metadata

license: mit
base_model: xlm-roberta-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: scenario-TCR-NER_data-univner_half
    results: []

scenario-TCR-NER_data-univner_half

This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1160
Precision: 0.8555
Recall: 0.8189
F1: 0.8368
Accuracy: 0.9828

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: 3e-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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.1189	0.58	500	0.0623	0.8010	0.8531	0.8262	0.9822
0.0469	1.17	1000	0.0640	0.8246	0.8567	0.8404	0.9833
0.0348	1.75	1500	0.0668	0.8335	0.8550	0.8441	0.9834
0.0242	2.33	2000	0.0734	0.8202	0.8538	0.8367	0.9826
0.0215	2.91	2500	0.0717	0.8455	0.8598	0.8526	0.9843
0.0142	3.5	3000	0.0802	0.8383	0.8424	0.8404	0.9836
0.0144	4.08	3500	0.0836	0.8443	0.8554	0.8499	0.9843
0.0103	4.66	4000	0.0811	0.8479	0.8590	0.8534	0.9844
0.0087	5.24	4500	0.0887	0.8364	0.8628	0.8494	0.9840
0.0092	5.83	5000	0.0876	0.8367	0.8430	0.8399	0.9833
0.0076	6.41	5500	0.1004	0.8440	0.8495	0.8468	0.9841
0.007	6.99	6000	0.1080	0.8215	0.8518	0.8364	0.9830
0.0055	7.58	6500	0.0988	0.8454	0.8358	0.8406	0.9831
0.0055	8.16	7000	0.0950	0.8485	0.8461	0.8473	0.9839
0.0044	8.74	7500	0.1001	0.8456	0.8414	0.8435	0.9836
0.004	9.32	8000	0.1084	0.8340	0.8495	0.8417	0.9834
0.004	9.91	8500	0.1175	0.8351	0.8505	0.8427	0.9829
0.0033	10.49	9000	0.1160	0.8555	0.8189	0.8368	0.9828

Framework versions

Transformers 4.33.3
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.13.3