Skip to content

Latest commit

 

History

History
27 lines (19 loc) · 764 Bytes

README.md

File metadata and controls

27 lines (19 loc) · 764 Bytes

CRFpy

CRFpy implements maximum likelihood estimation and max margin learning for conditional random field using the Scikit-learn interface. The core of the code is the CRF class which implements the learning algorithms. The user is expected to implement inference in the desired model. An example of implementing inference for a specific model is in linear_chain_crf.py.

Requirements

CRFpy requires a number of Python modules. These modules are listed in requirements.txt and can be installed using pip by running,

pip install -r requirements.txt

Setup for linear_chain_crf

Inference for the linear chain crf is implemented in cython and needs to be compiled. To compile the code, run

python setup.py build_ext --inplace