Unverified Commit bff61258 authored by Kiryuu Sakuya's avatar Kiryuu Sakuya 🎵
Browse files

Finish 03

parent 9a3eda24
mnist
\ No newline at end of file
# MNIST手写数字识别问题的单神经元模型实践
按课程案例,动手完成编码实践。
在不改变模型的结构基础上,尝试采用不同的学习率、单批次样本数、训练轮数等超参数,让模型的准确率达到90%。
提交要求:
1、你认为最优的一次带运行结果的源代码文件(.ipynb 格式)
2、作为附件上传
评分标准:
1、完成案例中的代码,有完整的代码,模型能运行,准确率达87%以上;得6分;
2、准确率达89%以上;再得2分,否则得0分;
3、准确率达90%以上;再得2分,否则得0分;
\ No newline at end of file
epoch=01 loss= 3.198278427 accuracy= 0.4850
epoch=02 loss= 1.872825623 accuracy= 0.6520
epoch=03 loss= 1.446830392 accuracy= 0.7138
epoch=04 loss= 1.227654338 accuracy= 0.7532
epoch=05 loss= 1.094918728 accuracy= 0.7818
epoch=06 loss= 1.002100825 accuracy= 0.7988
epoch=07 loss= 0.934927762 accuracy= 0.8102
epoch=08 loss= 0.881675959 accuracy= 0.8178
epoch=09 loss= 0.839958370 accuracy= 0.8262
epoch=10 loss= 0.804570019 accuracy= 0.8306
epoch=11 loss= 0.776990473 accuracy= 0.8368
epoch=12 loss= 0.751436234 accuracy= 0.8410
epoch=13 loss= 0.728626728 accuracy= 0.8452
epoch=14 loss= 0.709876478 accuracy= 0.8484
epoch=15 loss= 0.693064034 accuracy= 0.8514
epoch=16 loss= 0.678240240 accuracy= 0.8542
epoch=17 loss= 0.665371895 accuracy= 0.8576
epoch=18 loss= 0.651332498 accuracy= 0.8576
epoch=19 loss= 0.639376342 accuracy= 0.8608
epoch=20 loss= 0.629004300 accuracy= 0.8624
epoch=21 loss= 0.620211422 accuracy= 0.8650
epoch=22 loss= 0.610780299 accuracy= 0.8644
epoch=23 loss= 0.601341903 accuracy= 0.8668
epoch=24 loss= 0.593505204 accuracy= 0.8674
epoch=25 loss= 0.585814238 accuracy= 0.8698
epoch=26 loss= 0.578637898 accuracy= 0.8710
epoch=27 loss= 0.572517991 accuracy= 0.8700
epoch=28 loss= 0.565916777 accuracy= 0.8726
epoch=29 loss= 0.560769260 accuracy= 0.8726
epoch=30 loss= 0.554210424 accuracy= 0.8748
epoch=31 loss= 0.549185216 accuracy= 0.8738
epoch=32 loss= 0.543703794 accuracy= 0.8764
epoch=33 loss= 0.539210379 accuracy= 0.8768
epoch=34 loss= 0.534550190 accuracy= 0.8778
epoch=35 loss= 0.530402124 accuracy= 0.8782
epoch=36 loss= 0.525754809 accuracy= 0.8796
epoch=37 loss= 0.520848811 accuracy= 0.8810
epoch=38 loss= 0.517098606 accuracy= 0.8802
epoch=39 loss= 0.513703048 accuracy= 0.8806
epoch=40 loss= 0.509720623 accuracy= 0.8812
epoch=41 loss= 0.506256223 accuracy= 0.8826
epoch=42 loss= 0.502291918 accuracy= 0.8836
epoch=43 loss= 0.499232858 accuracy= 0.8846
epoch=44 loss= 0.495934039 accuracy= 0.8840
epoch=45 loss= 0.493119240 accuracy= 0.8850
epoch=46 loss= 0.490840226 accuracy= 0.8852
epoch=47 loss= 0.487688959 accuracy= 0.8850
epoch=48 loss= 0.485160112 accuracy= 0.8846
epoch=49 loss= 0.482358336 accuracy= 0.8864
epoch=50 loss= 0.480277777 accuracy= 0.8866
epoch=51 loss= 0.477062583 accuracy= 0.8866
epoch=52 loss= 0.475359231 accuracy= 0.8854
epoch=53 loss= 0.472532660 accuracy= 0.8872
epoch=54 loss= 0.469635159 accuracy= 0.8864
epoch=55 loss= 0.467903763 accuracy= 0.8870
epoch=56 loss= 0.465845346 accuracy= 0.8878
epoch=57 loss= 0.463988930 accuracy= 0.8874
epoch=58 loss= 0.461396724 accuracy= 0.8896
epoch=59 loss= 0.460114956 accuracy= 0.8888
epoch=60 loss= 0.457748890 accuracy= 0.8906
epoch=61 loss= 0.455717862 accuracy= 0.8900
epoch=62 loss= 0.453327924 accuracy= 0.8912
epoch=63 loss= 0.451605260 accuracy= 0.8920
epoch=64 loss= 0.450798243 accuracy= 0.8916
epoch=65 loss= 0.448727846 accuracy= 0.8918
epoch=66 loss= 0.446790785 accuracy= 0.8920
epoch=67 loss= 0.445448399 accuracy= 0.8938
epoch=68 loss= 0.443943113 accuracy= 0.8936
epoch=69 loss= 0.441590965 accuracy= 0.8926
epoch=70 loss= 0.440179259 accuracy= 0.8936
epoch=71 loss= 0.438848883 accuracy= 0.8944
epoch=72 loss= 0.436886221 accuracy= 0.8940
epoch=73 loss= 0.435540825 accuracy= 0.8940
epoch=74 loss= 0.435228169 accuracy= 0.8944
epoch=75 loss= 0.432687342 accuracy= 0.8952
epoch=76 loss= 0.431297570 accuracy= 0.8960
epoch=77 loss= 0.430142432 accuracy= 0.8950
epoch=78 loss= 0.428755701 accuracy= 0.8960
epoch=79 loss= 0.428802639 accuracy= 0.8942
epoch=80 loss= 0.426565140 accuracy= 0.8954
epoch=81 loss= 0.425424129 accuracy= 0.8958
epoch=82 loss= 0.423666567 accuracy= 0.8962
epoch=83 loss= 0.422743022 accuracy= 0.8950
epoch=84 loss= 0.421432555 accuracy= 0.8958
epoch=85 loss= 0.421201527 accuracy= 0.8958
epoch=86 loss= 0.419389546 accuracy= 0.8980
epoch=87 loss= 0.418214023 accuracy= 0.8966
epoch=88 loss= 0.417555571 accuracy= 0.8974
epoch=89 loss= 0.415932417 accuracy= 0.8988
epoch=90 loss= 0.415770471 accuracy= 0.8986
epoch=91 loss= 0.413418323 accuracy= 0.8992
epoch=92 loss= 0.412451804 accuracy= 0.9000
epoch=93 loss= 0.411382765 accuracy= 0.8994
epoch=94 loss= 0.410894543 accuracy= 0.8994
epoch=95 loss= 0.409242630 accuracy= 0.9004
epoch=96 loss= 0.408311665 accuracy= 0.9008
epoch=97 loss= 0.408004224 accuracy= 0.8994
epoch=98 loss= 0.406678140 accuracy= 0.9014
epoch=99 loss= 0.405874610 accuracy= 0.9022
epoch=100 loss= 0.404687732 accuracy= 0.9020
epoch=101 loss= 0.403431445 accuracy= 0.9024
epoch=102 loss= 0.403403550 accuracy= 0.9020
epoch=103 loss= 0.402171463 accuracy= 0.9028
epoch=104 loss= 0.401789904 accuracy= 0.9036
epoch=105 loss= 0.400487155 accuracy= 0.9036
epoch=106 loss= 0.399224013 accuracy= 0.9034
epoch=107 loss= 0.398142815 accuracy= 0.9042
epoch=108 loss= 0.398032814 accuracy= 0.9040
epoch=109 loss= 0.396699101 accuracy= 0.9030
epoch=110 loss= 0.396185338 accuracy= 0.9044
epoch=111 loss= 0.395714313 accuracy= 0.9050
epoch=112 loss= 0.394927740 accuracy= 0.9040
epoch=113 loss= 0.393970847 accuracy= 0.9046
epoch=114 loss= 0.392998159 accuracy= 0.9046
epoch=115 loss= 0.392011464 accuracy= 0.9032
epoch=116 loss= 0.391528994 accuracy= 0.9042
epoch=117 loss= 0.390255868 accuracy= 0.9052
epoch=118 loss= 0.389445812 accuracy= 0.9054
epoch=119 loss= 0.388657957 accuracy= 0.9054
epoch=120 loss= 0.388166606 accuracy= 0.9078
epoch=121 loss= 0.387895525 accuracy= 0.9062
epoch=122 loss= 0.387123019 accuracy= 0.9058
epoch=123 loss= 0.386067867 accuracy= 0.9064
epoch=124 loss= 0.385235637 accuracy= 0.9064
epoch=125 loss= 0.384074599 accuracy= 0.9072
epoch=126 loss= 0.383924544 accuracy= 0.9074
epoch=127 loss= 0.383630335 accuracy= 0.9066
epoch=128 loss= 0.382886171 accuracy= 0.9082
epoch=129 loss= 0.381875962 accuracy= 0.9072
epoch=130 loss= 0.381316304 accuracy= 0.9082
epoch=131 loss= 0.380469054 accuracy= 0.9072
epoch=132 loss= 0.379814833 accuracy= 0.9078
epoch=133 loss= 0.379486024 accuracy= 0.9076
epoch=134 loss= 0.379039854 accuracy= 0.9068
epoch=135 loss= 0.377738923 accuracy= 0.9070
epoch=136 loss= 0.377503991 accuracy= 0.9074
epoch=137 loss= 0.377217650 accuracy= 0.9078
epoch=138 loss= 0.376586974 accuracy= 0.9074
epoch=139 loss= 0.375774354 accuracy= 0.9080
epoch=140 loss= 0.374864638 accuracy= 0.9082
epoch=141 loss= 0.374930799 accuracy= 0.9084
epoch=142 loss= 0.373751849 accuracy= 0.9084
epoch=143 loss= 0.373181105 accuracy= 0.9094
epoch=144 loss= 0.372630119 accuracy= 0.9074
epoch=145 loss= 0.372064948 accuracy= 0.9078
epoch=146 loss= 0.371509701 accuracy= 0.9084
epoch=147 loss= 0.371103019 accuracy= 0.9084
epoch=148 loss= 0.371076375 accuracy= 0.9080
epoch=149 loss= 0.370037347 accuracy= 0.9088
epoch=150 loss= 0.369185656 accuracy= 0.9086
Train Finished.
Training set Accuracy= 0.90843636
Test set Accuracy= 0.9017
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment