i was using bert model ‘A KerasTensor is symbolic: it’s a placeholder for a shape an a dtype
ValueError: Exception encountered when calling layer ‘keras_layer_4’ (type KerasLayer).
Range of parameters of BERT models
What will be the range of parameters, or how can we can calculate the range (like max and min values) that were saved for inference after we train Bert models or any Transformers?
ValueError: Exception encountered when calling layer ‘preprocessing’ (type KerasLayer)
tfhub_preprocess = ‘https://tfhub.dev/tensorflow/bert_en_uncased_preprocess/3’ tfhub_encoder = ‘https://tfhub.dev/tensorflow/small_bert/bert_en_uncased_L-2_H-128_A-2/2′ def build_smallBERT_CNN_classifier_model(): text_input = tf.keras.layers.Input(shape=(), dtype=tf.string, name=’text’) preprocessing = hub.KerasLayer(tfhub_preprocess, trainable=True, name=’preprocessing’) encoder_inputs = preprocessing(text_input) encoder = hub.KerasLayer(tfhub_encoder, trainable=True, name=’BERT_encoder’) outputs = encoder(encoder_inputs) net = sequence_output = outputs[“sequence_output”] net = tf.keras.layers.Dense(64, activation=”relu”)(net) net = tf.keras.layers.Dropout(0.1)(net) net = tf.keras.layers.Dense(num_classes, activation=”softmax”, name=’classifier’)(net) return tf.keras.Model(text_input, net) intent_classifier_model = build_smallBERT_CNN_classifier_model() while running the […]