我的问题出现在如下代码段:
feed_dict = {model.user_input: user_input[i], model.item_input_pos: item_input_pos[i], model.item_input_neg: item_input_neg[i]} if output_adv: loss, output_pos, output_neg = sess.run([model.loss_adv, model.output_adv, model.output_neg_adv], feed_dict) else: loss, output_pos, output_neg = sess.run([model.loss, model.output, model.output_neg], feed_dict)
核心问题就是两者shape无法匹配,通过修改代码加入reshape变为如下即可:
feed_dict = {model.user_input: user_input[i].reshape([1,-1]), model.item_input_pos: item_input_pos[i].reshape([1,-1]), model.item_input_neg: item_input_neg[i].reshape([1,-1])} if output_adv: loss, output_pos, output_neg = sess.run([model.loss_adv, model.output_adv, model.output_neg_adv], feed_dict) else: loss, output_pos, output_neg = sess.run([model.loss, model.output, model.output_neg], feed_dict)