kietnt0603's picture
End of training
48eb25b verified
metadata
license: mit
base_model: microsoft/deberta-v3-small
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: deberta-v3-small-nslp-forc-subtask1
    results: []

deberta-v3-small-nslp-forc-subtask1

This model is a fine-tuned version of microsoft/deberta-v3-small on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2167
  • Accuracy: 0.6649
  • Precision: 0.6642
  • Recall: 0.6649
  • F1-weighted: 0.6595

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-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 Accuracy Precision Recall F1-weighted
0.3563 0.77 2000 0.3333 0.5035 0.4651 0.5035 0.4562
0.2443 1.54 4000 0.2647 0.5708 0.5598 0.5708 0.5484
0.1736 2.31 6000 0.2359 0.6152 0.6105 0.6152 0.5969
0.1404 3.08 8000 0.2207 0.6424 0.6391 0.6424 0.6250
0.1109 3.85 10000 0.2181 0.6581 0.6534 0.6581 0.6490
0.0817 4.62 12000 0.2167 0.6649 0.6642 0.6649 0.6595

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.2
  • Datasets 2.17.0
  • Tokenizers 0.15.1