def gen_coaction(ad, his_items, dim, mode="can", mask=None):
    idx = 0
    weight_orders = []
    bias_orders = []
    for i in range(orders):
        weight, bias = [], []
        for w, b in zip(weight_emb_w, weight_emb_b):
            weight.append(tf.reshape(ad[:, idx:idx+w[0]*w[1]], [-1, w[0], w[1]]))
            idx += w[0] * w[1]
            if b == 0:
                bias.append(None)
            else:
                bias.append(tf.reshape(ad[:, idx:idx+b], [-1, 1, b]))
                idx += b
        weight_orders.append(weight)
        bias_orders.append(bias)

if mode == "can":
        out_seq = []
        hh = []
        for i in range(orders):
            hh.append(his_items**(i+1))
        #hh = [sum(hh)]
        for i, h in enumerate(hh):
            if order_indep:
                weight, bias = weight_orders[i], bias_orders[i]
            else:
                weight, bias = weight_orders[0], bias_orders[0]
            for j, (w, b) in enumerate(zip(weight, bias)):
                h  = tf.matmul(h, w)
                if b is not None:
                    h = h + b
                if j != len(weight)-1:
                    h = tf.nn.tanh(h)
                out_seq.append(h)
        out_seq = tf.concat(out_seq, 2)