ratsgo / cam2 Goto Github PK

class CustomLayer (tf.keras.layers.Layer):
    def __init__(self , units , **kwargs):
        super (CustomLayer , self).__init__ (**kwargs)
        self.units = units
        self.W1 = tf.keras.layers.Dense (units)
        self.W2 = tf.keras.layers.Dense (units)
        self.V = tf.keras.layers.Dense (1)

    def call(self , features , hidden):
            hidden_with_time_axis = tf.expand_dims (hidden , 1)

        score = tf.nn.tanh (self.W1 (features) + self.W2 (hidden_with_time_axis))

        attention_weights = tf.nn.softmax (self.V (score) , axis=1)

        context_vector = attention_weights * features
        context_vector = tf.reduce_sum (context_vector , axis=1) ##attention adjusted ouput state

        return context_vector , attention_weights

    def get_config(self):
        config = {"units": self.units}
        base_config = super (CustomLayer , self).get_config ()

        return dict(list(base_config.items())+ list(config.items()))




def build_att_lstm(MAX_LEN, vocab_size, EMBED_SIZE, embed_matrix, LSTM_Unit, dropouts_input, dropouts):
    sequence_input = Input(shape=(MAX_LEN,), dtype="int32")
    embedded_sequences = Embedding(vocab_size +1, EMBED_SIZE, weights=[embed_matrix], trainable= False)(sequence_input)
    embedded_sequences_do = Dropout (dropouts_input) (embedded_sequences)
    (lstm, forward_h, forward_c) =LSTM(LSTM_Unit, return_sequences=True, return_state=True, name="lstm_1", dropout=dropouts)(embedded_sequences_do)

    context_vector, attention_weights = CustomLayer(10, name= 'CustomLayer')(lstm, forward_h)

    output = Dense(6, activation="softmax")(context_vector)

    model = tf.keras.Model(inputs=sequence_input, outputs=output)

    #config = model.get_config ()

    METRICS = [
        tf.keras.metrics.BinaryAccuracy (name='accuracy') ,
        tf.keras.metrics.Precision (name='precision') ,
        tf.keras.metrics.Recall (name='recall') ,
        tf.keras.metrics.AUC (name='auc') ,
    ]

    model.compile (loss='categorical_crossentropy' , optimizer='adam' , metrics=METRICS)

    return model