数据集和误差分析

我们的作业

任务说明：调研新冠疫情预测与分析系统。使用一个城市的新冠历史数据（自 行网上获取），应用一种合适的人工智能算法构建新冠疫情预测模型。根据构建 好的模型，能根据前10天的新冠数据，预测未来3天的新冠数据。设计一个新冠 疫情预测与分析系统，完成模型的训练，模型的测试（给出测试精度）和模型的
应用等功能。根据调研结果，可以自行设计更具有实用性和创新性的系统功能。 完成系统设计和开发后，请按附件1实验报告模板按要求完成报告内容并以附件 形式提交

数据

中国1.28~3.10各个省份数据

.csv文件打开

主要是发现中文在csv里面怪怪的，考虑省份在数据处理时候没啥卵用就缩写了

hb,3554,4586,5806,7153,9074,11177,13522,16678,19665,22112,24953,27100,29631,31728,33366,48206,51986,54406,56249,58182,59989,61682,62031,62662,63454,64084,64287,64786,65187,65596,65914,66337,66907,67103,67217,67331,67466,67592,67666,67707,67743,67760,67773,67781,67786,67790,67794,67798,67799,67800,67800
gd,277,354,436,535,632,725,813,895,970,1034,1095,1131,1159,1177,1219,1241,1261,1294,1316,1322,1328,1331,1332,1333,1339,1342,1345,1347,1347,1347,1348,1349,1349,1350,1350,1350,1351,1352,1352,1352,1352,1353,1356,1356,1356,1356,1358,1361,1364,1370,1370
hn,206,278,352,422,493,566,675,764,851,914,981,1033,1073,1105,1135,1169,1184,1212,1231,1246,1257,1262,1265,1267,1270,1271,1271,1271,1271,1272,1272,1272,1272,1272,1272,1272,1272,1272,1272,1272,1272,1272,1273,1273,1273,1273,1273,1273,1273,1273,1273
zj,296,428,537,599,661,724,829,895,954,1006,1048,1075,1092,1117,1131,1145,1155,1162,1167,1171,1172,1174,1175,1203,1205,1205,1205,1205,1205,1205,1205,1205,1205,1206,1213,1213,1215,1215,1215,1215,1215,1215,1215,1215,1215,1227,1231,1231,1232,1232,1232
hn,221,277,332,389,463,521,593,661,711,772,803,838,879,912,946,968,988,1001,1004,1006,1007,1008,1010,1011,1013,1016,1016,1016,1016,1017,1017,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018,1018
ah,152,200,237,297,340,408,480,530,591,665,733,779,830,860,889,910,934,950,962,973,982,986,987,988,989,989,989,989,989,989,990,990,990,990,990,990,990,990,990,990,990,990,990,990,990,990,990,990,990,990,990
jx,109,162,240,286,333,391,476,548,600,661,698,740,771,804,844,872,900,913,925,930,933,934,934,934,934,934,934,934,934,934,935,935,935,935,935,935,935,935,935,935,935,935,935,935,935,935,935,935,935,935,935
sd,130,158,184,206,230,259,275,307,347,386,416,444,466,487,497,509,523,532,537,541,543,544,546,749,750,754,755,756,756,756,756,756,758,758,758,758,758,758,758,758,758,758,760,760,760,760,760,760,761,761,761
js,99,129,168,202,236,271,308,341,373,408,439,468,492,515,543,570,593,604,617,626,629,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631,631
cq,147,182,211,247,275,312,344,376,400,415,428,450,473,489,509,525,532,538,547,552,553,553,560,567,572,573,575,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576,576
sc,108,142,177,207,231,254,282,301,321,344,363,386,405,417,436,451,463,470,481,495,508,514,520,525,526,526,527,529,531,534,538,538,538,538,538,538,539,539,539,539,539,539,539,539,539,539,539,539,540,540,540
hlj,38,43,59,80,95,121,155,190,227,277,295,307,331,360,378,395,418,425,445,457,464,470,476,479,479,480,480,480,480,480,480,480,480,480,480,480,481,481,481,481,481,481,482,482,482,482,482,482,482,482,482
bj,102,111,139,168,191,212,228,253,274,297,315,326,337,342,352,366,372,375,380,381,387,393,395,396,399,399,399,400,400,410,410,411,413,414,414,417,418,422,426,428,428,429,435,435,436,437,442,452,456,469,469
sh,96,112,135,169,182,203,219,243,257,277,286,293,299,303,311,315,318,326,328,332,333,333,334,334,335,335,335,336,336,337,337,337,337,337,338,338,339,342,342,342,342,344,344,344,346,352,353,355,358,361,361
hb,48,65,82,96,104,113,126,135,157,172,195,206,218,239,251,265,283,291,300,301,302,306,307,308,309,311,311,311,312,317,318,318,318,318,318,318,318,318,318,318,318,318,318,318,318,318,318,318,318,318,318
fj,82,101,120,144,159,179,194,205,215,224,239,250,261,267,272,279,281,285,287,290,292,293,293,293,293,293,293,294,294,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296,296
gx,58,78,87,100,111,127,139,150,168,171,183,195,210,215,222,222,226,235,237,238,242,244,245,246,249,249,251,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,253,253,253
sx,56,63,87,101,116,128,142,165,173,184,195,208,213,219,225,229,230,232,236,240,240,242,244,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,246,246,246
yn,55,70,83,91,105,114,119,124,133,136,138,141,149,153,154,156,162,168,171,171,172,173,173,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,174,175,176,176,176
hns,43,46,52,60,64,70,80,91,106,117,124,130,138,142,157,157,158,162,162,163,163,163,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168,168
gz,9,12,29,29,38,46,58,64,71,81,89,99,109,127,133,135,140,143,144,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,146,147,146,146
tj,27,31,32,38,48,58,67,69,78,81,88,90,94,104,110,117,120,121,124,125,127,128,131,132,135,135,135,135,135,135,136,136,136,136,136,136,136,136,136,136,136,136,136,136,136,136,136,136,136,136,136
sx,27,35,39,47,56,66,74,81,90,96,104,115,119,122,124,126,126,127,128,129,130,131,131,132,132,132,132,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133,133
ln,39,41,48,63,69,73,77,88,91,96,105,107,108,111,116,116,117,119,121,121,121,121,121,121,121,121,121,121,121,121,121,121,122,122,125,125,125,125,125,125,125,125,125,125,125,125,125,125,125,125,125
jl,9,14,14,17,23,31,42,54,59,65,69,78,80,81,83,84,86,88,89,89,89,90,91,91,91,91,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93,93
gs,24,26,29,35,40,51,55,57,62,67,71,79,83,86,86,87,90,90,90,90,91,91,91,91,91,91,91,91,91,91,91,91,91,91,91,91,91,102,119,120,124,124,125,127,127,129,132,133,133,133,133

湖北的

其他省份的

这是美丽国的数据

确诊	治愈	死亡	日期
5	0	0	2020/1/27
6	0	0	2020/1/31
6	0	0	2020/2/1
8	0	0	2020/2/2
9	0	0	2020/2/3
11	0	0	2020/2/4
12	0	0	2020/2/6
12	0	0	2020/2/10
13	0	0	2020/2/11
13	3	0	2020/2/12
14	3	0	2020/2/13
15	3	0	2020/2/14
15	3	0	2020/2/15
15	3	0	2020/2/16
15	3	0	2020/2/17
15	3	0	2020/2/20
35	3	0	2020/2/22
35	3	0	2020/2/23
35	3	0	2020/2/24
53	3	0	2020/2/25
57	3	0	2020/2/26
60	3	0	2020/2/27
60	3	0	2020/2/28
63	3	0	2020/2/29
63	3	1	2020/3/1
77	3	1	2020/3/2
100	3	6	2020/3/3
122	3	9	2020/3/4
153	3	11	2020/3/5
232	3	14	2020/3/6
324	10	14	2020/3/7
445	10	19	2020/3/8
572	10	21	2020/3/9
704	10	26	2020/3/10
1004	10	31	2020/3/11
1004	10	31	2020/3/12
1635	10	39	2020/3/13
2084	10	41	2020/3/14
2885	10	60	2020/3/15
3700	10	66	2020/3/16
4661	48	90	2020/3/17
6420	74	108	2020/3/18
10259	106	153	2020/3/19
14250	121	208	2020/3/20
19624	147	260	2020/3/21
26997	176	340	2020/3/22
35360	178	473	2020/3/23
46450	178	593	2020/3/24
55243	354	802	2020/3/25
69194	619	1050	2020/3/26
85840	713	1296	2020/3/27
105470	2072	1710	2020/3/28
124686	2612	2191	2020/3/29
143101	4865	2517	2020/3/30
164603	5896	3170	2020/3/31
189633	7136	4081	2020/4/1
216722	8672	5137	2020/4/2
245658	9311	6068	2020/4/3
278537	9920	7163	2020/4/4
312245	15021	8503	2020/4/5
337971	17582	9654	2020/4/6
368449	19919	10993	2020/4/7
399979	22539	12912	2020/4/8
432579	24213	14830	2020/4/9
467184	26522	16736	2020/4/10
501615	29192	18781	2020/4/11
530830	32314	20646	2020/4/12
558526	42018	22146	2020/4/13
583220	44319	23654	2020/4/14
609696	49966	26059	2020/4/15
640014	52772	31002	2020/4/16
672246	56236	33318	2020/4/17
706832	59672	37084	2020/4/18
735366	66854	39095	2020/4/19
760570	71011	40702	2020/4/20
788920	73533	42458	2020/4/21
826184	75682	45150	2020/4/22
843937	76616	46838	2020/4/23
870468	80937	50031	2020/4/24
907096	99121	52063	2020/4/25
940797	105823	54001	2020/4/26
967585	107070	54931	2020/4/27
989357	111587	56386	2020/4/28
1014568	115936	58471	2020/4/29
1040608	124064	61123	2020/4/30
1070032	153947	63019	2020/5/1
1107815	164015	65244	2020/5/2
1133069	175382	66385	2020/5/3
1158341	180152	67686	2020/5/4
1181885	187180	69079	2020/5/5
1206323	189791	71152	2020/5/6
1231943	189910	73566	2020/5/7
1256972	195036	75670	2020/5/8
1286833	198993	77280	2020/5/9
1312099	212534	78862	2020/5/10
1332411	216169	79606	2020/5/11
1351200	232733	80897	2020/5/12
1371395	230287	82461	2020/5/13
1395265	243430	84313	2020/5/14
1419998	246414	85974	2020/5/15
1443397	250747	87568	2020/5/16

一些思路

对于这道题的一些思考

• SIR 是基本的传染病预测模型 构建起来比较简单 考虑到 易感 感染 康复 三者即可 （精度不足）
• SEIR 即SIR 的升级模型 在上面的基础上 加了一层潜伏人群
• 基于LSTM的时间序列模型
• 神经网络算法
• 保底 RNN算法

另外借鉴了guyouyang大佬的一些思路，这是他的GitHub地址。