metadata

license: cc-by-sa-4.0
tags:
  - generated_from_trainer
datasets:
  - fin
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: fin5
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: fin
          type: fin
          config: default
          split: train
          args: default
        metrics:
          - name: Precision
            type: precision
            value: 0.9243027888446215
          - name: Recall
            type: recall
            value: 0.9243027888446215
          - name: F1
            type: f1
            value: 0.9243027888446215
          - name: Accuracy
            type: accuracy
            value: 0.9908666100254885

fin5

This model is a fine-tuned version of nlpaueb/sec-bert-shape on the fin dataset. It achieves the following results on the evaluation set:

Loss: 0.0752
Precision: 0.9243
Recall: 0.9243
F1: 0.9243
Accuracy: 0.9909

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: 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	129	0.0825	0.8327	0.8924	0.8615	0.9811
No log	2.0	258	0.0633	0.8593	0.9243	0.8906	0.9866
No log	3.0	387	0.0586	0.9038	0.9363	0.9198	0.9894
0.0547	4.0	516	0.0607	0.9357	0.9283	0.932	0.9911
0.0547	5.0	645	0.0656	0.9216	0.9363	0.9289	0.9904
0.0547	6.0	774	0.0692	0.9249	0.9323	0.9286	0.9909
0.0547	7.0	903	0.0716	0.9246	0.9283	0.9264	0.9904
0.0019	8.0	1032	0.0742	0.9213	0.9323	0.9267	0.9909
0.0019	9.0	1161	0.0748	0.9246	0.9283	0.9264	0.9909
0.0019	10.0	1290	0.0752	0.9243	0.9243	0.9243	0.9909

Framework versions

Transformers 4.25.1
Pytorch 1.13.0+cu116
Datasets 2.7.1
Tokenizers 0.13.2