langtest.modelhandler.jsl_modelhandler.PretrainedJSLModel#

class PretrainedJSLModel(model: NLUPipeline | PretrainedPipeline | LightPipeline | PipelineModel)#

Bases: ABC

PretrainedJSLModel is an abstract class for handling SparkNLP models.

model#

Loaded SparkNLP LightPipeline for inference.

Type:

Union[“NLUPipeline”, “PretrainedPipeline”, “LightPipeline”, “PipelineModel”]

abstract __init__(model: NLUPipeline | PretrainedPipeline | LightPipeline | PipelineModel)#

Constructor method

Parameters:

model (Union["NLUPipeline", "PretrainedPipeline", "LightPipeline", "PipelineModel"]) – Loaded SparkNLP LightPipeline for inference.

Methods

__init__(model)

Constructor method

load_model(path)

Load the NER model into the model attribute.

predict(text, *args, **kwargs)

Perform predictions with SparkNLP LightPipeline on the input text.

Attributes

hub

classmethod load_model(path) NLUPipeline#

Load the NER model into the model attribute.

Parameters:

path (str) – Path to pretrained local or NLP Models Hub SparkNLP model

abstract predict(text: str, *args, **kwargs) Any#

Perform predictions with SparkNLP LightPipeline on the input text.

Parameters:

text (str) – Input text to perform translation on.