SklearnProjector#

class SklearnProjector(config)#

Bases: Projector

The generic scikit-learn Projector step projects from one vector space to another. For more info see Decomposition.

Note - So far only svd is supported -> TruncatedSVD

Input - all input fields need to be of type list [ float or int ] or numpy.ndarray

Output - all output fields are filled with data of type numpy.ndarray with shape (‘n_components’,)

Parameters

  • type (str) – sklearn

  • model_type (str, 'svd') – Type of scikit-learn projection

  • model_kwargs (dict, {}) – Keyword arguments for the scikit-learn model

  • n_components (int) – Number of vector components after projection

  • normalize_output (bool, True) – Whether or not to normalize the output

Example

{
    "step": "projector",
    "type": "sklearn",
    "model_type": "svd",
    "n_components": 100,
    "input_field": "embedded_text",
    "output_field": "embedded_text"
}

Methods Summary

load()

Load a step

process_batch(batch)

Process a batch of documents.

save()

Save a step

train(docs)

Train on a step of a set of documents

Methods Documentation

load()#

Load a step

process_batch(batch)#

Process a batch of documents. If not defined will default to using self.process_doc for each document in the batch.

Parameters

batch (list(Document)) – List of documents

Returns

List of processed documents

Return type

list(Document)

save()#

Save a step

train(docs)#

Train on a step of a set of documents

Parameters

docs (generator(Document)) – Generator of documents

Returns

Generator of processed documents

Return type

generator(Document)