nerblackbox 0.0.14

fine-tune transformer-based language models for named entity recognition

A python package to fine-tune transformer-based language models for named entity recognition (NER).

Resources

• Source Code: https://github.com/flxst/nerblackbox

• Documentation: https://flxst.github.io/nerblackbox

• PyPI: https://pypi.org/project/nerblackbox

Installation

pip install nerblackbox

About

Fine-tune a language model for named entity recognition in a few simple steps:

1. Define a fine-tuning experiment by choosing a pretrained model and a dataset

experiment = Experiment("my_experiment", model="bert-base-cased", dataset="conll2003")

2. Run the experiment and get the performance of the fine-tuned model

experiment.run()
experiment.get_result(metric="f1", level="entity", phase="test")
# 0.9045

3. Use the fine-tuned model for inference

model = Model.from_experiment("my_experiment")
model.predict("The United Nations has never recognised Jakarta's move.")
# [[
#  {'char_start': '4', 'char_end': '18', 'token': 'United Nations', 'tag': 'ORG'},
#  {'char_start': '40', 'char_end': '47', 'token': 'Jakarta', 'tag': 'LOC'}
# ]]

There is much more to it than that! See the documentation to get started.

Features

Data

• Support for Different Data Formats

• Support for Different Annotation Schemes

• Integration of HuggingFace Datasets

• Text Encoding

Training

• Adaptive Fine-tuning

• Hyperparameter Search

• Multiple Runs with Different Random Seeds

• Detailed Analysis of Training Results

Evaluation

• Evaluation of a Model on a Dataset

Inference

• Versatile Model Inference

Other

• Compatibility with HuggingFace

• GPU Support

• Language Agnosticism

See the documentation for more details.

Citation

@misc{nerblackbox,
  author = {Stollenwerk, Felix},
  title  = {nerblackbox: a python package to fine-tune transformer-based language models for named entity recognition},
  year   = {2021},
  url    = {https://github.com/flxst/nerblackbox},
}