mrm8488's picture
End of training
40abe33
metadata
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: codebert-base-finetuned-code-ner
    results: []

codebert-base-finetuned-code-ner

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

  • Loss: 0.3522
  • Precision: 0.6297
  • Recall: 0.6417
  • F1: 0.6356
  • Accuracy: 0.9185

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: 64
  • eval_batch_size: 64
  • 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 1.0 191 0.4601 0.4861 0.4578 0.4715 0.8853
No log 2.0 382 0.3989 0.5806 0.5243 0.5510 0.8996
0.5081 3.0 573 0.3547 0.5723 0.6017 0.5866 0.9059
0.5081 4.0 764 0.3507 0.6161 0.6115 0.6138 0.9135
0.5081 5.0 955 0.3412 0.6299 0.6252 0.6276 0.9161
0.2299 6.0 1146 0.3418 0.6162 0.6465 0.6310 0.9175
0.2299 7.0 1337 0.3497 0.6288 0.6287 0.6287 0.9175
0.1618 8.0 1528 0.3474 0.6340 0.6397 0.6368 0.9189
0.1618 9.0 1719 0.3501 0.6262 0.6432 0.6346 0.9179
0.1618 10.0 1910 0.3522 0.6297 0.6417 0.6356 0.9185

Framework versions

  • Transformers 4.23.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.6.1
  • Tokenizers 0.13.1