为什么某些数组中的值没有标准化？

Question

我有一个数组，它的某些值需要从0规格化为1。当我尝试用minmax\u scale函数时，它返回相同的集合。你知道吗

[2.80756379e-01 9.47215085e-01 2.98665545e-01 2.71729701e-01 9.53844447e-01 4.09155122e-01 7.73782687e-01 4.04866838e-01...]

如果我将其切片为200个值（集合有400个值），函数将返回标准化值：

[0.28024599 0.94726674 0.29817026 0.2712117 0.95390169 0.40875301 0.77368808 0.40446111 0.0427401 0.98420903...]

例如，如果我生成600个测试值，它就可以正常工作。你知道吗

[0. 0.00166945 0.0033389 0.00500835 0.0066778 … 0.9933222 0.99499165 0.9966611 0.99833055 1.]

所以我假设我的输入有问题，但这不会引起任何错误，我可以理解发生了什么，为什么值没有被标准化。你知道吗

所以如果有人能解释一下我做错了什么。你知道吗

import numpy as np
from sklearn.preprocessing import minmax_scale

x = [3.4983787694629807, 6.8238036546277545, 3.5877400241270765, 3.4533384061573176, 6.856882141805892, 4.139049184316268, 5.958429090197188, 4.11765195576695, 2.314295679864955, 7.007979076325537, 4.591575092528983, 6.500165465269498, 2.8914665615540556, 3.543340276730918, 6.561891009978831, 3.6289541084867656, 3.5071777128342223, 7.056680192241884, 3.4788374494994114, 3.4464924916748862, 7.014214072992434, 3.4916155718006134, 3.5360046270365486, 7.086001703906676, 3.612050358553576, 3.896165015866581, 6.09922605707944, 3.3460547945703287, 3.910944920634269, 5.971772677577327, 5.178500485785331, 6.6090028576236675, 2.841807831193129, 5.149220347017544, 4.735677896132761, 3.8354921976610226, 7.038466835269509, 3.587747218864995, 3.555693227090981, 6.954475444783193, 3.437365707196104, 3.4325883202929894, 7.020900371706621, 3.4368379490007386, 3.475833945795732, 6.8688992156695585, 3.583554218078379, 3.520770727926182, 6.92133265995129, 4.044292336171232, 5.602564848428526, 4.411557629422499, 2.45924872382912, 7.086402867274765, 4.7372741724391965, 6.374350060657951, 2.6542178607167273, 3.508325152205329, 6.461025102348145, 3.6812426559272033, 3.5307800915694383, 7.066468458976694, 3.493276512558132, 3.4568759587746243, 7.0176506092750595, 3.478361485227409, 3.5201937331883713, 7.080979349338214, 3.5795505246747212, 3.8031183328096256, 6.245609213027018, 3.4165971380697018, 2.1723738636075023, 6.126159977852875, 5.013068437451677, 6.904030395006403, 2.7075477902159992, 4.895577020527206, 5.020958429692195, 3.9029736137362567, 7.005502917949908, 3.5789396294713756, 3.5676476014611818, 6.929131913377069, 3.4472779998622487, 3.4308686427915935, 7.018771879565203, 3.432009619429056, 3.457840956435269, 6.905877831610167, 3.5755167560389163, 3.5618219034900966, 6.97358473363722, 3.9600759412714175, 5.260824000465443, 4.692774833624142, 2.602129482124068, 7.035148649249085, 4.8915972720677825, 6.2369642563916825, 2.3878063740977975, 3.4610250988299063, 6.344643530169994, 3.745077609377894, 3.557344813692928, 7.0753891572282885, 3.5084174545346944, 3.4687392098349816, 7.018578929608326, 3.4657417965161157, 3.504674479158238, 7.073344198878297, 3.5505846727930717, 3.728275930579232, 6.374512212337401, 3.473738663253011, 2.455000320545692, 6.271460848569801, 4.853259564210921, 7.060300163815499, 2.5675976179261544, 4.625519502757723, 5.34193581659188, 3.9796698149770413, 6.961928522596542, 3.553928370489732, 3.577760192959362, 6.897549440276217, 3.4618068657264773, 3.4328376059178547, 7.019069392542759, 3.430955309222881, 3.4444356247624213, 6.935848551405397, 3.5648686589574554, 3.5823723667040626, 7.0144769584373465, 3.885706919437764, 4.947869760566333, 4.958988699542243, 2.740882010237675, 6.844670575188505, 5.052957262016582, 6.08929327319099, 2.1012153486394873, 3.4006432713737014, 6.211570326896137, 3.823951658625252, 3.5871290770943087, 7.082486008656476, 3.524032336877465, 3.481547522774735, 7.017048732257519, 3.454184925027273, 3.4896816808661573, 7.064127863850646, 3.5247589190924855, 3.6675470038864804, 6.486998043891378, 3.5179550925822936, 2.7150317741174828, 6.406130756762303, 4.700905963998837, 7.0786411289315705, 2.424024165788486, 4.340976405808333, 5.68871874308043, 4.066399458564197, 6.906772355480829, 3.5069849107646553, 3.584982183669687, 6.8586979941775175, 3.4809101839812273, 3.4385861350276734, 7.021776708035995, 3.433539453335249, 3.435646282061486, 6.959814531248965, 3.5525938762287192, 3.5880328691774226, 7.04501753482764, 3.8203341873417145, 4.672129393281222, 5.208396760624669, 2.873447040265478, 6.517042667648001, 5.219450800175008, 5.932960908282111, 4.071807320428156, 3.32681370883367, 6.060786017704781, 3.922140073470534, 3.6205701108946693, 7.086704780969265, 3.5398647239459082, 3.4949072952266786, 7.012978275363125, 3.4442572447205126, 3.4754567104312315, 7.054260858195003, 3.501860081901249, 3.6176350355084987, 6.584272251688416, 3.550139011669634, 2.9441212137915196, 6.528971419532436, 4.557509169353122, 6.971230117604573, 2.2788693414108083, 4.044384223064448, 6.045075381817579, 4.16380702973696, 6.8392444953122915, 3.4322587777495, 3.588015280869961, 6.811444662244569, 3.504571583212776, 3.448038447176144, 7.0267448030100095, 3.4394438861946495, 3.4313813610293606, 6.978663067682745, 3.539410445197603, 3.5836697468529928, 7.066336743389743, 3.763013423804524, 4.436678180031973, 5.439712464353549, 2.9978564390796802, 6.06598407921124, 5.38892971887227, 5.769885855777033, 4.7305353792164615, 3.2396389326858333, 5.891487403995336, 4.044576070907139, 3.658273834104114, 7.086908629948878, 3.5554434237079002, 3.508553789136474, 7.006154398318428, 3.4366257440898904, 3.462343969119664, 7.044561860711188, 3.4818481341401735, 3.576035272214573, 6.667629115969775, 3.5715246418069957, 3.1369130300148993, 6.639150212621939, 4.424195703276443, 6.759495560732593, 2.134069612887043, 3.7386525746914123, 6.388620586154337, 4.272297018941626, 6.7587843489580965, 3.324494111518356, 3.5853639214122595, 6.754568772053236, 3.5329516319997514, 3.4610002607293953, 7.033692354375017, 3.4481972797191154, 3.4313661148467935, 6.993158859550646, 3.5257969569299723, 3.5731476929367916, 7.07964197799761, 3.712767027132367, 4.2405670182736745, 5.652147101651347, 3.112331986409608, 5.515185072672031, 5.559088312172956, 5.602219157421699, 5.362655122965983, 3.1396940731308915, 5.703149958832117, 4.196548195898355, 3.700992351744346, 7.081897811873345, 3.569845801358458, 3.5223021303817146, 6.996232651966579, 3.4319885789163944, 3.450813363928631, 7.035729967483905, 3.464829989158296, 3.5409776462526583, 6.738403993391335, 3.583589535854453, 3.2912893051822354, 6.7362012208338795, 4.301697490919809, 6.470822564075378, 3.659773459738205, 2.0974903963829354, 6.692161174433521, 4.3919729995424195, 6.665106087684514, 3.1798600950044777, 3.5754169970262786, 6.686782075167818, 3.566547285071714, 3.47722596516465, 7.042206860585631, 3.459238171506453, 3.435134332592052, 7.003939734793729, 3.512049727586488, 3.559270908983708, 7.086176708003556, 3.6686375261134057, 4.080241002256502, 5.84537422915953, 3.215380734121881, 4.89528857023289, 5.727559855711503, 5.432265203060416, 5.934141424807943, 3.027998394822877, 5.495590110107973, 4.3831413048066254, 3.7495930218165427, 7.070434254364885, 3.5814467824887473, 3.535972176884382, 6.982737736524654, 3.43098967364103, 3.441422724579513, 7.028339316715845, 3.451013794888052, 3.5113341674962717, 6.797932388511367, 3.587942414572208, 3.4081496049032274, 6.82001018784346, 4.190357260509626, 6.134736264429092, 3.967315538885193, 2.2419098299323275, 6.926226800256952, 4.522587633421751, 6.558238517691441, 2.9967711785804605, 3.556550927861764, 6.606754741200447, 3.6063414285217674, 3.496491069547794, 7.05174661557621, 3.471997346906242, 3.4420693538840723, 7.011514159064076, 3.4983614083674994, 3.5438804663925714, 7.0871842215570755, 3.629734022986171, 3.9507935039967084, 6.019481038594121, 3.3058784855515917, 4.240129283449884, 5.892015134812656, 5.262390436269602, 6.413168227114006, 2.905958254165465, 5.269023998291438, 4.608362850137262, 3.8050205047007704, 7.051271296451422, 3.5877015700240324, 3.5493020337345564, 6.965064567979088, 3.434139087088657, 3.4347436167612146, 7.022835521591056, 3.4406465746514856, 3.4864976824344196, 6.847515844073905, 3.58620796770146, 3.4908273530643696, 6.890786354350282, 4.090156575701794, 5.77929297599503, 4.2665273948822655, 2.387152222353216, 7.062521670699347, 4.663509517553511, 6.438556051230168, 2.776539669077205, 3.52724382177749, 6.513147765418067, 3.6539307251684234, 3.5186578165204065, 7.061643896805739, 3.4859770491431257, 3.4514785509234778, 7.016259943745782, 3.4849078166839633, 3.528054396333257]
xa = np.array(x)
q = minmax_scale(xa)
# Not normalized
print(q)

d = minmax_scale(xa[:200])
# Normalized
print(d)

za = np.arange(600)
z = minmax_scale(za)
# Normalized
print(z)

Answer 1

你可能做了一些错误的假设，我可以告诉你：

>> import numpy as np
>> from sklearn.preprocessing import minmax_scale

>> x = [3.4983787694629807, 6.8238036546277545, 3.5877400241270765, 3.4533384061573176, 6.856882141805892, 4.139049184316268, 5.958429090197188, 4.11765195576695, 2.314295679864955, 7.007979076325537, 4.591575092528983, 6.500165465269498, 2.8914665615540556, 3.543340276730918, 6.561891009978831, 3.6289541084867656, 3.5071777128342223, 7.056680192241884, 3.4788374494994114, 3.4464924916748862, 7.014214072992434, 3.4916155718006134, 3.5360046270365486, 7.086001703906676, 3.612050358553576, 3.896165015866581, 6.09922605707944, 3.3460547945703287, 3.910944920634269, 5.971772677577327, 5.178500485785331, 6.6090028576236675, 2.841807831193129, 5.149220347017544, 4.735677896132761, 3.8354921976610226, 7.038466835269509, 3.587747218864995, 3.555693227090981, 6.954475444783193, 3.437365707196104, 3.4325883202929894, 7.020900371706621, 3.4368379490007386, 3.475833945795732, 6.8688992156695585, 3.583554218078379, 3.520770727926182, 6.92133265995129, 4.044292336171232, 5.602564848428526, 4.411557629422499, 2.45924872382912, 7.086402867274765, 4.7372741724391965, 6.374350060657951, 2.6542178607167273, 3.508325152205329, 6.461025102348145, 3.6812426559272033, 3.5307800915694383, 7.066468458976694, 3.493276512558132, 3.4568759587746243, 7.0176506092750595, 3.478361485227409, 3.5201937331883713, 7.080979349338214, 3.5795505246747212, 3.8031183328096256, 6.245609213027018, 3.4165971380697018, 2.1723738636075023, 6.126159977852875, 5.013068437451677, 6.904030395006403, 2.7075477902159992, 4.895577020527206, 5.020958429692195, 3.9029736137362567, 7.005502917949908, 3.5789396294713756, 3.5676476014611818, 6.929131913377069, 3.4472779998622487, 3.4308686427915935, 7.018771879565203, 3.432009619429056, 3.457840956435269, 6.905877831610167, 3.5755167560389163, 3.5618219034900966, 6.97358473363722, 3.9600759412714175, 5.260824000465443, 4.692774833624142, 2.602129482124068, 7.035148649249085, 4.8915972720677825, 6.2369642563916825, 2.3878063740977975, 3.4610250988299063, 6.344643530169994, 3.745077609377894, 3.557344813692928, 7.0753891572282885, 3.5084174545346944, 3.4687392098349816, 7.018578929608326, 3.4657417965161157, 3.504674479158238, 7.073344198878297, 3.5505846727930717, 3.728275930579232, 6.374512212337401, 3.473738663253011, 2.455000320545692, 6.271460848569801, 4.853259564210921, 7.060300163815499, 2.5675976179261544, 4.625519502757723, 5.34193581659188, 3.9796698149770413, 6.961928522596542, 3.553928370489732, 3.577760192959362, 6.897549440276217, 3.4618068657264773, 3.4328376059178547, 7.019069392542759, 3.430955309222881, 3.4444356247624213, 6.935848551405397, 3.5648686589574554, 3.5823723667040626, 7.0144769584373465, 3.885706919437764, 4.947869760566333, 4.958988699542243, 2.740882010237675, 6.844670575188505, 5.052957262016582, 6.08929327319099, 2.1012153486394873, 3.4006432713737014, 6.211570326896137, 3.823951658625252, 3.5871290770943087, 7.082486008656476, 3.524032336877465, 3.481547522774735, 7.017048732257519, 3.454184925027273, 3.4896816808661573, 7.064127863850646, 3.5247589190924855, 3.6675470038864804, 6.486998043891378, 3.5179550925822936, 2.7150317741174828, 6.406130756762303, 4.700905963998837, 7.0786411289315705, 2.424024165788486, 4.340976405808333, 5.68871874308043, 4.066399458564197, 6.906772355480829, 3.5069849107646553, 3.584982183669687, 6.8586979941775175, 3.4809101839812273, 3.4385861350276734, 7.021776708035995, 3.433539453335249, 3.435646282061486, 6.959814531248965, 3.5525938762287192, 3.5880328691774226, 7.04501753482764, 3.8203341873417145, 4.672129393281222, 5.208396760624669, 2.873447040265478, 6.517042667648001, 5.219450800175008, 5.932960908282111, 4.071807320428156, 3.32681370883367, 6.060786017704781, 3.922140073470534, 3.6205701108946693, 7.086704780969265, 3.5398647239459082, 3.4949072952266786, 7.012978275363125, 3.4442572447205126, 3.4754567104312315, 7.054260858195003, 3.501860081901249, 3.6176350355084987, 6.584272251688416, 3.550139011669634, 2.9441212137915196, 6.528971419532436, 4.557509169353122, 6.971230117604573, 2.2788693414108083, 4.044384223064448, 6.045075381817579, 4.16380702973696, 6.8392444953122915, 3.4322587777495, 3.588015280869961, 6.811444662244569, 3.504571583212776, 3.448038447176144, 7.0267448030100095, 3.4394438861946495, 3.4313813610293606, 6.978663067682745, 3.539410445197603, 3.5836697468529928, 7.066336743389743, 3.763013423804524, 4.436678180031973, 5.439712464353549, 2.9978564390796802, 6.06598407921124, 5.38892971887227, 5.769885855777033, 4.7305353792164615, 3.2396389326858333, 5.891487403995336, 4.044576070907139, 3.658273834104114, 7.086908629948878, 3.5554434237079002, 3.508553789136474, 7.006154398318428, 3.4366257440898904, 3.462343969119664, 7.044561860711188, 3.4818481341401735, 3.576035272214573, 6.667629115969775, 3.5715246418069957, 3.1369130300148993, 6.639150212621939, 4.424195703276443, 6.759495560732593, 2.134069612887043, 3.7386525746914123, 6.388620586154337, 4.272297018941626, 6.7587843489580965, 3.324494111518356, 3.5853639214122595, 6.754568772053236, 3.5329516319997514, 3.4610002607293953, 7.033692354375017, 3.4481972797191154, 3.4313661148467935, 6.993158859550646, 3.5257969569299723, 3.5731476929367916, 7.07964197799761, 3.712767027132367, 4.2405670182736745, 5.652147101651347, 3.112331986409608, 5.515185072672031, 5.559088312172956, 5.602219157421699, 5.362655122965983, 3.1396940731308915, 5.703149958832117, 4.196548195898355, 3.700992351744346, 7.081897811873345, 3.569845801358458, 3.5223021303817146, 6.996232651966579, 3.4319885789163944, 3.450813363928631, 7.035729967483905, 3.464829989158296, 3.5409776462526583, 6.738403993391335, 3.583589535854453, 3.2912893051822354, 6.7362012208338795, 4.301697490919809, 6.470822564075378, 3.659773459738205, 2.0974903963829354, 6.692161174433521, 4.3919729995424195, 6.665106087684514, 3.1798600950044777, 3.5754169970262786, 6.686782075167818, 3.566547285071714, 3.47722596516465, 7.042206860585631, 3.459238171506453, 3.435134332592052, 7.003939734793729, 3.512049727586488, 3.559270908983708, 7.086176708003556, 3.6686375261134057, 4.080241002256502, 5.84537422915953, 3.215380734121881, 4.89528857023289, 5.727559855711503, 5.432265203060416, 5.934141424807943, 3.027998394822877, 5.495590110107973, 4.3831413048066254, 3.7495930218165427, 7.070434254364885, 3.5814467824887473, 3.535972176884382, 6.982737736524654, 3.43098967364103, 3.441422724579513, 7.028339316715845, 3.451013794888052, 3.5113341674962717, 6.797932388511367, 3.587942414572208, 3.4081496049032274, 6.82001018784346, 4.190357260509626, 6.134736264429092, 3.967315538885193, 2.2419098299323275, 6.926226800256952, 4.522587633421751, 6.558238517691441, 2.9967711785804605, 3.556550927861764, 6.606754741200447, 3.6063414285217674, 3.496491069547794, 7.05174661557621, 3.471997346906242, 3.4420693538840723, 7.011514159064076, 3.4983614083674994, 3.5438804663925714, 7.0871842215570755, 3.629734022986171, 3.9507935039967084, 6.019481038594121, 3.3058784855515917, 4.240129283449884, 5.892015134812656, 5.262390436269602, 6.413168227114006, 2.905958254165465, 5.269023998291438, 4.608362850137262, 3.8050205047007704, 7.051271296451422, 3.5877015700240324, 3.5493020337345564, 6.965064567979088, 3.434139087088657, 3.4347436167612146, 7.022835521591056, 3.4406465746514856, 3.4864976824344196, 6.847515844073905, 3.58620796770146, 3.4908273530643696, 6.890786354350282, 4.090156575701794, 5.77929297599503, 4.2665273948822655, 2.387152222353216, 7.062521670699347, 4.663509517553511, 6.438556051230168, 2.776539669077205, 3.52724382177749, 6.513147765418067, 3.6539307251684234, 3.5186578165204065, 7.061643896805739, 3.4859770491431257, 3.4514785509234778, 7.016259943745782, 3.4849078166839633, 3.528054396333257]
>> xa = np.array(x)
>> q = minmax_scale(xa)

>> x_set = set(x)
>> q_set = set(q)
>> x_set.intersection(q_set) == set()
True

正如你所见，没有交叉点，它肯定不会返回相同的集合。。。你知道吗