孤独症项目（3）
今天一刷朋友圈，一个小学同学订婚了。老实讲确实是羡慕了，到我这个年龄，游戏的吸引力已经下降了很多了，我需要性和爱。可是我身高，帅气，money什么都没有，所以我怎么办呢。我把那个同学屏蔽了，哈哈，事情就是这样，我抢不过别的男生，因为我个太矮，家太穷，自己又没有能力，所以我只能视而不见，活在自己的小世界里就好啦。开始今天的学习吧。
1
这个npy文件我知道咋回事了
{'timeseires': array([[[ -4.671351, -12.931799, -20.238678, ..., -4.6999 ,
-2.881232, -5.373659],
[ -6.610331, -10.146871, -11.706947, ..., -8.545325,
-4.824956, -3.549801],
[ 6.539966, -3.924838, -22.075667, ..., -10.810907,
-9.507657, -11.782674],
...,
[ 23.815056, 14.159346, -4.235137, ..., 2.318333,
-0.419391, -6.08656 ],
[ 29.358294, 29.569822, 12.97839 , ..., -35.478521,
-32.452544, -38.755226],
[ 12.700474, 10.523634, -1.103658, ..., 16.980012,
18.994251, 7.554179]],
[[ 15.055096, 14.30167 , 8.35954 , ..., -2.320052,
5.738047, 18.929523],
[ -2.650435, -5.070064, -6.582098, ..., -6.500024,
-6.520322, -2.209336],
[ 7.958244, 7.113133, 3.146248, ..., -7.816352,
0.160806, 12.985252],
...,
[ 0.216612, -7.946295, -14.338741, ..., -2.661149,
-1.293657, 4.052661],
[ 0.595963, -1.462514, -3.352045, ..., -7.601621,
-1.38041 , 7.066522],
[ -1.754612, -2.342497, -2.100473, ..., -4.888682,
0.817308, 6.089231]],
[[ 10.699614, 12.640328, 9.107096, ..., 25.303575,
14.631509, -2.431361],
[ 8.946028, 11.992252, 10.768465, ..., 17.12695 ,
4.901006, -9.559137],
[ 12.502747, 17.729441, 17.117685, ..., 13.455708,
-13.292004, -35.746319],
...,
[-13.968082, -18.558227, -20.739389, ..., 30.461445,
48.840466, 47.846905],
[ 7.775516, 7.945796, 4.77154 , ..., 35.356228,
42.096015, 35.434606],
[ 3.752828, 5.027397, 4.62332 , ..., 22.339197,
34.736423, 33.95411 ]],
...,
[[ -9.594532, -18.276965, -12.977462, ..., 22.482481,
9.076281, -3.665357],
[ -1.753394, -3.048763, -0.66247 , ..., 3.479453,
6.685637, 5.947072],
[ -7.053816, -4.420122, 6.336983, ..., 23.162804,
14.313564, 2.746504],
...,
[ 14.337799, 18.14827 , -0.895313, ..., 37.246694,
58.276117, 34.074897],
[-15.100426, -22.201254, -24.41196 , ..., 60.251682,
66.325714, 32.992202],
[-13.52109 , -18.166892, -16.496382, ..., 38.908534,
41.846779, 15.183713]],
[[ -4.960212, -10.467057, -2.350238, ..., 8.175621,
-3.594708, -13.050947],
[ -2.188337, -5.102677, -1.05976 , ..., 3.500581,
5.308049, -0.665579],
[ -7.689325, -4.299846, 4.479113, ..., 7.190843,
-6.158385, -10.379174],
...,
[ 4.838299, -1.461336, 3.449214, ..., 22.214853,
10.277305, -20.902231],
[ -9.16024 , -6.62222 , 5.643113, ..., 1.50843 ,
-7.206676, -21.635724],
[ -7.149388, -5.05326 , 5.05269 , ..., 12.605102,
7.582936, -10.857408]],
[[ -8.999973, -8.265457, 0.224464, ..., 2.243862,
-2.492217, -13.048096],
[ -1.115053, 0.311383, -0.91307 , ..., 3.98221 ,
1.695672, -5.227182],
[ 2.39671 , 2.200741, 5.486765, ..., 13.607921,
7.737062, -13.571547],
...,
[-24.200021, -15.751269, 3.40317 , ..., 3.005727,
3.171528, -1.583004],
[-12.074557, -7.229352, -1.982399, ..., 4.310723,
-9.359018, -3.962187],
[ -7.750716, -6.153427, -2.874191, ..., 4.213236,
-2.779235, -1.473658]]], shape=(1009, 111, 100)), 'label': array([0., 0., 0., ..., 0., 0., 0.], shape=(1009,)), 'corr': array([[[ 0. , 0.97663709, 0.82530072, ..., 0.13740499,
0.35407144, 0.08548723],
[ 0.97663709, 0. , 0.81736156, ..., 0.12545135,
0.31653123, 0.08523937],
[ 0.82530072, 0.81736156, 0. , ..., 0.32199278,
0.48073326, 0.23490678],
...,
[ 0.13740499, 0.12545135, 0.32199278, ..., 0. ,
0.91433765, 0.76759212],
[ 0.35407144, 0.31653123, 0.48073326, ..., 0.91433765,
0. , 1.1123392 ],
[ 0.08548723, 0.08523937, 0.23490678, ..., 0.76759212,
1.1123392 , 0. ]],
[[ 0. , 0.62647921, 0.68386934, ..., -0.05088815,
0.60446506, 0.41812769],
[ 0.62647921, 0. , 0.54538175, ..., 0.09402108,
0.39961469, 0.28481799],
[ 0.68386934, 0.54538175, 0. , ..., -0.06276508,
0.30605398, 0.22979597],
...,
[-0.05088815, 0.09402108, -0.06276508, ..., 0. ,
-0.01265819, 0.08046417],
[ 0.60446506, 0.39961469, 0.30605398, ..., -0.01265819,
0. , 0.85968089],
[ 0.41812769, 0.28481799, 0.22979597, ..., 0.08046417,
0.85968089, 0. ]],
[[ 0. , 0.52387629, 0.39586813, ..., 0.49770721,
0.77686881, 0.67761376],
[ 0.52387629, 0. , 0.43492339, ..., 0.08623291,
0.30371602, 0.28660143],
[ 0.39586813, 0.43492339, 0. , ..., -0.10309543,
-0.00903498, -0.06115111],
...,
[ 0.49770721, 0.08623291, -0.10309543, ..., 0. ,
0.69001642, 0.72559831],
[ 0.77686881, 0.30371602, -0.00903498, ..., 0.69001642,
0. , 1.37230362],
[ 0.67761376, 0.28660143, -0.06115111, ..., 0.72559831,
1.37230362, 0. ]],
...,
[[ 0. , 0.28196335, 0.62367273, ..., 0.31654042,
0.30184585, 0.24758647],
[ 0.28196335, 0. , 0.52368336, ..., -0.05335835,
-0.01349599, -0.08051841],
[ 0.62367273, 0.52368336, 0. , ..., 0.07688247,
0.15630406, 0.09915298],
...,
[ 0.31654042, -0.05335835, 0.07688247, ..., 0. ,
0.33613925, 0.35266004],
[ 0.30184585, -0.01349599, 0.15630406, ..., 0.33613925,
0. , 1.26734374],
[ 0.24758647, -0.08051841, 0.09915298, ..., 0.35266004,
1.26734374, 0. ]],
[[ 0. , 0.3318541 , 0.3479959 , ..., 0.17348234,
0.33035729, 0.28096653],
[ 0.3318541 , 0. , 0.13110274, ..., 0.16131239,
0.18152441, 0.0901034 ],
[ 0.3479959 , 0.13110274, 0. , ..., 0.03690945,
0.19735146, 0.23887414],
...,
[ 0.17348234, 0.16131239, 0.03690945, ..., 0. ,
0.21702285, 0.29280098],
[ 0.33035729, 0.18152441, 0.19735146, ..., 0.21702285,
0. , 0.63961068],
[ 0.28096653, 0.0901034 , 0.23887414, ..., 0.29280098,
0.63961068, 0. ]],
[[ 0. , 0.61490946, 0.86683146, ..., 0.28751356,
0.35820952, 0.36030001],
[ 0.61490946, 0. , 0.44664105, ..., 0.39767059,
0.28708429, 0.39182893],
[ 0.86683146, 0.44664105, 0. , ..., 0.18048986,
0.38597263, 0.43650839],
...,
[ 0.28751356, 0.39767059, 0.18048986, ..., 0. ,
0.22311195, 0.35032753],
[ 0.35820952, 0.28708429, 0.38597263, ..., 0.22311195,
0. , 1.03944121],
[ 0.36030001, 0.39182893, 0.43650839, ..., 0.35032753,
1.03944121, 0. ]]], shape=(1009, 111, 111)), 'pcorr': array([[[ 0. , 0.0845738 , 0.05283968, ..., -0.02852326,
-0.05030299, -0.00900934],
[ 0.0845738 , 0. , 0.05794923, ..., -0.06134557,
0.02918709, -0.04988144],
[ 0.05283968, 0.05794923, 0. , ..., 0.08585561,
0.0933945 , -0.01767182],
...,
[-0.02852326, -0.06134557, 0.08585561, ..., 0. ,
0.0742375 , 0.11123407],
[-0.05030299, 0.02918709, 0.0933945 , ..., 0.0742375 ,
0. , 0.26601118],
[-0.00900934, -0.04988144, -0.01767182, ..., 0.11123407,
0.26601118, 0. ]],
[[ 0. , 0.02515851, -0.02687615, ..., -0.07865261,
0.03765896, 0.03533854],
[ 0.02515851, 0. , 0.03731194, ..., -0.04081472,
-0.02018316, 0.02902786],
[-0.02687615, 0.03731194, 0. , ..., 0.02341885,
-0.0237432 , 0.07630574],
...,
[-0.07865261, -0.04081472, 0.02341885, ..., 0. ,
0.05092006, 0.03885538],
[ 0.03765896, -0.02018316, -0.0237432 , ..., 0.05092006,
0. , 0.08766317],
[ 0.03533854, 0.02902786, 0.07630574, ..., 0.03885538,
0.08766317, 0. ]],
[[ 0. , 0.0345411 , 0.02792029, ..., -0.01667542,
0.01530761, -0.00780056],
[ 0.0345411 , 0. , 0.05455556, ..., 0.0643729 ,
0.01796368, 0.03800115],
[ 0.02792029, 0.05455556, 0. , ..., -0.01042823,
-0.03121966, 0.01426741],
...,
[-0.01667542, 0.0643729 , -0.01042823, ..., 0. ,
-0.07791113, -0.04875181],
[ 0.01530761, 0.01796368, -0.03121966, ..., -0.07791113,
0. , 0.13051308],
[-0.00780056, 0.03800115, 0.01426741, ..., -0.04875181,
0.13051308, 0. ]],
...,
[[ 0. , 0.0855516 , 0.13597267, ..., 0.16457852,
0.04648458, 0.03031205],
[ 0.0855516 , 0. , -0.02231855, ..., -0.01696728,
0.06332429, -0.05170068],
[ 0.13597267, -0.02231855, 0. , ..., -0.14505349,
-0.05634801, 0.06301302],
...,
[ 0.16457852, -0.01696728, -0.14505349, ..., 0. ,
-0.09211864, 0.03218952],
[ 0.04648458, 0.06332429, -0.05634801, ..., -0.09211864,
0. , 0.56874401],
[ 0.03031205, -0.05170068, 0.06301302, ..., 0.03218952,
0.56874401, 0. ]],
[[ 0. , 0.00369714, 0.01803972, ..., 0.02636135,
0.01498399, 0.00107738],
[ 0.00369714, 0. , -0.04215888, ..., 0.08585435,
0.01525622, -0.0044203 ],
[ 0.01803972, -0.04215888, 0. , ..., 0.00435903,
-0.04385825, 0.01943377],
...,
[ 0.02636135, 0.08585435, 0.00435903, ..., 0. ,
0.08473985, 0.14461459],
[ 0.01498399, 0.01525622, -0.04385825, ..., 0.08473985,
0. , 0.1207058 ],
[ 0.00107738, -0.0044203 , 0.01943377, ..., 0.14461459,
0.1207058 , 0. ]],
[[ 0. , 0.02612483, 0.03413983, ..., 0.06152281,
-0.02745578, -0.01738546],
[ 0.02612483, 0. , -0.04195744, ..., 0.01392757,
-0.07636841, 0.00512999],
[ 0.03413983, -0.04195744, 0. , ..., -0.00615086,
0.00395361, 0.02567149],
...,
[ 0.06152281, 0.01392757, -0.00615086, ..., 0. ,
0.01919661, 0.07781511],
[-0.02745578, -0.07636841, 0.00395361, ..., 0.01919661,
0. , 0.16214421],
[-0.01738546, 0.00512999, 0.02567149, ..., 0.07781511,
0.16214421, 0. ]]], shape=(1009, 111, 111)), 'site': array(['PITT', 'PITT', 'PITT', ..., 'SBL', 'MAX_MUN', 'MAX_MUN'],
shape=(1009,), dtype='<U8')}
它其实是一个被处理过后的数据集，比如说ABIDE数据集。这个abide.npy最后一个属性是site，它是一个一维数组，有1009个元素。为什么是1009呢，这代表这些数据可以分成1009组，它的site有时候是相同的，这代表可能在一个地方收集了多组数据。你比如说这个PITT吧，它就可能代表匹兹堡。
这个label也是一个很简单的属性，给数据打标签嘛，这个数据集都是ABIDE数据集的内容，所以没有打标签，也可能是后续要划分实验组什么的打标签。
我总算知道这个coor是什么了，我之前看过一篇论文，里面提到了皮尔逊积差相关系数，是计算两组数据之间线性关系的。比如此处吧，coor的shape=(1009, 111, 111)，意思就是说有1009个二维数组，依次分别是[0][0],[0][1],[0][2],…,[0][1008];然后是[1][0],[1][1],…,[1][1008],不断递进，最后是[1008][0],[1008][1],…,[1008][1008]
[ 0. , 0.97663709, 0.82530072, …, 0.13740499,
0.35407144, 0.08548723],你看这个吧，0代表第一组数据和第二组数据之间的皮尔逊积差相关系数是0。这个肯定是处理过后的数据，因为一组数据和自己的相关性为1，它这里是0，肯定是去除了自相关。再看 0.97663709，这代表第一组数据和第二组数据的相关性为0.97663709，相关性快接近1了，这说明相关性很强。还有这个0.82530072，它代表第一组数据和第三组数据之间的相关性也很高了。中间有一堆的省略号，这个很有意思，这是因为数据太多了，所以它显示的时候把中间这部分省略掉了。这个0.13740499代表这两组数据之间的相关性很弱了。
我说既然有了corr，也就是皮尔逊积差相关系数的矩阵，为什么还要有pcorr，原来pcorr是部分相关系数的矩阵。
我懂了，我之前在论文里面一直看别人是怎么处理数据的，你像这个
'timeseires': array([[[ -4.671351, -12.931799, -20.238678, ..., -4.6999 ,
-2.881232, -5.373659],
[ -6.610331, -10.146871, -11.706947, ..., -8.545325,
-4.824956, -3.549801],
[ 6.539966, -3.924838, -22.075667, ..., -10.810907,
-9.507657, -11.782674],
...,
[ 23.815056, 14.159346, -4.235137, ..., 2.318333,
-0.419391, -6.08656 ],
[ 29.358294, 29.569822, 12.97839 , ..., -35.478521,
-32.452544, -38.755226],
[ 12.700474, 10.523634, -1.103658, ..., 16.980012,
18.994251, 7.554179]],
[[ 15.055096, 14.30167 , 8.35954 , ..., -2.320052,
5.738047, 18.929523],
[ -2.650435, -5.070064, -6.582098, ..., -6.500024,
-6.520322, -2.209336],
[ 7.958244, 7.113133, 3.146248, ..., -7.816352,
0.160806, 12.985252],
...,
[ 0.216612, -7.946295, -14.338741, ..., -2.661149,
-1.293657, 4.052661],
[ 0.595963, -1.462514, -3.352045, ..., -7.601621,
-1.38041 , 7.066522],
[ -1.754612, -2.342497, -2.100473, ..., -4.888682,
0.817308, 6.089231]],
[[ 10.699614, 12.640328, 9.107096, ..., 25.303575,
14.631509, -2.431361],
[ 8.946028, 11.992252, 10.768465, ..., 17.12695 ,
4.901006, -9.559137],
[ 12.502747, 17.729441, 17.117685, ..., 13.455708,
-13.292004, -35.746319],
...,
[-13.968082, -18.558227, -20.739389, ..., 30.461445,
48.840466, 47.846905],
[ 7.775516, 7.945796, 4.77154 , ..., 35.356228,
42.096015, 35.434606],
[ 3.752828, 5.027397, 4.62332 , ..., 22.339197,
34.736423, 33.95411 ]],
...,
[[ -9.594532, -18.276965, -12.977462, ..., 22.482481,
9.076281, -3.665357],
[ -1.753394, -3.048763, -0.66247 , ..., 3.479453,
6.685637, 5.947072],
[ -7.053816, -4.420122, 6.336983, ..., 23.162804,
14.313564, 2.746504],
...,
[ 14.337799, 18.14827 , -0.895313, ..., 37.246694,
58.276117, 34.074897],
[-15.100426, -22.201254, -24.41196 , ..., 60.251682,
66.325714, 32.992202],
[-13.52109 , -18.166892, -16.496382, ..., 38.908534,
41.846779, 15.183713]],
[[ -4.960212, -10.467057, -2.350238, ..., 8.175621,
-3.594708, -13.050947],
[ -2.188337, -5.102677, -1.05976 , ..., 3.500581,
5.308049, -0.665579],
[ -7.689325, -4.299846, 4.479113, ..., 7.190843,
-6.158385, -10.379174],
...,
[ 4.838299, -1.461336, 3.449214, ..., 22.214853,
10.277305, -20.902231],
[ -9.16024 , -6.62222 , 5.643113, ..., 1.50843 ,
-7.206676, -21.635724],
[ -7.149388, -5.05326 , 5.05269 , ..., 12.605102,
7.582936, -10.857408]],
[[ -8.999973, -8.265457, 0.224464, ..., 2.243862,
-2.492217, -13.048096],
[ -1.115053, 0.311383, -0.91307 , ..., 3.98221 ,
1.695672, -5.227182],
[ 2.39671 , 2.200741, 5.486765, ..., 13.607921,
7.737062, -13.571547],
...,
[-24.200021, -15.751269, 3.40317 , ..., 3.005727,
3.171528, -1.583004],
[-12.074557, -7.229352, -1.982399, ..., 4.310723,
-9.359018, -3.962187],
[ -7.750716, -6.153427, -2.874191, ..., 4.213236,
-2.779235, -1.473658]]], shape=(1009, 111, 100)),
它记录的是1009个时刻，然后大脑分为111个脑区，在固定的时刻，固定的脑区，还有不同的测量维度之类的，我感觉大脑真的是很难估量的。
2
现在我看懂这个代码是什么意思了
final_fc = data["timeseires"]
final_pearson = data["corr"]
labels = data["label"]
npy文件分为五部分数据嘛，只用到了其中的三部分，然后分别赋值给final_fc，final_pearson,labels
3
_, _, timeseries = final_fc.shape
_, node_size, node_feature_size = final_pearson.shape
这里就是要把1009个时间点赋值给timeseries，node_size为111，代表111个脑区维度，100代表不同的维度啊之类的
4