philschmid / distilroberta-base-ner-wikiann

distilroberta-base-ner-wikiann

This model is a fine-tuned version of distilroberta-base on the wikiann dataset.

eval F1-Score: 83,78 test F1-Score: 83,76

Model Usage

from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline

tokenizer = AutoTokenizer.from_pretrained("philschmid/distilroberta-base-ner-wikiann")
model = AutoModelForTokenClassification.from_pretrained("philschmid/distilroberta-base-ner-wikiann")

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: 4.9086903597787154e-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: 5.0
• mixed_precision_training: Native AMP

Training results

It achieves the following results on the evaluation set:

• Loss: 0.3156
• Precision: 0.8332
• Recall: 0.8424
• F1: 0.8378
• Accuracy: 0.9193

It achieves the following results on the test set:

• Loss: 0.3023
• Precision: 0.8301
• Recall: 0.8452
• F1: 0.8376
• Accuracy: 0.92

Framework versions

• Transformers 4.6.1
• Pytorch 1.8.1+cu101
• Datasets 1.6.2
• Tokenizers 0.10.2