distilroberta-base-ner-wikiann Open-Source Named Entity Recognition Model

Distilroberta Base Ner Wikiann

Developed by philschmid

A named entity recognition model fine-tuned on the wikiann dataset based on the DistilRoBERTa-base model, used for identifying named entities in text.

Sequence Labeling

Transformers

Open Source License:Apache-2.0 #Named Entity Recognition #Multilingual Support #High-precision NER

Downloads 16

Release Time : 3/2/2022

Model Overview

This model is a lightweight version based on DistilRoBERTa-base, fine-tuned on the wikiann dataset, specifically designed for named entity recognition tasks.

Model Features

Efficient and Lightweight

Lightweight architecture based on DistilRoBERTa, reducing computational resource requirements while maintaining performance.

High-precision Recognition

Achieves an F1 score of 83.78% on the wikiann test set, demonstrating excellent performance.

Multilingual Support

Trained on the multilingual wikiann dataset, supporting named entity recognition in multiple languages.

Model Capabilities

Named Entity Recognition

Text Token Classification

Multilingual Text Processing

Use Cases

Information Extraction

News Article Entity Recognition

Extract entity information such as person names, locations, and organization names from news articles.

Accuracy approximately 92%

Social Media Text Analysis

Analyze entity information in social media texts for user profiling.

Knowledge Graph Construction

Knowledge Graph Entity Extraction

Extract entities from unstructured text for knowledge graph construction.

🚀 distilroberta-base-ner-wikiann

This model is a fine - tuned version of distilroberta-base on the wikiann dataset. It is designed for token - classification tasks, offering high performance in named entity recognition.

🚀 Quick Start

💻 Usage Examples

Basic 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)

📚 Documentation

Model Information

Property	Details
Model Type	Fine - tuned version of distilroberta - base for token - classification
Training Data	wikiann dataset
Tags	token - classification
Metrics	Precision, Recall, F1, Accuracy

Model Results

The model distilroberta-base-ner-wikiann has the following performance results:

Evaluation Set:
- Loss: 0.3156
- Precision: 0.8332
- Recall: 0.8424
- F1: 0.8378
- Accuracy: 0.9193
Test Set:
- Loss: 0.3023
- Precision: 0.8301
- Recall: 0.8452
- F1: 0.8376
- Accuracy: 0.92

Model Index

Name: distilroberta - base - ner - wikiann
Results:
- Task: Token Classification
- Dataset: wikiann
- Metrics:
  - Precision: 0.8331921416757433
  - Recall: 0.84243586083126
  - F1: 0.8377885044416501
  - Accuracy: 0.91930707459758
- Task (en test split): Token Classification
- Dataset (en test split): wikiann (config: en, split: test)
- Metrics (en test split):
  - Accuracy: 0.9200373733433721
  - Precision: 0.9258482820953792
  - Recall: 0.9347545055892119
  - F1: 0.9302800779500893
  - Loss: 0.3007512390613556

🔧 Technical Details

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

Framework versions

Transformers 4.6.1
Pytorch 1.8.1+cu101
Datasets 1.6.2
Tokenizers 0.10.2

📄 License

This model is licensed under the Apache 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご