khalidrajan's picture
End of training
5cbc615 verified
|
raw
history blame
2.4 kB
metadata
license: mit
base_model: FacebookAI/roberta-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: roberta_crf_ner_finetuned
    results: []

roberta_crf_ner_finetuned

This model is a fine-tuned version of FacebookAI/roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: nan
  • Precision: 0.8044
  • Recall: 0.6309
  • F1: 0.7014
  • Accuracy: 0.8064

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
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.0 1.0 85 nan 1.0 0.0 0.0 0.7707
0.0 2.0 170 nan 0.5105 0.1412 0.1284 0.8910
0.0 3.0 255 nan 0.3443 0.3458 0.3346 0.9210
0.0 4.0 340 nan 0.5898 0.5990 0.5930 0.9423
0.0 5.0 425 nan 0.5650 0.5795 0.5606 0.9421
0.0 6.0 510 nan 0.6261 0.6867 0.6515 0.9470
0.0 7.0 595 nan 0.6874 0.6640 0.6724 0.9457
0.0 8.0 680 nan 0.6825 0.7224 0.7011 0.9549
0.0 9.0 765 nan 0.6744 0.7224 0.6972 0.9551
0.0 10.0 850 nan 0.7020 0.7062 0.7035 0.9552

Framework versions

  • Transformers 4.44.0
  • Pytorch 2.4.0
  • Datasets 2.21.0
  • Tokenizers 0.19.1