metadata

license: mit
base_model: microsoft/deberta-v3-base
tags:
  - generated_from_trainer
datasets:
  - nbroad/company_names
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: deberta-v3-base-company-names
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: nbroad/company_names
          type: nbroad/company_names
        metrics:
          - name: Precision
            type: precision
            value: 0.7739696312364425
          - name: Recall
            type: recall
            value: 0.7962863774326013
          - name: F1
            type: f1
            value: 0.7849694196330357
          - name: Accuracy
            type: accuracy
            value: 0.9769126125154315

deberta-v3-base-company-names

This model is a fine-tuned version of microsoft/deberta-v3-base on the nbroad/company_names dataset. It achieves the following results on the evaluation set:

Loss: 0.0693
Precision: 0.7740
Recall: 0.7963
F1: 0.7850
Accuracy: 0.9769

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: 8e-05
train_batch_size: 48
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0752	1.0	2126	0.0664	0.7416	0.7979	0.7687	0.9757
0.0484	2.0	4252	0.0652	0.7725	0.7903	0.7813	0.9768
0.0415	3.0	6378	0.0693	0.7740	0.7963	0.7850	0.9769

Framework versions

Transformers 4.34.1
Pytorch 2.0.1+cu117
Datasets 2.16.1
Tokenizers 0.14.1