2范数归一化错误记录

最新推荐文章于 2023-04-23 22:31:24 发布

SKY_yiyi_9

最新推荐文章于 2023-04-23 22:31:24 发布

阅读量263

点赞数

分类专栏： python与深度学习、神经网络

本文链接：https://blog.csdn.net/SKY_yiyi_9/article/details/108673779

版权

python与深度学习、神经网络专栏收录该内容

24 篇文章 0 订阅

订阅专栏

#2范数归一化
def l2_norm(input, dim=0):#dim=0行，dim=1列
    norm = torch.norm(input, 2, dim, True)#torch.norm(input, p, dim, out=None,keepdim=False) p (float)：范数计算中的幂指数值 True:保持维度
    output = torch.div(input, norm)
    return output

看齐的对象：

tensor([[0.4980, 0.4824, 0.4902,  ..., 0.5176, 0.5059, 0.5020],
        [0.4745, 0.4941, 0.5059,  ..., 0.5216, 0.5216, 0.5176],
        [0.5373, 0.5059, 0.5255,  ..., 0.5373, 0.5137, 0.5451],
        ...,
        [0.5373, 0.5059, 0.4745,  ..., 0.4314, 0.4157, 0.4314],
        [0.4118, 0.4235, 0.4431,  ..., 0.3569, 0.3412, 0.3294],
        [0.3725, 0.3647, 0.3804,  ..., 0.4000, 0.3725, 0.3804]])

batch_y,failed
before

tensor([[-12.4335,   5.4173,  -1.6632,  ...,  -1.2565,  -7.1197, -11.5201],
        [-12.9857,   5.7168,  -1.7476,  ...,  -1.0469,  -7.6428, -12.2431],
        [-13.0031,   5.5625,  -2.0829,  ...,  -1.1548,  -7.3446, -11.7510],
        ...,
        [-13.4243,   4.2034,  -2.0059,  ...,  -1.8164,  -6.5609, -14.5081],
        [-11.6454,   3.7870,  -1.3004,  ...,  -0.6608,  -7.6010, -11.0177],
        [-10.2765,   3.7000,  -2.0770,  ...,   0.1464,  -6.1413,  -9.3848]])

after

tensor([[-0.0289,  0.0293, -0.0224,  ..., -0.0214, -0.0293, -0.0288],
        [-0.0301,  0.0309, -0.0236,  ..., -0.0178, -0.0314, -0.0306],
        [-0.0302,  0.0301, -0.0281,  ..., -0.0196, -0.0302, -0.0294],
        ...,
        [-0.0312,  0.0227, -0.0271,  ..., -0.0309, -0.0270, -0.0363],
        [-0.0270,  0.0205, -0.0176,  ..., -0.0112, -0.0312, -0.0276],
        [-0.0239,  0.0200, -0.0280,  ...,  0.0025, -0.0252, -0.0235]])

[00/06] Run time for data\Set6\Boats.png is , PSNR is , SSIM is, 1.8950, 5.44, 0.0126,runtime is 0.0040
[01/06] Run time for data\Set6\fingerprint.png is , PSNR is , SSIM is, 1.7406, 4.68, 0.0030,runtime is 0.0035
[02/06] Run time for data\Set6\flinstones_DL.png is , PSNR is , SSIM is, 1.9958, 4.42, 0.0145,runtime is 0.0040
[03/06] Run time for data\Set6\house(2).png is , PSNR is , SSIM is, 0.4211, 5.01, 0.0234,runtime is 0.0010
[04/06] Run time for data\Set6\lenna(2).png is , PSNR is , SSIM is, 1.6724, 6.21, 0.0216,runtime is 0.0040
[05/06] Run time for data\Set6\peppers2562-3.png is , PSNR is , SSIM is, 0.3775, 5.72, 0.0234,runtime is 0.0010


CS ratio is 10, Avg PSNR/SSIM for Set6 is 5.25/0.0164, Avg runtime is 1.3504, Epoch number of model is 300

batch_x0,failed
before

tensor([[0.4881, 0.4879, 0.4881,  ..., 0.5091, 0.5073, 0.5038],
        [0.4959, 0.5054, 0.5127,  ..., 0.5448, 0.5428, 0.5423],
        [0.5261, 0.5209, 0.5226,  ..., 0.5163, 0.5115, 0.5121],
        ...,
        [0.5045, 0.5056, 0.4878,  ..., 0.4403, 0.4423, 0.4546],
        [0.4088, 0.3948, 0.3913,  ..., 0.3445, 0.3658, 0.3876],
        [0.3316, 0.3277, 0.3249,  ..., 0.4756, 0.4760, 0.4601]])

after

tensor([[0.0287, 0.0287, 0.0287,  ..., 0.0298, 0.0298, 0.0296],
        [0.0292, 0.0297, 0.0301,  ..., 0.0319, 0.0318, 0.0319],
        [0.0310, 0.0306, 0.0307,  ..., 0.0302, 0.0300, 0.0301],
        ...,
        [0.0297, 0.0297, 0.0286,  ..., 0.0258, 0.0259, 0.0267],
        [0.0241, 0.0232, 0.0230,  ..., 0.0202, 0.0215, 0.0228],
        [0.0195, 0.0193, 0.0191,  ..., 0.0278, 0.0279, 0.0271]])

[00/06] Run time for data\Set6\Boats.png is , PSNR is , SSIM is, 1.3832, 5.88, 0.0555,runtime is 0.0045
[01/06] Run time for data\Set6\fingerprint.png is , PSNR is , SSIM is, 1.6955, 5.06, 0.0118,runtime is 0.0030
[02/06] Run time for data\Set6\flinstones_DL.png is , PSNR is , SSIM is, 1.3922, 4.78, 0.0375,runtime is 0.0030
[03/06] Run time for data\Set6\house(2).png is , PSNR is , SSIM is, 0.2055, 5.88, 0.1371,runtime is 0.0010
[04/06] Run time for data\Set6\lenna(2).png is , PSNR is , SSIM is, 1.1541, 6.68, 0.0781,runtime is 0.0030
[05/06] Run time for data\Set6\peppers2562-3.png is , PSNR is , SSIM is, 0.3008, 6.66, 0.1175,runtime is 0.0015


CS ratio is 10, Avg PSNR/SSIM for Set6 is 5.82/0.0729, Avg runtime is 1.0219, Epoch number of model is 300