3

当我使用 spearmint 优化 Keras 模型的超参数时，它第一次运行良好。但是从第二份工作开始，它总是会引发以下错误。

<type 'exceptions.TypeError'>, TypeError('An update must have the same type as the original shared variable (shared_var=<TensorType(float32, matrix)>, shared_var.type=TensorType(float32, matrix), update_val=Elemwise{add,no_inplace}.0, update_val.type=TensorType(float64, matrix)).', 'If the difference is related to the broadcast pattern, you can call the tensor.unbroadcast(var, axis_to_unbroadcast[, ...]) function to remove broadcastable dimensions.'), <traceback object at 0x18a5c5710>)

我正在使用以下代码加载预先创建的训练数据和测试数据的 numpy 数组。以下参数由优化 python 脚本传递。但是，如果在没有留兰香的情况下运行，这组参数可以正常工作。 

def load_train_data(arg_type, params=None):

    X_train1 = pickle.load(open(arg_type+"_train1","rb"))

    X_train2 = pickle.load(open(arg_type+"_train2","rb"))

    Y_train = pickle.load(open(arg_type+"_train_labels","rb"))



    model=combined_model(X_train1,X_train2,Y_train,params)



    X_test1 = pickle.load(open(arg_type+"_test1","rb"))

    X_test2 = pickle.load(open(arg_type+"_test2","rb"))

    Y_test = pickle.load(open(arg_type+"_test_labels","rb"))



    loss = model.evaluate({'input1': X_test1,'input2': X_test2,'output':Y_test},batch_size=450)

    return loss
4

1 回答 1

0

我用 spearmint 设置的变量，必须使用 float(),int() 显式转换为基本的 python 数据类型。这有助于解决这个问题。

于 2016-04-14T05:13:47.343 回答