AdaptNLP Embeddings Module

class EmbeddingResult[source]

EmbeddingResult(sentences:List[Sentence]) :: SentenceResult

A result class designed for Embedding models

Parameters:

  • sentences : typing.List[flair.data.Sentence]

    A list of Flair `Sentence`s

EmbeddingResult.sentence_embeddings[source]

All embeddings in sentences (if available)

EmbeddingResult.token_embeddings[source]

All embeddings from the individual tokens in sentence with original order in shape (n, embed_dim)

EmbeddingResult.to_dict[source]

EmbeddingResult.to_dict(detail_level:DetailLevel='low')

Returns self as a dictionary

Parameters:

  • detail_level : <class 'fastcore.basics.DetailLevel'>, optional

    A level of detail to return

class EasyWordEmbeddings[source]

EasyWordEmbeddings()

Word embeddings from the latest language models

Usage:

>>> embeddings = adaptnlp.EasyWordEmbeddings()
>>> embeddings.embed_text("text you want embeddings for", model_name_or_path="bert-base-cased")

EasyWordEmbeddings.embed_text[source]

EasyWordEmbeddings.embed_text(text:Union[List[Sentence], Sentence, List[str], str], model_name_or_path:Union[str, HFModelResult, FlairModelResult]='bert-base-cased', detail_level:DetailLevel='low', raw:bool=False)

Produces embeddings for text

Parameters:

  • text : typing.Union[typing.List[flair.data.Sentence], flair.data.Sentence, typing.List[str], str]

    Text input, it can be a string or any of Flair's `Sentence` input formats

  • model_name_or_path : typing.Union[str, adaptnlp.model_hub.HFModelResult, adaptnlp.model_hub.FlairModelResult], optional

    The hosted model name key, model path, or an instance of either `HFModelResult` or `FlairModelResult`

  • detail_level : <class 'fastcore.basics.DetailLevel'>, optional

    A level of detail to return. By default is None, which returns a EmbeddingResult, otherwise will return a dict

  • raw : <class 'bool'>, optional

    Whether to return the raw outputs

Returns:

  • typing.List[adaptnlp.inference.embeddings.EmbeddingResult]

    A list of either `EmbeddingResult`s or dictionaries with information

EasyWordEmbeddings.embed_all[source]

EasyWordEmbeddings.embed_all(text:Union[List[Sentence], Sentence, List[str], str], model_names_or_paths:List[str]=[], detail_level:DetailLevel='low')

Embeds text with all embedding models loaded

Parameters:

  • text : typing.Union[typing.List[flair.data.Sentence], flair.data.Sentence, typing.List[str], str]

    Text input, it can be a string or any of Flair's `Sentence` input formats

  • model_names_or_paths : typing.List[str], optional

    A list of model names

  • detail_level : <class 'fastcore.basics.DetailLevel'>, optional

    A level of detail to return. By default is None, which returns a EmbeddingResult, otherwise will return a dict

Returns:

  • typing.List[adaptnlp.inference.embeddings.EmbeddingResult]

    A list of either `EmbeddingResult`s or dictionaries with information

class EasyStackedEmbeddings[source]

EasyStackedEmbeddings(*embeddings:str)

Word Embeddings that have been concatenated and 'stacked' as specified by Flair

Parameters:

  • embeddings : <class 'str'>

EasyStackedEmbeddings.embed_text[source]

EasyStackedEmbeddings.embed_text(text:Union[List[Sentence], Sentence, List[str], str], detail_level:DetailLevel='low')

Stacked embeddings

Parameters:

  • text : typing.Union[typing.List[flair.data.Sentence], flair.data.Sentence, typing.List[str], str]

    Text input, it can be a string or any of Flair's `Sentence` input formats

  • detail_level : <class 'fastcore.basics.DetailLevel'>, optional

    A level of detail to return. By default is None, which returns a EmbeddingResult, otherwise will return a dict

Returns:

  • typing.List[adaptnlp.inference.embeddings.EmbeddingResult]

    A list of either EmbeddingResult's or dictionaries with information

class EasyDocumentEmbeddings[source]

EasyDocumentEmbeddings(*embeddings:str, methods:List[str]=['rnn', 'pool'], configs:Dict[KT, VT]={'pool_configs': {'fine_tune_mode': 'linear', 'pooling': 'mean'}, 'rnn_configs': {'hidden_size': 512, 'rnn_layers': 1, 'reproject_words': True, 'reproject_words_dimension': 256, 'bidirectional': False, 'dropout': 0.5, 'word_dropout': 0.0, 'locked_dropout': 0.0, 'rnn_type': 'GRU', 'fine_tune': True}})

Document Embeddings generated by pool and rnn methods applied to the word embeddings of text

Parameters:

  • embeddings : <class 'str'>

  • methods : typing.List[str], optional

  • configs : typing.Dict, optional

EasyDocumentEmbeddings.embed_pool[source]

EasyDocumentEmbeddings.embed_pool(text:Union[List[Sentence], Sentence, List[str], str], detail_level:DetailLevel='low')

Generate stacked embeddings with DocumentPoolEmbeddings

Parameters:

  • text : typing.Union[typing.List[flair.data.Sentence], flair.data.Sentence, typing.List[str], str]

    Text input, it can be a string or any of Flair's `Sentence` input formats

  • detail_level : <class 'fastcore.basics.DetailLevel'>, optional

    A level of detail to return. By default is None, which returns a EmbeddingResult, otherwise will return a dict

Returns:

  • typing.List[adaptnlp.inference.embeddings.EmbeddingResult]

    A list of either EmbeddingResult's or dictionaries with information

EasyDocumentEmbeddings.embed_rnn[source]

EasyDocumentEmbeddings.embed_rnn(text:Union[List[Sentence], Sentence, List[str], str], detail_level:DetailLevel='low')

Generate stacked embeddings with DocumentRNNEmbeddings

Parameters:

  • text : typing.Union[typing.List[flair.data.Sentence], flair.data.Sentence, typing.List[str], str]

    Text input, it can be a string or any of Flair's `Sentence` input formats

  • detail_level : <class 'fastcore.basics.DetailLevel'>, optional

    A level of detail to return. By default is None, which returns a EmbeddingResult, otherwise will return a dict

Returns:

  • typing.List[flair.data.Sentence]

    A list of either EmbeddingResult's or dictionaries with information