Attention: This MediaPipe Solutions Preview is an early release. Learn more.

mediapipe_model_maker.text_classifier.SupportedModels

Stay organized with collections Save and categorize content based on your preferences.

Predefined text classifier model specs supported by Model Maker.

AVERAGE_WORD_EMBEDDING_CLASSIFIER <SupportedModels.AVERAGE_WORD_EMBEDDING_CLASSIFIER: functools.partial(<class 'mediapipe_model_maker.python.text.text_classifier.model_spec.AverageWordEmbeddingClassifierSpec'>)>
MOBILEBERT_CLASSIFIER <SupportedModels.MOBILEBERT_CLASSIFIER: functools.partial(<class 'mediapipe_model_maker.python.text.core.bert_model_spec.BertModelSpec'>, hparams=BaseHParams(learning_rate=3e-05, batch_size=48, epochs=3, steps_per_epoch=None, shuffle=False, export_dir='/tmpfs/tmp/tmpkuq2qujc', distribution_strategy='off', num_gpus=-1, tpu=''), name='MobileBert', uri='https://tfhub.dev/tensorflow/mobilebert_en_uncased_L-24_H-128_B-512_A-4_F-4_OPT/1', tflite_input_name={'ids': 'serving_default_input_1:0', 'mask': 'serving_default_input_3:0', 'segment_ids': 'serving_default_input_2:0'})>