tflite_support.metadata_writers.bert_nl_classifier.MetadataWriter

Writes metadata into the Bert NL classifier.

Inherits From: MetadataWriter

tflite_support.metadata_writers.bert_nl_classifier.MetadataWriter(
    model_buffer: bytearray,
    metadata_buffer: Optional[bytearray] = None,
    associated_files: Optional[List[str]] = None
)

model_buffer valid buffer of the model file.
metadata_buffer valid buffer of the metadata.
associated_files path to the associated files to be populated.

Methods

create_for_inference

View source

@classmethod
create_for_inference(
    model_buffer: bytearray,
    tokenizer_md: Union[tflite_support.metadata_writers.metadata_info.BertTokenizerMd, tflite_support.metadata_writers.metadata_info.SentencePieceTokenizerMd],
    label_file_paths: List[str],
    ids_name: str = _DEFAULT_ID_NAME,
    mask_name: str = _DEFAULT_MASK_NAME,
    segment_name: str = _DEFAULT_SEGMENT_ID_NAME
)

Creates mandatory metadata for TFLite Support inference.

The parameters required in this method are mandatory when using TFLite Support features, such as Task library and Codegen tool (Android Studio ML Binding). Other metadata fields will be set to default. If other fields need to be filled, use the method create_from_metadata_info to edit them.

ids_name, mask_name, and segment_name correspond to the Tensor.name in the TFLite schema, which help to determine the tensor order when populating metadata. The default values come from Model Maker.

Args
model_buffer valid buffer of the model file.
tokenizer_md information of the tokenizer used to process the input string, if any. Supported tokenziers are: BertTokenizer 1 and SentencePieceTokenizer 2. If the tokenizer is RegexTokenizer 3, refer to nl_classifier.MetadataWriter.
label_file_paths paths to the label files 4 in the classification tensor. Pass in an empty list if the model does not have any label file.
ids_name name of the ids tensor, which represents the tokenized ids of the input text.
mask_name name of the mask tensor, which represents the mask with 1 for real tokens and 0 for padding tokens.
segment_name name of the segment ids tensor, where 0 stands for the first sequence, and 1 stands for the second sequence if exists.

Returns
A MetadataWriter object.

create_from_metadata

View source

@classmethod
create_from_metadata(
    model_buffer: bytearray,
    model_metadata: Optional[tflite_support.metadata_schema_py_generated.ModelMetadataT] = None,
    input_metadata: Optional[List[_metadata_fb.TensorMetadataT]] = None,
    output_metadata: Optional[List[_metadata_fb.TensorMetadataT]] = None,
    associated_files: Optional[List[str]] = None,
    input_process_units: Optional[List[_metadata_fb.ProcessUnitT]] = None,
    output_process_units: Optional[List[_metadata_fb.ProcessUnitT]] = None
)

Creates MetadataWriter based on the metadata Flatbuffers Python Objects.

Args
model_buffer valid buffer of the model file.
model_metadata general model metadata 1. The subgraph_metadata will be refreshed with input_metadata and output_metadata.
input_metadata a list of metadata of the input tensors 2.
output_metadata a list of metadata of the output tensors 3.
associated_files path to the associated files to be populated.
input_process_units a lits of metadata of the input process units 4.
output_process_units a lits of metadata of the output process units 5.

Returns
A MetadataWriter Object.

create_from_metadata_info

View source

@classmethod
create_from_metadata_info(
    model_buffer: bytearray,
    general_md: Optional[tflite_support.metadata_writers.metadata_info.GeneralMd] = None,
    input_md: Optional[tflite_support.metadata_writers.metadata_info.BertInputTensorsMd] = None,
    output_md: Optional[tflite_support.metadata_writers.metadata_info.ClassificationTensorMd] = None
)

Creates MetadataWriter based on general/input/output information.

Args
model_buffer valid buffer of the model file.
general_md general information about the model. If not specified, default general metadata will be generated.
input_md input tensor information. If not specified, default input metadata will be generated.
output_md output classification tensor informaton. If not specified, default output metadata will be generated.

Returns
A MetadataWriter object.

get_metadata_json

View source

get_metadata_json() -> str

Gets the generated JSON metadata string before populated into model.

This method returns the metadata buffer before populated into the model. More fields could be filled by MetadataPopulator, such as min_parser_version. Use get_populated_metadata_json() if you want to get the final metadata string.

Returns
The generated JSON metadata string before populated into model.

get_populated_metadata_json

View source

get_populated_metadata_json() -> str

Gets the generated JSON metadata string after populated into model.

More fields could be filled by MetadataPopulator, such as min_parser_version. Use get_metadata_json() if you want to get the original metadata string.

Returns
The generated JSON metadata string after populated into model.

populate

View source

populate() -> bytearray

Populates the metadata and label file to the model file.

Returns
A new model buffer with the metadata and associated files.