经典CNN轻量级网络之MobileNet v3 对厨余垃圾8分类数据集的分类任务

[train hyper-parameters: Namespace(batch_size=16, epochs=100)] [epoch: 1] train loss:0.1167 test accuracy:0.1211 [epoch: 2] train loss:0.1048 test accuracy:0.1211 [epoch: 3] train loss:0.0999 test accuracy:0.1211 [epoch: 4] train loss:0.0963 test accuracy:0.1226 [epoch: 5] train loss:0.0924 test accuracy:0.3028 [epoch: 6] train loss:0.0900 test accuracy:0.4372 [epoch: 7] train loss:0.0864 test accuracy:0.4815 [epoch: 8] train loss:0.0820 test accuracy:0.4919 [epoch: 9] train loss:0.0806 test accuracy:0.4978 [epoch: 10] train loss:0.0766 test accuracy:0.5066 [epoch: 11] train loss:0.0738 test accuracy:0.5111 [epoch: 12] train loss:0.0673 test accuracy:0.5214 [epoch: 13] train loss:0.0658 test accuracy:0.4742 [epoch: 14] train loss:0.0600 test accuracy:0.5022 [epoch: 15] train loss:0.0586 test accuracy:0.5037 [epoch: 16] train loss:0.0554 test accuracy:0.5170 [epoch: 17] train loss:0.0504 test accuracy:0.5140 [epoch: 18] train loss:0.0468 test accuracy:0.5037 [epoch: 19] train loss:0.0474 test accuracy:0.4948 [epoch: 20] train loss:0.0426 test accuracy:0.4993 [epoch: 21] train loss:0.0392 test accuracy:0.4682 [epoch: 22] train loss:0.0357 test accuracy:0.4904 [epoch: 23] train loss:0.0361 test accuracy:0.5066 [epoch: 24] train loss:0.0349 test accuracy:0.4978 [epoch: 25] train loss:0.0297 test accuracy:0.5155 [epoch: 26] train loss:0.0308 test accuracy:0.5081 [epoch: 27] train loss:0.0276 test accuracy:0.5052 [epoch: 28] train loss:0.0264 test accuracy:0.5111 [epoch: 29] train loss:0.0230 test accuracy:0.5347 [epoch: 30] train loss:0.0221 test accuracy:0.4579 [epoch: 31] train loss:0.0217 test accuracy:0.5052 [epoch: 32] train loss:0.0218 test accuracy:0.4919 [epoch: 33] train loss:0.0221 test accuracy:0.4047 [epoch: 34] train loss:0.0198 test accuracy:0.5037 [epoch: 35] train loss:0.0195 test accuracy:0.5170 [epoch: 36] train loss:0.0180 test accuracy:0.5007 [epoch: 37] train loss:0.0157 test accuracy:0.5347 [epoch: 38] train loss:0.0175 test accuracy:0.5480 [epoch: 39] train loss:0.0180 test accuracy:0.5037 [epoch: 40] train loss:0.0147 test accuracy:0.5258 [epoch: 41] train loss:0.0160 test accuracy:0.5288 [epoch: 42] train loss:0.0152 test accuracy:0.5199 [epoch: 43] train loss:0.0134 test accuracy:0.5052 [epoch: 44] train loss:0.0132 test accuracy:0.5199 [epoch: 45] train loss:0.0129 test accuracy:0.4889 [epoch: 46] train loss:0.0134 test accuracy:0.5244 [epoch: 47] train loss:0.0120 test accuracy:0.5229 [epoch: 48] train loss:0.0111 test accuracy:0.5111 [epoch: 49] train loss:0.0106 test accuracy:0.5155 [epoch: 50] train loss:0.0120 test accuracy:0.5140 [epoch: 51] train loss:0.0123 test accuracy:0.5037 [epoch: 52] train loss:0.0113 test accuracy:0.5007 [epoch: 53] train loss:0.0101 test accuracy:0.5303 [epoch: 54] train loss:0.0099 test accuracy:0.5126 [epoch: 55] train loss:0.0097 test accuracy:0.5007 [epoch: 56] train loss:0.0090 test accuracy:0.5066 [epoch: 57] train loss:0.0111 test accuracy:0.4742 [epoch: 58] train loss:0.0088 test accuracy:0.4934 [epoch: 59] train loss:0.0092 test accuracy:0.5170 [epoch: 60] train loss:0.0082 test accuracy:0.5140 [epoch: 61] train loss:0.0072 test accuracy:0.5081 [epoch: 62] train loss:0.0092 test accuracy:0.4697 [epoch: 63] train loss:0.0075 test accuracy:0.5096 [epoch: 64] train loss:0.0090 test accuracy:0.5096 [epoch: 65] train loss:0.0104 test accuracy:0.5081 [epoch: 66] train loss:0.0070 test accuracy:0.4963 [epoch: 67] train loss:0.0065 test accuracy:0.5185 [epoch: 68] train loss:0.0092 test accuracy:0.4904 [epoch: 69] train loss:0.0080 test accuracy:0.5052 [epoch: 70] train loss:0.0094 test accuracy:0.5126 [epoch: 71] train loss:0.0084 test accuracy:0.5244 [epoch: 72] train loss:0.0068 test accuracy:0.5214 [epoch: 73] train loss:0.0079 test accuracy:0.5170 [epoch: 74] train loss:0.0083 test accuracy:0.5332 [epoch: 75] train loss:0.0082 test accuracy:0.5037 [epoch: 76] train loss:0.0063 test accuracy:0.5303 [epoch: 77] train loss:0.0066 test accuracy:0.5022 [epoch: 78] train loss:0.0089 test accuracy:0.4742 [epoch: 79] train loss:0.0087 test accuracy:0.5037 [epoch: 80] train loss:0.0067 test accuracy:0.5199 [epoch: 81] train loss:0.0058 test accuracy:0.5421 [epoch: 82] train loss:0.0067 test accuracy:0.4934 [epoch: 83] train loss:0.0094 test accuracy:0.5391 [epoch: 84] train loss:0.0054 test accuracy:0.4742 [epoch: 85] train loss:0.0045 test accuracy:0.5111 [epoch: 86] train loss:0.0057 test accuracy:0.4978 [epoch: 87] train loss:0.0071 test accuracy:0.5199 [epoch: 88] train loss:0.0061 test accuracy:0.5022 [epoch: 89] train loss:0.0049 test accuracy:0.5229 [epoch: 90] train loss:0.0045 test accuracy:0.5391 [epoch: 91] train loss:0.0063 test accuracy:0.4978 [epoch: 92] train loss:0.0054 test accuracy:0.5199 [epoch: 93] train loss:0.0067 test accuracy:0.5377 [epoch: 94] train loss:0.0068 test accuracy:0.5318 [epoch: 95] train loss:0.0051 test accuracy:0.5377 [epoch: 96] train loss:0.0060 test accuracy:0.5495 [epoch: 97] train loss:0.0060 test accuracy:0.5081 [epoch: 98] train loss:0.0052 test accuracy:0.5081 [epoch: 99] train loss:0.0050 test accuracy:0.5155 [epoch: 100] train loss:0.0047 test accuracy:0.5244