使用MLP对之前的房贷项目做预测，损失函数采用MSE。
运行结果：
Using device: cuda
Epoch 10/1000, Loss: 0.2133
Epoch 20/1000, Loss: 0.2080
Epoch 30/1000, Loss: 0.2049
Epoch 40/1000, Loss: 0.2032
Epoch 50/1000, Loss: 0.2022
Epoch 60/1000, Loss: 0.2015
Epoch 70/1000, Loss: 0.2011
Epoch 80/1000, Loss: 0.2008
Epoch 90/1000, Loss: 0.2006
Epoch 100/1000, Loss: 0.2004
Epoch 110/1000, Loss: 0.2002
Epoch 120/1000, Loss: 0.2001
Epoch 130/1000, Loss: 0.1999
Epoch 140/1000, Loss: 0.1998
Epoch 150/1000, Loss: 0.1996
Epoch 160/1000, Loss: 0.1995
Epoch 170/1000, Loss: 0.1993
Epoch 180/1000, Loss: 0.1992
Epoch 190/1000, Loss: 0.1991
Epoch 200/1000, Loss: 0.1989
Epoch 210/1000, Loss: 0.1988
Epoch 220/1000, Loss: 0.1987
Epoch 230/1000, Loss: 0.1985
Epoch 240/1000, Loss: 0.1984
Epoch 250/1000, Loss: 0.1982
Epoch 260/1000, Loss: 0.1981
Epoch 270/1000, Loss: 0.1980
Epoch 280/1000, Loss: 0.1978
Epoch 290/1000, Loss: 0.1977
Epoch 300/1000, Loss: 0.1976
Epoch 310/1000, Loss: 0.1974
Epoch 320/1000, Loss: 0.1973
Epoch 330/1000, Loss: 0.1972
Epoch 340/1000, Loss: 0.1970
Epoch 350/1000, Loss: 0.1969
Epoch 360/1000, Loss: 0.1968
Epoch 370/1000, Loss: 0.1966
Epoch 380/1000, Loss: 0.1965
Epoch 390/1000, Loss: 0.1964
Epoch 400/1000, Loss: 0.1962
Epoch 410/1000, Loss: 0.1961
Epoch 420/1000, Loss: 0.1960
Epoch 430/1000, Loss: 0.1958
Epoch 440/1000, Loss: 0.1957
Epoch 450/1000, Loss: 0.1956
Epoch 460/1000, Loss: 0.1954
Epoch 470/1000, Loss: 0.1953
Epoch 480/1000, Loss: 0.1952
Epoch 490/1000, Loss: 0.1950
Epoch 500/1000, Loss: 0.1949
Epoch 510/1000, Loss: 0.1947
Epoch 520/1000, Loss: 0.1946
Epoch 530/1000, Loss: 0.1945
Epoch 540/1000, Loss: 0.1943
Epoch 550/1000, Loss: 0.1942
Epoch 560/1000, Loss: 0.1940
Epoch 570/1000, Loss: 0.1939
Epoch 580/1000, Loss: 0.1938
Epoch 590/1000, Loss: 0.1936
Epoch 600/1000, Loss: 0.1935
Epoch 610/1000, Loss: 0.1933
Epoch 620/1000, Loss: 0.1932
Epoch 630/1000, Loss: 0.1930
Epoch 640/1000, Loss: 0.1929
Epoch 650/1000, Loss: 0.1928
Epoch 660/1000, Loss: 0.1926
Epoch 670/1000, Loss: 0.1925
Epoch 680/1000, Loss: 0.1923
Epoch 690/1000, Loss: 0.1922
Epoch 700/1000, Loss: 0.1920
Epoch 710/1000, Loss: 0.1919
Epoch 720/1000, Loss: 0.1917
Epoch 730/1000, Loss: 0.1916
Epoch 740/1000, Loss: 0.1914
Epoch 750/1000, Loss: 0.1913
Epoch 760/1000, Loss: 0.1911
Epoch 770/1000, Loss: 0.1910
Epoch 780/1000, Loss: 0.1908
Epoch 790/1000, Loss: 0.1907
Epoch 800/1000, Loss: 0.1905
Epoch 810/1000, Loss: 0.1903
Epoch 820/1000, Loss: 0.1902
Epoch 830/1000, Loss: 0.1900
Epoch 840/1000, Loss: 0.1899
Epoch 850/1000, Loss: 0.1897
Epoch 860/1000, Loss: 0.1896
Epoch 870/1000, Loss: 0.1894
Epoch 880/1000, Loss: 0.1892
Epoch 890/1000, Loss: 0.1891
Epoch 900/1000, Loss: 0.1889
Epoch 910/1000, Loss: 0.1887
Epoch 920/1000, Loss: 0.1886
Epoch 930/1000, Loss: 0.1884
Epoch 940/1000, Loss: 0.1882
Epoch 950/1000, Loss: 0.1881
Epoch 960/1000, Loss: 0.1879
Epoch 970/1000, Loss: 0.1877
Epoch 980/1000, Loss: 0.1875
Epoch 990/1000, Loss: 0.1874
Epoch 1000/1000, Loss: 0.1872
Training completed in 2.32 seconds
