distilroberta-base-ner-conll2003
This model is a fine-tuned version of distilroberta-base on the conll2003 dataset.
eval F1-Score: 95,29 (CoNLL-03)
test F1-Score: 90,74 (CoNLL-03)
eval F1-Score: 95,29 (CoNLL++ / CoNLL-03 corrected)
test F1-Score: 92,23 (CoNLL++ / CoNLL-03 corrected)
Model Usage
from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline
tokenizer = AutoTokenizer.from_pretrained("philschmid/distilroberta-base-ner-conll2003")
model = AutoModelForTokenClassification.from_pretrained("philschmid/distilroberta-base-ner-conll2003")
nlp = pipeline("ner", model=model, tokenizer=tokenizer, grouped_entities=True)
example = "My name is Philipp and live in Germany"
nlp(example)
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 4.9902376275441704e-05
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6.0
- mixed_precision_training: Native AMP
Training results
CoNNL2003
It achieves the following results on the evaluation set:
- Loss: 0.0583
- Precision: 0.9493
- Recall: 0.9566
- F1: 0.9529
- Accuracy: 0.9883
It achieves the following results on the test set:
- Loss: 0.2025
- Precision: 0.8999
- Recall: 0.915
- F1: 0.9074
- Accuracy: 0.9741
CoNNL++ / CoNLL2003 corrected
It achieves the following results on the evaluation set:
- Loss: 0.0567
- Precision: 0.9493
- Recall: 0.9566
- F1: 0.9529
- Accuracy: 0.9883
It achieves the following results on the test set:
- Loss: 0.1359
- Precision: 0.92
- Recall: 0.9245
- F1: 0.9223
- Accuracy: 0.9785
Framework versions
- Transformers 4.6.1
- Pytorch 1.8.1+cu101
- Datasets 1.6.2
- Tokenizers 0.10.2
- Downloads last month
- 3
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Dataset used to train DidulaThavisha/distilroberta
Evaluation results
- Precision on conll2003self-reported0.949
- Recall on conll2003self-reported0.957
- F1 on conll2003self-reported0.953
- Accuracy on conll2003self-reported0.988
- Accuracy on conll2003validation set self-reported0.988
- Precision on conll2003validation set self-reported0.991
- Recall on conll2003validation set self-reported0.992
- F1 on conll2003validation set self-reported0.991
- loss on conll2003validation set self-reported0.056