We use the embedding function, , for learning the embeddings of the support set. We use bidirectional LSTM as our embedding function, .

We can define our embedding function, , as follows:

def g(X):
    
    #forward cell
    forward_cell = rnn.BasicLSTMCell(32)

    #backward cell
    backward_cell = rnn.BasicLSTMCell(32)
    
    #bidirectional LSTM
    outputs, state_forward, state_backward = rnn.static_bidirectional_rnn(forward_cell, backward_cell, X, dtype=tf.float32)

    return tf.add(tf.stack(X), tf.stack(outputs))