Module: mediapipe_model_maker.text_classifier

View source on GitHub

MediaPipe Public Python API for Text Classifier.

Modules

model_with_tokenizer module: Text classifier export module library.

Classes

class AverageWordEmbeddingHParams: The hyperparameters for an AverageWordEmbeddingClassifier.

class AverageWordEmbeddingModelOptions: Configurable model options for an Average Word Embedding classifier.

class BertHParams: The hyperparameters for a Bert Classifier.

class BertModelOptions: Configurable model options for a BERT model.

class BertOptimizer: Supported Optimizers for Bert Text Classifier.

class CSVParams: Parameters used when reading a CSV file.

class Dataset: Dataset library for text classifier.

class SupportedBertTokenizers: Supported preprocessors.

class SupportedModels: Predefined text classifier model specs supported by Model Maker.

class TextClassifier: API for creating and training a text classification model.

class TextClassifierOptions: User-facing options for creating the text classifier.