metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- wnut_17
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bert-base-multilingual-cased-WNUT-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: wnut_17
type: wnut_17
config: wnut_17
split: test
args: wnut_17
metrics:
- name: Precision
type: precision
value: 0.5913669064748202
- name: Recall
type: recall
value: 0.3809082483781279
- name: F1
type: f1
value: 0.463359639233371
- name: Accuracy
type: accuracy
value: 0.9500726682055228
bert-base-multilingual-cased-WNUT-ner
This model is a fine-tuned version of bert-base-multilingual-cased on the wnut_17 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3832
- Precision: 0.5914
- Recall: 0.3809
- F1: 0.4634
- Accuracy: 0.9501
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: 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: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 213 | 0.2791 | 0.6008 | 0.2817 | 0.3836 | 0.9427 |
No log | 2.0 | 426 | 0.2697 | 0.6520 | 0.3299 | 0.4382 | 0.9479 |
0.148 | 3.0 | 639 | 0.2846 | 0.5783 | 0.3661 | 0.4484 | 0.9492 |
0.148 | 4.0 | 852 | 0.3032 | 0.6248 | 0.3642 | 0.4602 | 0.9500 |
0.0413 | 5.0 | 1065 | 0.3355 | 0.5729 | 0.3568 | 0.4397 | 0.9495 |
0.0413 | 6.0 | 1278 | 0.3343 | 0.5714 | 0.3892 | 0.4631 | 0.9501 |
0.0413 | 7.0 | 1491 | 0.3522 | 0.5877 | 0.3818 | 0.4629 | 0.9500 |
0.0182 | 8.0 | 1704 | 0.3844 | 0.6120 | 0.3698 | 0.4610 | 0.9499 |
0.0182 | 9.0 | 1917 | 0.3847 | 0.5986 | 0.3828 | 0.4669 | 0.9504 |
0.008 | 10.0 | 2130 | 0.3832 | 0.5914 | 0.3809 | 0.4634 | 0.9501 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.13.1+cu117
- Datasets 2.9.0
- Tokenizers 0.13.2