nlp_primitives.tensorflow.UniversalSentenceEncoder

class nlp_primitives.tensorflow.UniversalSentenceEncoder

Transforms a sentence or short paragraph to a vector using [tfhub model](https://tfhub.dev/google/universal-sentence-encoder/2)

Parameters:

None –

Examples

>>> sentences = ["I like to eat pizza", "The roller coaster was built in 1885.", ""]
>>> # universal_sentence_encoder = UniversalSentenceEncoder()  # normal syntax
>>> output = universal_sentence_encoder(sentences)  # defined in test file
>>> len(output)
512
>>> len(output[0])
3
>>> values = output[:3, 0]
>>> [round(x, 4) for x in values]
[0.0178, 0.0616, -0.0089]

__init__()

Methods

__init__()
flatten_nested_input_types(input_types)	Flattens nested column schema inputs into a single list.
generate_name(base_feature_names)
generate_names(base_feature_names)
get_args_string()
get_arguments()
get_description(input_column_descriptions[, ...])
get_filepath(filename)
get_function()

Attributes

base_of
base_of_exclude
commutative
compatibility	Additional compatible libraries
default_value	Default value this feature returns if no data found.
description_template
input_types	woodwork.ColumnSchema types of inputs
max_stack_depth
name	Name of the primitive
number_output_features	Number of columns in feature matrix associated with this feature
return_type	ColumnSchema type of return
series_library
uses_calc_time
uses_full_dataframe