训练cifar10的两个接近85分的版本，哪一个更好？

做个记录1架构是3个残差块layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 5, 32, 32, 32, 3, 1, 1));layers.emplace_back(std::make_sharedBN(cudnn, batch, 32, 32, 32));layers.emplace_back(std::make_sharedLeakyRL(cudnn, batch, 32, 32, 32));layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 32, 64, 32, 32, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22(cudnn, batch, 64, 32, 32));layers.emplace_back(std::make_sharedMaxPool2D(cudnn, batch, 64, 32, 32, 2, 2, 0, 2));layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 64, 128, 16, 16, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22(cudnn, batch, 128, 16, 16));layers.emplace_back(std::make_sharedMaxPool2D(cudnn, batch, 128, 16, 16, 2, 2, 0, 2));layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 128, 256, 8, 8, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22(cudnn, batch, 256, 8, 8));layers.emplace_back(std::make_sharedMaxPool2D(cudnn, batch, 256, 8, 8, 2, 2, 0, 2));// layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 256, 300, 4, 4, 4));layers.emplace_back(std::make_sharedLinear(cublas, batch, 256*4*4, 300));layers.emplace_back(std::make_sharedBN(cudnn, batch, 300, 1, 1));layers.emplace_back(std::make_sharedLeakyRL(cudnn, batch, 300, 1, 1));layers.emplace_back(std::make_sharedLinear(cublas, batch, 300, 10));————————————————方差和均值爆料情况时间: 27535.343750 mstrain Classification result: 97.72% ok (used 49984 images)时间: 2174.266113 msTest Classification result: 81.56% ok (used 9984 images)learn rate0.0001轮次24rb均值: 1.6140453815,rb方差:5.120904445648均值: -93.1453933716,方差:16.6951541901均值: -89.5411300659,方差:13.9249677658均值: -91.6143341064,方差:10.9254045486均值: -88.4864883423,方差:17.3460960388均值: -92.7311325073,方差:16.6858062744均值: -90.4656372070,方差:6.9966602325均值: -91.2220077515,方差:20.6955375671均值: -92.4978637695,方差:28.9232978821均值: -95.8759613037,方差:15.1841077805均值: -88.2818069458,方差:13.0533695221均值: -84.8050689697,方差:11.1592187881均值: -90.9559707642,方差:26.2159500122均值: -90.2095413208,方差:21.0307292938均值: -94.9991760254,方差:18.5383968353均值: -88.7679290771,方差:15.3791704178均值: -94.6937179565,方差:24.2541046143均值: -87.7234725952,方差:7.5897364616均值: -89.1307373047,方差:18.1437034607均值: -89.9150695801,方差:13.0529365540均值: -92.5738677979,方差:17.3618869781均值: -95.2071914673,方差:20.6090621948均值: -92.9055862427,方差:11.8797712326均值: -91.4030151367,方差:24.5985851288均值: -90.6616897583,方差:18.5187110901均值: -89.8819732666,方差:19.0706024170均值: -86.7467422485,方差:11.9170856476均值: -88.7020034790,方差:14.6680479050均值: -93.5892105103,方差:21.0500984192均值: -89.5009918213,方差:15.7073564529均值: -88.4498672485,方差:9.8511447906均值: -92.2534332275,方差:11.2799034119均值: -89.2119445801,方差:14.4264163971时间: 27513.873047 mstrain Classification result: 98.23% ok (used 49984 images)时间: 2177.084961 msTest Classification result:84.59% ok (used 9984 images)learn rate1e-05轮次25————————————————25个 残差块架构layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 5, 32, 32, 32, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22(cudnn, batch, 32, 32, 32));layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 32, 64, 32, 32, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22(cudnn, batch, 64, 32, 32));layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 64, 128, 32, 32, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22(cudnn, batch, 128, 32, 32));layers.emplace_back(std::make_sharedMaxPool2D(cudnn, batch, 128, 32, 32, 2, 2, 0, 2));layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 128, 256, 16, 16, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22SE(cudnn, batch, 256, 16, 16));layers.emplace_back(std::make_sharedMaxPool2D(cudnn, batch, 256, 16, 16, 2, 2, 0, 2));layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 256, 512, 8, 8, 3, 1, 1));layers.emplace_back(std::make_sharedresidualExt22SE(cudnn, batch, 512, 8, 8));layers.emplace_back(std::make_sharedaverPool2D(cudnn, batch, 512, 8, 8, 2, 2, 0, 2));// layers.emplace_back(std::make_sharedConv2D(cudnn, batch, 256, 300, 4, 4, 4));layers.emplace_back(std::make_sharedLinear(cublas, batch, 512*16, 350));layers.emplace_back(std::make_sharedBN(cudnn, batch, 350, 1, 1));layers.emplace_back(std::make_sharedLeakyRL(cudnn, batch, 350, 1, 1));layers.emplace_back(std::make_sharedLinear(cublas, batch, 350, 10));————————————————方差和均值爆料情况很明显稳定很多增加了两层senet也稳如泰山轮次17learn rate1e-06时间: 74121.484375 mstrain Classification result: 98.05% ok (used 49984 images)时间: 5473.725098 msTest Classification result: 84.42% ok (used 9984 images)轮次18learn rate1e-06时间: 77872.156250 mstrain Classification result: 98.12% ok (used 49984 images)时间: 5715.910156 msTest Classification result:84.48% ok (used 9984 images)轮次19learn rate1e-07时间: 77190.242188 mstrain Classification result: 97.97% ok (used 49984 images)时间: 5709.939941 msTest Classification result: 84.47% ok (used 9984 images)轮次20learn rate1e-07时间: 77821.828125 mstrain Classification result: 98.12% ok (used 49984 images)时间: 5641.894043 msTest Classification result: 84.46% ok (used 9984 images)轮次21learn rate1e-07时间: 75851.140625 mstrain Classification result: 98.09% ok (used 49984 images)时间: 5531.998047 msTest Classification result: 84.40% ok (used 9984 images)轮次22learn rate1e-07时间: 75607.789063 mstrain Classification result: 98.13% ok (used 49984 images)时间: 5596.937988 msTest Classification result: 84.45% ok (used 9984 images)轮次23learn rate1e-07时间: 76939.070313 mstrain Classification result: 97.98% ok (used 49984 images)时间: 5557.229004 msTest Classification result: 84.46% ok (used 9984 images)人生路上我们有时求稳有时激进只有回过头看才看明白这两天又学会了标签平滑的确有效比上面更平稳上3个senet块也能跑只是3060显卡不够用了开始变慢一轮上80秒了得分test83.65还行