Open nomorecoke opened 4 years ago
context = tf.matmul(attn_weights, enc_outputs) dec_output = tf.squeeze(dec_output, 0) # [1, n_step] context = tf.squeeze(context, 1) # [1, n_hidden]
I think dec_output shape is [1,n_hidden]
I think dec_output shape is [1,n_hidden]