metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - conllpp
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: roberta-large-md-conllpp
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: conllpp
          type: conllpp
          config: conllpp
          split: train
          args: conllpp
        metrics:
          - name: Precision
            type: precision
            value: 0.9971177780689113
          - name: Recall
            type: recall
            value: 0.9968043586452576
          - name: F1
            type: f1
            value: 0.9969610437242934
          - name: Accuracy
            type: accuracy
            value: 0.995003768708948

roberta-large-md-conllpp

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

Loss: 0.0457
Precision: 0.9971
Recall: 0.9968
F1: 0.9970
Accuracy: 0.9950

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-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0748	1.0	878	0.0309	0.9959	0.9962	0.9961	0.9935
0.0111	2.0	1756	0.0346	0.9974	0.9967	0.9970	0.9951
0.0057	3.0	2634	0.0348	0.9974	0.9960	0.9967	0.9946
0.0031	4.0	3512	0.0434	0.9976	0.9964	0.9970	0.9951
0.0017	5.0	4390	0.0457	0.9971	0.9968	0.9970	0.9950

Framework versions

Transformers 4.21.2
Pytorch 1.12.1
Datasets 2.4.0
Tokenizers 0.12.1