From 5af12ffde4f438e69587a1db7ef522db35f91079 Mon Sep 17 00:00:00 2001
From: mhjensen <morten.hjorth-jensen@fys.uio.no>
Date: Fri, 19 Oct 2018 06:47:47 +0200
Subject: [PATCH] perhaps last typo...grrr

---
 doc/pub/NeuralNet/html/._NeuralNet-bs033.html |   2 +-
 doc/pub/NeuralNet/html/._NeuralNet-bs095.html |   2 +-
 doc/pub/NeuralNet/html/NeuralNet-reveal.html  |   4 ++--
 .../NeuralNet/html/NeuralNet-solarized.html   |   4 ++--
 doc/pub/NeuralNet/html/NeuralNet.html         |   4 ++--
 doc/pub/NeuralNet/ipynb/NeuralNet.ipynb       |   4 ++--
 .../ipynb/ipynb-NeuralNet-src.tar.gz          | Bin 88051 -> 88051 bytes
 doc/pub/NeuralNet/pdf/NeuralNet-minted.pdf    | Bin 626143 -> 626150 bytes
 doc/src/NeuralNet/NeuralNet.do.txt            |   4 ++--
 9 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/doc/pub/NeuralNet/html/._NeuralNet-bs033.html b/doc/pub/NeuralNet/html/._NeuralNet-bs033.html
index ab60941fc..4b6741d79 100644
--- a/doc/pub/NeuralNet/html/._NeuralNet-bs033.html
+++ b/doc/pub/NeuralNet/html/._NeuralNet-bs033.html
@@ -390,7 +390,7 @@ $$
 where the superscript \( l-1 \) indicates that these are the outputs from layer \( l-1 \).
 Our cost function at the final layer \( l=L \) is now
 $$
-\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(i-t_i)\log{(1-a_i^L)}\right),
+\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(1-t_i)\log{(1-a_i^L)}\right),
 $$
 
 where we have defined the targets \( t_i \). The derivatives of the cost function with respect to the output \( a_i^L \) are then easily calculated and we get
diff --git a/doc/pub/NeuralNet/html/._NeuralNet-bs095.html b/doc/pub/NeuralNet/html/._NeuralNet-bs095.html
index e69a3c7b9..d2f9e0e3c 100644
--- a/doc/pub/NeuralNet/html/._NeuralNet-bs095.html
+++ b/doc/pub/NeuralNet/html/._NeuralNet-bs095.html
@@ -369,7 +369,7 @@ c(x, P) = \sum_i  \big(g_t'(x_i, P) - ( -\gamma g_t(x_i, P) \big)^2
 $$
 
 <p>
-In order to minimize it, an optimalization method must be chosen.
+In order to minimize it, an optimization method must be chosen.
 
 <p>
 Here, gradient descent with a constant step size has been chosen.
diff --git a/doc/pub/NeuralNet/html/NeuralNet-reveal.html b/doc/pub/NeuralNet/html/NeuralNet-reveal.html
index 9ec93ee5b..30803e519 100644
--- a/doc/pub/NeuralNet/html/NeuralNet-reveal.html
+++ b/doc/pub/NeuralNet/html/NeuralNet-reveal.html
@@ -1367,7 +1367,7 @@ where the superscript \( l-1 \) indicates that these are the outputs from layer
 Our cost function at the final layer \( l=L \) is now
 <p>&nbsp;<br>
 $$
-\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(i-t_i)\log{(1-a_i^L)}\right),
+\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(1-t_i)\log{(1-a_i^L)}\right),
 $$
 <p>&nbsp;<br>
 
@@ -4048,7 +4048,7 @@ $$
 <p>&nbsp;<br>
 
 <p>
-In order to minimize it, an optimalization method must be chosen.
+In order to minimize it, an optimization method must be chosen.
 
 <p>
 Here, gradient descent with a constant step size has been chosen.
diff --git a/doc/pub/NeuralNet/html/NeuralNet-solarized.html b/doc/pub/NeuralNet/html/NeuralNet-solarized.html
index f5c9407fe..95c8c283e 100644
--- a/doc/pub/NeuralNet/html/NeuralNet-solarized.html
+++ b/doc/pub/NeuralNet/html/NeuralNet-solarized.html
@@ -1362,7 +1362,7 @@ $$
 where the superscript \( l-1 \) indicates that these are the outputs from layer \( l-1 \).
 Our cost function at the final layer \( l=L \) is now
 $$
-\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(i-t_i)\log{(1-a_i^L)}\right),
+\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(1-t_i)\log{(1-a_i^L)}\right),
 $$
 
 where we have defined the targets \( t_i \). The derivatives of the cost function with respect to the output \( a_i^L \) are then easily calculated and we get
@@ -3909,7 +3909,7 @@ c(x, P) = \sum_i  \big(g_t'(x_i, P) - ( -\gamma g_t(x_i, P) \big)^2
 $$
 
 <p>
-In order to minimize it, an optimalization method must be chosen.
+In order to minimize it, an optimization method must be chosen.
 
 <p>
 Here, gradient descent with a constant step size has been chosen.
diff --git a/doc/pub/NeuralNet/html/NeuralNet.html b/doc/pub/NeuralNet/html/NeuralNet.html
index 67f62dbb7..37461c5ce 100644
--- a/doc/pub/NeuralNet/html/NeuralNet.html
+++ b/doc/pub/NeuralNet/html/NeuralNet.html
@@ -1367,7 +1367,7 @@ $$
 where the superscript \( l-1 \) indicates that these are the outputs from layer \( l-1 \).
 Our cost function at the final layer \( l=L \) is now
 $$
-\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(i-t_i)\log{(1-a_i^L)}\right),
+\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(1-t_i)\log{(1-a_i^L)}\right),
 $$
 
 where we have defined the targets \( t_i \). The derivatives of the cost function with respect to the output \( a_i^L \) are then easily calculated and we get
@@ -3914,7 +3914,7 @@ c(x, P) = \sum_i  \big(g_t'(x_i, P) - ( -\gamma g_t(x_i, P) \big)^2
 $$
 
 <p>
-In order to minimize it, an optimalization method must be chosen.
+In order to minimize it, an optimization method must be chosen.
 
 <p>
 Here, gradient descent with a constant step size has been chosen.
diff --git a/doc/pub/NeuralNet/ipynb/NeuralNet.ipynb b/doc/pub/NeuralNet/ipynb/NeuralNet.ipynb
index 4a8251538..097b9083c 100644
--- a/doc/pub/NeuralNet/ipynb/NeuralNet.ipynb
+++ b/doc/pub/NeuralNet/ipynb/NeuralNet.ipynb
@@ -1549,7 +1549,7 @@
    "metadata": {},
    "source": [
     "$$\n",
-    "\\mathcal{C}(\\hat{W}) = - \\sum_{i=1}^n \\left(t_i\\log{a_i^L}+(i-t_i)\\log{(1-a_i^L)}\\right),\n",
+    "\\mathcal{C}(\\hat{W}) = - \\sum_{i=1}^n \\left(t_i\\log{a_i^L}+(1-t_i)\\log{(1-a_i^L)}\\right),\n",
     "$$"
    ]
   },
@@ -4352,7 +4352,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "In order to minimize it, an optimalization method must be chosen. \n",
+    "In order to minimize it, an optimization method must be chosen. \n",
     "\n",
     "Here, gradient descent with a constant step size has been chosen. \n",
     "\n",
diff --git a/doc/pub/NeuralNet/ipynb/ipynb-NeuralNet-src.tar.gz b/doc/pub/NeuralNet/ipynb/ipynb-NeuralNet-src.tar.gz
index e098aee2cea53b714650130aeb81215630aa51c3..8412488c556d384038cd86efbbc8594c8af5b866 100644
GIT binary patch
delta 20
ccmeyop7ry3RyO%=4u(^Sjci-l7~h5i09Yyqxc~qF

delta 20
ccmeyop7ry3RyO%=4u-1<jci-l7~h5i09b_w!vFvP

diff --git a/doc/pub/NeuralNet/pdf/NeuralNet-minted.pdf b/doc/pub/NeuralNet/pdf/NeuralNet-minted.pdf
index 108191e10b5a296cbe8727a651a9de1d78012191..cbe373bc08c855cd74a6caf56f6166bfee552590 100644
GIT binary patch
delta 6176
zcmV+*7~kjLn<nO)Ca_+o0y!{~@v<m??OR!M+_({b*RNonGOFFd;y&#1gOlv0YAdxl
zR#lsn?Nm`Sl!o>k+9AiX_t&SJ04ah3Iitgyq>>kLNP<9jqw)3E4Txsf8O{E9M#J@Y
z7iZsm$4usk(_Cj4R~aLe33x{{mN99GVLa2y5H3`9vB}=dZdc{voZH#zd@i_uo9&C`
zVNq26rrzPz&gxa+AFOxRWmUrm&Eu=Xc2SqRt$$Z;2aPYwZNC5Pe=pA2EUyZ3K9|DH
zF1r584>)$QE>|!VH#7WeXG`DH?&|#A#rGh15MKylw9T|sgj+9ty(%kg^v!qhq^Gqb
z5+2a(VhN3Y?^{H6F~kbdHkdDe)_H|G3o&bcmT#$$#oj+D!<+5y93;H;H`nFIM)Iw{
z&8rs)F$^Qx*laEsVOsmK{u0`WNEp)z`<5!|n+VoxB8WAzZSw~hl<L|0L|4oZYFWmK
z=FT_%XF{+!B^Jyv?=^c%IZyN_iBqyC8^dX`vsa<1%QSIDSZvBcOX)Iy#)(wIf3OXj
z$o|-ZI4W>#XT#Y4Nc09GQVX_&UO5Zx%Fs^tM+Hq>E8r`Z=!r36q{+A>oZ84E1n)J|
z@LJqyN@-#+Xv0n0hNYBalqjc8(M~F_9d$vdMPfifl<t(HsrhC^YKa#Of;8>kBqp3U
z^GzS>f>9rl3w};0C5g3vpgKqd7Vs0cNDSqOmLTz6t5@Eq7`oTHlw)ylEa;sv5e+UV
zx~IjbI5?GPR_F9!?7j{fM0Dta49w9qdyxQ#IhULWi%zQH8s_o!bzX+$e%q{l{4Hm7
z{-L`Q0sY|d77(wv&TGU8)4m1#oDVS1_An2>-}&pZKnU7~?{<fOfJKX4Rr~%s@MWzE
z5neEGcTI7L!*XLRa{FjD6tQ<M-OfB%Q@aL5#7Q6`o`4~-*oH7=6RR+GAW6F=4<=0o
z07(?H4|n({V(yo*){&){_F$$V*5hgfMO$R}v537?^aDEh)V{dh7vNA30Gp@|kJiPa
zKER`W!wJyucEJ^YbAP!ytk;ptmUZ_r1n_vc)_q0izQl)kLMb1x6eVEA^T1Lr+GTZ_
zH*DpD@8xE{yUB5pwIdckADX*1WEaIUEXN&wb{pivRkQx=Z*qUBHhHUv=R^7Sy7)SL
zoNtP@IqH&PB??i>cxi{05S=BO!8piuJ8qD<lydgdxfLycYUKOEe^##$OAeL4+ZXxr
zbBr2?YB<3ai0*%ZX^@0@>!1b{oP&Y01r*MM%`O`_Dqx<A-vrlVnQzF^wcs<3>?vT%
zx1aLO&ANDTK3A09aXzDZOlr~25Ilt`yIdEWgwKK#@SahH%cKVIa=`^&qkN9fa_q<f
zxcY(p!^XaUjt9NrZGIW}B>eFR409=|1*@%#kLOBZKk&Hr-+8Zw??Ajp|M$2K7U3-q
zBL@jQ`mfia$8`~!9m0DgyI|yU*)T}_uyt}K2VNr;CloYoG>ix{8ZXf0*MW{N%W45*
z!1`>wy5rm~#D@VvcYZy$iy7bIV%}q5s=oG5Zrk90gx!ab6Z+1BAn2hfhnEr=n)IUg
zQUPSl;aV5_O*H}qH~Ai16IvwsyRt-DcZY&ZCL9LhfaPtQyh<SWoPnz<fXX>{V14t@
zFbIi;lSj!kP#Hep(yhGF7#gww_QW_C5nBLA82KpSmJk5~;SN;u?}^?huw?#7z8#H+
z@;D@aZI<0dx)h;Q1er2G3sr8KRZ7n)Wh~YpYC=5kHJacB-8|nU9l<Be&&7menFstr
zfGoX6?~}~k=uY5IiLn$=wX}Wg)L1Xmi$K~h>h7G+zVeq5(?aI%An4O^tAV~!Csu}y
zkiltjY=VwE)jgwmEtocqdjYp{T^86E5kQB3OZyNR$2^gr;EC!{ivDkV;(gTeYq}3c
zfLbZ;ZHR_^DuS(-DF!YmB%@U04`Ktyx?L6f;(iLb%9nB0x16BqqVa{Qyx#1}rFY1N
zmqS=@gK)yxC}e}8VQxbFdf=zV*OqR`L!t_hiiQfBb?fP=zNt>}P4(|Y$(9sk9297O
zic}`7mEzNl$DfhO;Sj}@zR#PY5DvBw@dVMNVA_hlExtMiVkZxmT2_)s?fSz#Y!=1T
zEVfJj?Awd8pU+w%iwl%mlL@7WFeY1U&fdJE*%Dqtas>H@&2F2fo6HKrEIR!<`|<3r
zzRVsE+4d#DuUj$}2fSlbCk$~+o^DHjdCJtuXPY{C%GAln1c&1hH$@$(!D60}69~)|
zpq_J070<FZ&D}0j#t=j+6DP#KMoc~ekSYO?bPOQPF@Utk08;!{0pw`7;F1U>{?q8!
z|6}1}dD_EDpTN3ouT&vf#vK5>a=l6-*8=98&4x9IsHz(};Zy?HU@yie(3?kps5vGf
z2fz_V$0B-axNqd-eNKpCHwB<_d+qOXf7z7Va#Q{i%53-*sI}@g7O+XX?ciTu@AG9@
zZ1G9m4pS}K2dQ!ohZrb<LN*e%HQ*nL+j_U1|1G~+RlDtPLj|!nwlDAEuFSB!%MF-I
z26O&!SDn>8W~^rW;%1#Mid3S1*jdnl<p%vdC;$!TrW2J5zkeHbeOuNmFI!;edX_J6
zmq%$|T%C|rU|9zOQPkltNm<e3ASSV}nYC{0h`WZ+V2)dr?lc_)x44BEjTlP!%2%Ux
zyqB$J>m2VsdncUGMp6bdY2v=uN92*zg7Aas!k8SJba$w44t3Rfj~zOHqjE5f!5Q;v
zLP#cI-q8U4FCJeJ=*yyjW^oz!{Ax-R4WrqCD0y=Bxz_h*pQa{$R61iJ)6)--&RSR?
zCjQ{+$RJJ<EeGF>6?VFBEKK~sY|pX=#-!v_6J^9bDcLb-h8$p^TT?YRfB}HP00R@K
zBy6)u`TSariYpz=ACE_WyxaBS-NsH6JMcdDm;Nc9uu;X^(gSbn6y^2m91jFllXxJB
zp#zt@P3UaEG-;i=9_ajjT4~8Su(ZL7YQO~bE%U0F6j(dL=m1SFli0w4=AE`6M@2P|
z#m7UDdw~|d;?fW$0x<S>0apzyf9#(;SQW`=!y#M6tPSW?i)G^*Fw;9$(bGy9eT=Sh
z$ofP}R})guex?0cxp7mIk+30uQFVY7?V3XjQ&oqiFUWY9R!^@wOj$(={sgNSV~!{E
zz>dsRNtH8*QMx(PqFT>x_*iNYKx%?#@A?`2n0oyFDrnSmCoBc7Q%fP<5W}UQal2B>
z@%=YQ0Hluh-`=O!m7%IR)Yhssj<G4VxE`qWK3U|(-O`?3Qvd?#LvM_K!Ld3{lzzmi
zy7;slqzu+%bL02Q`=$|t)Y$ZW?P%F+9PCeo#rp|!&t@-2B7%t1Q&^oqkY#%uH|!9&
zmJGXDb-3}NdtWW~<xM@sJGdhdW8yy7$z3#I#gTJXeObB0^JFUSm>p4&D^4uu@kix$
zSuXP0pC>{s^SaZw3L58s_iWH61kKIWez)<zx?IH_$yI=?OZEI_k&wn1_dv5C6+o&M
zeMFL3Hzpa~`H(*+P676Ief(*rda!lW6c6%nvI)&)rcyhFyQ|&Nx@5t}iMtc2V{yN0
z@u$=|9}29iEA#YESKAiK28TEp9s^>JJ;b5$q#wW#n<`OpzgO3PYxYwLVE}xc_(DfG
z<Y`AZ-1HF+aoQ0Md%_V8=D?nEvWhAJ-3{V14bxI5#KLH8PCCJ1PCUWUNk7eNWzPv)
z05{_jSfC%a58%C{3duS%gSE*YQfD5sjy-8(L?1bdFeWGd`vE_t9?rjrSphO)jx@0Y
zud+K=0zp~w2sei!A3c|A!ff|1FaxmBc(fFp4|pza@OX-$qc4D-{|hDA(=CI>vWLd9
z0k_7o0`1NLHIwnOD3{!U0}g*i3@K5s6g@O(Hcb!BBJiO=S{Pi5wuQ)&p-1fG*LOxF
zWl7eiaqOg<gKa*B!+Cx~($rd8)OvYC%`b6%ixQuzuM$RwBX3^bJfGh@y=9DB5s8C<
zTl1C0LYHuE1%60e&$s4@_16Aj=J>8}|5{beR#vK$=DVg=&4c>6muY{#R!xzs-y)Yz
zl5S;QrA00>Jz19G!|b2=ADyTugV&RRT29+hyXVbz@m^#N?A$IQ92GEMtzLr>(93$;
z?)1xG7d@Eo&R}Dc5evIYKgz$UPY>|LaB?K=;VMENE|Rg7)kf@EG0)-S4ETKJA~xRq
ziSqK%YsOHY6_Mj`1ml0Qki?!h5s9i@Z$xcc@+3Ah@j;fVnG_~=^P=8#v04`ouRfd+
zriX=Ecun_TZZB=H-_s<KdCw%zM!x><ogQ*UlIXNnKV`AkxxeZ#pc+~_&6in|Xlvi6
zqWYh#d?6Ani+m|+nZpM&4L=uhL{k}7=P)|W!D05{N2jH+U)z7_jf+EiA-tZkkdEc;
zV`)FeJ}x#Ya36uYx1!!Q8M<b#r5c*_p?Yvl6{>W-Ez$&qFs8XGy6T*M-PA(>y&77)
z8mIoVvvU9OM8kKQ{F`1K_ztqaP=WZ!vR_Fxvgw2Ae?=We2BFmC9M#MQBRb#xhx5IZ
zbyMb_%s|g2vG0Gv)QL7vZ(U|F2GcMPlupP<;Bw1xLlQ-iTJT&fXCAZP?aE>&)}n@M
zfLhB!D~3(npuKYtid#?%7<wKFgUI5Plh9+<d<$IuEM{(K!?}WUs?U3zGj(Fap_P=$
zswnp}H-g(mqGp$P3BlR>8Hi*{HJU)0f#-M-ndQvR<)MF8q*X0<ss;I=_9#?8k;^pE
z+ii~GC?U}8Y@~n}yCe()rNB;<Vk>J|0$0!T>}pdqS)$rYlzJobwZuNRt458eS*5LN
z$q$NRo6bP!bPG2THPC9(@+vA#y=w9$TAL{0IEmp>=O}k6*SwKSk!8bOkQMOOY?T~E
zg@7^5UH5;A(kxE3g5dA_Om0<2VGG1d)bn`|5gtLXcqC>_Nwm~tUXP3uk9FZ^pmD^?
zBT>k^5dC@PupqSWCYWLD(>^zbp%JNMdwT2fAuaQ97anMX{iQj^RzTv=i;&7EVHAKC
zjJPaTWa$($D3HL0#Uz3Z(Yw<rT{!UpIsidvsOW#znw>k?mC+bTv6#0+bV4+T63-7U
zCm?>P!Kj|Fi3fNQS#akM+%t?kX^J4m5F<S1151q9c#!%%qEETYB`%nu?=EEv){i4X
zqxeYE8MV)-ZJh?ufDh{DPw?prACi#w_F_JifWUZYU<`0L54?dj9yWfD=#d#enlL^;
zk?VgLtY9}y2J=VhbWpp526|(6q6#=Jr!7ARijE)a8udzpOi?DX)M!;096W*`pbS=I
zY=~S<7dwEm8YzUqDp(_HfNv}ai5@_ZYYebqt%z*dM`px7(j8BF0Q$QurQUK~iX;V?
z1bU65P%D^SEpaTgHS>L1Yg88oQ1hx5{bhfu0JSBeQ*0xufp8rhBWr}~o`fWhOdaA8
z2?07i8W0Y@Z5?_p0jrTM<#vW<i0$HC`x1g9Z4IENoKXL$R@}-5BZ*WxbpvQm6rtsn
zpg9<`GpT2rN^z{DX5H-6gx-RingT<s2CLfA91%_msvC2{0nl1SL{_v?_pI70gJplZ
z-gHLAZlWjfO=3X#`nMM*XmmW(C5VxVO#7qAny!p~(G?3ZpCb~$kQ)y!Lv!=uRF|oF
z95oxNUQ->PsSI%<F+|!Ut!zrOE|>MbDBq1QYF~6j>>Xbdxtih#Uu?+HjJ!ELLW1Uz
z+rj2~p|m}{1JJ{h_MBAdll7W@vnziIo4k`eyX2hRGve9D?DIAM>EcJ^AIBC@(Y4Tl
zh5pgrIb8JQ;P?#AOl<OXpi2l$VEz73rRKT7JScXV0Lgo`cn`l?yiZ5E4sn$XrYE*r
z@y%l1{l+2xIih_!Eu68$=NjxZ$OmZjwW3+oIBc1F)8mI}lzGuz^Cm+0&b)tr^uy*_
z|L?rO=POamQNs7?<nacIJ#_{c&@P*@l)1Kh|Nf2ryE9%LAw;h7c%ws2rT^Vm_*3jQ
z&6AU}LYcmY)9-uT-;<II|Fk^;)rmvJ8OBPmUf?6qRZJ!`xdk8Y(@HMzEZV-Z4$9J#
zi(X3g^C-mf;nDhID&_~Ne)Jqm|1#yqKHSRS`j4k)^FO{H43@r_-~0<B6oFrdeTV~x
zeTW2yeTW5zeTW8!eTWCQeTWEL{R1~LHkaIm2_OSEGB&sWg$cz0QZqL;F*GtTI6*i>
zHbO!{Mlm%;GB!p*IWRCmLNhZnK|UZnGdDIdG%_$aK{!M<LP9}CF*QaqHby}?Ffc(v
zGcz+mK3xhgOl59obZ8(kG%z!h@v<m?CDuu7RaG3v@$-N8z4pD*mO@8{4z!fg8QRjh
zg|^Vrd7v$wCt^g1i9_5G6B9%ho)|QqL`@`U99SrkgpImzW7wb@H4)>2C|eU_qKO)Z
zet&E}Ik*3FZ|{5e`}*pfbN|Jh`5x$n<1yzF&fK-%94mu#b8e>&V-Uq$EF%wpa-jyw
zpc)DxpM7yiKpH<QhAP+(1yBjKY}pA(bIDiqi&BsQJ0KIXAe%FaSQS8yx!k)tV;Ag(
zT*w36MhO&}EBROZbnYJ53;SR{6hk$Xnyda%o6Fz;l=FK_p#l!WK?W6|RV5sPDmV<y
zP-Cw7cU?~}wGQgJfG(;t8lVw>j=)i9f(~dg*Y`jdI0miI2JKv6AMAm>pv!kdr@7I0
zby^qn!YSx6H`%Fy9`_08gMJu*lQ0CMFlcV6R%e`sVHkla7&G^Hm!{(|0h2HX)8?Lf
zLem*I1GC&|3FzHjgL!k8*0t{}EWjcx!7{ACDyMC~e=bBEuA95^w@%xCfKAwfb8KM{
zs5r#T#+G%Mgt)mk3Ux@LL5Y%2NC$!oDRUnTX$vI=N(__}3L(qf?VmKw1_BN_K#4(@
zqQpRnff57723iY6=03ZwbBQnzVW728Z0>GC1F{VYjy3{?g#+fkeqYmaAiY4bq0-#_
zCJm@5&|jdxptotezlTeIn$lk&kU(Lf!Q3yV0qF(O3&a(g%>A)j0|E{$KzgAadZ5kB
zUe*>84V^${p__Zw^++o8nq}Y6K3WU37KkwPn-%`50l@`=3j`M^EKpq-G%Nl~TL>-?
zTp+G6W>(dx0l@{53X^7aefk+qhG`(UKuBTM>{za*G#TdMEO)AZ7rhFLX5Htth2R3g
z1%eCfpqkdq`fq6qtp!>O+pSrC0-V__A1li@h(QE#ixzHbkc2d|rCuGT!wyK9t$(SX
zWxy`j37KY(eXKzi=<nZAR=Xh|a?CD1uPwQd2ZgX73e2AQM$<j87xqCB9Dri8OAoZA
z1WKU{4neuu)$^KvR=`21gj%RFdo5Sf!%z)1P!A1eA0E`S8S2d5uGe8B909$UCSF#P
z#lCn`({^aF*q!fncnn&h4LYF<x}gs`EGqj{oB0~cKMp5=ueSUF7=eC^s;_C^NjL?A
zFa)Pz7$#xVqLwGMc?`y30%l>#qMklYr(p)pz*(5HXrNYq(|KN-Uj8yHSTyvs28*x+
z8?a*0c!dV5um<aVC#$ASi)OynRKcBN;4H1RhM##d^Kr<8m_-}kX+W4h4QS2NpHBgq
zdD8RQKx;k+D92}6bm5XNK#88#JaKvA@>J&wfTTRp_#%s*dvpecc?$FN=k>kc!m1of
zEqdWS?V~V%Pk)|he1%0<U)O-bJcW7s^Ho4xz5!}1dUaF#=+D;yt@$QswCJrmO^M4B
zmnRzEY|)Jk4RkLg<w?rZkZ-r>=A+X(fWmwiH?DmrSm_<2HQ&!6pGHqto{&5t`9TJH
z6RF9c=1}h+NqPG7qd-lbn*0P?^Z*FXPXRS~YVy>7<Y~>%!8{O`-)`vE54zn&i|$?4
zYRTfgiq$V${F9HgTCw=Oqgt(6B7R4!HA`f_qt&`4%Jt#&8<rUUN~=vv%>AR)mL-0E
zv^r;LBeRY>b(VT|V!OJ_pX<;p?O~TzF-sr2u2p2|xBt{CZpo-ttAr&B3$#kJWcij>
z>6Wa2zpqu&l5O)^r7U?pp;d;Z@`6@7EOlW<t4wHxHqie*9ncASTKewkfnGQcdU$=H
z=QaQ*;S}hR4S}B4FpR(`jKMhQK~2IGOv4PE0sT8O2l}w<i)8^8VF{LD1@!;R8mz+x
yY{C|7f3RaS+wb!g{z}DW%u+8-Zy&yN(ftnu#Lhk=m$r=wAPG4!3MC~)PeuwgcF4Z~

delta 6193
zcmV-17|!SBn<n3zCa_+omyl`+Du3-;S(Dqi5q_Uv!8}D%yTal=?DB(??51ifwK;aG
zHYwYw;z*nky*g$^$z$)YPd5N^2o6MP4m(LDFGG<8f$m1*>#rLaT3i>j_~RLk*WX>7
zee)eNg(FULU0hrhj8G=v9W7YFq$P&&LMubKP{qZncr&?O)U$JLCyVo`;D2tit>*h#
z)r6bf2CsIqTU6n}a&uibJNUpqzS^&6yLz(@@9OoS@nyX(x1YlA**Tk(O-0V9Qn<-Q
z*I)fJj$JM51q{W_1pnH}JoL1=I)8WZJqRAf7lIgV3oRAlHb}o))D1TJ<~w-O)7lXU
z4`^{Qhep2-EfTvJVuffMOn+y~vO%4NnE0RdTPkF=4NvO$X1zHF32(#Ab^XChz7Dr#
z^CBaLVMH5SOa&uM>oC?|VmpxtV>)5qQbj`(!Fo*uu|~FS{s4ngJ$aw$iWx#JD>%{I
zg~tEP2sWj}f;pzWW^XCynf@ekO7>)9I8Ao)DmHb6Ce8?pO*v>OU4OwikxGOQ)=?AL
zA6pPd1&-}(9Qz-c-atfZ!FJFqXR%!!+v)zOpowb*e8n<7F-D9u1($?Vn|Oray=EF-
zi#tsz%?t)@xM|z4oN|m3<<u$KNforCE($fv3@C`wopLlc-;78t@uEQx-|kIj!nvPs
z`cM~4`bb>xQ${ICtbYa7K_aj~n6O!9C`YseiSJsy3O>cqz24;<i-Ti9?~I9La6!>M
zEk4D;sYJ6nrw3#Ab<iN9Ll<OV4&UrW1{|hbaw04`sfKHu$Jf_o9hdv9U;Fr5PIl$b
z-JOW&2anf)c-3{eL!2<}Tfooh0P}1Q^YHskxXvqtpmqFivwx3RG}|=0(0>QMtW_bw
z3kL44DK2qXZj2>vpUj3L_Rgi-nFniX=TStQ1R~-I7?O%@2vat*3R4G?v|I9E(p&(L
zL^1nthkqjGejaNbS&DfNW*TBWu0~L_MTQ@%)JsJ_po33stLtqA4g~?wM|F6#tY*7C
zJlc9rfPS}$u76mD%hi6lOkB3!bss|jPls#WS9I=6LWn1n@&QXx0#-baEajqIR+pt`
zD?f)`Znm475*JxJV)e`3-?bsTs^)PyZt%0)C>O4p<)?5{hD)<5TSWpND%aQ5*YV?W
zRkh7gmmDimh*Bm<JG6x8Ea3;^AlL0-gG{BAlb_D5Xn#?o+*aYU-2$;>--NquRn9-9
zsIhN`6WoC4;TM<&Nm#HBYCyp`7&u!*;WFCn(!)^&^IZKVx*jV+LyoQmpK)YQ5mUbX
zSgvlC)r<3~qV$gQ8P#J_vv!8yDNNbrvRY+)7My_hj4E6qHGr3kF7O)Vb9_=_M-IR>
z3>+SM`+qtf^@g|QW#p6aCm=A)rKA?DwyZv!D~0{Q<6ZbJcr|<n;(7hw<2sndw>*v<
zB?#!hT*e-kRcyA8?~&|+k?Xl<koaNi<V+5{Mk-DyXzDdg2s0Wl(&g8Yj?e3624leb
z^j_U@ZWrUjfS|jup4-KYZ*ehiF)%e>hbOmfaDT$?W5|hpmr)S((450di408!(R--?
zGUj;QRohi_019r(Ex0DM$ntk(iL~wx1)0n^48#G;+cssBLGUR9S5*L&Q|`d}=AmH_
z5)CJhl4+nae88n!d6O|TWC84naWEma0FW^9QNk@D0z|?cspj7^y-{Gv{E>V+84u-g
zNPpTayNh%=La7KcWq=l{+&HV0fmO;>tU=U-c;0I?!V9{2zR5a5NSL3C38yj-_=Nyj
zdX3&^nY+=Qz#lVXDWGb3``EFuUgj5pw4c@8IiGwLE(c5tnY)9aPY+uS^p!fZGHikj
zPK#p`blj=#87*kRwDIl*+{$%XU|&Q49e*zELu5SUiTnglRF6{hf7=u9qgGhceJ}#l
z%5iT)G~`nWY`x4ea6us%r5b+_8#&hPqS{vXQ^-X*PqV)H2u&B>7n=HdwW;U9Aw4gL
zu-*pYjI&Y528CyCLVP*!Q-{}<_T(W`1;|B11<ktk^i<zer}(D&ccNrV3JMMi_<tgm
ziEE|$coXpFz~pd<;!5A=O-Tp`TZnXmXjU+7Mc)=*9RsnGhs!N1Nu+lD;T|@NVr&-M
zC4ctq#n~@sEs@0qO06k`QbZV2%vNV_-qB(XFCjUC{KFQvzUiv4f-s9tzbt+{`)eq(
zr$e@VN$~5IjHLnZ*whI_9FwQpQh%N@b@JJ!PM$J#@-e~Tc*M<7M{2N`C*%kMa|Nj9
zoUh_p)~31ZdCnMuXl2rb_)Em(BLJx~07(x4q&WtV_835l|0;ky7%sRZLW%!0`t|=<
z_*kCyu+k^6F54?rNLFwM0IyuHlE}4yIcJk$4I-)P#!fhu05;f*2?_M-5r1kPl8^)7
z2%}RG12x=wIeDKGqS(FwRIjhYT^TN`dR?#TUt^gKzXG)u-NphoX}2B1>+5Yfud6ja
z*|ozov-Ux*+`}OTN}!Nl!nOwdLv_2`tfzk~Zx+pF{o7bU?2R4DySOVeEbnpy=91Bz
zKipMkb&naV$+o&#ma{6CD1UYqbYQtbe-8>k!@221<s$6gCSBjwyG4*KvU5EtXSmCw
zbSSQl$SSa`1A!>%@Yk%Y=y4E}SlrB7dOPB-AvBocR;4>l2f;0FAtocn621!6Xr1n5
ztI4v&yHCLhN3@ZY!AzR8@8uDBB()&?V7f3SrzYL(cQ^Z8(|V5`I)9^bFpbd}%VtDK
zCSl&u0R1l>UlHibqJU;`nfCl@OcV{H*?}l!cJ{f}_h+BxCjOvw#zLm2A0VBzus}@u
z!PSvLoFrNfzByFb>AtZr@dL9x%NjT&C8wGwBkoDbjzKfz00Z58)!YCE00sjLjG&UR
zev|U~wLB=UbTEHB9e?p|*Nb->J3e;ceHkwOQ#^5_inpZ)-qtDV>%}=92x><0KoUa-
zE_a*J*??*CI&(eH`Te-kl5=2bgB9h$1of@UrWzGkJHqGyO)j(8z=7tSwjd9RY9NbG
zha&d^EqukLAxcDG?C%0DJS>0cpFCJp*=WNdTg9vm=v0ekhZ|w0cdVkPl`{GmUFDGV
ziI%Qrq@evu`?GRmUz3rrA%9YJfEDfhA%?N4gYOG69;VgPs}5sUk%B+LD#n=O2|chQ
zGgeaN3}Te_XIfP2*`AN(76GItc=oQJ(H~Ne-(Lk@J$J%V;5xSy(hV_O3L3X7wLHB4
z1_^-F>HgdM{JJt!`9p23T8A+<rxw=(wcclo+_+oX(`yVsAbseKF@HE#$BEL9I8~RP
zmV=bRnqq3gUU}a%Vvrh}zONlEdmRS*6Jhaw!rZgj%L5TXr0FTFP9Vs#J&qf92wY2s
z-K5#ygwVZhX50E^H^w`-BM@WKK9|{DG-1V&b5?y>xy18iD(#pZQIIQ6Ea&M*^?F{<
z%3U~5gj$xnPTwYKTz}rPL7Ncxo2%_+6@GQOiaV0409lvn`OP9BjWO<lW<e@|R4w_4
zWV3EeGP?62f6kl&?Ctu5(@eX;){!qBl<{N}n#+u(b_#b_yQ6j4f{zn-CsGf^{jSBI
za_4*~u&%Dm(?4BpTPPbG;$Xmxm_F?ghrV}+12}>|ou}AXjem-V&AMK*pK=fb=<CF1
zI>RAPJHz3=*clGyz@T!pj4A=%H3Bq^^Rml`h2h$sbcDm4c!Z-f>o~EMJuhq#+#Hs`
z0{^gm1V3_C0U4Zm*0TzStY@C(tODwht109ECvg4{IiJ<+fGH`78r`D5b2Si@WzTpc
zo#?rwkHg*n$QTUJNHY%)Qy{{)0~9!jPQC#C{x2gH+RKB`vWL*J0k_by0`1P1;aCGI
zm*;^44S&WADN(N!Jv3-GO%Kf?@S#9j7+j0Ch3H}E5j*+yoe@b{lC^0ZJL%?Nn~&je
zp5Kr(wKf*DUfxjiOB~Ol#G~q~gwf&1o0m7w7dKCD8RJ$+BH!oMVr{X&A)H&j7ZAtw
ztVL|SwSSn+J;$?ut*R!ImFgt<uBlb?pnmRUl7DYhQ{?Km$mKjvGMQINk&9GMmZkVG
z`)Bb-Co0O|_GF;eyzQvn^Cnxq7ij}Kw~Giz1<Y5g*I)$nvfj2k{W91^59Yfw*w|#m
z!mg5!@^9+V1AH-@97%h)ijaqkWGrR16}wi<bNI{$d_Hp#8*lzZdHLuyW2n!H$Z<G=
zaerArBG;XWMAfdhqBbpg5}T>`AWPMZ3lqC}QE$6gt&4|OAI=EV!%{80ru#3qmp0h%
zNgT_(XOd?lU;p<`54j?aby};Rve@g~Uw0T#4K1DItF(!=weJ&A{ZCfD5D8XAz7n;}
z;e(lmp9?vnsSK-g7@g+eF#GVM)6&?l?SJ&f#UZ&6Ue8!a$MW{Ev>#(1ms=IMkHFoT
zsIw+T*X*@aLz6yK53Z>~m29#iiBSk+nyaF#&gs`pJp|CJp~b6l>OVUx_a9F*e5c93
z?bU(rAo~jyh>tA$l~f~}KA8Sj)L~>0N=?pD&1^8D^WA$m-%D9HW&X(wbR80T4u4FY
zX!G>eVHRUB4f8-b4;b+sZq1#5gkh)_JQu5(%j|c%ve=1@sNou**0R8gU=zo0?;M2U
z7SsZUu1f+xv^eD?aGABpfXko7%n58bS8z`Cd5?3ZPHZ@|mNH%!<$mUbaJz`r><TX-
zID0<>k!-0(V@Nab9QQ-BoY}cN)PIV!s^w0#ARp8oh3dz0mBf0x?NJ;h1iGD#6mTPl
z1c9#<*ojhPvX&)qbzRr4wndZ1s=Y#~w<6z2>~p(n)QFl@+NzfPpeV9r20|wp+(gts
zt4YhNs5JGu$yaD?qJ-llf=k_8xl6g`ja-Q|9qxjxfVXC=<R~fxjA`z=SAUddaiSFj
zf8VDvQyqmZ5HDfR=YB|d2*Ki#h%qJ6N|U)gGEQ99g`a`O5i6I30q;We=h>Y3fqgf@
z3}c`6xiJilP$k>bTbB=MnTNY@UmNT%%`vik5(RFER2~UJAGBb^VUZ$BrkFv21U4)p
zA!LZ&olfb(i3iXD2torzw|~*>oWZV)MnH<iyd|O&qRlCBy}+9L#0xYS)e|;x0WU%e
z?)<)UhLI;t;YS!^ghzZ}i4hwYQol#^DR()<0W<X7rEJ0aaYSes9cenF_8GOU(;yn~
zLH)c5K7HW>67b$$%!3l(8xIYP0S@QBJFv#X#_th5GUJC6#^)z;9e;xr?54?J{wSRe
zYKKr?Z|qD|fw{wJ%Wn=v$BT50dZj_8C}UY_v?>e^9zhUL1}ic)M6RZb9Y9%~WFk#I
z3WHX#O4b12SQ27AfH2n>Xv1m|+Om($h=8O!p!6X0cUelk>82EM0ze7$8cCsAFuPje
zSYT`B`=r*mE)1mRRevq|%Tx(!i$tf|R#XGwIzUF&2-rOdNfeq&#6uDQc)HXl9DdtM
z^jrd3BU{St6wMIX#k=+`1V`E`KutlR-chZnl@CT@sdVZF(4Htl%PU56FlMJx&o-6f
zSV_&g*{KP=1vga%hSm+%wUs#}oRm~2;)DaFwTg(eXr=C1wSQL#%Ve|djEdb@PvEP>
zfb#WkFHG3zc&JMdBN3_gN0ByNA^oCD7NR~!EPx>=8eE4K=FO=t6Z1T3HdMW)IzChx
z;#gvcw1-;RlxAHn>wQta8(-DF=#bbuz{YYt#Sy;PkfRxSb9#sb%|oXH&dpM3dwL6?
zhbQexsnTcb4S)S+ml8I4D|vRwNxNsnvk%(mYyQ*4&&WTHEuf-nsRIlBqrG#u=+R;B
zF*r4`$@hUSAuxgU3q+Ne2Ltn@*kuAF@73Zx_-gS!9qBs6RWg{K$TIQGV&3`2A^$m|
zeL5}XV~Ni-*lCat(CBMLv#N2}GB>Bk57Q{~vb*R_gn#gzc>(E%&9(mFd4UgDqL!nC
z@72lU9Ta=&3^1TwHDxJtZTJ2S9Q$`?ygNdOT;=&jhnh<NyRY!K*lm)>CufB+c@L-G
z_qx9)B^my0djhHxhl(?d6=S`?XQHc^Or|meAMTS%F7YhdzOxR>(vyqcO7#ON#PZ?M
z`eQ2Q2Oz0_^h^Ib<wPFb%L2Gf!RgujlkfS1r7spY{{n<Rf-HxFhy#a%hy;g&hy{m(
zhz5s)hzGZWhzMN$CNnTMD=;!AFfb=63NKA>WJF<fWgsX-Ix;XZI599WIWRUfGcY%|
z1%?U50a8OjML9(>Mm8`vI5{&$FhW8(F)}qcLNYioI7Bo<L^3`gJVQZ6IYlx?HZV6h
zIWtBuLP9w)GBr3tGB_|eL^MN0GCo}jFHB`_XLM*FF*GnTmw`M1D1Rl^OKenC9LMqV
zpF7jel(rQ5F7!=lOG|;a(08FNw9t1;Tl#*`5F&02>W-Lb6iCboKH@P1f-!2)6-uHT
zHm)QpVWTmcxG<5pa$z*e!bI%%$L5oB`#<+`@7!;hI_KPzm^0r4y>K+<T*8@q^H0ah
zAl;nXuEQ8aF&E3ogMVDu2Nh5cg^<s_I3ysApOr!_?1loUfd;m0horgWHT^>=$bfB-
z30aWM8AYrLAje$peVwrbc0w-Xfo`J=3eA=Mr+qrN7)oFl?1oaPhjMfEKWlRZR6-U1
zZaGv#9qeIH4O-Q}UZ{mSXn}p^TK?Ac^r(%np9|=sI->~=z<)tF1kKO^t>*fE(FG2}
z5om*UF0c!Vp#*gKZs;^O@_|n4f?hZQJ?17lHPGum27S;E18^LMU<3xuEjH+klW+=#
zVG>5oo$u0g48~ysW?{<QbI)ix4X0s-J1ql!x~ni}?$Vm}oq>5+fJIn>Wmw^~t<TSe
zh{H8=m;cpi>wmBTXJL~q3<4F0nAzx*4wDc!cfC-DBpQ?``Gj;JxR5e;dq`U-F;HTl
zoKOf^=I;KeX*LjW$N@?Wx)db_N(_`3C^pbqC^C2N9i2;rfd~Vwg;H~0>0tud1_ehO
zfx<$ixraA3tpd^u6dP*H{Zyv`H3j+$^cVCoZT0tPQh!tW3j`7<EHs(>{jmn57f3G<
zS7<i(Pf`N{4y{0Xp&fdl&CFiW77`7eKxUzvd)D<xD)gFVzpH(;7HBOHVdyt2{7nOb
z3j`MkE>KvYx-e)~`iHg<Tp+kWTw&C#wnhVj3nUdL%o^MDGnx!jKyZPO!i?GBl%_Np
z=HLu>s(*)Gfd#YfRc#@-KyZQJ!WyWiRkQw&w1w6Jt%a@DEI$FxZ04Y2We|f1;ug)n
zr$G|Z%oba9m=4<@Wwv%tKg)m}upKhZ&V8ss7U=)KL#%c}KIE7^`+~OQLLL;tZYVIj
z__e0RPy)N42r8k}?9va~QU>Kv0ehj!?CPqf)qk)DYM=pX&E8IFS_k#85B5Wo*~cZC
zwm_rVt!f<}fP<iiY38w-E%xObnzlo$#Xf(i!^3a{+MpA<pd0$2!=j2$wVAK6{G)IT
z_-e}^fMMvjsQ!xf9fuPz2t#lZPQe6>Sk$_y&7&{|<1ho07WK4gIt9~k8qUD1MFW+Z
z&VTW2di*7rw`l0R1`DtV>#%ImSg{5xunKE@C#$9ni>ANPRKaaBaF%vt4?pu{=HrkF
zF^kr})qpU48qk`jKc50J^Q7mqf!2HuP>#>C=)y%^fD%2edE)ZK<*CjW07-eG@kJKB
z`1A}4^AzUk&+B`?l~om#TlDe`?V~VHe}A56e6>YauW3MGp29r+`C1?@-vs+CdTUAh
z=+8F-t@&m+V9|}eni7{ME>ASR#iI8YHPF3~lqV@qL%!XjPoAFE0TkxDxN+?}#!8<M
zt@(Zq`EB%c<q63Xk{@KC50RSuNe=b-k(8%DKLXU`smYJCMK6Hh{3KA5rzTHLo`2T-
zEX)CM`K^ZTe6QPGu;{@{S}j_<S6%rfi+^@ot7VHnXwYiK67f4)ty&`crdDf~s4CNH
z-4dtnYqepC*~eO)wZyMaSDThLJnpztXQ}5;ZdLc{7dkXcdvsW<n5B<i*DA8~yMJjF
zw`5eKRl<^mIa;MzvT9kYbW864PJgSUCEF&oN?G#O6Rk2VmG_2L+bnfqRI5xl0&Sqb
zeLA2M^tSZf(*wP56!h}?K<{k;j>8GiD;om6ty3@zBQOeMpcgd(lQ0F-a2oXI%q-}a
zU0*EoumFp&1k0emUshoa)?owA!qzW#Y<lZ+zWh25J#ChHZDi~4$~pHx_5)fzoR__g
P2_OkLH3}sqMNdWwZ!@{P

diff --git a/doc/src/NeuralNet/NeuralNet.do.txt b/doc/src/NeuralNet/NeuralNet.do.txt
index 5c0da3c04..0735a92d2 100644
--- a/doc/src/NeuralNet/NeuralNet.do.txt
+++ b/doc/src/NeuralNet/NeuralNet.do.txt
@@ -995,7 +995,7 @@ where the superscript $l-1$ indicates that these are the outputs from layer $l-1
 Our cost function at the final layer $l=L$ is now
 !bt
 \[
-\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(i-t_i)\log{(1-a_i^L)}\right),
+\mathcal{C}(\hat{W}) = - \sum_{i=1}^n \left(t_i\log{a_i^L}+(1-t_i)\log{(1-a_i^L)}\right),
 \]
 !et
 where we have defined the targets $t_i$. The derivatives of the cost function with respect to the output $a_i^L$ are then easily calculated and we get
@@ -3281,7 +3281,7 @@ c(x, P) = \sum_i  \big(g_t'(x_i, P) - ( -\gamma g_t(x_i, P) \big)^2
 \]
 !et
 
-In order to minimize it, an optimalization method must be chosen. 
+In order to minimize it, an optimization method must be chosen. 
 
 Here, gradient descent with a constant step size has been chosen.