将张量列表转换为numpy然后将数字求和

Question

我想将张量列表转换为numpy数组/列表，然后将其求和并乘以历元数。我尝试使用MSE_for_all_epochs = sum（np.MSE_for_all_epochs），但收到以下错误消息：Traceback（最近一次呼叫最近）： 文件“ train3_Eval.py”，行177 MSE_for_all_epochs = sum（np.MSE_for_all_epochs） AttributeError：模块'numpy'没有属性'MSE_for_all_epochs'

        mean_test_error = mean_test_error.detach() / len(loader_test)
        #store_MAE += mean_test_error
        MAE_for_all_epochs.append(mean_test_error)

        mse = math.sqrt(loss / len(loader_test))
        #store_MSE +=mse
        MSE_for_all_epochs.append(mse)

        print("Test count error: %f" % mean_test_error)
        print("MSE: %f" % mse)

MSE_for_all_epochs = sum(np.MSE_for_all_epochs)
MAE_for_all_epochs = sum(np.MAE_for_all_epochs)
print(MAE_for_all_epochs)
print(MSE_for_all_epochs)

model_mae= MSE_for_all_epochs / epoch
model_mse= MAE_for_all_epochs / epoch
print("Model MAE: %f" % model_mae)
print("Model MSE: %f" % model_mse)

模型输出如下：

tensor(1445.2251)
1602.6452224985646
Model MAE: 534.215074
Model MSE: 481.741699