Edit model card

Cybonto-distilbert-base-uncased-finetuned-ner-Wnut17

This model is a fine-tuned version of distilbert-base-uncased on the wnut_17 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5062
  • Precision: 0.6603
  • Recall: 0.4682
  • F1: 0.5479
  • Accuracy: 0.9355

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: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 107 0.3396 0.6470 0.4269 0.5144 0.9330
No log 2.0 214 0.3475 0.5948 0.4539 0.5149 0.9335
No log 3.0 321 0.3793 0.6613 0.4253 0.5177 0.9332
No log 4.0 428 0.3598 0.6195 0.4944 0.5500 0.9354
0.0409 5.0 535 0.3702 0.5802 0.4571 0.5113 0.9308
0.0409 6.0 642 0.4192 0.6546 0.4459 0.5305 0.9344
0.0409 7.0 749 0.4039 0.6360 0.4610 0.5346 0.9354
0.0409 8.0 856 0.4104 0.6564 0.4587 0.5400 0.9353
0.0409 9.0 963 0.3839 0.6283 0.4944 0.5534 0.9361
0.0132 10.0 1070 0.4331 0.6197 0.4547 0.5245 0.9339
0.0132 11.0 1177 0.4152 0.6196 0.4817 0.5420 0.9355
0.0132 12.0 1284 0.4654 0.6923 0.4507 0.5460 0.9353
0.0132 13.0 1391 0.4869 0.6739 0.4436 0.5350 0.9350
0.0132 14.0 1498 0.4297 0.6424 0.4769 0.5474 0.9353
0.0061 15.0 1605 0.4507 0.6272 0.4626 0.5325 0.9340
0.0061 16.0 1712 0.4410 0.6066 0.4793 0.5355 0.9335
0.0061 17.0 1819 0.4851 0.6639 0.4523 0.5381 0.9351
0.0061 18.0 1926 0.4815 0.6553 0.4563 0.5380 0.9346
0.0035 19.0 2033 0.5188 0.6780 0.4420 0.5351 0.9350
0.0035 20.0 2140 0.4986 0.6770 0.4698 0.5547 0.9363
0.0035 21.0 2247 0.4834 0.6552 0.4714 0.5483 0.9355
0.0035 22.0 2354 0.5094 0.6784 0.4595 0.5479 0.9358
0.0035 23.0 2461 0.4954 0.6583 0.4579 0.5401 0.9354
0.0026 24.0 2568 0.5035 0.6667 0.4595 0.5440 0.9354
0.0026 25.0 2675 0.5000 0.6599 0.4658 0.5461 0.9355
0.0026 26.0 2782 0.4968 0.6697 0.4738 0.5549 0.9357
0.0026 27.0 2889 0.4991 0.6545 0.4714 0.5481 0.9352
0.0026 28.0 2996 0.4936 0.6508 0.4769 0.5505 0.9353
0.0021 29.0 3103 0.5005 0.6535 0.4722 0.5482 0.9353
0.0021 30.0 3210 0.5062 0.6603 0.4682 0.5479 0.9355

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.10.0+cu111
  • Datasets 2.1.0
  • Tokenizers 0.12.1
Downloads last month
2
Hosted inference API
Token Classification
Examples
Examples
This model can be loaded on the Inference API on-demand.

Dataset used to train theResearchNinja/Cybonto-distilbert-base-uncased-finetuned-ner-Wnut17

Evaluation results