653	0.9190	0.9343	0.9266	0.9844
0.0126	4.3333	18733	0.1845	0.9176	0.9343	0.9259	0.9843
0.0133	4.6667	20174	0.1855	0.9174	0.9322	0.9247	0.9837
0.0119	5.0	21615	0.1782	0.9168	0.9329	0.9248	0.9843
0.01	5.3333	23056	0.1892	0.9173	0.9366	0.9269	0.9843
0.0083	5.6667	24497	0.1800	0.9251	0.9343	0.9297	0.9845
0.0079	6.0	25938	0.1868	0.9237	0.9352	0.9294	0.9851
0.0059	6.3333	27379	0.2073	0.9178	0.9350	0.9263	0.9842
0.0068	6.6667	28820	0.2061	0.9195	0.9379	0.9286	0.9843
0.0062	7.0	30261	0.2011	0.9215	0.9377	0.9295	0.9846
0.0037	7.3333	31702	0.2100	0.9209	0.9373	0.9290	0.9846
0.0043	7.6667	33143	0.2145	0.9202	0.9389	0.9295	0.9847
0.0039	8.0	34584	0.2070	0.9256	0.9377	0.9316	0.9852
0.0024	8.3333	36025	0.2138	0.9218	0.9394	0.9306	0.9851
0.0034	8.6667	37466	0.2159	0.9229	0.9394	0.9311	0.9849
0.003	9.0	38907	0.2156	0.9244	0.9377	0.9310	0.9846
0.002	9.3333	40348	0.2201	0.9252	0.9402	0.9326	0.9849
0.0015	9.6667	41789	0.2217	0.9245	0.9393	0.9318	0.9850
0.0028	10.0	43230	0.2205	0.9248	0.9402	0.9324	0.9851

Framework versions

Transformers 4.41.1
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1