# Copyright 2022 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Text searcher task."""
import dataclasses
from typing import Optional
from tensorflow_lite_support.python.task.core import base_options as base_options_module
from tensorflow_lite_support.python.task.processor.proto import embedding_options_pb2
from tensorflow_lite_support.python.task.processor.proto import search_options_pb2
from tensorflow_lite_support.python.task.processor.proto import search_result_pb2
from tensorflow_lite_support.python.task.text.pybinds import _pywrap_text_searcher
_CppTextSearcher = _pywrap_text_searcher.TextSearcher
_BaseOptions = base_options_module.BaseOptions
_EmbeddingOptions = embedding_options_pb2.EmbeddingOptions
_SearchOptions = search_options_pb2.SearchOptions
@dataclasses.dataclass
class TextSearcherOptions:
"""Options for the text search task.
Attributes:
base_options: Base options for the text searcher task.
embedding_options: Embedding options for the text searcher task.
search_options: Search options for the text searcher task.
"""
base_options: _BaseOptions
embedding_options: _EmbeddingOptions = dataclasses.field(
default_factory=_EmbeddingOptions
)
search_options: _SearchOptions = dataclasses.field(
default_factory=_SearchOptions
)
class TextSearcher(object):
"""Class to performs text search.
It works by performing embedding extraction on text, followed by
nearest-neighbor search in an index of embeddings through ScaNN.
"""
def __init__(self, options: TextSearcherOptions,
cpp_searcher: _CppTextSearcher) -> None:
"""Initializes the `TextSearcher` object."""
# Creates the object of C++ TextSearcher class.
self._options = options
self._searcher = cpp_searcher
@classmethod
def create_from_file(cls,
model_file_path: str,
index_file_path: Optional[str] = None) -> "TextSearcher":
"""Creates the `TextSearcher` object from a TensorFlow Lite model.
Args:
model_file_path: Path to the model.
index_file_path: Path to the index. Only required if the index is not
attached to the output tensor metadata as an AssociatedFile with type
SCANN_INDEX_FILE.
Returns:
`TextSearcher` object that's created from `options`.
Raises:
ValueError: If failed to create `TextSearcher` object from the provided
file such as invalid file.
RuntimeError: If other types of error occurred.
"""
options = TextSearcherOptions(
base_options=_BaseOptions(file_name=model_file_path),
search_options=_SearchOptions(index_file_name=index_file_path))
return cls.create_from_options(options)
@classmethod
def create_from_options(cls, options: TextSearcherOptions) -> "TextSearcher":
"""Creates the `TextSearcher` object from text searcher options.
Args:
options: Options for the text searcher task.
Returns:
`TextSearcher` object that's created from `options`.
Raises:
ValueError: If failed to create `TextSearcher` object from
`TextSearcherOptions` such as missing the model.
RuntimeError: If other types of error occurred.
"""
searcher = _CppTextSearcher.create_from_options(
options.base_options.to_pb2(), options.embedding_options.to_pb2(),
options.search_options.to_pb2())
return cls(options, searcher)
def search(self, text: str) -> search_result_pb2.SearchResult:
"""Search for text with similar semantic meaning.
This method performs actual feature extraction on the provided text input,
followed by nearest-neighbor search in the index.
Args:
text: the input text, used to extract the feature vectors.
Returns:
search result.
Raises:
ValueError: If any of the input arguments is invalid.
RuntimeError: If failed to perform nearest-neighbor search.
"""
search_result = self._searcher.search(text)
return search_result_pb2.SearchResult.create_from_pb2(search_result)
def get_user_info(self) -> str:
"""Gets the user info stored in the index file.
Returns:
Opaque user info stored in the index file (if any), in raw binary form.
Returns an empty string if the index doesn't contain user info.
"""
return self._searcher.get_user_info()
@property
def options(self) -> TextSearcherOptions:
return self._options