小编Ell*_*lla的帖子

我读了这个https://www.tensorflow.org/guide/keras/custom_callback，但我不知道如何获取所有其他参数。

这是我的代码

 (hits, ndcgs) = evaluate_model(model, testRatings, testNegatives, topK, evaluation_threads)
  hr, ndcg, loss = np.array(hits).mean(), np.array(ndcgs).mean(), hist.history['loss'][0]
  print('Iteration %d [%.1f s]: HR = %.4f, NDCG = %.4f, loss = %.4f [%.1f s]' 
                  % (epoch,  t2-t1, hr, ndcg, loss, time()-t2))
 if hr > best_hr:
     best_hr, best_ndcg, best_iter = hr, ndcg, epoch
 if args.out > 0:
     model.save(model_out_file, overwrite=True)

正如你所看到的model，我需要hist和model.save。有没有办法在自定义回调中使用这三个参数？这样我就可以将所有这些写入自定义回调中。

class CustomCallback(keras.callbacks.Callback):

   def on_epoch_end(self, logs=None):
       keys = list(logs.keys())
       print("Stop training; got log keys: {}".format(keys))

如何添加 Keras dropout 层？不幸的是，我不知道到底应该在哪里添加这一层。我看了2个链接：

• https://keras.io/api/layers/regularization_layers/dropout/
• https://machinelearningmastery.com/dropout-regularization-deep-learning-models-keras/

例如，我见过这个

model.add(Dense(60, input_dim=60, activation='relu', kernel_constraint=maxnorm(3)))
model.add(Dropout(0.2))
model.add(Dense(30, activation='relu', kernel_constraint=maxnorm(3)))
model.add(Dropout(0.2))
model.add(Dense(1, activation='sigmoid'))

据我了解，密集层是用循环创建的，所以我不知道如何添加它。

def get_Model(...):
   
    # build dense layer for model
    for i in range(1, len(dense_layers)):
       
        layer = Dense(dense_layers[i],
                      activity_regularizer=l2(reg_layers[i]),
                      activation='relu',
                      name='layer%d' % i)
        mlp_vector = layer(mlp_vector)

    predict_layer = Concatenate()([mf_cat_latent, mlp_vector])
    result = Dense(1, activation='sigmoid',
                   kernel_initializer='lecun_uniform', name='result')

    model = Model(inputs=[input_user, input_item], outputs=result(predict_layer))

    return model

如何在不更改顺序的情况下删除重复的二维列表？

list_j = [[100,2,3,3], [4,98,99,98], [5,99,98,4], [5,99,98,5], [100,99,98,100,100,6]]

list_pre = [list(set(i)) for i in list_j]
print(list_pre)

[[2, 3, 100], [98, 99, 4], [98, 99, 4, 5], [98, 99, 5], [98, 99, 100, 6]]

如您所见，它更改了顺序。我想要的是[[100,2,3,],...]

期望输出 [[100,2,3,], [4,98,99], [5,99,98,4], [5,99,98], [100,99,98,6]]