From c135ddb856a6ea0ed4665618e7fcf8f8a5f8561f Mon Sep 17 00:00:00 2001
From: ltcptgeneral <learthurgo@gmail.com>
Date: Mon, 30 Sep 2019 13:37:19 -0500
Subject: [PATCH] analysis.py v 1.1.1.001

---
 .../__pycache__/analysis.cpython-37.pyc       | Bin 6879 -> 8414 bytes
 .../__pycache__/regression.cpython-37.pyc     | Bin 0 -> 6985 bytes
 data analysis/analysis/analysis.py            |  28 ++++++++++--------
 3 files changed, 16 insertions(+), 12 deletions(-)
 create mode 100644 data analysis/analysis/__pycache__/regression.cpython-37.pyc

diff --git a/data analysis/analysis/__pycache__/analysis.cpython-37.pyc b/data analysis/analysis/__pycache__/analysis.cpython-37.pyc
index 312a3c477055eb5751e888931443aebd26980fba..c23175fb52f7da346d7986a31ae2456c9d76ccff 100644
GIT binary patch
delta 3493
zcmbtW-ESP#6`${&o%MR{_2)WqvPr;XLLkQZpt4ZDNC<)uR2=be`B;X%*PC(XV>vS$
z#qLHUfq81FqRu~HOQ0gP>I*znl|bSJUXUvFg=!vJwLFD?0Evf|bMJb`YbS0=JEOUC
z=XcLJzjN=obI$$y>Azh_E|$xdgxAaqzj@=E$0X@n{?WgR*g20+^lMp?s6-?x6PYSR
zp(@p=P7PY1CM{BnmRd!k(ji**G@^;OPIP<^d%ABBLzV<qAf|X1NfF-_&+?NJDT%=m
zZ^$o`vgnUuf0zt!DOE`2ibSh)Y+a&bY~($ajM8zcu1loK3_3w4KalAZ)z+o=WHOc`
z(^MD8c#hPlA&`k2nW2S5K$AJNkD3CR>LaYOGQCmTl<0msyR6a!^x(RTL^J(yTBk*!
zu`fpsQA;5EbL22B31n6vN6_I%=%WHj4&=}reM~?HbL1$s1yawEWAt%>9HQkV$v(bT
zeCkPl&(F_0=PHuayx{~b=7p_AD<4?2>Ucd-cPM4F4y*+%irg@0u%P7z2|H@<5O6$i
zKq#uS+fCMv>+R564MN{_JYjhrDWC3J?zE{BqwR(pxN(EBTW*tE6^^0YnZ6t_KfHx<
zorrnpLp_GodJ@H`>$4z^a)r~o6<og^qR(%8tR8maX6UnR;ZuF#-u&7OmP7EJpvme!
zi@|M1^`r3S!tuHvvEy~{qix4dqUtBVl`F%Vt*p(af2_<ZYX{SBDhJc=D=*GxMQB9t
zbR1wbQ&BlnQTZPKGp^Zkrmna#zO3!_OkW=9%#7hM@aBh$CwxX-=Y#|CiAxu*Ur7Ht
zaw@?PNgv@8DcG1Y@b4x*;=*4{;`p=pM3dOW(u%yHY)UXp<v^jz9jPl})YWTiMzcem
z#d)GO+Z|iY^f-j(jZEVx8FHhBbIWl(2iFT-G<pHyUd{G=gK;ztobF>zrC}q<RXP1-
z_4Dz30zDz_&+pSW#yn#m_F#Yd=P|ZhM#Yg&)91%8p5)55M)Y>*1~DAr8joT_^r#sE
z6YJ7zgae%Si}bVcH_!KMbMav=UJQ5yJ$H@SD(puj(;R3;nTqXhBRUR$mA*Dn8|H3s
zlVbT^Pv4#R`75!}#ebD*HECJFh10Ly<CC^!%uMkr?i8N*?r}Z40J#sD>ahS0BZ`U4
zU0%+#D1;b<n3_H@Ig{-6!9dx+;1dmDBWad0saVPgKJEx6)YdP?|8V6WYwPEl1wM=y
z&#ERq!qbn@a7(&VpmJBbY220JyI9)5{-%uSPwA>R4fqvq1O-tw*7$sqO*z&#aB7?K
zMj_T$OnetNC4nieU1(SD8a!ElVB9ggg$=7~QoSavl<;lTFs0*NcwfaHqAf*`nu7FS
zTckrh(gDuoKGL0z{122xPKk4eEz<IRqywBuZ>Ik@sU=8N_;8DK_&<>ja3=YZKm1JZ
z%E6y@$M>!RZ|#imz+!LBdw^a~c*Gx`nu@q0+1GnU_5oxM;7WeVChA30g`fSmhz`!|
z0b~#0O77dSKRJ>QPqv$@TUbgYyL1hBf+mLN2F!u$wn{8FX5hqFSR+#Q5DeU))j))i
zGO)M<LBqo$#v+k*;25gk`!`;gZo(E(+3hw_Be!K&fM*e6MXX;+iMe#;(sgj$fK;v`
zU*f<kto0&;fO&8ng<(SY60-m^6;ZGl@a!c#%nYa9W&tHu+W`j)B7=wwT@72g1u}H`
zc6<A4ddY41p-UYv$1Xs8qXTvkrUZ?f1no{7kph-t-hrbW*7)Q|sWGq$iU*_w`FSHE
zx*t+>d1r~qMz=+JM-ui=G#D*nBdHo@QA;tE=j1tTFJWsbbNDhW{5XcG)>57kVBX>g
z@54emJ6cQ6O&xx@%>6TAq0y^Cd7`(KtklNp?lwi#vO<jYh2eyDItV;%AHu>RingSt
z_0dUg_a$yOtNP3d`saiO0|;Tel-LTKL?JlEn_X6U0hk-G`pMJ_XSqc<&6@!y_cp=n
z$m0Vh&QG5?B@~a~fGq<SpZjR}(EPTJDAMWG>0>>`NA_0aL#GE6`FE~}TQ{bAF!B??
z;(#Lm=89*ytC@Psfjt!GaQxhWBLB7zr(f0<UOj+aq6Z!SZ3jyIetWgIY-bS2Oeb)>
z)yR!{zf&Q)DgH?iOUzFh;-|^1R~@gzE{m{L(#4rWH;57hr0~+x<*T@%VS!8X4FEh|
zz*oF?uHn-cha2E-4Fp$&c^t9Kd<OS>$75&tF(Y~&8*NonHC>VUo&Ij7c0?YMO=Uzi
VwW?VySB+|^I%3MxXRG@5{UiBJAwmEE

delta 1944
zcmaJ>U1%Id9G}^*+xtrHF4xP~rP-uubumqoREvb-N2&-Gf*Q3tpkaG6xo*3+o6hV7
z6H*RLUVV@S!MFCw#|plR1s{A9eBK8|1)+%GiwJp8|1;NY&q&;5{_~&T|6~3iGxPn6
zKRnZ0Etd-jzWbl|-a}1<ewK@)&jRKJ7~)Ng5Jd!0Ofc1mMs;dXlUg)K?Y2d9nx}=p
zAcmSvV!~Vu%#KAYj1-n5wwm)K4|6Ffbc&=1vR<B+VNoI_j9x)>hz=hhI?T%ZIvJuB
zsvjUS%q&`^wU00zp~eB)$E1=Wb!sZ4njxpCrI1>NG-ysCBN;MEZH3gC)qCW6jE*bl
zRHigRCl%7jkS3i{$f!b2LqOBiQOH<^%+OhdjAzIhI;W6{3^_~BDP)os)~AsB#BP3R
zVQFD;VR7-IcWwxw)(tOgvmk1(eDRys9MX25Ps-IPZQe=dtJ8QsxmvwiekfbC=HgVp
z>Bq@8)oy(l04~Yl<qJ2W4!hv-IC-n~biD}JCoqHt#Kv5%lfRmc-YzhE(9&BPe`#kf
zMt5<HZfbY+dx$>?xHgJnV-N4+Q4~VM@8NZX)gPPG*vB8@chT+bcTl(tx;pp>cM8(S
zF->zo(*{dRt|?kaZZEf%8m%t%4)F@nTU*`TW-ICj)Cr^5@r3Z(p(Djk#2qQE|5pm=
zwAf-Tf755wp={f4F^5J>C?D(tU&O9O%sAq$8^r3%c@@$oC0|_I@U}fa@U90e)kVxz
zVO)*m1dC%oY^S=%+p3~{QUN;-_W*qgLo|ROY-0`2VGCCjSCE{RoE`b<aW}X!lcoPG
z`Ew+&#(?AFiC<^Ob###2PkySeUX)#Q4PtIZei(B{TFe3=rf5Z+9S_L>(U-}!Q}4I?
zzNNan(j*^7yyFG_E^~E04oGT1MuOL+Ms{5K0h)k)4MW&KkY3OVT1{)oY~?4Va+@nK
z1WlTa*U23{w&dZU#yGlT?CE>PrUp$l*B(kE7sDG6?`$!igVjTsnw#TUe5oF=kWT{&
zPlms_nHnPEF;{A4@}M#C8X4{|FWjJv``cdZZ!=O6F*I-hIi7-itiFXIWW|wD&>E0L
zMJwPEE@^P!pCx&~iOrWc`gP30akaF##ToTmu^)vZ&BahB7K>v$q<Ov}Zg)M<{88>B
zn~HRP9;kHxhIF6SRNRl$HW=p-f1J0g@uxtE&r9;}4y1ZVuwMW_a(e;zWPj?_E7C1r
zkYpg>|M9~+;@1H`OVUsA^ra<b?ZC>#d=92#0jP8KJK#qFILUXX&-blnPPUSwKL)I3
zGAja?ik`5SJkEy$R<qLTl8i9bw>^G>)ftfgJzyo*ZZr9KdU<scSYn3V&UKH=BdqwX
zoqov{C6E)h0_O3Mt9r@{k`~_ff-ZYm^{AMPJ2O4iOfE~$W!?IBPo}6YAeS2sludO|
zp0?D!+=*y6V9!avVg-m18oFU>@ILf^%eLW-D8Z1lq*ru%sAkv7HLF&v)oW(|SJOX9
C6OI}H

diff --git a/data analysis/analysis/__pycache__/regression.cpython-37.pyc b/data analysis/analysis/__pycache__/regression.cpython-37.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..cfd0b73f501db7e02f505de40c76bcce59f105f6
GIT binary patch
literal 6985
zcmds+%X1t@9mnUjGqX?2ifqd^#FG$0HrSSJUP(wLa_l(ZI3lqOp_WPwqwUse*0Zzg
zo>^NKqXXqsPNYKR!lO8>a-;YQxbrV?=_6bSXHJ~>{(5E~dO0dcs(?{-Pj_!mzkZ+o
zb^m^PtzNHac+Q;t_`N?rscC<u%=oDxv5b)X9TC%78q-@kGg<~SnZ-)XW)5>%nN?Vo
z)mWW1I<=O`j<LDOYFXUsTHVsf-qFM|J02TsK5*C*%zR&Kxom-z@Gi3x%)-0EPBI(s
z>Kht6#hmvw<~%Z5HRKkVi=2yG9l6u2hFlG~26AUu8M!iY$B;YAs>oH5n?vqPtb$wx
zx#Jke{OE}tjeVIt$<D1?%wzTU_13~n`MGtIJ;fR*J;CNywdT`X<;xc@(ch&jL9A<9
zWs-fV;^Db)l`+n|USDu;5Qw11Q!WxO?R!zbbKc{dgMQ58Gz_Bi-k={njQc&LXFpvH
zJH39$g6KByyvaq(qpKp_7>e`W8gB=~gnN133kK)_ZAA~gm~)n>M$l0zf-v^d4ellF
zAPVD-7sSj<Hp2CkGW~Tg$-77@lhI!;Mjyr@jm;B0;z7dCdk?~N!$TuHzyKaMh4XnY
zdCT5iHSl~$7lKIefIal?^LE-7;TG2CIlkHEgVbB^i<xEcIy~kgNW*^IRLx!c;q|K*
z`hzs=(N~p9Q{OswGidkMyj%P}2IalnjZ|Ve>?OfP&W5kZ@_L}U2_jifgS*@hqR1aS
z5Q%KW!=6uz?n5TML{<`>`V8N<xjc4D&2h5ka(=%YWbIbJGp$j78%<5KwbwTX)7;|z
z&vUU!clC|ycSuCCdTp4b{hrEvihZQ5Y}xH03tqY;&Gul}243o+w}o&GA@LAp*lu?X
zQDep`(kbPQMeUKkpsivv0#kEE)KTwK`m(8u{3Z4H_nJ74qJLgr>TwnZOMyt2u3x=#
zbtx5BE?v2DAptun?>t;$K^i~;LG&;Qlch;Q@Q&a~0{L7VJXFno6U`+yB29O7%V-;v
z#6#~Af*SK{NMw33W_=YSPW22Wr;2eSr$5rqs?vNWXEB@9pgJmVoSK=HIELn$rZmGi
zRbvn)6&TVb`Q8s8>C&M8fD1L(Jeo%hRqaJ2rsLHo<E61W2x`1{kVv(z8Y9C9J~X$D
zR2vzYF*2bp%+uPAzHM%vL79~qJNk9)z1p^wnVI#Xwr$?K_%6O4l`=E6)eJ1i#@K<Z
znx&OH7^f~;bF{B~!3R;$<~?j;C9iyDjx1rDv^Qj}-S4f1vDy!bw4gW;C)f#6PeDkK
z3R>l6Nt)PMEfb8locxnSn*Cua^>FVJ&<C#|Bucd#bNakKr`Pmds1Yq{8C)bXEy$75
zSDL!C^WD<aTa}nUDE7|YJ_ywt)CYB@8EuNk4;3vVB+nt@8qN%@Wpp&04SLH$C^4KM
z(Dyc-B6vFtxHN4+;#rvpi&~k$^=)L7umD(Ukgzh%H>m_?(X_<Mv@U+>S{-2Jm((_~
zlsM{w-y4SD+V6;f2@mb%g0RIYq@^XGKUJ}G9`LZUkwOe>VE_d()$kCjmGEd?EYRn)
z$)?~**Y`=W`u+;a3;O1<^jnL2w9dtG#mQ6&?a5T1M+Q^SMwn`*j|{M=pVm4c5PX_o
zQR|j6eFvPX{bH<)=kW|als0hVfaVM``ONHIkoJa=_zFszrg)ZWnZ0yVmZN^$@oA2G
zr{X|kKL`Z7jk*a*TGJQxty6pFQPgU>H1`;%pzq6y@Avv_7*V?B`}c-Hl>Z{Wie}Ux
z1PxStjiU1uP1jMS)GnsgxQU<;hR#ucu5~bGS5WaF%(&fkM(z$>G;-i71#r~@I8td}
z5OCrtw7m~A@${sW#F4KAElkr;f;k8*wNgJpomr;N?mu}B#)_tRjXu7F=rF7@ZDghv
zl$qAG$7t=)o*cvF<Rhj1tjLm1al256i6)iL@Fj?0@pX!(!c}r5f%^zks?HoRlQCE|
zae>;nNYN5Sdl*tX*N9{36paX)<!7KpEAk_Rgs!(^v?}^xaJ__D2f+nBW+OP<?UqLs
zKvpdvs}+#d3&=<b9|GALemaU)+o$JB`$}*K<(f~a2gtDlMd3SB4Y(IbHy7P*Af*$=
zN+ZZJT`PU)ja0fgWpXuf6jHmGQ-1J3_vL?EPwkg+eoxxZENPp`5l#B%aAv#a0V+~y
zfACyN?U93T-6_utT4wH2nPu?}bRZ^Fi*HeuE;~6#gx3t)R=!1vZCO*gbG%q}m76Tn
z0n88ZQL%p)Ntb$uuE?1E#^LO%QLAC90jptNrm>PDJjxazBuSMfbBM-UfOIUjflWV5
z9y)}V5tRCP+%_oScMJH-1^gt+{}uRi8m8-d6xx4AYeUPaO=&+dYe0NKgr~5q5&i?h
zmF_fQ>>W(Fpc|L?4n+sx{BxAg;=HhH%438pM@IPZmjQxkNIf<X&Eonw!&*l3|Ie`g
zg4S>^tg{B~e>tr1tNc5#eoWJ*Q=e|J(7g-0ru=!sN)UZOJsuTSx)L3#S@V0VQ)pGU
z(5hs!t+Y+V^w5!6Lb!&I+(49(@k4gDo*9pfZ8J4+H6Sy|Dh#I*l_<k$Tte3eb^TL@
z(_^m<C(whkvJbjI%-m>;<8pzj(>JnAmb-Gw>{ThW{eOp0z-i+gQkK)nUCx*Nbdpdh
zsF^b)I@-3e`3&rG%w1XRC$~-5^RSy>x5F%ig|6$`y*XvO>&kvd2`nqJs&HXY78fjr
ziN%GB)~Jl%Dwzd~tF-I-P|X+MEc%r;{B#@!B%PP}Y)eTRwD0b@H_zZ}&)g=gc0WqQ
zIego4$mE7~cIfT-6Vy^ySIw|s)}KTG(<gSX)EMgW;Tl*OVM>;dSH8sBuf}~&m9J}u
ze*=J>-~NW4KfZdSO<I~B;+~X@63Rc2pu1`pNCLtV(?2w}9hkLNTEeos5Kf*dXSVn?
z#8=t;DI6-+BXYOg*{)*w9R?A7w5)5HGUo6mk+}G}2BFrnG7zt4l~E<DkE%Ox+Mq^#
zRNK+i7j=5S3-87|+HLK}+PxRv#rIjI14qr~?MY8JW_r5v89g1|_9xW##5o5~iV=xB
z&4$86mXwFHWx_G2%ou6F|KnkKWC^Z^4wrM}s3hlv&*NY%;!KvexajBp3Q`i{yHwi>
zQ{I#A8sH)m9S(_<51I{W=iXRpkylUJ9nl}g_%-S$aADw^+~b<aI()t&hL6}MFQ9~9
zBpr&_r+PdQuTmFfYQiTcrL5-d<QcoDM4(D*ATUG&-0ry_RPpN{5E2?QU5hICqswIj
zJ}6h^8hD@7U1M58`3XVf)tnIY7i2&Q)EOl1WM+|a!OI8Wr3_xMF$-QQ<RD?Fsd&*x
zB}IWYvdMwBUCpelnwj7Q)Zp$^Bj<qPr8X+7nmSo>yLfq4@lq~$8TWK!d)gE6GTW0g
z=?VY0Ku?Fa{mU6%40x%kVi}Xood#DarEd1GQHnNG?kyn*#C3{Zr-<}@vo7ACGzm|9
zkD~8Wbd#btDSC^d6^ed9(c2V}5XCzbk(ISqh&7b`2_bnNk#=MeHc!f`@88EiL*PAz
z2dM7*?G2(E_srZm6b#dizChaXmCxSKcZa2DtA&Xl+z-Nt_|MmY{FM1=qg1O!Svu?l
z>3u<z2vR(Pj$J{zN{}TaowrtBzXhZP(I{v(Gc#1c&Qmq_<=k2I3VoiufXJ#_rezuW
dL4UUG8Vi<N!k_C}jasALm~R}%-@@YQe*p&5kud-O

literal 0
HcmV?d00001

diff --git a/data analysis/analysis/analysis.py b/data analysis/analysis/analysis.py
index 17c2cd55..1ed658ee 100644
--- a/data analysis/analysis/analysis.py	
+++ b/data analysis/analysis/analysis.py	
@@ -7,10 +7,12 @@
 #   current benchmark of optimization: 1.33 times faster
 # setup:
 
-__version__ = "1.1.1.000"
+__version__ = "1.1.1.001"
 
 # changelog should be viewed using print(analysis.__changelog__)
 __changelog__ = """changelog:
+1.1.1.001:
+    - regression_engine() bug fixes, now actaully regresses
 1.1.1.000:
     - added regression_engine()
     - added all regressions except polynomial
@@ -157,7 +159,7 @@ import numba
 from numba import jit
 import numpy as np
 import math
-import regression
+from analysis import regression
 from sklearn import metrics
 from sklearn import preprocessing
 import torch
@@ -224,7 +226,7 @@ def histo_analysis(hist_data):
     return basic_stats(derivative)[0], basic_stats(derivative)[3]
 
 @jit(forceobj=True)
-def regression_engine(device, inputs, outputs, loss = torch.nn.MSELoss(), _iterations = 10000, lr = 0.1, *args):
+def regression_engine(device, inputs, outputs, args, loss = torch.nn.MSELoss(), _iterations = 10000, lr = 0.01):
 
     regressions = []
 
@@ -234,18 +236,18 @@ def regression_engine(device, inputs, outputs, loss = torch.nn.MSELoss(), _itera
 
         if 'linear' in args:
 
-            model = regression.SGDTrain(regression.LinearRegKernel(len(inputs)), torch.tensor(inputs).to(torch.float).cuda(), torch.tensor(outputs).to(torch.float).cuda(), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            model = regression.SGDTrain(regression.LinearRegKernel(len(inputs)), torch.tensor(inputs).to(torch.float).cuda(), torch.tensor([outputs]).to(torch.float).cuda(), iterations=_iterations, learning_rate=lr, return_losses=True)
+            regressions.append([model[0].parameters, model[1][::-1][0]])
 
         if 'log' in args:
 
             model = regression.SGDTrain(regression.LogRegKernel(len(inputs)), torch.tensor(inputs).to(torch.float).cuda(), torch.tensor(outputs).to(torch.float).cuda(), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            regressions.append([model[0].parameters, model[1][::-1][0]])
 
         if 'exp' in args:
 
             model = regression.SGDTrain(regression.ExpRegKernel(len(inputs)), torch.tensor(inputs).to(torch.float).cuda(), torch.tensor(outputs).to(torch.float).cuda(), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            regressions.append([model[0].parameters, model[1][::-1][0]])
 
         #if 'poly' in args:
 
@@ -254,7 +256,7 @@ def regression_engine(device, inputs, outputs, loss = torch.nn.MSELoss(), _itera
         if 'sig' in args:
 
             model = regression.SGDTrain(regression.SigmoidalRegKernelArthur(len(inputs)), torch.tensor(inputs).to(torch.float).cuda(), torch.tensor(outputs).to(torch.float).cuda(), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            regressions.append([model[0].parameters, model[1][::-1][0]])
 
     else:
 
@@ -263,17 +265,17 @@ def regression_engine(device, inputs, outputs, loss = torch.nn.MSELoss(), _itera
         if 'linear' in args:
 
             model = regression.SGDTrain(regression.LinearRegKernel(len(inputs)), torch.tensor(inputs).to(torch.float), torch.tensor(outputs).to(torch.float), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            regressions.append([model[0].parameters, model[1][::-1][0]])
 
         if 'log' in args:
 
             model = regression.SGDTrain(regression.LogRegKernel(len(inputs)), torch.tensor(inputs).to(torch.float), torch.tensor(outputs).to(torch.float), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            regressions.append([model[0].parameters, model[1][::-1][0]])
 
         if 'exp' in args:
 
             model = regression.SGDTrain(regression.ExpRegKernel(len(inputs)), torch.tensor(inputs).to(torch.float), torch.tensor(outputs).to(torch.float), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            regressions.append([model[0].parameters, model[1][::-1][0]])
 
         #if 'poly' in args:
 
@@ -282,7 +284,9 @@ def regression_engine(device, inputs, outputs, loss = torch.nn.MSELoss(), _itera
         if 'sig' in args:
 
             model = regression.SGDTrain(regression.SigmoidalRegKernelArthur(len(inputs)), torch.tensor(inputs).to(torch.float), torch.tensor(outputs).to(torch.float), iterations=_iterations, learning_rate=lr, return_losses=True)
-            regressions.append([model[0].parameter, model[1][::-1][0]])
+            regressions.append([model[0].parameters, model[1][::-1][0]])
+
+    return regressions
 
 @jit(forceobj=True)
 def r_squared(predictions, targets):  # assumes equal size inputs