Colorful Image Colorization Trained on ImageNet Competition Data

Colorize a grayscale image

Released in 2016, this model recasts image colorization into a classification problem by dividing the AB color space into 313 bins. The final color for each pixel is picked by performing an ad hoc modified mean of its probability distribution over the bins. The resulting method produces vibrant and realistic colorizations.

Number of layers: 58 | Parameter count: 32,241,005 | Trained size: 129 MB |

Training Set Information

Examples

Resource retrieval

Get the pre-trained net:

In[1]:=
NetModel["Colorful Image Colorization Trained on ImageNet Competition \
Data"]
Out[1]=

Evaluation function

This net takes a grayscale image as input and outputs the A and B channels in the LAB color space. It needs an evaluation function to merge its output with the luminance of the input:

In[2]:=
netevaluate[img_Image] := With[{lum = ColorSeparate[img, "L"]},
  Image[Prepend[
    ArrayResample[
     NetModel[
       "Colorful Image Colorization Trained on ImageNet Competition \
Data"][lum], Prepend[Reverse@ImageDimensions@img, 2]], ImageData[lum]], Interleaving -> False, ColorSpace -> "LAB"]
  ]

Basic usage

Colorize a grayscale image using the evaluation function:

In[3]:=
netevaluate[\!\(\*
GraphicsBox[
TagBox[RasterBox[CompressedData["
1:eJwsvPWfFGfaN7qf991ns9kkSJRAgAAJJMEJwTW4uwcbbIABgrszyPj0SLtL
WZdLl3W1d0/3KBLbZB87u+c9P5w/4lz3PKeBYaSnqy77yl139biqC9tP/q8/
/elPl/8KH7Yfvb780qWjN3cMgy92nb98+tT5E8fXnb9y4tSJS/Oq0NP+8///
Fw6HI9FIwOUO+ENRLOzzur3BKEbi0WgwFqdILByOxoLBUDjo7mxubm53BiI4
RjI0FQn6wzEcJ0gSC0UwnCRwMh5neE5gWTZO0wzHsZwoq4Zh6KqqJTRDtyzL
TFrJpKkrkqQkdMM0NFVTdfQMSeAYlpcSZjKZziQNNSErsigwNEMxgsCzGMaw
rCCyvMDGqVgkHAwGfT4//I3FSSKOYzgWoRiGpRiW4XhRZGn4XgSLU1QcR+GF
Q35vIOgLhGOxaCyKEwRFEXg4SsUZhohFMRzHwjFvW9Ozpy8aOtzBGMWxNBX2
ejx+fzAcI4gYvD6OR+EzmmPjjCDyLBWPQ3I4UdHgoahqQlNVI2noZjqbTpmq
oiiaYSUN0zR0DX6kCCzN0oIkaslkJpdJG7IioVfBYhjFiwIfpxheTChw6hwd
x7BoNBz2BYJ+ry+IE7FILIoKQvOCIMkSJEkRBIGGJzIczVEovEjAFwpBOkIx
VAMWTpIiY1ESPqUgVihRLBbpaHj+uLauucMTxuM8xxBBr9vtdnk8wRhB0yQe
iwbDGBFnOZqWZJHGMZQXkhYVOaEkUJVUzdQ03Uxl0+ZgfVQdCggPLaEokoDS
AdnnZMO0sinLkHiGImJhnz+EQ2SSxLKcpMi8IIscQ5E45D8K5xTBCHTgQDga
xnESKg2ZkhUNXh1KzXMcD7mJwiMWCYYjQZ8fZwWWEUWBg/ODmpAkCYHGoQmx
WMDeUlff3Gp3BmM4RccpIuR1Ol1OhwPihWJR0UgoEMLgwDQryiIVDgYCwUgk
Fmfh7KB+EB/6q+mpdBLi01VZlpXBkqpw7jwL3Y3HoMMh+zr8XJPjsaDP4+p0
+SNxDpUFkpbQ4ORlnmcZhuGZOMfRDATB4KEg9CCGYRTHKSidErSLBFHAoeE3
g4P9F4EqBmO8ooroORILR0OpwaCDKfhxLBxwt9s63F6/P4JBU0NLhgIer99r
t3V6Y/AkaONoBF6JwClekgUcMu/z+LzwbIYTeRlKqMAQQrcaSfinabLACaIo
67oi0nBmBEnh0C0kwUAwqgH9inkdtpaO9k5PiIKGUGBeVejkBCriYMokVkgk
ZGh7hY+FILURNFgUK/ICz8swzdDrLDxLUYIhqBMOz4kScUABmGpRFuJkDDKC
voceOIwltKPLH46GIpBlGIBwMBQM+IKejlabK4Lj0DJQcOgRiqJYnqOg4b0+
Z3u7w+mH9mVg6AFR1EQCKog6SJYBOlgG4pZ4mghGiDgBX6jQODwcHR58POpq
a25qd7q8UUAYBeKC34QI4Vc5KCjUHpJlmskUjCodC/uhWQBFCOgkBG0stBE6
pgLxw/yRBAaIhNMsjIkkQN/GSTwCDRuIAHIgxIlgsVAwGMAoEhUJoguHIWGR
YNDbYWtzBdH3SASYDPQtTDARCfl8Hre9tdVmB9yCX6AQ6siqpsAwor4RBRr1
ImAIDQCNx2kuYVop09QlWUbNKFCYz+Vwuby+MLS4AGVQ0ARDctBo0bwsJwxd
T5qmlUlpIgcYE8UImCJorEFwJAiYZUFGNQfojxFYyBcEbADokTiagkmAaQpC
CBAb9C9MMiBUBJqHIRBuwNeB0GCcfkenG4FoBDBVhI5gaPhtiM7j9nk629ts
tg5vBPLkD2EAQiIMviILLBmHMnGA9ZIoQCARnGElM5XLF3IWQh4YHUARKhYe
zC4AKMSXgNJruiqKEKgkDRYZRZtI6EnLhJlioW5UnI0jIIdqxSgSWlVE6K0C
t0WgPX1hEjAKYhbisXAoGPIHAtDUIfgUgAoOhJMkg9gHhhGGLBbyRykKiur3
B8IheGAUQCUPh4FJDHkcTng4bK2tHQ5vKBJ0O90oQgbmSOEZGroIDg5Fgg8i
w1BxQdStdL6UL6StpA5cqAB/oNgRWiI4FyUYNAMgF8qGsBEhowCII3OcBGAC
3QADh8IGGAYwJHCACAAtoFTIlBdoHceiwSABrUXDN+iIz+0D9Av4AhAcVC5K
QluT0NcA00Cc8Dw86vcD2BFhQOhoKAR8G4boAdgo4JUwisft9jja220tDq8/
4HV2dDo9gRABPSpAfRGPwDlBpCwECP9xkgasnk3Bw9QBhRIoGkWHmskwLYge
QB0kDQA+AKQEQAhQG8NCnAB9OIFQVBAYoH/0g1g4BqRJkPz/cCUdd3sGFQnS
KDFIMS/EI14XnE04HAiESZAlMaAJGFk4G1ASHA0ICZAa8MVI4CdoYOhgfzAA
LAgCgqEwLOgB0opG/B5nZ2tTfVOHy+vuaGu1O1weP84MkhdqCYaD8kcoToBZ
R/2qgIoxkLjREghfFIgTSAKi54C0AfRNwzABjBRDUzgKUQkB7UZhiIQDUYqm
aRgbGDxIvM8TCMLAEDSiuThODmoQisb8LncQ8srRsYDX7fSHI3g0RDJQOIpB
6aKQOOFFxPxxjoUWJ0giBKOLYYg4EcVi0HgEIi0vFCga8NrbWpvrnr5oc7kd
ne2d0LAIC4HFcVRugsJgBKJACkj9ANoB3KtmUoculBH6AxBpKtSWQiiEvgCh
o4hSwtASDAwHHDGKcMwHQ9AO9aFIUGwBfwCO7ui0Oxx2dwAlkwDYAKLyhygC
c3Xa/TjFoKP7Pe4QNCWBxQEvCAYgF8m1CAAQyj+0GdSd+B+MBfEXhhdA7BmC
UkY8drvdDXAb9tqhas3Pn9Z3eDxO+Ka9va0DDoDYJgIkCqIOCh2BlMO044iN
SBZO3lAEaEcB0IeFEeMZIowN0rsC9UXaRAPqp0MBt9PpDQElhX1OGHI3ii/s
cbs6OtFINDe1tNo6HT4C2iwUi/uB2HyxWKijvd0VRiwdjoT8niAeJyjQVxT0
NyhfKganQTDQCAROsAJoxThHgioDqsdDiMVxDARvIABD19rS6Q9FAnZba2Nz
a1Nds90LUNrhcDpgCAODOAzCHI+TkUAQEQMBEp4gMYApnJdFYGYgKDSTMO5I
1sRAMQKfAGmgoQT5Y2oSEQEPAMINcA5aDdoOVFM4GnS4Oto6W5oaG+vr61rb
O4G2AB5ijNfZaXeHgt72ltZObzACSiYUgPJDSkgMBEkMdCvHg6YFiY0xMo1k
eRy1GRA56l4c83tBqUK7eT0Oe2ebrbnVAZThbm9psUEFIZMu+L7T5QE57vH5
QEoF/F7QuSSONFWcQyIYXjSAAhzUvvJg+eClgRYA+KFdYPABjSTUv4aVSuoC
FwtGgoDYDE0C8EcAHeJhLOr3+NxuRxMkFTLbbPO4gXpB0cBsdLj9kPiW5pZO
ZxToA7RMOEogi4H4MkYB0AoowggOHYPD7ACFA/DQAJigc4JAwaBRQQ74oVva
gPPanX4gv9Z2p8fdaevs7Giz2d0w5ABCkAOX2+XyIJ+ClAIgMQll4kks4kdU
jsFsgyKXB+kRxCJwHRgeiJqGcEG3GYOeSkHASNICEsqAKHFeUQRAUvBz4VCg
o62t0+1qa7WBegTlQgudHbYOj99tb2/vAJDD4gDyBAZ8DS6OgP7BAXsAfsGL
gQqBVo1AY0DsFOiPQbkSA/SMIZEDg4WDieqAwFx+v7PN1m53e0BAdnbaPX6A
AOQoAy44GtTVDezjRy4zFojAKQjQKVBW0Fhgg0BjAJ5IMqJBkKcg+OMEom0g
7EGfBRoNqBHQH+g7IYFD5GVQlQIUHMmSoLPNHoqFPe1tTm8EgIPlne1tdp/X
1Qnp9vowEvwp0rkgxkE6gcoBxMeBtxSRwWLAcTA2QAkUgWCUQgocSuJHNEgC
mfEsGXC57KBTQZg2t9lddoAyB1BfGHnfOBkCu9HWWPsC+mSwVUMxwF10FiBi
wmG/GzgHeA1C00A6I0ENsYBRwKOgNSCBNEgu+C7icwHpOBByigjsryZTAKsi
CJMYHgTdS/EM6ekIRKFWQMlAvh6/1+mAeGMYnKSYSPCAl4gSAF8YEo0ZBy3D
AIqDT0CMB4YQThcHCwYCxuFweaGfIb0JRRTjIHigywDO2trsgNQgY1xOP4gm
8FRgqbyQzWeP69pAWYIVCSDrCT2CVAcYjqAfJD5w0aAcQ6YRfD74XgAwkKII
msIYDVwvImkHxAjEASHCfwkQoamkmhDi0PHgCFkgUCIMcpqD8OJATP6QDwRU
jJWQaFBVYFWehQmAKGjASVnlGQlZXjCHcDRoFEDTQeQBhvAB9jucUDEwpyJI
J4iaFiC3VMgb9LuBfV2dnZ5gBApDUsiEQf/UPnja4QARgNiFQfQKtAO/FkE6
CAYZWIlF1gnFCGmlqTB0e7vdHw55YHBJtHSAtBT6CHmQOAAkI2OlQfmAaI2z
OCmY5f4eDUw6CDxQpA6Pd1BqQtimZVnAqoheIRss5BJQBDSqhGaREliwVjDQ
0BOYzwMcBuYeg0M73NCSXlAkoAoh/DiQVMZMcBQy4BEC80V5LWlapjHo3xjc
72iGrnV7cEguBxmBeoAnR76GxAe9RzBKUIPsALJSEZCWAMEII+zvbLN7gyQU
CU0KyFhIPSAL2Hs1ZaYzmWQCKitruVe//+v/+edvWRP0G8sypA/4LoiIVjJz
2VRSk2HaVC2pwsQOHlU0VJkhaSgbSjRPx2VRDLvc7iALyY36/C4n+LxOVxjg
GoxJJEIrJpgyq2hZua5cLtuVz3UVM6aRzhVL6ZSRLhTzaY2gYvGswYVw4Dk5
MbjWgsS3iI5AAIxGwaoiDmKBDiWehUEBFQyy3RMKwIgFkOACp0lLyEsg/a2p
Blq1ApbUu37/v/8F8f3nQBrCZUBP4lBpkgTXrqXSaVOTwb5Ah2mqQKLWgT+Q
ByA8+C4VxWmOJEGkBb3OoJDQFQqIzu3uAA0BxEIg7xAMxJUkzDurZru6SoV8
vpjRGR7gMA72AnLP6FY2nevpLfYVko5njiZbe0w1QDOLCR05BwgUwSAI5MFl
KxKsrYzchADVhbKhBRewoIPCAGhZQVDKIQgFJ5JIgGlKmN3//n/+9d///V+/
9SQVGD4wuYi+BzEoaSYTssTFCZKCmYUmjEUgwjjBAVZCdCQR9fpBWEKpyVAw
JqhJg4243b6gu7OlsRN+Aq0JLOmnDMvkmM52wbDQOoPE80yEIiXIC+aOBt0k
Gi89k++t5NznLl+5VOvCZB1KqKiGCY4BSU4M/AoJWjYM4oxFHhy02eAoAi2K
AO8M0lScCL8D8Ao9ySJTzYgJoE41Wf79n//XH/2lAjQieAGKVUQefJuoGSnL
0GASgLFjvARwCe0WCmNoDQywAWIClHa6QN6EAOxJLgFEK5BBF1iBoKPVHkAI
CehHUVzSklii8/b1GI1ehgVxQscxBmRDPBYADwa0HoKRlKxckrm8p+rojXo3
xtKyBs4OWAEsAiAHjA1YTToO2MohFoTkW6ZpDi4yqqjMMJrQlQnwUqDcBj0N
qFcQdpAKq/z6dVnBwQqAFQPhgaQCgD10P/w2xwCr4eA3gXEZYDyk8CGZLHIk
GChYD85EfGEYFdCBpkzjgUiIiGPAzNE4DU+C1IppQ4y1tj/fV9Xi9YSjYCw8
TrT+AYMZE8l4DNLkaweFESJ5JtDyaOPiQycawOBgtKRraAEKgcL/rOvBvzgi
b0nR9GQqlQajjlbbNBOwUlcHFwyB7kWEhKA1gBgBi5D9VZIpjQgQYAr/x9rF
yShafMIIwVBAklE0QXISGcYUVcQDAbB1PHQ4sEEk4AlRAhMDEkqY0M3APOAl
QkAzQCIUwCB0Ivi4BBdyNly9tLu6pb7Jb29usYGU94aQvAyxXMzr8LGhkL+5
xWF3BexPri9Zs3zDrpqn3qir3YUBkkF1oIYCWj8Ccgd0FcHiQutaEBQECPZQ
1QzT0hUeCXAQqKBhVAHggYfM0CRIVo6GJgSTTYI4jYSwQb/tCULPhOISj4Nq
jYPLZfAwkdAksNrg7RmBBXkR8vv9GCeyMS8Bcw09A6/Lg+wjWFBJ8AAcD3h9
oB0wZ+P9q1U7jp28dPru82etra7WzmZbyAni0wNM5gsQtMfbbne117d1Xjl0
YM2s2ZMXnn9uD9Q9tbkwCVw6EIksqbqRymQzQNogGkDQIF2NlvehcCKaRQHk
AqJ5tGIEgQHOg5nHQEoCDIHOiWOhCLSczxcBGEaL10G/yxOD/IYIMFWhOJhc
gudpYB4AZAJIPQYV8AZISaGDXrScy3OJNFhtFcZA0vQEA0as89nzR/ftAZ+v
/UHNsZPXj2zaefbE3eetbr/X1e7w2gPeQKvb6QJ3FPe1eUF/tzyuvVe9e9v8
8e9P31D93N184+HTx52ECTCAVlp0K5PPZwHPDR3gM2GmUqYFESZhhAZHE5w3
EjCcoAKA0ED4LLjtKIgPNFMk2ApQ7367OwiAQEJpgiG3I0zEo2H4UTiAbE8M
ei+Klv+BaUClxSJBry8OGg9ESDgcjsUzGctIJE0FghRD7Y3Nbc8eXT91q83v
67x9au+mFZs3rNm28djZB622Ok/E0Ql1i/ja3aEWewxUXTQUdNe3vqi9tnvj
0sVjxoydsvuWveXUxftP7jwJqCm0TggDkM1nM5k0GrwEmHcrk0argaYGc0nH
4bSQ4VcAbtHEQqtSZDTgDaJFcA/IXSyM1uM99g4XJbOYx+kOhvwg6jG08BSP
RWkZrQOGA4EIJcg8BcPFgI0NhGlOQP4IpGCEtlKGpKTgqKqGNzy69+j2nXvH
d79wu20nt26YM2XshG8XbD1y4OqLO0/bO4JYGBxw0NfuCIL7AKdP4XTQ0VF/
//6lnctXTpk04aNp6574nxyuufDk0RMimcmgpaYMaINcPlcsFnMQmQX6JIVK
CXApcMBV4DoUqK3Co6VGJNfiVBSmjSXDAS9anvQH0LJWmwcXqZAD7FnI3dDp
9oYwcEzAqQBFNFpCCQvA82BwZTBH4WgUreFS4VAYSionTQEUTyalK8TzyxfO
Xzp76dDGe87O1po1Mz///KMR0+Yv2n3m1JknDmeHD+vEfGGoe8Bpw/CgJ+QF
2YmHwvV3nl8/e3bP8vGfDP968QmX78auvUcvXY5Z6RyElkEfcsVKuae7nE9Z
WfgqDQIFWQcBWRwOiEMzgKUBZxA9g+aloXqAhAGYkfYO0MSujuYmeygeGbzQ
5Glvb27rRBKeAegCeyLSUb+fAlXAgRUUQfpDXzNSQoqHgjjIBcOQiZgAcKAJ
DUc3bz90oubQxr1P6hy3Vkz9cMSIEWNmz166c9PRm7W1tjYX4H/A2xnwej0v
Ov2drS6HZ3DF39HadOfuiR+qln4+btS3O2zBp1vmLFyzrz2bykHVIJx0vtjV
39tTLhcL+UI+m8sYSMcBPIJmgQ+SljTAMmBopRP4TdUVDo+CKAAD6GpqBF9t
szU1tXrAlIKk9TgcnfAVWlyjZHQ5UgIKDEV4VQaUxVg24nV2uiOkYllCOBwH
3WfoUgzjFVNTJMeu9Tv3H9mxe8fVx8/urJ362dCP3/988aotB344eO1eW6sH
tFuz29fhfdJuAyff3NTU1tru8XlcWNDdXHvu3PULuye+N+yzuacj9XvnTl6+
4jHASDqbz6UtM9PV09/bXa50V8pdpVIeQjYBAOLI4IO/ULSkrooMKSQSoD6T
ybTJk6FQwNlSV9f47FlTc0tzSzuytugiZ9TtcIDPGVxLx9HlGbTgIZCUpPA4
zCS4dneH0x8mJCujAcGAndckpN+BfUW9dunShStWrVi7t77+ye6lY996e+gX
8zccOnfjyvWnLd6Ozlirs/0BpO/Jw8dPG9pb6huaWmHum9wERgZsj67er6lZ
O2Xsx18tuEVcWTp/+Yo7ACPpTL4A7WgVewZ6uopdPT1dXVDFfCZfNBUG1Ieo
KKKsG7omI+WVyhXyRcBajYl47U7b8/u3Hzx5/KIeLb16WtpcYEqwgNsNg+KM
ggeFeIBZADMVMc7KAhmOwdgRQZcjAoPLJJBGBNUA1MPQLMuCiaYvrl29fNGy
ZWvu2ltOf//FO++9PWbhgUt371w6d70zEPQ0NPma2xyd9tZn9x9cfnj1oe35
k2a7r7PZFQK8I7xt9288f7h23MhRXy+tE87MnLvqbNJKpVO5fDpjpgrl7nIh
VyhWKl2FAoxjoQDpBVsH/hWtiCaTIMgTVm6wuhaEHna22xqf3L9x+9GDZ/V1
Nq/bb2tsB+UY9bvcXpfTi/EiyE40sgTGKDyFMzA4oHKpiBNihywE40gII+JL
oIs8DMSbjJzetnnlhrXLT3t9NzZN/9u/jZ25av+FZw037tx6Yvc3u+z3Gzta
Ohte1N998PDW7cfX79293dpoa3G66usCIZ+36Wlj7ePqVWOHfDR9TSC2aeH8
PQmEn7lMOp3KF0uFXCoJ9YEBhIbN5AopXeSB7VUZqCmZBibRtFSmWCiU86pA
xcA8tzQ8fnD3zsP7tc86AEedLc8aSZg/wFMI0U+AFgVVJymgFIE7Y+EYFNbj
C+JhVzhOE+DHB6/HonVITZETYpySdS1QtWHVwtWrDraHb+xeMGbIOyOW7T5x
p/bZ86eNDS3OFza3HYintunJo4anl2+fe3D88r3Hz2/cftHQ7HK62lvsDk9n
8/Mft48a8vEn3x7EWrcv25koZqHdIKRsV193MaMqBgRaRoYAqUJw9Tq6CKMb
JhgeyzBNRCVAlBIWDAAvdDS8ePz4yf3bT+odwLId9bUvkLL0gskNRkgOaIGH
xGR0JF2R6Aw47Q5fkATdRjLALAGAGhqcCLw60sIs+Ai5de+aZfMX7Lb5n/2w
YcInH09YcfTKnYaGx7daWprrGtttz1tbW9z1Vx8+eHT31JWr1249vHX38qkb
5x7eb3X48HA00tnuaK6tWfrx28M+W3KNvLf6UKILqlWEniuWevoK6SQ44xx8
WUzpg0dVEX9YYCLAPRgAsoZu6tl0Uqc8nQ6Pyx10vXj+4vnj2tZ24PRgS0Pt
Y3R5JBTByLgIfkZkJSOZscwEKCEJvAcRQIv3bMyH0USEikZAo9KsDHIXXT+X
aWB+7vnqBas3/fAi0nTx4JQhfx23aGfV/RetT5uaO+of1ja62h7fvX350v3q
ozUndh88euZCzcVTF/bvq9r36PajBk97W1ubrbWxo+3K5q/f/vPfxq2pix7e
KfdkM/lST093qdhVAYOcyXYBhpbA9UN8qp7MILawTEtTFB3UnKaaGmiNBOF1
25s73N6wv74eYYvD5+u029rqHxOxCFrdYvnBC+RIzIKg1JOAI6l8UgMZCzgS
jcJUArFygizGeQ2BFcg+SQAXyD9Yu3XbhTZv272q79579/2ZO4+du/msoa6l
saHD1my796ju5tUTm/cd2X1477q1O9bu3HXk5L4tB3bs+fHC+ZuP7jfabO2N
T+tar+6YPOytv340+YDPtinemzZTuVJ3OZfOZAtAfoVKd0+5CPIlCf2ZyqNH
oZQHKyEl0QCijTRagiODXjLgcAWCEc+LuvY2h98Jh+9oa6kDMEHbXGheNCWR
JSgmoWmSDrZEzxXzOiJDQ0YrDOh/VkJqdzA+sJOSAbkUrm8/eL65s+3Oobnv
/e39z5fsrrp2x2Z7UXvnTm1Dy+PHjy7fuH5+/bqVG7dtXrNuxcr1G3fs27ht
9aZNu6rOnrp0p77R5u6sb3xw7+BXQ/7tz/82evJl50l/r5VQzUyplLVSWYit
mC11deUtK2OBy82kC3kYtlylrwSlQ1tJwBijS9YYWqITom2uSDjmqHe4XT5X
Z0d7Z0ebnWLADsRBQ4PbJiMBP6WaBhKugmqkkzDcqcFtOCZU1VJ4CV0TThbz
aXQdy8j09GWk64cuPrO1Pjq6aOjbH01Zt2vP4ZoHDx/Xv3jytK7lyf0XD69e
Ondi1+rvv9uwdtPybxduWLJlydK1875fuXPvyv1HTp668Lixqf7epcunFn7y
1v8e9cGYFc0PsCJURtDS2axlZQv5rrxuwKdGMgtiGy2TZUxZTRd7f+pNWZaR
stBlapmj8EgkzrO0r9MD4+YFw+ALgRtze5x+oPfB3T4MR+M4zBshoUVjhiMp
RkBYqavJbCFjqCY0OS/r4FjSOZQ2Wc129RTTiTvHbjU33Du/YcLbfx3x3e7D
O4/X/Hj1xp0r1Zfv3Lvf8LTm9NlN6zcuXvDt4lWb5s1bsHjxnDmLli5cvH79
so2Hqk5fvFFz+/md+8+fXFo45C9/rq8ZOvZiI53WEyzNq6lsBq1KoD0hYAIB
OzP5nJaED1lL09PF/jc9aCuUpUEFNIEh8HAMrTC2tHj9YZICEQ9WIRoJBJko
EYdyoW1PNMfTsSglJDiKB/nCIVusg8WDQSzCzOcNZGQ1K5PKQLiSOphW03h6
/FrdizPrZ733b38ZvmDnzt27T1299eOJU2eu33/47Fb10cM7tmxbOm/ON5Nn
z5k295tvFsyfPnve7MXrFq7YenB/9aWrN57U1d58/Pji4jFvvX/Btei92U8F
U5Fh5tV0oQtAM6mK4KZVSQfOzwAZZgulru7uUtrq6usvQNNm0iB3DE3iGIJk
4yThdXlCGCMJkTDo0jhPYpIvSkjgUIkwJqqmREZZQRZk4G8gcCYuGVKcToBN
yHdVimlDgKdkC9mUlc+pCdOUecnI1+2/03Bx1+LRH7/1v6cdPXH65MlrD29c
v/Pk1rP6RzcPbNlwaNe2bYumzJwyddKsr6Z/NWHWzCnTpn81b/Gchet37N1z
7P7VE9fvXzhdc37P0o+GVQe1ce+viyUVWRAlEyCmG4Yvo8maYcpGOp1DlikN
2q174FV3JlMs5k3NTBcKSbRBDwZKBlnMEZFICGSKSMUwjuF4MInhcFTQZMzv
j4p6JgluSFRUU5cFcB8kJSgsLZmpdKZU7urKGJKkw7QX8z3dBQO6RhDUTOHJ
4Xt3T20aN7v6f78zb/feIyfOXr1z9fylK1evnz978eTR09VH929fMGPs+Ilj
xo37asL4LyZ+8fWUyV988+2CpeuP/njl5rUzF2+ePnHt7tmZw4bezP/89E/v
Xc4lgNLMbLGULwFydqdUAeQwDIWVRDwn83Kq0t9bBnuR1QUtC0IA8b0moz0I
msRTJIazInQ4TsdpBgqFQUsmhEAgGiKETEYCfQ7PF5g4eAM8ytARMlEoF4CR
CuVs0tTVdL7cVekt51QhAVCqp/KP9l65tW3WX+ov/a/xGw8cq7585caNS2eO
nbtw/lTNiaM1V88cPL556Zxxo0ePGfvlpFFjxn0x9oupX4yeNO3rOVt27T95
9cH1E9UXq6sv/bh94hj2ZfdvV/40uZiGSHJdlS4EoJWeos7RMVpLpS00+zLL
caycKxXAE4IhVLOlUrFYSoGeAlwHQw+SmyEJzhA4ChQXk5AUCBjjiADBhTvD
KUtkuEwlL1AUCfGR3pA/wJqlvi4k6POFTNJKAoSBLASwNkBDKHqu8mjfw2NL
hq/Ln//L1FU7Dhw4f//y8VPVZ86eO1V9/Oj1u2ePHT64csmcqWNHfDp67Lgx
YyZM+Hz8WPj42ehpy5dt2PrjE9eDPfv37T50qWrmrER/X+8fLz4l+zPJVLGn
r1LSQD3n85Yk0qRomoYGUIm2x/E8GGsTLTmpWhK0ZzdoN7SQzAkwqwBOJEXK
AkOSYBVNlUMbG2PBACMTbr8kxCmhXEnTFMsphk74QxFGTuZzKYBMAGtoD7TM
ncubJhjqNFqAKQ3Yqq5umvR+ov/au1P37dl78OTJ6ovnzpw/cRxCvH3/2plD
21Ytmv7N1DEjP/ps9NixY0eP/2LEZ2NHffj+p599M3vuin2XG8Lt1WvWHKo+
OmOJ9aqvv/efEtUH1iFfedlT0mmaA3FimTpPA56Z4EoBwY0EQ4MyS6qQX+B2
K9NV0tC2U4ZBF/cMhY3hOEtTgJGKkUuDyATz4AkzEhX0or2SbD4DGeLBefA4
VFXUNV03k5BGYB4T+tHQwbok0yCS8ilwMK9ip47M/9PlN4VbQ8auOX7kTM2J
Yyd276+uOX3qTnNr7YX9O5bM2jTrq/FjRo/8YvyYcWPGjh4Jfz8Z8sEHI8Z8
PmXBqjUnHsciV5Z9v2Pf7F39b17190MNAciMLEjqNAAAKyg6yC+R5hLJVELg
OUbUdKgfEJ8K7SNDzRKGoTDRUISEEkqcBL2Hk4NX2QDr02mKjgc89jBF4FTA
7vQFMFkD66+KCYkDzsTj4GSB2vVcJgnGU0sC86lwPAOJpXQKxuRVbPuiP+0c
yOceffTFqo37frx0aM/O3QerT524Zeuov1GzZ9W8OUsnjp0w5pNRo8aNGTdh
3Ofjvvls9MfvDhv+yccfjpp76MTFm3YKO7N28caFzf/1998Hensr5WIKML9g
qTrILpZmBT2bZClGSaYNVROgLAlVNa0EN7hIKpIksEGcDjsDMd4AzJfAAVMC
FsN4HpRk0iLB8gddFE+F41FQ4FHOUHk1jTabxAN+jI6LgCKqZYBusbK5PHCr
hPSeJMkKuhKUzg94Rn2wsZTNFDpHfLZ+1w9njh45cnjPgf0nOuh4sP7yoY1L
Js/9dtTIieBgx06cNGbiuIlTJ02a+Ol7Q9//bPjfhg0b9c2S3cduBbznd65e
Kv3XH7+/7i73VkCZodVPtCxvMiB5Ab9ZnOEUC46ucDKIMnRxALDfSKhiHAeB
GYmGQXqymqWDAEEb0SL+ECHGMYZjIlJCiHkoOhwRSLfLHVWSGqtYpppOsYEg
wXNg85SkqbAkn+4FnFEFGS0L8CJoRNBLVrovdKC921STeWPjR8v27Nm2b8/2
vfu2VzWG7LXXD22cN3n08OnLxn8yduSoEeMnTps4cdLUGV9Pm/jhsA/eH/HO
22+/O/LbjVs2r9975+TOzdv7//mPXwe6ewbe9IJ4zlV68slcsYA2xGayOhMF
uBAMC9RMttKVM9HVSY7mGLTfU5DQGlHQFRIyeWipjG6obMTjxSgGbVcKcpqE
dQZC7UEKc9vdhKoLGCPKQi4jh8I0UJzIJzQpThBSttJf0DheBvoVhEwZVD7Y
sExPz+vXWQVJ7R8/nLG7avOOPRu27j321PXkxpXqs1Ub1n372bi5Yz78eOSo
zyZ8OeWbydPnfDN96pRRQ4aPfH/I28PWPSTN8NWdC7/ftWHD+d//64+fBnpf
vvnlNTBR7tWr3nwmnwWvW86leLSHi0J7muRU/8tyUh68Ck9TaG84l7AsCcIN
xORCfxnYOaWIVMDjDmAYxYqcj2IZf0t7Z7vdFwxHopwoYP4wTjCaTgfDbEJG
m1BZEgemVMH38nFBlThe0Hped6s6zGH5t99eV/LgiXPdjV/O275/x/Z9uzbv
OPXk7OFLtS4+nbSf2zV99rChw0eNnvjtjK9mzpnz3aypMyd/NvzDTz9+752J
TyN3n/naT8+fsnzlxsu//vHLQG9/T++r1wPFTH7gp/58Mp3KlyvljEiEcAot
1tGYlOsuWyJYUxAeJDgfmhb1XFdaF0Wa0Qq9L/te9WZE8PIOgP4oA0Is6vb7
OpqaXL4ORzDOg8zGQqC9AxjPRj1uUgQfwRChCJiKhAx9CT5RYNBez/KrUiJd
SGX7f/npZSFb7ivnK/5ZK7cc2Ldl14ndm48eX7/nvF1LEH5/pPnQgtEffjxx
2tffzpo9b/6CuSsWzl0w5eNPRn/x6bCF28Z88ufRG29uGT113vKabhi+MhSg
5+df+rvKL3/uKYFkATublmmfN8YkzByMjZLJgywE6FdBPfIAOwnTAo7UwJpL
iWxXpf9lRedFDg8E0Wo2r2qR5uam5sYWn9frjnASQ+LwiARDUPlQwBeihMHN
BCEpl9J1XcSjYYyJeV1hwsgb6e4S0ERXMZe0+oCsuqPfL9u+b+fOQwe2b1m9
ftWqRu+R+dM//Wx2bf2yKVMmzJg3Y9rceXOXrVi4ZPHyJXPHjf5m1tSRy6dN
WDL6o798tmnO+Jkbjqe7AaXAtJdf/frrz29+/qm3UkgbZtYUyYjLRangliBA
pME4ZMYTKs3JGhvXMmnd0sFGJAwNxGg+rQsk6E2Gxggog26QLS+evqhv90fQ
bSs4Fg3jPIuHoows0eg6Bi+Tfrc3ohTzqazJRbxejCZD3gAGpc9WcsCgmmZB
l/YVUoXny9bs21+1beuRYxvnTZl+qGr8p6vq5+z6fH7jkbXrFy9bNG3y6oVL
lm5fvXbb+vXfTx73zfzvZiyYuXLSyMlz/rRg6cz1G07xPQXIF8zby1/++P2X
X9/0g7rOdvekuTjgBwx80QSGMkCDAXYjrckgty0bxZxkIqdrGulM2gKPx/Oc
qssiQ5EkK6elzhdPahu9GOZ3uhxou0CQF3GvP66BMtBFRkqQHm+Y1IuFQjrB
xiMhCuCXY0EraIUcdIpm6sl0b385Zao3F289XH3y5JE9Z06tmbGufvP0xdOm
fXtt+6ptT07u2bT6uwnfrl6ydt2BE/sObN6zdd28abPnLP9i6KKZoyZ9NH/X
lzMP7K12vSrldD2VypYH3vz608Cr1697in0vB8rgVLKWXuytpEUWQqM5JQmS
M6lyGCXoCbP8JqcqQgK0GlgL2eeJ0LIsgoXlCHAhUGfPs7uPvBS6wyPg5xh7
nVdSsE57WDSsYncZxAoYDFEHE5/PyJzEEySP7qGRWB76AVSOYVlm7y8/F4wC
f2Dl3kNHz144evrq7ap1e13evauWTd9++Ns54w/X3a1eN/2ruXPA8x25fPZ0
dc3x4/tXz/5+1bxhQ8dOeG/GrdlfLDh56Ed/F+h4VZW1dKW7v7+3d2DgZWXg
11c50yh0defykGAhRssJwJgMiOqUIdNkVE0mCy9/UokwwTC8AaKA8gZJTgLc
ZCmMM1ISFQ89v1VPsp4OQpHiqh5p8iUMLtjhiyfSpd4ey8wbUKliCtDT5BiY
TwyLcZrGEbTIElGcAro0Kn09llqOb1yx/+jJWz8eq7lxbPeh3fvs2furF6/8
7Os5VYdvXju+4fvVa7dsPXj2xpX7dx5fu3bxwoXDu5YdWTF79jcHG7Z/8c3K
Y/tv2TLdXSDhZSSr8oVKoVwuVF6+7CuYRiYL42EkNYlhQXzGcA2kr4IWsXkq
mUxWXubxYCjOUnLazGQ0OhoD3qNoKozJpoSHeceTdpHzuAXwQ3GddvlkKAxB
4Ixk5KyEnta0Ul9ewAb3E6JNlmipHo9FYgyHB/0B0kilUhlTETPY/s1HDh67
cOnMiUunjx/ZvWqP/6XpffGwOW4ptzZ8t2b/oRNV1RceNzxzt7xoeH7t3v1z
R7ace375+oMnPywdP3bj8U2XXWBZkXdOWml0cVPVkYsYGAC2H5wsdHsAKxp6
PILhCnwaZ7SkwRlJScsXJRwDw8eompkvJQmPL4yTFJxdjGWZuEwB69GhIIHT
MQqQNUyIMsdxNAYGUeCAxFO5tIx53b5gIOC2t3d4/a4OZzBKxmgRTD+uZFKm
KIOaYQ/tOH7s1JWaqhPnfjx1cuPS1TtqCUUTMK+3Ye/K1Xura07+ePtRm9fu
9Xk7Gute3Ll45sCZ+nNb1q2YNG7CpIPVq886FLQOncnmy8UM2HZRSGR6B7p7
eypl5McyqZRB46wisXgkHFd4hiDVjCHwShx8uRiPgG+PygbYKF3wNztDMQr8
LBbCYnFN5ehI0OtyOgM0i5GySOKRGCuhzasExeogjoAB/Y62dkeHrbm5sclL
UFFflMSiYjqVSCTEZCaJLlDlAkc3Hag6dvPEDyd+vHb8+N7lK5asWb3z8OEL
tx6eObS96sdbt27euNeMU85QxOf1NdXW3rt4+Nz1o3vWfjdu5Ihvqk6vPOZN
FNMpxHjFQiaX08FoZ3pf95SK+a6BnnQSAFKNeKMgWkQabQUnSFlPqXQ8Egyi
G2Tcbn+Hk04aLGjtWKfNGQgFcI4DVRoGiUIG3GgNO8zJOJFM8zGPBxNUTebA
I0oG0p4k4W5usrW/eNLq9ITQxW6KJOLJSnelqztvZpI6utIaOrjzyNEjl09X
V9f8eKVq5/a1C+ct2bD1h5oL924dOnwXLZLdfeINyTHErH5Xa2vt7bPba7Yv
WD5jxPDhU0/uWXrQaRbT2UxGlS1UxbQEgqTc1wXBlnrKOdMqZOmQL0hgMRJd
WyajjKEp8UjMF4zF3H5np8tr93BmlsGiZNjvabc5vRFaCDbZfDjJoh2tMV6X
QHsxckpiOAld2WMIgsQYA7wDuErC9sJmb25wUiwWwWmRjmKcku4q9wy86gPW
zUIPebdtPLTv3PkDh8+erK65enD3mkWrN2/fe7Dq4qMLJ27dvdlc32zz85Sq
cAmB9LPEi47Wxyf275wAbnDY8G/3b//+gF0v5krlsgWOTzOyWS3BCya8bne5
iDboZQp5kcBIhooEQKyRMUpIKkQo7PUDdAZ8dputze9nrLRCxYI+l8vW0NLa
6XG1tjRHnTFeTeoilSgYrCoRRELmzILKiYTL7g3GpHQ2a3BqVonYbG6XT7Fk
yh0SNBrdDMCANAJuqGSSqVKl3Lph7Q+naqq27T95uvr0yfMHli9auWbz1v0H
rtysuf68ts3hgK5KyJKeSAjQVCFfe3Pd9fVrJw0bMnz4qKUHVi7YGUCrub39
XRbawmSl0R2tyXy53POynLaSppEvygJYBo0IOJyROM0rqoB24IXCJBYOeFvr
G9uDmCCDnPS1NTfUP29seHjv8YPHTcEIp2RTukDLBSXCcBzB8wG51zLTTGuL
n2IYzbBEXMskKU+Hy518WdaZcExWOU5Ey9yymn/9qpLL5EBV2Tcv33Pk8L7N
h44eOHj4UPXeVQvmzvlu5cZjN8/efFz/HPRfDIlFoGGZZ6hw2N3WWXt1/+Iv
3/vbu++PWbH1++UnwrmeYr480A1ThTNGOp3NZQFP8+W+St4y0WXABJA7m9Tx
js4w2uPI04SkCnSMxMOhcNDRafdEKfBLuK+tse7p07q6Bzev3Lj2xKWKomWi
JcekEAE/hOGER3+VS5azTCRuGFJcVBjatAzO6/CEu15XQDeQslUoVl793GOI
erqYzxQqXYVc3rP5u9WHTx374XDV3n37fzh8ZPfSWZPnLN9+vubKo9rnjS4P
uvOVUeREMqdjFAj8hvqn12u2TPrwb29/8OXaxQsWncbKL/NqvqfHBPGod3X3
9FcyMIbJfKkMPhptilFAbwrpgo6HaIllCBzqpYhg8Al/OEZGQkg0U6CQ/Z7O
hsePHj99cPfGg7r7dSAI4mjdjUkYgsBGbJ5oJNNrSoW+TJy2NHAYIhnXcjmF
pnEqU9E1VVDS5f6e7oHXpaQEeJxIVl71grJt3TZnXdXRg0d+2L9v26a9u/fs
XDF71oLVh6ur7z56XOeMxikywYZiIpwouKygw2FvqT19aO/MUR+8++GsDYuW
f3eM6H2T48xilwnEnen76dWb3hRwYDKVzlqqZaK7xxK6ouZ6snEykeCYeMwT
wAU+zsSjLj8hMGi1iBcICNPnsD17cOvmresPWgPuF6FIJMCmkjSRKGXiAWeA
YtVSRmCtvgzL6qJiKOCx5EK+lDc4TlcBVwFXwJOBwNVNiVMUNlEC/C71eDfO
mrF+146t247s+H77wUOHD+2Yu2TV7sM1t+8+rGuNEEEsREZpVuIkSmLASje2
1NVd3rZr/ehRI0csOzj/u0WXwl0/FRUtV9BlPZntqaBLRCAvMyCqkwZYchZt
rVOlTFcJYjY5VqT9gRDa2hdydPgIQZHZMOAOEQn6PXa7rf7RzQd379hIytHi
9gfCalYK80UzGjCLKT3bndfVrp4UalorJQsw55lUMSMrOmB2qgKmOpM20R6N
BI/sImWgC3KGbde3szbs2bB53YYtW7dt2VO9d9v6ddu3H6+5cv1BsxsP+wJe
ryAHA/UvwEZHSDbsdjY0n9+6/ruRH09Ys+XbBRuvO/OvKlYinTH1XB4IMAOC
E/RWLg/StzeboHldkRWFEq1cFzgnSTFlHN3/wrrbOpxRjOIFlqS5IBTLbe/s
cDkcLfVP7jUTpL/dCV6XNLWokOb82qteUzUymXz5VS8CHSmb05Pd3XkroXFM
JsVGw8mBnjxam5fNNLoxNNLZHmSTuXJFaVo/Y9zsNZu2rduwYfOOw1VVB/ft
3r1916kzt+49bYpFop21zR0uj7ujtsGPdfgJJoqzUUfr2Q0rZ46dvnbbt1M3
XXSk+otpK59KQ3/k0O5BS9WKfX2FpF7o70syvMzTHBcNcZlKuTsjwzTR6K4W
0t5gD5IxjGEwUtEIP2hGt7/N1hGIeG1Pm30gRfzOAIYJmpAqxPlCysjqLC1Z
RfAlYISUTCGTf91fzCjQiuVSIuSRc/l0Ig4UD3pVFEXS3RHQewf6klz7zslj
xk5duW758nlr9x86c/rs6WNrdlcfu3Lh2sOOjoYbVy7ecrR3BJ7cexFsfPDA
iWNehzfmrD20afE383ftGj9p75lnAiBx15uXWSBBtOGllNGSPW9eAvlVerrT
HMvjOOCDjzJSlp5QCALdqx8niIjfG8aiuCiRRMJgwxF/IEqEWxwxivI2N7nB
xvtiMhmLK2ap2yhm4omczkGvW8W0aapqOgWt0l/OWoYuW71dUsAdF3lQDUhs
M6LACXLcG0u9fJ03lbp9s8eNnDj/++83Llu2q+ZiVdXpPft/rD538fqjuocv
bmxeta3q5rPnZ44dP3flzrWTlxsjHtujlvaHP+6a//WcnRvHfrd1500HKKKX
v/8xUOmvdEFIlWLXwOvXRasEseYsARAzJghMCE8CictMO1hyCicgDEGg4iSr
6wzJK0w0EvGGMAL3E4pMeNvafbEwZlh4UDASuZc9BZWkNcPghATHm715SR9c
zs3KtKKrvFVQfDZ7lCDCnhAnUKEgSUUiDBuLSl3doMib93332XvDPv9q2sqN
m9afuHXux7NVF0+eunT7ad0LR1vN8rnrNq46cuowMP/dZ9d+OHyjte7hlZom
352TC76csXXxhAXz15xpjyR7B/7+60+9veVcrtLX31V6/VN/WpEH711FGwdp
QQCXpCeZOIf7QugecJYMEWKCoghJlsDwyTTNiRJ6IwtwtjKH+TyBaJjWuaBH
q4iJVxWTJflclwk06XOrb9I0lyoagqFGo+Ck4poWsjuCJKMIaN9v2ONjuXBM
0SRazqUEOnDv4IKPQYx8NGrq+q0bfrhx48rJmqNHqx811Nc9s9leHNi4btHK
/WfOXrpyt/7xjZMnbzU92b9l27VQ3c39U+dsmzPu08+W1zS72O7+VwNo10sy
W3k5ALMC9iErcYlUvpDREVcpMHwmS/MCL6roFl2NDsdwBg8RtAKOIGHIWhLA
CdR4UucTAscA9QcE0ufrEHs4qphPM5jUWwIUCbXakm+seFzvzepJIRzXLBlG
2eEKxOKCbhCh/9lQorJc0lJZuaDz3oar6+d/8f67733w4aiv9x7bce5Wzb6D
58+fftDe9PBZS/31Q7tXzF+15/yFKzfu3H50+9qFq/dqf1j+w1MA0aPTvts+
ZeLn09YdvNzqF7M9vWCPLCNZHOjXaSU/8HNPwUikukpZQM90SjcrBROsg6an
coaqZTSeQne3SgVTMnUDvZOHmQaZh3YCWdmMwlJRMFO+gM9H5+WAmLfiAdyQ
YtGwrz2S7zE0ibd0ReMwtZxOYDjmsbu9EZKX6FgsRqP7KQVJU5VkVovU37m0
f+2M8R8Neefd94Z+MO3k4eMXqnbvP33u5qPOhw/an166tW/pjJnf/3Dm8bPW
Wq/H9vzG1Utn9mzeebHTVn97/c7No0aP+HzB8j3XngXUUgltyTKNXHc5Iwm5
/u58pZzOdvdkVKtQLqQUM5HgednKZdOqwCV19E4hlFb56Sc9nU0pIk1xrJQw
FR5duevLCNFgxO+KUhESkxQXneFD6L7oOIVueKdlLZfimWiMitP5XlPjKJHo
tHW6fX6MA9knl7OGjq5iq4XulK/u9rkzBxZPGPruX996d9iwIUuu/3DixMGd
x58/flh76rTj9q6Dq2Z988W8Y7cILZ/KGDkxfOPKiTXfbjz5zP2i/cgP8z95
f/g3q/YcO3zBrybTSSlh6JU33V15wIT+spp/2ZMp9BRL/T/3lwqmkgBfJGf7
wAOn1VSXATzMm6WXZTVXKigsL6M7lY2EwHJmpSdrxEOE1yXwnAiI2BHjQbv5
KEkRSJ5scwvgj6mw3xEguXSKVy2JCza0OH0dTS6ukFUyAyVQFZIgKbluPeJw
PH1+dcmkj997++1333t36OeHj566VnXkWvB2re1x3YNLOzfOm/DFZ8t3Pci8
+f0/fv/tj4GScH7d+A+mH3oIr9Z0bubIEWOXb9135vqDgDJ4papQ+u9/lXt6
ckZ2oDehVHphEgt9AwOvfh0AesxkJLn750p64NXL1z+VdNWyAOwVo1TJp8SE
ZcgiJ8moUQulXFaLc0Fc4WklaRItPgwPNDniHHq7GY+d4YlIyO7xucMEJ3Cs
Yop0sLk9RriclN7dW8iV8qYKFli18gXB3XT3UtX2uV+OHP7Xt/727ocjp2w7
dfTq+cvtraeePb3XcO3yru/nLV/y9bxtN8u///vvv7/p70/6ji3+ZOSMTacb
muyBq5P+8m/j1/2w58iJg3Z4uXJfz+tf/9//0/3651d5oPoEmbAkWrDKxfzv
//1HX77S06vLlZ+L5sAvP/36a182V+kd6E/xVimfyaIL5pbEiKJqpTO5bMKy
VCkSJjk8nkrzopnLM+1tTicgLu5m6Y66Fy3oxjcyFor6gxR4FiZAKWyITOYr
EFy2aCnodspkRg+3Pj1/9uTa5bMmjBkz+pPRI0ePn7x514XHD50/rrt8rubm
w2s7N25fOmPd2UtVz6J+27Omu4fWzJizfuqE5Vv23n7QEb85e+7Yb5eu27xh
1fIbYFtf/v3Nyzf/8c/X//6vNwUrbQmkIMRJzsiXir/8482rnmL364qR709x
+Z7Ky59edZf7f/3HT1aiqz+XLViGldFZQYP/C+CvJDmXSViBAEfFk7qUVPSi
7LY5A9FozOllcE9rbaOjw4kRztYg5scIMROPK0kFziAHCKwXK6Cf0HsOZLmm
WzefnD+6d+XcCSM/HvLO23/725ARs7dX3X3RdnzG4ceXd9ac3L5+7bx5Sxz1
s6fNnDN17ISvx4wa+xVU9PvDNVeudERvfD1pytxFK1csWLb8OM5ar3+uFEr9
//Xbv//zVcaURBpjeJID+Ch29fUW+stWsqs7VSh3MVKp2NtX7ioP/Pof/Wb5
598GcqVKT19ZZhNW3jJyXT1lLZHJK3mWEWkuKUWCbZjg9/kpcFlBuz/Kq6Hn
jS0tTo+vvckV8dOqoUdw9OY+spZNgm8pdCWT+ZyVLeaY2prjB47vXDfrq/Ef
DBvy9nvvvff+519NP3Ctzlb13Ya7J/YfXLth96Y1Gzc+efbpO+8M++Cj4cM+
eO/d4bO3rtu48/iFO7XO6nFfL9q0afPG76cvPO3Revu7U1ZX129//+nvlXwe
zCdJMsBPyUw621XM9VRSOujCVKmbp5PZUjYFMPn7H93Fgd9+7svmB377qcTS
RgascK4ftIGkK0IOciQGuAQedHHhehdalhFjfiJE6Jyns6mx3eNtamh8VEeq
ghClYXgBg40EI6XzuVKppzub6zIidecv7d67acWSyRNGfTDkr3979+3Rnw2f
uPjwg+eHFyyrPvugatuGLZvXbj1QfX8l8Mf7Q4cNH/Ln4Z/P3Lp956adRx/c
b9n20ZcTZu/eNH3lytU3uN5X/d3ZTFf55T9evoRXT8maKspa2sA5zUim04Vs
Ksn6g5yVpUNkAix+vlDu7gMR97IbjGL3L69NMWHmi5aWqbzuLYBn5AsDKYmL
SSbl7gh5mjs6Gpsc/o6OiNujmrirsbYu4KurfXi7mTU4luZkEA6kj8BwrVzI
9Q10d6dT5ZTv3tmje1csXLdywZQxI4a++86QER++N+rTj2b/+PhW9Z4zB4+c
P7lv8+qNG3cf3rNt4Zr1I4cNHf7hZ5MmLNi5cfOWE6f3XbDt+NPHk2fMW7tx
6dwVdenuNz/9/KrcVRn446feblNL54uVIpC1igU5RUzlLRlt6/DHZSkhK7me
3t5KKZcxWUYv5LV0rvyyZORyuXwxlcoXe1+VTCslpX79JS0JphWre2EDU9H2
7EFdW21tc60zbflsrXWdRNBWe7fRIZaEwZsmwWeFXV7GTBtK96suKF9e9N7Z
u33VysVzJk8Y//nHQ4a88+mwd0Z9OPTjz/c9eXZxxaI9R9bt3r9r/Y6NGzZu
WTrk/Y+HDn3ng4/e/XDOmqWrN+3avPfarnu7Pv5g/HczZ82bO2XLLTLX+/d/
/cfPIEBLvWWTx1mFR9JZTYmEAtRZKSRMg/JHaTIm54vdb/q6cinTUhiWBZwr
ZYGK032lYiGTTL981fe6nDYTdCHX/aqsKAk27PSAmXC13b/ZXP/ggT33SyYa
waIxIoZHGtw4k5MDYQbtyGDCnRFV5Gix/LrX7OrLcrbze7ZvmT114rhPR34C
k/XRkLeGfzDkw5EzFh54eG3vmpWHt6xfs3nr+sWLFqxZsvCzoUOHDRv6wcjp
i5eumr9m+7wd+3Y9nDj0o0+nzpz21bdfrbnhMvve/OM//1FWE5plMiE/q1CU
FCc1mPhCz6CN0XWAQyHcHtN5SbFMKzF4PwQjWoXufO+rfNfLnnzKzFd++73c
lzcAfBOs1l3Jg3hNZmSSigZDtsf+SFus69c3XIRXeb+fYCUsLjKyEnNjskwT
Ya+L1wSWV7NdOaurnFdazp6o3rZ8+qRRn4wa8cHQD4eChBk+bPjnE7+ef+re
6aptKw7sXrdq2cJ5382cOWfploWfDP1k4tdzZs+ZumjenNW7tqzdf/Hzjz78
cNTnX48b88322lChu7vvt9/6u7rL2bQM+h1zsUIUlxJqsaJxIhmjwb2xrCm1
eEQiHOGTlilLip6U9WK5UPr5dd7q7imZkKQ/fu0u6eguDk+IySbFBEMzBUtT
BZEjPTImVjKa2NoWZENt7oAdN1mwk6IejidZXyDqDGmaoKcUOSEly5mUZD++
ec+6hXO+mDDqkxHvD3n/rbf+/Jd33vt89Jczvlhy5vKZfYd2rtyydcnc6XOm
TPzy68WLvh6+aO13U2ZM+WbKvPkrluyourB/+PsjRo6b8PnID7/cfadOLGey
PW9e9f8E0J9NJBR3iI0BbitWOkkzMEdZAb3ZhZ4N2EUTvVkd2kqvJQumVezK
FQb6k6Sczheyr39/092XFhOGRHh9lK6HojSOKYmUgAsKiICoITmCrmZnazTQ
3ul0SWWDirnCOZnvV124xCR0SUgX0PuQ5QfKSa5u/9p1S+ZOGzMWqgfY+O7b
f/nzX9/9ZMSnX4z+dNv5UxeO76s+vnPHqvlLJk2esWjevFmfrfl+0ZRZM6ZN
nvz1d3t+2HNm2btvvTN6/KQpX4+bWHXjBVfKmNlyqVAo9XRl8qUUHRfIqB/j
NVMShGSxryKoKY03UngI5LFOs+DBiwV0H3pXrtjz5rVFESKQfPkl2kNtJGWa
9Af6K2rExdDBvpIiYKLZVTLYqMOHd3TW1WJ+dMeeUHpZYF1tekYYSLSRktmV
NtPFYkLmhVRPT0loOLN98cwp3/x/HL0FVFRr3/7v8xyV7u4Owe5AsEUxULEb
uxVFaemG6e7e092dzABDo2B76vH0OcaJ933e9f9vfnuYNTAza8F3fe/7uj7X
7H3f5ETEh0eHhvoH+PkunB8QHBKRkhefe/Lm7ZPHTu85uG3butV5mVnLVi1L
jVu3KjczNzdvUc7KNVv3le5JARU3KiclMyNt3eVuKM1h1A8M2a324XG7RmFw
aOVWD4vSD2NJwGmmtowOu0efjw8NmhQqg0kh1Xs8oGoOe6yOUVBHZmZsagFX
olDp587Ea/Q2M58lROB/nBKzMVwpc8xIk0vMg68nTVqWAIQieheSj4VRhAKJ
SD+kJWO4IpFbS2Jw9XYQJmx2vVoAKJ2DQ4q+6wd3r1+cmxIXlxgfERnu7+vr
4+sfHBqTsWRpZmbRiZ0Hd61cHB8Rn7NsUWZcWHhyRn5qUkJETHxcQmZ8bu7q
tNDgoNjEzOTU9JR1l55CcZYR68SkzTI0YtNJUeBPE6/e2rRqLFmg0NmHxl/O
et1gPnSBv12jl0vtI6PTUyMTz9z2cbdzcHLYYFYAMrnS7HQO2lV6h0YoUmL7
htwSej+Gzzab6Qah1jMyMjasmTuXKOeweTgCAUcFC3FbJWwBT2ixyMVChUrI
FYESIxKINBaP1wC/ebp05aJ0UD3DwiOTw/z9fEBE8w+KTUtbkxUbl5q3bXlC
UFBIxpkmUl/1tdMdbTXHtyyJCQ0LiYqKiU8MDgwMDI7PXb4sOz9j6dWGhlbO
2Oj0jNs4Met2GRn8ibezL98/n5yySyRS3eCzWa9GrjGq1QYwoBvMepXFYbG6
Jt59O2r3WnTuZ073oJ6pMMhUBjP4otFpN+os3F7ALiQjkEwMf1Rk4+kdPJPX
oTOp6VyTWcblifgMlNxsG/EOzO0LaHebjXKxQTn3eRUbbKTJZHdpEA03D27M
S0pNCg+NjYsOC/Tz8Vm4YIFvaELSxuVZmcEhsZn58aFhK4rv0uR8KpqD7646
W5AXHxUe7AvOVP/AgKCw6LicZSsysosuVNY1scA88GzM9fLrUY938vW309ax
N29fTnndJoXOBJq4UGcwOYaGdEYDiL5qtUUrlGgcQ06zRqnwvvYY9Q6dxsCi
KZRMkdpocTvUZj6SqiWTxViARBxgyoW9RJ4IFJpJA0+lMurFIolBxtI5B2zD
XvPIhMfhsioFIqtFKlCLmELnoNVos6thNTdO71udlhgdFRkXExkc5OPr67vQ
z2fhwpjcvG0Hd+SnhGWtyc/NW36qBV5FQHYhnzw4XLx1cZz/v76av8DX1y8o
NDAuMSc3e1H22uP369sJU2OO0UH32xnn1OTM29khk/v5ixGVa8iqVIDMolS5
nM6pV7NOsDYTKANWLVcE2p/aKNUbn78Z1HO1HpNOwFLzCUyt02G3qI1MLJFL
wOrQWiVeQlWb+jsYNJ5c+nxCbZEbTBLToFvvGTPoTaMTIy+9GqmBT+eJbA6d
QsRk8h1g4nzpEkBqKq/tXL0kLT42KjEhNCjQPxCcgX5+/uHRMbGrHvR3Nh/J
zlyyavXee3AsEeByWJD+R6Wrkvzmf/Xv+QvBNwaFR0YnZuclpC5ef/T24xrs
6Oth76DnpdcxPT01aLMOeEeMUpVeNbedn0AE5mzT6Mu3w1arYW6zjkGHijt3
3afSYrK6nk9ajDy506w12jRCClvjHBk2mCxsPIZFhfP6OSYyna4xUTv6RICA
b3znMbrBHGR7Net9MasV6R1KzxunlGvlsxQam9WoBUlAqDO5Jkf1fET7g0s7
lyzNS0+MjogICQsL8luwwN/HD9SY9MWFZRdYDpMAdqlsy7X7XRw6S8QiYZ/e
XR/j4/sVeMwH3xYYFJkUl5iYkrp08/ErlQ2wwTevZiY9z9wGy8Sk26Aw2Wwq
nlYhFmk0Qol+wKbVecfder1l2CHXmmwO0PSFMhHfMmC2jnrMOqXL6zTYQPFk
iYw6k1au0GsAFIcP0VF75ACRolbz2/utKoDHNL0dGxlzey0e9+Rbt1hp0bI9
r5/bAYCNl9vVAFejojOUoMu7jBIqpLb5cdme7UtyF0WFxcSEgYDpFxjgO9fD
tNTYpdUG19T05JCJrxd2dLRQpVwBsbQw0n+h74KvFoBHgL9faHhMQmRMfPbq
zQcuVVQ3ul6/m5ycfua0uZ32sSGTnqcE7daoV1gkgHFkatDlGhrQgjPI63Fo
xVK1Wg+im4ouNLptE7PPXRbv9NiAXa4z6O2eARFXKFJPPVfxRHC9qVMqIAIj
aikTRRUDDD5n6PuR6UGHxy7yvrSZva/fT49MjPLoMDSar5KzkAKXTS6zWT1O
pZCBbqu/WLJr3crs9NjkuNDwiKCAsJBAUEJ9wiN9k7becjybmpieHpvwujUK
jcGq1vBv5EQu9Pn3gq/mL/Tx9ZtDgghQlhIWb9i0d/+Ri3eUszPDA8NjDs+0
Veucmhm22vQisVElUrBYCsfEiHfQ7naZNO7BYa8KECgGxgY9LhufJbNYRkYG
bBb3yLjXyAVU7qGh6SG7zWKcfalikZGawXYaC8V/6ZJwOlEDYgagGHw28Xxw
cNJMU424J6yeN28mp0bwCAgawxHrlTS2cmpIMzxmV/GFAO7J5fIzh9YvT0+I
jwkPDo0KjooOCwjw9/3KNzF9c8HTqeczs6+mp6bGhkZHwKZY9Ub8xvAQcPrN
B7sXEhoRGh0SExUVF7NyX+n+LSXHzxpmJz0uh9nlHlbrLCPvXw15Rkafz1ik
VgMgd7hsQ96xcY9cBXrggEapsQzbbR69ms0BJENuBRdQeEDxmTQKtFajc3Rk
wGA2jdm5HApKYeroRqEoE0NGCY7soZGppu9HZ7/Rm4e1Uud3JuOoVaaxjo6L
OAwCnae3yUU0pNji9rqUUoBP775dcevKyT0rs2Ijo2LDokKjk+PDQwP8Qhdt
y08vKeC6X797+/rF7Ivnk2PjQwMjQ3aXoTItOHr+AlBdQMwJCYpMSAwPjM1a
vWr34YKDx8uVs26HZcBkGLDotPaRyVcvhk0Gm9lqmv2PSm/Vagamxz1COuiG
c2dIDGaNyGxnkzEslmrKJZOItc/evPFOTw/bwWgEzkm5SqNyqiBUDJXV+7SH
gjRbzS4lA8D0EPTvxwff2KxjOpnmh2nV2KBWxxVavXYNwBA67BqThCpUOB1z
H2pzME9r62/dKi1cmrVoxdLEqLjwqLiosNDQpNz8pICinMM41zffvH375u37
lzMzM1PjIG6MTfIPBYcEfLXAxz8kMio4NDg0JSokftnWlUkbC/YdPiF6PvFy
ympT6+0K3cjMiPuVlcaRGcwj332rkWkkysHpUZC0RWqXBQRrkwEAzNa5/aMA
tXNMKbJ4J2fG9KaBaTcY5BSmAY1cCwgcEAwUR+pphyLauAquecSMwxBAXRlx
Dwyb3U493fjtqMsxNQwIRGqnFuBphkZGXsyMDJv1OhGDxGNiW2pvXyg/dLBg
aVZqbHB0ZFR0ZGiYX2RsUHBM9taCY8hX37z+5v23H759/+bNm5ezL148n519
1pPjHxTouzAgJDoqPDTQD4xHuYuXb1u9YvWW/cdp30+OjqtsDrVAaHW7tArn
tE7u8ng+fPRwJBr12PvJ2UGVTAE6n9VpG3MLAZmUxSJDWTqzRyvQjD1zDdi1
psFBndakkBs0ajkB5aCTYb1UFIHe0s9hEHlqiUTANevEIpXBqZB4h7iUmXeD
tlGjwKEQ6xVqx8jLMQ8IqU6LXsnn8Nh4WM+dE5fu3zy9dePq/LS03KzUqPDY
4Iik+MiI5SXrlhwRzs6+/v7rH3788OHb7755+/rN23ev37+bqs5aGBQWGBIR
Hh4K+klwYs6y/GUbNuQvX7vlCGLQM/tKpTVoNXqQVQwK6cDUxJRn6uN3AkCi
tg1/PTFu5ov1dr1zyCo3OtwGk1bEwaEZejVYst45rFNqtUqjx+m0a2g8pdLK
EHnFGAiEBSfyEGwy+BSlj83tqOKauGSKSQi4LBKOfZynHnar3EqL2ep0D4+Y
XDKTy6BTi/hMOgfTcP/O9crK66dLtm3fuSJrcVZ0WFxEUGRsREzi8aK8mHP9
3sGv//Pdrz//8uMPH77/9utvvv3mmw8//GeosdAnNCAIJNaQ8JCIiJi0FYty
lq5Yunh5yckmyzf/vBbJLVo5SwUKuNHz9U8ff/72u188PClPKtVMjrmEHO3g
uB6kD4Fu0AuCqMauYHIETKVDZTFbdBydgcVRGQxqrcw05BlzGG0sZA9EQ4Ao
OH0wusAu6WMYqM1QvV4K51hEfJVGaXFT+L+MUvQ2i05tMlqsertNB+ZfgYBJ
ZDMx3XWXL96tuHLu6MnDhWsXL12emxQfHR4eHpWycvHGzBUPu0ie7/7zw8ff
f/n5px9++M/3Hz58//Mvv/72teB6fkRoUHhYXGRIXGJmVnzqkrxFy1YsX7X9
QuvQr7+Nmwx6hYQueT1jU7gmh6e//fnTryYpm2NWiLSDL0wKz3OTYUiglVvc
FpuYx1YrqTQeV2q1uOxem1hm5RFZfLF5Ymbm5Q+/qIkiXH9nt5aN51Mra1E4
KYPANNJJKJleCLD0AIYrZuq0WOErUofCIeLpjHaL3uB0WVV8gMZnkrBECqT6
/t37Fy7cunNyy7Z1y9fl5SWnxEWFhmfmxCem5Wy8CEGZ33374+c/fv31p59+
+vDhx59/+PXTxy+/Tlxelx4VEhYRlhwFvi13dVbupuU5m7fs3nO4c/ibd3a1
xjBipgufWwxvvDIG3zX86o/vTXjRkFxoHh4fGX35wmJ6Pqo2O602u4pMYknZ
VECgHnk+OfluGvRDKcfonX337QubTOoQk4Xc/o5mFgNLluJoRgYCL5SKOTQa
IGPQSSwqhsRhcek01ZgYweGy9R6nQeEYclj0QjqVwqahkEx0T0XFzQvnL10t
239yT8GR9UuXpmWmpkQkxIXExkdn5O3sgYunJ3/489PH33799Rfw+P3n3z9/
+fLXb/TVGxLiIsMTYhOiMrPT89blL965acm6bSXljbZxuwEQDr5S4RBUhXHQ
ALqVxml59fk7ud6rZA+4eGy7S8XQDX54+9w1PPt6TMGTymQAT6kfnnQPDdun
p0F89Y5PjYw6lDIumdbPYPA7GmHELgJj4P24CIJh8LQjI3KKVEZVsNhkNI1C
QIl0pndGNII27DCITV6r2qTg0kk0AqwfziB03Hr06NajKzfPnz9xoHzfxgNr
V6WnJ6UlhkcnxYbGJK/uvMAG+D9+/vz5t4+//wYen37/9OeXP//59Ork/tTo
1KjE9JTUhIylOdnrVmzYtq64eF95Jc/sMBksRg0e3ied1YGDhCriaXWamd9n
zFI8zWaRDnlBimXxn/88MDvtGvfaFBy+XMhXCEUOIZWhAv1dp7U5BwZtjqkB
HiBh4TulYjSxrwkOeJ0UMYJMRdCIIpvJxDQa53YHEwpxUDqVIxvlk/RGrVhn
A1OXjseiEBg0WBOURuqqb3pw9+qV+9cP7jt7auvBkt1Ll63ISkhISckEnS1w
S0NJTf3D2V8+ffr48fffwfvnT3/8+eff/3z+nVu8MTQ1My41Iy13Uebq5XmL
lhQtX7VpxebrDLPJKBHZ1AgIRDqrlfBpJCGVJBdY3v40zmNQBZ7XQ0NGixpH
s86opgdJOK5OSSUQaHSukKIGoIB30jW3AxgYD9US20s3qYMOgcAMVgAFQzB1
QB1bwWKSUF1MKZ/PsNAZVAqSIhdBATGjE7AYppRGs1wiU5vEHC6VyMS3NXcQ
8GhIVXXV9fuXTp68fObw8dK9O4u2rshOTcvITAgJDwvfe2rb0Xtlzh8+/vH7
J/Dr0x9fPn/68vdff/31w1R5UVx0Gjj3chZlpy8q3Lp28eai7JzVRfvhxrnt
l3SkPiiJMywXOzh4gIoVSKcnnv8s9fCEP7ioHLFMBCiHpi0qLa4Tq9EKSDQG
TcinSzVM6/RzG5YvVRk1SjlPoTVy2uAeBoY0qMWSIFi3uIco4uFQJDmrrRnK
mtvVlobpJPOFMnZ7K5KmnnUa1WqhVK/iE3B0Cgne3tiJw0AbHlffvnP39Nkr
pUeOHj1RemBXwablS/My8kAoCcssLNp4dO9N/Yfffpur7+PHj3+Cs+/vv//8
8qMddiotOishK2tRdkbWktzlaxcvyVmVnVR8Ba0fHB3QsKE9JBlHxFRbMDwG
g8fWTAkFY/rp1+/1ZDIRDe2jCo2jKoADhVFofJV2SEYm8Ogsz8DYy2mDiM1X
yBhEBiAWcdidhLdyKpYxKnj0pFtJbOLLBd1whsRDbEIQWBQcnsoicIxCeX9d
B5ZIUOhkBocNjPUkLAZPoyIbnnahMYi21kd3rl4+fvbo4cMnzu48dLBoa9HG
FStWZcbFxESvWJJTUrKlgv/6l18/fvzj06fPH//68uXzX//89fknS9edrcEZ
2TGJ2elZi5ekJi7JSctOXrp042Hoi7czJgW1t5clYzC0Zh7LKOHQqU4zhUi2
jv6ohckFhE40kULmerVsaF0/BYMhy8c0eCiLQnJ6p98MajSAyGyg9mEoNBKN
9KTKZkZRhQMDpOYquYUGZmGRiE20j3KoNC4eihSxOtCDNCoZpqLiSDTJiFPt
1CuYeDwaTWF2V7b39JGp3W0Vd+9eu3r+2MGjh/ceOrZ7+9b1i9dvWpmfExcX
k5qVsyV/8VGU7bvf/gB79wm8/fnXl89/f/n8s+JWycHM0MyMiPD4jKTk5NjM
ldlZ2elLCg8+nZjwaIR8NpYOEERGpcpoBs1WP6Xkc/Gil+8oZLsc6KMohAQW
GPE84Bij4TCAS4VnOHWAd2RqTC6Vc0DoljKZLDFPIFA2dg2iGELtMz3G+O6F
QWHzDpoUZNyQBcV3yGAQJe1xNZWBFxMmyBgJkTfhNlnlQjYOicIgIZ2trU09
LFJ7U0PV5RNHDx05XLL/6LGdxaVb1yxftSovMysxMSEqfmlqQuamho6x3z5+
/Pzl45c///oTHJxfPn38hXxvf0ZmSFhqQnxwWFRkekpSekZmbkbW9uIqgVwr
FRgUVBJLIFYbFBK5Wqp0z9jo6DrB2++HuGDJc9sLzV0taxiQMqkQGKAaEJGl
r0Z0g+4hLpJhnJ50SkEX0FjNVg3JSdIhK53ct073O47SxjY7iFQUEq7TYWgi
gARj02sft+BIQsIMHRQUqljnMsi5RDQBiYSj+xset/SjCbC6qksnTp65cunk
6XOHiw+Wbl63Zsnc2ZaYpISs6Ijo3KSkNTcqLb98BKXly99gdWCJX778/Dvl
UUlqeHCIX2BoWERiYkpmek5mfPaqrKXFD1RiLgEFx7BwPbKXTg4eh2eygbmt
vcmIAd2zGSWBSufIlVoeXSNB4mlsJJZo14sZcu6H12r9gAyFcn79duYbGZKh
sNj0WlkbRiOEskcJ9lGT0y5WiRljgklKGx4mZBJhvVo6gQptru3DCSUzABZC
JFANepWU2QPt7UdBEb3N3R39ZFx7TWX5uduXrly5euVI2dEjW4rWZeZlxyVG
RMZHJ4YFBqTkJWVsKhd/+8cfnz/PKedcgZ8//eDgXi9ODw4NWuAXEByTujgj
I295fuaKgjWr9t01cwksBh4Pa2/sM755+XrINuHiW9lMj9dh4JqdU163S6fQ
ath0k2Jukx40SWgSYaFs2oufFGNDVp3lu7dKYHTWDXBVFiuVR6bzhCqnBcud
NNjGZ9kEhldlIsC4WgCO6cFYqL1t7I5aHEXmmmIhCGo9Q69QAsh+ZG9/P4pY
U9VaBQOoba3XLt+ruHP12vmjl8pP7t6xbdWyrMjE+KiIsJCwsMAg/5TohMWb
YK4/PoIm/+V//v7nzy9///F5jNlZeSAqKMI/MMA/MTF9yco1y/OXbd5VULjj
QKV8YtplcaD6aVg+0/lBJR96P6UwEBhs8xiDpbXZDSouloTFI7CAWvRsgAvj
jumEBA4ZJXCzn3nHJiafaWUUkve1lEBmmw1CIxrGpXBdGJLJ7DZa0HSSVsIg
A1QlraOnFsElt3diCe0oNuAaJvYK304JtVIuqb+9sb29C9P1qPnxEwgD+6Ty
wu1b986cOn/y5NWLZUeOHirakJOVFhMLRteI0GB/v6CAqDU5t+W//PHpy19/
/c8/YAs///HTNPXUttKQiMDAqJjI2JTkpEVL1xSsXrJx+47iMw2MMa9QhYFg
STQeT/+ay7VaHV4zgcNkGXRWN13e2kXl8wgy/GO4iv/1l3E9y6sk0RSMPpiE
4Z2xWb0anlBpfjtmVAJ0k1P6QssXGqxmFtpiVUnMIpxYgYQgpSIckkBveEKi
VGIazOxmiUTCJUAY3w1I1QCJCGtuaGrvxLU9eHy3GYlFNdQ+rL59sfzSieM3
Lu47sP/EsV1rVuXnJsRFh4F/fyh4hPslJB5GvgPB7O9//vn777+/fPrtP2/h
W2IKkkJDw6LDEuPTk1LT8rNzN2zdsL9477kuQOLQ06BiqVCsHx9+adJ6B969
1FPoDLGU7p2EoIT1os/ufrGSwxPBXF8mnDS3nc5kQDsZQxLXu1GPSW3XawYG
5aaBAY3VyR6sxUjezaoGyZppBgoYAUGgB0lzfM3jOsStLURSOwEx7e4DQxJd
xWG9cDt1dHrfk7qG+nYYorPuwd26Rgq8tr3x4ZnjV64dv3D6QPGRQ4cPr1u5
cfmG1OSUiLCI6MjA+KjQ2NDQgivKHz7/+c9///s/f4Gs/eOPrztTIhetCYiK
ToiOT89Jy12Rm7d03dbtB8/cetgtMXicVJFa6hrgiY3DnvGXL+xqySCPoKR3
8ZwKLJ88+yuLpHMoFWrZyCcShaF9buUCPc1Ml1TscU7YlN/+ZBbC8SKu1gai
D6XKMD79gue0ap+NEqA8h4pkHydLjWqXvpFKaa6sJ3HfuoUSFvfVCx5TzDeJ
cKBcttS2NsAQ9VU369t7yb1NLY/PnLlw+fLJU/t2Fxau271ldeHmjTuTUvLj
wiMiE8EsHx6elLZkf9uLj1/++e///fMXmAV//2Hyem5c7OaI6ISkjLTUpJzc
vIyExZt3lxwuPXQdqpl2q112g0AsoqOYg6aJCR1Tb9ayJSIKyMVGlURms5jt
Lj3PKBN7P9EI9H66Tkfp6uLKFIDePOWlWTUqAMkHoUAMraM8bn4xzbOrOE61
7d2sR+6xsCQAETM8zW2iuBkiGkMmfPmMwtFMO7QsnZcsBPq7O6srqiqqIZjW
yutPmvuI2LpHd48fvXf70J7Na/LzszYW5axev2nLkvRF6REx0bHRibGRsenJ
63Yc6HL98eXv//7f//z5+6+//PadrWZTQtTyuZXx8Sn5KzMX5aXFpGVmLy/Y
tuk8VjGtpeo93IqnPJvCODVuMOkMBruEbeLxTTKuSD5gG3K5jEoBi8ZUjE4J
WTTi2ItxXBNSw5ebzSrzsMYmVXNpAAcgclXsnjq4jkMUTijHuZTRsaEpg24A
z+UQ+WzRQ/L0j7+NmlXqqddm2ySHJpCYbEoNltBTf7/6yZMOKKq96t6jZhye
2Nv89MG9K/t2bdmcD5p0dkru4lVLF6VmZmWkJGQkp8TEJcekbD5SRTK9+Pnj
n//7f//98umX337+zovbkZOUlJWdEZ6SnrxkcXJchH94Ukpayvqy6z2aMYNG
IidzB0ZtJq8ZDgFUZs+gg6+nakTEuz1s/ZQZCaOTVN8NcAccbBaXjFYNSDqb
cHqTUCmjKCaZTIlIRJURcGjF7+Ncvkyvo6KFJiuf7nzNoXCVP/djRIgu+ihP
8fk5j8ZjMQYn7N7hd26u1KmxWGkwVPP9OxefIhjkjqf3b9QgEUzokyunjx8u
WLtqbW72kvSEuISICLAVS9KTkuOTU1MSYjIXxW6skw49f/PDbx//+b//fv71
1z9++mEGtn1VVkZa2qLUhJxFyYnRoaGRCSnZa3cfuVsFF9lso+NC1wDAlg5q
6rshzVytfmaEJcB9AwCUOq5cRyPWdWPU77+Wq0xGAZvaD1WL6pR/KGRs0A11
7wQckVAosIAGgtEQABSEbtQPKmhEEmFgUstTy59rWFIrVy76ZnTGCiZ3JNQ2
M+V5N2mzTY5RJDZ8Y233g/LT5Q/a8cTGy/erqTxm+4VLR/bt3r5xRW5qYkRE
SGBwUERick52akxCTGR6ambm0pUrd/U7pt9++83Pf/79v//98tsvP3//nfxO
0bpFyzJSMpIyM1OzE/2Dw+PSlmw/dKq8sr6TrB3QekUi5G3WjK4FquRTiRC+
RkCmITUdXDvYMMMY7X4dg2d691xjefGKiyIi8dh64u9OOEyr7Ge7ACZfwgNM
1I7WZjyGUtdwBzr8u1XNhmIBjkWk9mAHmL0ynVZkeeFS6BU6hVrx7OUHu5ID
YgORLOure/AEd/fE4zvd2Pb6lvs1SBqp/cq5k8f37N+YlxEXGxsZGRYcEgI2
Iy48PDUiPj8xJjW3eMORNt34++9/+unT//zv/3789fcXM1hU4ZKCvM0ZkXlp
0SlJyZG+viEpSwv3X66+f/XSjSbegAteg5J4n3HaGQAZh6FAYAg2hQpHdpFH
v51k88XUuqr2LvW0Wzs0MiXqb8fC2x82ab16rQrEMQUVrtXwJNbnelJvc5e4
mkPge5+LmJQaFA8BBskhBhPXwOA4lSKXenDEptOrBqanZx0O26gZzJGU/ppb
h+9iL5272IvuLm/putvJ5TRduHC6dNeadUvS4yPDw8LCQsNCI0JCQVPPCMlb
HLYocdWiUydvP+pSz3z74Y+P//3y8zuA23Xt7KK0xSmLVwZEpgYHhfqB7UvM
WLNj16Zjj6+dvVVDFKkptwgqcbvaqUFVVbZLbWwCFglpqpfymzsH2BXdODT7
KVEmlQukXhWGhaaqDBQkWmWHsolP+a/FjGcKrVY68sVu6kB93UiVz5Ltso6u
yhoCDSNQ6+Awk8g28nJYIOS+HMLyVRrvzNC4VfhaKbSZaNjWJ+VlBx/Aztxt
662tutNRVUfldFy6f7u8ZMPaJUkxERHBIeEBgZHR4SG+0RH+frF5AclRa9Pz
dx45uP4oymTzfvzzt5na0zeKzy9PiMkIDVgT7Ofjv3DBVwt8Q0KiQAJdveHY
7Zv3K3vEGIj9rQCHVRN4AL58X6fFSenBQvsIJLkGDVcI5dMENlPv9Azx5Dql
3MJtbyUKBR1QI98CbTdNge7Gl9Ldw1Ov2bxnAzCRHtJKx3bhO25ChTK10EAn
87DtPCWWY0LPvhlhQeCu50M2t2RmUCQT8DBdt/fuOb6vtetSS29t3e0njfcg
yM57VdePHt2+elF6WmIEODZDwqKi4iPD/P0DgwMjwyKi/EMDgzLjlm3c+BBH
/+Gv3/mrypZu2xAVGRfi75McMH/BVz4+/gEBwSCsxuWtKdhWdrP6dptEODEj
6JmQtKKb+Iza0icaNbOPjOpnYPAAj8GkiK14ltZgtxgFNplTwJejnl6rwmGv
U7xeCGmCyhSzmVykasg7ZRr8247iMuuFusYuhw0mFvEhHRoDuo/aScU38U3c
ieFf9Cw6SzE4OWWxu+U8DhfSduPYxYMniruaajqbqsvPVt9r6u3qbrl6cVfx
1lUbVuekxkRFgmkuKiYqIioiODAkODA4OMB37nKKwLUrom5Yvvv0AbJ7dciS
0PCYQD+/hQsWLpi/wMc/KDw5MS4kNj1/7bL1u0/cutVs/H6GTBuXA7Tudsaj
1rvNNAqP0t+NIEsBMgEhFLaiFFiaQiVh4rVChV3OcUwzH1Y/vVQrHkf0tNfS
lSwGHMnlCEYn3v1kMvR2UiYZ3Y8JY3YSgwLr50oQHVx2D3pkdmh2wjIxjCL2
KczWd++4XIFWTO6uuX3zXPGudedxjW2tt8suPKzpgTI5zdcfXD5QvHNv0fKM
5MioKJA4wyIj4/KTw0MCQ0JCAoICAufWDaRFJhF/+eN71uGQ0JjgyEAf34UL
fXx8FizwCY1IWJyaHJeenbVy95EDJ8rONxmfqSwToxpuSxmWfGPLrnqqlEQj
dVJJTBmPhOvXPYO3w2FYrlpDxfcRXHqYzuIU1N+/9kTutsOrOnlqDA9AwuAc
l2ZAqTWrcOPVHVdw9XXa3g4unkdnEmB4C5JotdotIvrIcw2MN21VMvQDZJra
APRXlF/bXrCxaM9dyMOGS2ePXe/FMKR0yJObl4+dvrx3Qw44LkMCQsJCwsND
Ypfkx0aFgKWGBgcG+PiCthGaBXv74QV1dVBIaFhYgO/CucsLfBf6BYdFxOfk
xSWmJqXnrt5QsmPNnpvc718NOl02Do4tIfCfdhkmHGJ8fxse2oVTKRhN7c6h
tgf9FDKLRSA0dPMBVNPT6kZW7e2bWoaM2ohXAwgMnozsVaP72VJ3p+NnzKmO
Www8mt1bRgDkRBYdAReozI5hHVln977Ss/WmQSuZPKQS6aTk+ivnd25Yt3pz
4d6u1nuXT9+93wUOF0xLY/XNy9f2r1gEOnpUUEh4dFRQaEhSdkJsVHRETFRo
kG8AOBkDg8JzkW9fj0Bz/IOCwPp8FoD98/X38wsODo/NyE9Oys5csrFo05Zd
pRceC4b18nunsHyqiEVsBxhM1fCwyDC3X5+Bz+Xee6AY1xIE3b0CHAPV3d7Y
zKR1tCIl6JaHVGY3GyViYvgDAiKFJ61pgLmGccPfSVuwFPPMwKjbwsGj6XwT
DNcpezagw3PsfNUbpxbgWjwkuV1h49qwlVf2bt22YcXa9TvPIi5fuv3wVgNO
QEB09zQ/uHBw/RIQmVMz5k4ARqZkgKEnKiwsOj45LjzALwicgKEhsduBtz+N
PE3xCY8KDQBn30Kw6tDAQLCb8bnp6Tl5uSu27Nm+p/Ts4xaGRKMWyVQsAFeD
bL5IJGKVKqeKq7a5vB6LltmNkpLFtF40vgbOpJglGGQLTcYVyIRorMSqI2MA
x5CORET3drb31+lGHZPqZ1oeEaaiw0GcnbRzlA4kBquUA1ZQSIVq74ieybGw
GIBQTyJT+h5dLCnKW7xqz4bCwoa6C3cuVbQw6PjO5v6HJ3ZvWrI4NydnUXZe
Qkz2ln0r4+PjIyJiEtISo4Pn9MU/PDwiv37smx8/dCZGRoO5N8Av0N/fzz8k
JDIiPDZpUVbOkrXLVq7Zcqjk5Llr9QyOQiankBgQSidV3luPwqmU0K4mAq2/
lyAzOa1SJpokgFBYcMtHLqzzKbaLwOqqrelAYqVUiZ0jAEQ4CLG66d7TjhsC
r1c6oJHCR5Rtz5VoNI4glcn4SLEQppI5AJhqeNLldHMVbjpDTRBDtdSOihul
S5PSC9euzV17vvPamWt9eIDQ3dlXdbps26pFuasKCvfvWbWt5GjdjYLs5NjI
qJSMzKy4kMAgUF9CYzNPab/5+sNU9bKYqPDoEJByAvxDg0MjIyMT43NyVy/K
LyjYsm330XPnzl65262zCHDUToYaB+nrpyOeILUyyP0GMpGL7oCxhGyd3qFX
MiFU05/vaN0kBQkDPDl1/ti+2/1UOcLAh1OUnb1PO0RPe+5cAzw61wu9VTA5
2cX63onCw3qBCSaCj8SaFAbAOKbhe2yv5AoHINPjeDSjve3uldLF8fmr8lfn
b95y8d6dhkYYF4OD1d4sv3Bo+/pNO9YWHdu949C9q5eq7u9enZEaF5OSnBQZ
EhA8t8o9/pTo+es3717URyekxkRGhYIsEBwYOLcgJD19Ud6GdRu37CrYevDM
sROX7jf26czmAYNEDLgAZBeegu5yU3ueoPAURGtfDVJAIgOA3jlg5an+o4H0
4eo7xhAXn96//LC9gzlibuPyqGJEfS+Tgnta/lTqJHFfDcrdr5VuFPmFjiTD
kdlUDJPPVTifKfWDRquVND6itAB8PdfLFZpqzh3dnpOTGJu9ZGV+ykNR081W
OBMKb6+uuHDsQNn+4vXLtxbu3LvtyoWth+6UFy7JzkxOy4gLD/L3DQwIjkm6
Pfnjh/ffTjPywzPj4sBoOCdDEXHxyVm52XmrV2zYXlx8cNfmogP79p+u6xPh
SBIWlY1miihiNILdRmxr7gDoeHQ/8lGXnovh8XFog92t4nIanxJrK58Dda2X
Hz1ubVH87mxGyVhQyGMeHNN+7xpWMAQQJa89z/4jZYo1bAkaMMCpcDTvO410
xI6k8LUzJsD5QqsARHzzBAPdXHf60u5F8Ss2pSXF5zQ3QTpvNXZ3NDTdu37m
6KHyS6Vr9m9bsX7X+nVHju3YVly6fVleVnJySmwQ2KWAoKCsfW2DH3569366
7nR0QkJsfHx0YlJkQEBMYkbWopSclWvX79ixfffuA3v37i07fKuD4ZBBuSwR
iyLhYEEqIz+tb2IrcGha53WRXMpHt3Wz0GapcNYoQz9ldNWN4p5Ar5252yH9
Rs6XUVgyVi9exkE9uHahijjCkMNpb/+jJ/S1/gfZiZHOUnAIFN6pBIwKLkr+
etw5KZw748TEj+sQLEhL6Z2thVUPLkCbIP0lsaeNtmELpfHxzfJzZ09eP3po
34EtG3bt2LRi++Fd6wpK9q7JX7UkOz48KBjUyOCwArzZ++b91y8HN69NjYmP
jYlPSElMWhgUHZeUnBqfs3xFwZ5dWzbuKy3Zsv3ApTt1uslxk0MoV4q4PJKe
I5RUtTT3kdu7qWSESmkWohHl3TIt3vKCxuXQiP0dtO6Gx+XnbxG/Jh2uIvJo
BKlQLCHBr504VMGcgnK4bcSXJir1iZeDVdC5cAQSiRCriVgSmj/ztcumM4/N
eAFY7ytuG6/3cfHZVV0XQubhvvvP2uiWCaOQCOl4XNVd1VRfU3/xxrHinYUl
h7cWbivZs2PDnh2Fq9LSshLCQHoJDgmOOScZfvfNN9/brRfCElOiYuKjEsEX
/QIjkxIiIyIyl+YWbtuxc3/Rll07So6cu1zNdwgEBIDL5DApbDabIXrSXNfX
0XsdwRDKuoT6HoDZx+DghzydkMYWOrm1p6GqktAC9Qjvt7R3t4jYMp2or5/V
XrrjTvdoB1QlxE0bsY7BUQPExOhVIXvgEDy/h0ChT/w641FzXrhHJoCnhLd0
pEx17eTuU9cjY489/LFvHnTk8pXHp25XtXQ21DQ/fvrw0rmzp3asOXriYMme
g0f3Fm8+tG79yri4SLC+oIDw8CVd7pH3H75/LRNWhCcmz62USM5MT/QNiYwK
D46MSMxeuX797m1btq3dvLf4wLErNSKTzEusJDgcOijVpKRBnqqQrc29d1pI
UokKTugls0USKXtS0YmG3T2LJzR1P1bSKB908JqKflIDcUDHbn5qknbs2n/3
BkPbi2YOvOPaLJoBfreSQBnF4nB4JOOJcMLzdnbAw6W47e4RZbf6hYjllRde
OXkrLz3lBv3b/Dua019trbh1r625svJpXfWte3dvXDyxr+TyyePnjx8sLT24
Yk1Bel5KUFgA2L6w4MzbitHpt1+/fOYU3l2ZGJsSEx6aGJ8Y5xsO2kNUQnha
zqqt2/YeKNm+YfXmwydOn6kmY5mqUZdKJLDIYUgUubUdMHZDIVA4FgKXMOpg
aKaNx9GN0+rRPT3t/Rg0DKW+PyHpwlQfaLEOm0YmDCQGu5Nw89yjapIKBcFP
jj2b4kkV0jaeEG4yKenjMhow7JkSyQdovQyeyG4XS+imbotdffRay/Lqbdnw
Gd1VWlf2vJ2dD2E9Vddu3L935+Gdi2f3l+0ru3j45tWTh3dv2bg2OSs7OSrY
HxyegWFBu0WusWeTE5Muh6JzZ2xSTHhIZExIVNDC0IjYiJTMqKzlS3duLysp
Kd29ubB4f8m5GmI3kcURqUgskYKPqK68196pNBOhHSgcHc/CwwEiWczt5g9g
6pCQZhQHCiHp8BIzjFzRQhwEeB661KGS2z0UJrLR/ZtoSoik2FSWGVEfkSY2
9tO0EIhUSUToXg/QFMYqmEvu0PMFgO6pxIMm32guP8NMOKkdhx671xp9Dc6U
s+ruPrhVATretfP79+zet3f/8UvlZbt2rly6FFSR+DAQL4NAdw8vlZuHJhye
QZmo/tFmkMGDw8JBTfXzCYyIjlmSmZiycdX2HUVH9uws3rtzT8mBCw0As79b
CeV7Zl9oNOQeWDceBTNqWZQm8pBTzzCMGLVSJZfKJtahIO2P63raWXoLj8xB
XdcM1HdzBe2PeJi2RhLMZrb+fxNsgYsAY8j4lNeTRjFNRGQQaO01dahqxphV
xFX1O2dmnzGICpIT46G3ER/V7mLX3jPqLB2Pe+/g+jgu1JXKmoeXT5Vfvnul
ZO+hvUVFe44fv3bi4MbUrIzkmOjw4LkLrEPA5LCzk282Kh16ROOZLUvDosLD
gv39wJd8w0JCMpalpSav271xz84dJdu3FOwqO77/YBsLSsOwpBgIig2GoKdE
QMCjCRQKWSfaqB8x6IU0LJkA8CmEZmhvf/2jR01KAhdWD9CMtnNXmQT6taNq
LQwH6+rtML4gojB4XFs/Aq0xCbqv3OFzMSSbRUIXtHb1sLVivoCHoApGZwdZ
PK2tD8KtbKMI8IBdKaAh+tidZPGTiOvw6gdX71wvv3nu/Pkt2zdvP7C/9Oy5
42tyMuLjIsAMPxcQggNCAwuPPGbySALM/fK92bkhCRHBAWB28PUJCAoNz83K
yly0efP2jYf279i1rWjZpp37T8MMgl5WH47T1tqMostxvddKy+tJHBZNCEfA
VTatlE0ndaLpVDwKR4I3VF56oGxCsZBOuab7NhJF4J46Bpmth5P6ultVlTT2
mApX29ssmiQ3tnTjeuBwElNidjsYVXdFRhmut03UdZ078urloLylE06gEvCw
ZjkTxewmoQEq5nRZh7Kt4mHl7Wu1jx9dLdm9dcOesl0HDp48vzwtLTYqOGKO
qwMCAoJCApfuL+3raai4eqpgeXJyYERISJCvf6C/j19AUGxuzuKcxesLi3cX
7z92cFfx7l1Hz54naFhiajeshwGHt/YgxLy2ij3l1VA2CfQLHamTLHUr2cge
CEnJp7JJAthFFLYZQiMbB8FUhULzUczqxskblQQMrhKKIppMIiGrphtgk67C
+to6sH3kPjA6GcgtvVgpB45CGWAUF6/NNirrqm0i4Wq67zUbJVhieysVcuTa
7WbV3d7muzdv3716/ODOrYXrNhTv37XnwOmyNUlJoH4E+S30CQwACwnwy9td
WHHpYuGO/ZkZ4RFgVAoN+H+FBwYFpaXn5Wfkbt5TtPvYtr2HS0qKdpRevYug
aTkMCUFjUuNp0JYmngJy9e7+ajaHz6Cz+UyaitZN5+NqWuU6s5Sm/I0lxwB4
CBnLFyhpeKOECJP1dx/qQPR0dkqcSB7BMSkVAdReXB2kicCD9eKRKKpJhe3s
7uOw0FDEyJvBjssdEgEL04TvrurpboGLm7vwvTDRpu09NC6xB1X/6PqV2xfO
HS8r3r95966ibQV7SsoKM9JiIuZiO9g/sJAgn4TFBWVla7NWZUcnBPy/OOE7
t04+3N83PDkmfWnWkvXbikHdXL35wN49pUeuPGS1dHMpdBqH6tKSiIh7jzB6
uuTeYQRGymX1YBkqEoGj03Kr6vtaWk0ignFUTABQDRCUoA9k6grhFA8Qt96E
8BrqIUxA09OKMJoHFVwaGfa4duI9tg1DEfKc3LZespDYj4AKf/r0x5NjHUwm
Blf3pLX1yelW2NGCbkgzgYU1Svr0pJuXam5du3HyxIkzR4rX79y2bPuGFVt2
nFi9LD85NiLA96t/zZs37yv/4IDgiNwtq5NDI8DoDioqGAjnqgwO9fMNjUzK
zl+5csOGHXsO79u2feuWw+eOnXkihhT2UUmINgVrwIFtxbc+6NNZYCUVjXAp
0IMgAzg4QUviChgINJnGAHFF0NUBb+6owuH4RPjN5kqWyyAi0xQQYm8HlEVg
YihStZfAgzR3P0Ka5fW4vl6xC2iECIDeig5IveLzX4LLl9toZDT1VhuEXN2A
uNWMoYpFwBDj7i0+8sLDppsnyk5fOH+0+MC+4sINK/JWbS4+vGLDppTIEL+Q
pPwNBeuyFs6b7x8akx4X4u+70DfQ3xfM7v5zrhEcEeobEpWWsXjlss1r1+05
VXp4z/btm3aVlD3Aik420nolaoWcOvXeMy6o625l2EQochdWjOxUs1HQTmVP
JxbQcGAiTH2PWi4kEjr7SRSpHNN4Ye/Bhw/6TJquFi6gx1cScHy+2EziMbBY
IqS6qUkgQEF6sViYCAqR8s81YNs7vL+MnjmHpPJ7Na3tTxh0SktVz4BFSCD0
Xb6JECn59dXlV66fPX+4cM/JnQWblyxZs6Kg9My6zetyEiKWLd+4a+/pBx3V
BxP/tTA4LGDhwgW+C+ZW7/j7BfgtDAoKDo8ODApPWbR4WcGW1eu3lR7aX7az
+OSRc3c7lDT4+W54T2sjnEvA0Edc+D56X7+Nf+dRJSDuYKqJkLr2J2ieTKnn
iVXoDtmAQWgRkOhCvZBIYbZABN2dRNesAH9zPwaFP3EC1oXWkAC2sJPQdr+d
MEDugXa3dPWAD0OPrzPJLfjnr6rKyTghUUBswtHrKtY9oiMRbDYWw2JplBIo
vOnqxZPHL545e7x05771y1etLFy798zG9RuXJ4QG+cz/ChyeYYv3lCwOWBjs
u2BhwDwfHx/fudvcRg4hYWHBQZFZizJXFW1aXbj50Nm9h07t3n3l4iO4jtZ5
8hAdZfnAkjJaypvVtC6KgvBYaGLzVBo+Uwxgejp7m5QOochlkNKpQrYKChHo
VUKV0e2eYQm7mvEdHWyzyohG42nN127j2iEiFb/rWhce3tfGo3UgEI3Qvv4+
LMx+p6+7t9X0TENoalYSaQoS9UwdFa2AY9pIJLrOTLtNgjZaeNfPHD594ey5
0qKSA4Vrlq/ZsHbT8aK1K9ITAxeABu7n5wvWOC96adBXvgv/ffz+PN+5Ahf4
zC0RjFuUFxYan5aRt3pTweoNa3ZVnDl48eqenYcv9FKEAEiFIhdHDaez4A/7
ABNbS698KJhyCbthRBJLyJBPmHgmQT1eIqBTuIQ+DQJ1p9uswUkH30jwBAi2
sbeixeAZt1VUSwQGOVugQzUJzhzv4IllIkI3Ad3QS2qBQJ7ogBpIVwvLJRE8
KRdjCFJC49YntbSrdf3qAQYZ2tHfRXuw/VT9jeNlR8r2r1t37OjeLetWrl6z
Zt2RkjUrkhJDfX0WLpjvM/cp7sJ5gSv9vlqwcOHNa//y8wGf8/P19w1KzCpZ
5heblLZoWcHaTWu27ig/X3pkz5WSnWc6HPSRMQWl92ItrLeFLADY3ShT5Z1W
gNrlnCHDQFHHKXQSjV2jp8EocKKYpZRIx95zSBi9tIOptAgxFIGYA+2zv9M6
pf2n2y0WuYxWVc9ob6pBs7UAqhP2pAbaf73xPgoYb+ltbepVKmjIq0QGDUVX
8NrYzTf7kRgKiwB58qClu7n84IXq82eOHi8rPXfvZMnGbQUbly1bW3SsYEVW
ItiluRVGC+cvWLjAZ17YkfngN/OO7Z/nBz76LfgqujA1tnBrYHhCcvbaDRu3
bjm4vfTu4ZNle48dPlbjFZJlj1v74NQbZS00CR2CgqCgN2EMtrxHMdIPRSKJ
QksfgyQ1GpVsIhNgWd367yepsH61W8visHRyzNx/0hS3kUdEcqFsYnbAMiRh
9t2HXSq43ni9/2kXq64GAr1edaelh4dp7GxuoGhazmF6yFQYDHkWKYDRex5V
sxiM/nu3G+vauzpJDeUXy0HCrrhw+cDu3Ws2LN+6YcOxvTkpsQvn/7/6Fnw1
R5pfxZxY/G+w2HkPUv7t6xfg4xeVHxuet6Yw0S8yfun6gq2bNhce3nL6xPkD
Owp3ltwRswkWSnl7H6mHA6urJ7fU96Nk7f1Uvhlo5ro8o1a71FVVIRR6jBoe
B8DySPiJr/FwEZLCQ4u0BuLAwID3hUshBYQOiWbYPiIgCMTsp6ea+zbUVlxq
bK61KmvRdXcq79y4U4Gr7WisaRO3rGrnc/E4gNXJYxA4InQnICK1UOEERn/D
08bb505fOn/p4aVTRw+XHdu9qWjr0vx9JYvioiLn6lkwf3546Ly0+HmLzuZG
zf9q+741Vf/yDVwYlB4amRKdf+xackBk+uI1BZty1xbdayBxnhwsXr9130W+
nC1j3Giq7SUj7z1sayMyuA6HXW9QAEwDCyNhExl6r5GDwEgcLrNQiCbXiYd4
fVgestM5AqClWqtdwZyaJbW1it8MW9sojJqK2nYlk/igv7zq5P1qyCOYVE09
V/Ho1umOs7Lu5rbaPgHsOIEF6Wxs6qYwKWKVxmTSmo06vWHU04Zsrbh/ruTM
qVvXLp4/cuJY2Z6tezcVZew+sjw+KylmwfyFPv8OTl8wb93mf6XAIvPnLfDb
ds+cOi8wMHvtrhv3Dp8rajjlG52Us3Hj8v0YgdcmenJq25YdRQVlRCaqtxuK
gPV0U/FiAx5KkA6IAQ68ortfoO5owmL7eg3PJwjXGjG6SQWZxDaoccguugZZ
yRx7xxGo3cphB1PA4jbTplXKRoGu7/rjjtlaBKkJ8eDU4yZU2+MWvAp+6/7l
45d2c1XNLU0oMpvS/gDaX1/RjG6404XANj9k0REQno5MEzLoRNLNQ2fOXz51
7vLtC+ePHSsp21KwbMuhtRGn9kQtmr8w7N85uf9OWrz1Kz9Kwhqfr4LmBYoP
z8vZe+LkzbL7k5JdO85EhcTE722zvH0mbzp64MKZsiNFq7evaINjaltbWvsZ
VquW0NJQfaP0LAkvBto727BmIZtIqLqCGfU6AGZLK81lh9HEYjIVRTYyHjE8
U5POni4SY0TVRWXrTBNSeu95nbz1wVPio/oO/OOaM/erIKgnT2u7RaSzx4or
77aPIx/WsQAijHQKz3jaDSHiVRaptPmqRNXUZ2/reHjmLsmEciNKL1y4eP7c
xUvl1w6VH1m/dvmasi1JaUvSs/zn+82Lzk+NyS7P94MVhfvPzzgRuAez8Wxi
Zsi8HdPqrbtyVsfNi7ttfW5GlO/eceT8zfPHVuy6vmHt9b62/ocwfE8LB9by
8EknDNcK5cv6ASWrCScwa5SAgCt++Y3HIibyif0ODV4lkaJBaRTUds4wh2at
+GaCYdpMw+MUigFMz40HUsSVahgRdhDT/OAQrLkVVMZ+cheiet/Bg5tuqAQP
GTQoucch6gDZfExFYvFlDCqDpe+gsXeVnKqtqOttuPHg9KVjpy9evXzh3N4T
p/ZsW1GwuaQo6asAH9D4UqsPrcv3O/wwMqvp/jz/+fOid5fbV59f55Ny5/fp
XWFLSy4v39pcdWzT8rWFO84cOFq6veRgEYm++yEChehq6IMSCWLP+5mp6RGX
TdJXxxCgeipQEt3QpMtpnnj/rUGgIlNwNA9ab6TTO1kcAYqjV45qzN8z60mK
QSaRgBMrkZVdNxoYvTWEyp4rtbTaYzee1JAx7Q0dsOq2iv3b1l7okbNglXfw
uAfEN4PPxpgXjmPQTzmtj0kUCaOr7hoSeb+l7Nz1uUsHL9y7cf7ymQtnThwo
3nRgRcHemIDEfcEZVyj8tlsRC0Rd83zP4eb7/SsoKvjwSF7Aw9vSH5yDP9su
FldDy+PWFe0s3Vu249Cx0u37i4o2ZEEVaDa0p+NxJ4SO7uN7tAqRQKa3qIcd
DL2dRGZNv3FPiNVmwvC3lOGRVhxGa0ZL8HSJV0GT9XB+NVm5M99wa5swXJtc
JOexTrc/fvAUCcPgsbdutXLEHddrKH0Przch79xovX7r2p5KbndlzTZ4WwXv
ezPF2ddwvRf1hNNXTKL0wcprSHev3Ll15Hj5pbJzN86dunvz+vWLl85dOnhg
77ptJ5P85u9s6us5lbSkbFvAtS+n5mUzMv8VFun7r2xX4bx1tvfT7XCJfeBh
4/30rP2FW4q37z9x+MShHUd3rdi0JKVl0CQBYeJmTT+to0fKo9V1tFL5RIeX
KZdpOToDjcoY0PHaeu3CzilLL0WuZdSiEIqxWRsHTdUhJHKdasbYi2BahVQ+
hdalaaH01VJoVW7RrSsQk5n+lN54/XED7PLe/7+i9wySG83z9HQRp9uQYrUX
K2m1utvdmV71dfdMT3vLdhw2m2R3k2x6TxY9q4quqljeV2VWem+BBJBwmQAS
NhPpvavMyizvDcmibbbZnpnT7molhRT6opyviAAiEO8fz+/5BRB4r43cPnFq
3NqrPffhxGBmVqO1zGuG7goCGAJbfHP2ngmNQwfcM9y+1XT06OlLx5vu3brd
23b1xtHDB3fteff07w5oDZ1HX/nv/tvff/vVK/b8quJ/cLT8u3/fmNg36z27
sH/a/qFa+r9KI4ErJz78YOfHX+3b9e3JPfsP7v9k74433vnkXXPRPpKiNIMu
PMBSjggyhGVAY26O4LKJQMWpD+P+3DJoDQRG2B9XcpOlNDHUG7y/mCWCLO0A
BK7ww8wSTXCoFJLS43MQssaaimLLzNrNNiQXdYV1XYPDau25E5e7jndrseG2
K7+9U62aDrXidbb/dlRQhZ3X+BrRMagA7NjYrZOdLcdPnN9//G7bnY6bLTfP
nD+996tzHac+a3J8+le/3v2//d2lvbc/ZsDr/+/V04H/sL+j/Vhzem5pIxLO
y+1A0AiMDdz98JO9X+09dmjnnn27937wj6++9p9efaWJme7ERIPaEqjmZQYR
0rlkJhyczoYjYZRmsXA5y6QTo1NSdktqFPKsPxaY8FLVNZpAcSeN0jEozMoE
iuBSKtWJKrWJlTJVAb+Zq8BmXYInfD39rYA0fHTXzV5NGzS4X3klgLxtRMek
yNRgezbR6TJcDGZYZZv6ry87vtzX3dx58vj+M5dOXbre2tZ07kTTwWNs2Xf0
4rvv73h7rO/DHVe+ajvHMid834+XSui5d0+GisUycvyda9oD310512y4tOe9
9xrO8vVX3x3Zt+P1nedvD+uoSCm/WBtjnQZb20guUUwzQGDKriBzFBaLiYnS
1EIpMROYBjzlVYXd5UXiXheSKAb59EbI7QMRT8APByYEoCACkVQ8ZNbjHJb5
IVKI352av20wW51ee8dE9PS9jn1d6jFX3rL7SiZ6yVOOg0FfWXmnJna5tZ2R
QLx4yehRmzp8Z9/p+OqLaxeOXum9fq256dKRo31roeYvTjbtPHOpNbT1ZLLv
0NGW1YkjE+yWcL3t398I+oNh4ap+x6WOkdPHD3Xe/Oj4629+tGPPF7/fd/DQ
e9fWnj1//vj586cvnv0yPeaywMCdjolAuiRZaW4MiUa4QiRdT0VlHUx5YNhe
DQwaOKNHYDFEKJE6KLQ+xXlgHygyMAnZQ3XZ76ttGbU3IiEtcz+/klRMa3rc
bsRE6uWhsyesV1RjneDsEiPIw0lbxNPBybyrux7uGVU4RCrweOIuZhi1BpHx
9i8PHPnyxLWTF86fO3P2IrYyuPPoscPnr93d8/qXtx2rz1MXzz4fPdQwe+k/
H855rCDhH2/WXG9z3Th9/PKtYYvxjZd27Pjgsz1f7D2052JvR+vFC+cu3BWc
rrWUfQIbAvsujZ02NvAfWYiITCImpiWtB4UjlIu2iZQE9POg20P7cHw2J3Ci
GBf9UhjGBJaT19MzPh/sX/Z3XrHnx0YGspENlphSU6DR7Jr0dba1j5qGRyfa
NXpHupK9qOT0I1Yb6dXV4z3d9/Qsa90M9ugUoLVX3/Xt/j0fffHdiVOHz1w7
fkqSDuz/9vCRby9cevvw8fMnRyzOfPj6Sgb2d37FansSfk7PtX/b39zUp1O3
vvHG+WNnbu166f23L1//tHFaoz7+7q3fvPryS7t6O/f7Zye9N49dAU9q2k41
aXjDapQJmzFTTEpBYzYNzUdjSaPdOgZjTlIcGPPm1AlAhYc4SSZ5Mcrywe0E
iHgZS3pxvEcXnrhxcbzOrcQlrq2r06wkb1y5OUrYhvnx8eG7bYgZ8dGGsX7D
KOW2LMcPDF4axilrYF6lGj3cqModTb9/b8euo2cPnPruzHfdjs+/Pbv31Jkz
Z67fajneW5/DxvoFaXZeHPz8pR6ncUgAR9nh4503R0fbHX71vn88p9PsfPWj
l4Zm9+zd/fsv3t+1892P3nzvtc9GLM2n/TWJHByyX7urazt+KdKlmQtImMaE
rIY9CdwIVRKBCARO9IBZu8MHAiXjVSrFMzzD5ZwwIohBYjLDFflYefrFyuwi
KgnYmJ+zb0r+7vYR5wQ0caP56ng3QO7p6uxVavf3QJevS/qBUUHfvxB46+6h
CVtAoYnqVcqLl21R1+53X93x5Sd7Dx04vO802fzdmcPfnTl96tjJ7v4ugt1/
rzHmeJgmtUcP3dVeb50cUYd7z3cHyet3bWCu9cvxC7/6u9989NrZGcUHOz8/
su+bE9989fXnH7535XLHrc4civCj6i4N8c0XF8DQiJOBvR5Xokp5aE9iFXcH
XBMu9VV1MWq0gGmh42gflU2zTjw9VzSqOYwrLs/f5xLzhQePshAx9+SXp4k8
yG+x9gv3JrSouvNSj/k2ALe2jA6YWs5rL/zFy2wyUEjYbbPuX3ecNquI5i6Q
UfTc3vmrD47vfed3n3++e/fOnX1DHfS+3Wf2HTmw/+C+b8+0UdSN/3Uf+I3F
x6AE0nmwpa8zgFrdeMDTa7+pG7ADsq3pm9373/3bt1/5PJ4/seP9fUeP79t/
4OCxz9853nJhUNEFq3E7AJR9qguWkSlCZeZSmC0eJiNlQzDqdaO2uzrglLJc
KXgFzAnfOj7O1KwgmJFinM41F83GaNaPkaapx2GfCMOBwnxqLruyGVFhY/c8
1ot9xqFO7/jFLp2+87bK6Xa4qSCOEkBl2fPtxcMKhXNi0JA1Osffee1//Kvf
/u6dnZ/s2nP0m9bJpe5PDuz7ZufXxw/u2nsACl689s57/pBtHERFqkc9bBw2
H9ZDZLRf2wG7D8Hp9qt9H372xWu/fvndty3Lg7/56OMvdp/4tv27g/t3fWVW
Hj9xsMmoTUzefxQR8jVBDvJ+T5a1Bxw2UDk+5O7u4+82hfK3b8SEQhIwURjZ
019LqazqxKQjJMRnNsNB0cN4ubPCdp0LcXj/5QuXTeX6oyeiqbdF4b9mMigv
X+s/pRy1nb4x2tdluHnhjvumbdxRzcOHjzqNIyOtYxIdvX1b57r62lsfvvfB
zpP7zhxpms2d+OzrvTt3f3Fo7zc62nqwX9kaRr2+oN/JBjTpuPrWKbBNd/f2
RVPPwXMK58VDdy/seu/NT2+++saR6WT7hPr0vsNH5P4jzdcPNrvdlouDkIOZ
fvaHp5H091uhX2Z8ZDJvGO4bs2oVRJ6mw3Rl5ifXKX3ETQ87EySM9ANwl35Y
mM0aVCGq8nOhKLEYdBpZ2TAoFzLo8KW9wxBQ2sjcOXljwu+watrOtQxfH5rQ
H2xrOnng2MFb11o19pGxhA4l7hnd53uuXYJi1D14JmA58fobH7+3f/fh82da
8Vjlzsd7D37++YHf7+Jys5l4nB25h3ByIEHbnZS1v9sAjhqPftdr+vjitZuW
Ay1e+61Pf/cXbYc/G45npzhn9NSZA3S2qeXy+aZLFyycwcNDC48ePy7j208y
2fkMaYVqJh2gtIxHS0pOwCMMp7rXwkIe5HYXkcLbIdBqNbmnMuPXhuzUi2Uk
z8MO4+XEs9rG47ZGlrc2B6yWTJ100pyD1Ljbj7X29SqQjstnL9w9dfnTr64d
ugoNG9PtX13inPqLluGr/SG8WvLv+lT369+88+Gnuw6ePjOMfJN4MP7Z7pMH
v/5GVdleefyTu61nwoVhNC7zNlfEfN011ONw2K+/+sl3+1TGk01K/sE0dfJU
rL4SrI189NEF89nP+pbONHf23mtVaTTD+5V0eUGM+9XgLAkKeQQ0MgThViET
cVdLDSc9V3WIRpir+Z3QifNDE23mzIhRbS5PG1UjYHD55wyQcxlsp1VPV39h
zn/ZMzR84xZrQ5LRjnEZc5iMd07fbr9kga4broyNtF3tbj506LzVMJob+Prt
4XjXDSd5TU1LkNvxnfbjl1/58Msv93997Ryo+WBf7WmQZnx9O28pe0Fue/q7
sTsTPgXvNHtZr2BoHTtzT8FFsAs7XvvvLdoTTTvfNwgL2+vV2Inm5MB7Oz88
5f707Mq9cz0DCtX4sGpS1xep5H1QwKqG3JgoWNCJEZPK7k74nITEAgCq7Ldi
/gQ5rImZDl27cbHfB6smOvhpEiZctXThn8ZUnLZhCvmtxdtXzl3rO37jktEV
4qNjJwWIpLBdn/T2DPKnzxLNI11tfW3jxz46hchArP3Se9e8HrcWu6WVO3fD
WqbnL3/34ZffHT554dJZ7NZNiK78NNO6Y/enr7fcGCnGNj0Owl3Mlj229gG3
b+Rq84GbI/zFSxn6zK/16JlDez/DZbHLV6V/O44Nvbbjjd/dNb7DO870D9zu
1mq0/VDT7RLPY1BUAS8tE7M47BuCGE7hkgkqRHggfHR8yNlhsN4dhCI3jl/4
rnmI7FET47V8UMbjgO350pDBoIxSuae5u2evXuza+8ZutRgMzVruUmoPWlS8
dEZlsF1oR/eP6e4ODvY3H/lYV8hFod5vriizpXaludvde8rmlG989srrXxw4
/u2hE99ypPPO7/fxL4pDH3zy7qWJcTfR63CQ8tM5wOZ1KToB3TdHznQqmV2/
2h8qj4xG+5vsMjpwu9Xmmc14nJY33nrpN29GhpT2k+0To8MabV+TQnMvAiME
mzFb8nFZzkIkbFD1XDdCbFCSKIToPrEEabu8ll6tyzl46sumwXG+1yWWt8kS
5bc4U7/MBtEkzE1uRzvbTn/d3Hp0XObhaA7Ww3qvLUZ095nb7x0Y5w7qhned
aFac/eQzE1LkLf2XLgcTIK8Ahl06Zz8dUH/++lcnjx/bc/7AcPTye5/s/eR9
4MUGfuLVi6quk5d7X9fNLmWdzmgASskQ+ebHFzqU0Kef/9bCl1JZNmbbd+jY
9YzqFjNfP+V6683fvP5Ke66z7/D19vFRpd5iHO3XB9wuBJ1MqnyXDggxv6eR
5WyfA2JjdDlAkYaJmXGlUacdgDV6u/HuV/cm7OCgsrJd6MdBeK2W+UEgH9jH
SvnZ/svX3zzWCScpko4rTQLssBPBkdvB8e5rp4bAq8jFD9//Zl/X7w8hXUya
gKUi58FOGJFxxKmELUrLiV3XL/XdOHleMTP8wf69H7z/1stjDx9W7zSprzQE
9nZx2SnqxsZ7Tnf9OMPvePfqaS1743/56wOCmUj5TJjm8pEmW0+HZ2gmQn24
d9dbr30gypd/s/v41ZYuhUbX+s4p1mg3OqMLqLHl8LjGj0UnV54+KQdSkg5M
8V477cccuLbNgvSPu0y2RiZrR8fUwMxag6kgsyYnwoRwy14PlujdzddQKUbR
vhLXdnGMMI+FEMdhzVDf5Vs9Xbhi/+6/Pf7V6Rv9qsswtpCVebsxpbwXB82o
XnvG7kh2NZ0cmLh1V+Du7vl2x1tvvP3pP9zaerwYKOJ6J/czedadiAAnT3Ye
qN+PHH2r/ZQTULw99MVbDoGUQG0gcuPkGHBdvWssl/l2T+jNX//90Y22V3cf
P9JjuDdEO661EF0KtQ0MzsWLhaSQRL2x7e0Hzx6kSdxqdHMEHeD5AMbBWMdE
ENYrBodNI8NmiNrYiIcjdi4pEBiwB8pH/JmRY+2aiGDyE7GS5AUIo55RGFQD
us4L42ch/uz1I//zy1+Z2nSGQb8rs76UdKLhSKhQpvt0+p3o8VTlVge43Mnq
saZPP3zrH1964/VXXzm88P3S6vPadPnGOXsuKdo+O9N+BX9eaPn7K62Izq0p
nG7tic0pWkn/48iVCRzVnSLCuba/srO/fvXv3FNn3v3k433WkZPdBakPNFAK
K8hE1398Nj9bwAlPrDC7VfM4IrpRJixI1XSyGJt3mlg+v7I9ife1d3QNaNyz
G1D05yROpypl690QnaTSMbTXpvajhl4uY9KBtKPf7UC091QtLcPNsaPdsfOv
fduVbht1wn4+ufX9NiFNxpNI8gFwzxi8+/qpLWw0Wze5+tiv3vloV9u46o7e
dOXu9OPF+YfVfeerc7XWzsjNw7dbO5YmJ46aOLEoZZGx8OQJwwODTqw/9CrJ
ZKy/uhyS711bTIWjQqzc/MVv396hOnVAxDFMjfr1aDYd21isL06JHmKqEIuV
QEeUMXq8/qm0jeXKzyOF9Qcby/WFH6dxh7qn9SL0wOeYe5BhRFiKR0UoHYrm
I2542MFw4UDE23NhoIvwqG0UYFM5rQLWGyldO35oKP1W58DVCKUS6/88DW/M
EDpyac1hsh/7yy+zc9hiih3ArW++8bmmde/nHx9rdodccz9uzm/PW8qr5/6m
lfedutTRGilbIsX6fDY82zmmIcI3TYsLwfg0h9aKBmDEEYqnZ4opjTHjDq9P
nXr5pY+sgRhcYMPRMkln6lI+T9djaPTnOll0GuCoKHIGVy7jLtSMvgezEBpJ
5Qsy7XGAMdFx/tLykifoD9ZyWTaXjviS2WRxiiFEJ6E+pcSME1dajitdXlgg
fDTFecb8sZniwD8MBt8Yu6cisUhaWf1TfvW+ut8Gbd23mYbePjPxuLIS6rXn
W371X/7LTWTfa2/87bvfWWxH/M+eLGy9WLj3q481ftIyfKyJqufjYZGbKm7p
WrTjmFEpJrcfTWXRTF3EEXkh7DiNFzK4DChTzjdePqkvLqdyHOCQfThP8qHS
/SezxdiTQj5MhiioPC/5lEAZ9dS1g+T6osXUj4AaPUL7BSkxX7TFH4sAS24j
zmSsWI2IcS4u2cxmSh46ryetipYbfbZBnBiO0naQsLtcA+Xon2bC+l2a4NAF
Gw9x7JP1/HScg/HyQjwhnvj7/5j+U8J/TlXf99cf8k+312ay3v6WLzU3/kax
/v36+sMb/7iDYLy+xI391Jz9SlT0hNLbGeNIG7jGKVTDyMKUu5wk42uPHnrx
QShMS8mcCKIdcMTkVIBNlyxalUbO+PhQ/f4fNsqL64liSqiuE2hVsjqs1ZzH
2jGMTD9Nxtt0FeBq67nxIaPBGVxdelAF8PCazTNGp2rpQKYhzR7QX4r4yAmo
kcUcbtaEiQNXcJvJhTcAFbl+szjv7sBzJ+8SSO8928JCaH6jVlmulFeClFvx
1yf/iZ1vR6qvvZR8oTr4+b6LtvL9KfvEf/OKd/oP9x88xV2pBuEsQac+H9F4
yVJALtY3OWTBxw/s+fAQuRkM8NFwfmML5abt58Y7ZTZZqS2meU27GtGBLidu
9GNejEtWllf9XjJWr6f9cr6Ppx28JTcXBYIBLPxTJZtMBhMu5eVOlQ5Wj07A
2el4GZK0wwMQnZn0hgmPBHji8aSYczpoKRYuiWAyI040yOwEGVFymS5dCi1P
xmOfNAFGNin6ZrbKojzNRBY2E2Y18h/fWF3acBbkl6H100eaf3X613/xm4vA
3HPbiF8i1h9u/euT4h01abdGkGwCN55gc2ERm91egtsTYOvli12R2TjMgcLa
DMemvztB4yKKUSYvlw7iNKE12Lmi3kB5TEKEkQoIz2KolMtn5khXvBCwc6VJ
0jbehmdL8kg0UgiR0dhiNRtmrVp7uD5DmjCznxamFySRZ4gJo5soxqu8CYKB
kWHCH8D5lF1jtxvsFEmRVy8HXevTzrtYICbTW0w6iQXCEBhd2sRi/9/61+2l
59uphSl5DfryvHPY9up/+u2H+z+/s7GVd97STG483HwC3MH9IilVyri23yWw
TCiUX6lEJbnrRme/QyrnPBOhWtTPSUGgFYiGGhD0u2sFN+pyy1FsvTGcLhXI
BYWIW2ZB2o3GJhefhuEXMYCMZ+mJ8T6ccoZr1UJ+vmKyRVNxRGfRGYQQPWEA
5pZ92cVChI9hZt2EL0OnCjG7Y0Ct3H/2xD0qQphNJofP64slxYEO1iAXIomQ
U+8QN0UvTtULdl989d+eP/6Xgh8p/vCksri0VQkXXUDo5l/+zUuvnRr5n6yw
IZzR3Zx8tLEZY9woDzi4LBdU2UCu21uWg4WsFLJfdFrHqeAkjFApHxkIRVUw
G/dQqaLRFd+IpxiPh0fiYSeLQ25aCEoMkvJk8kaA5Gt89E8rGC8k83MbSTjG
ehcf1mdW84g3Vp6fqUb9Do8bDni0DghLT3oZb9BhMuiogD/jT6EDI2OmnrMX
xzxwIMPo7WjIS0spEihkogygsvNhbnWNmawJ81OpGMn/8//+0x+XAPLpk+2F
6RTZ5qAmM8nTO956+/3fZ2/eUn36uTTLba8t3F+sSWYOdyVqQmI2G0RwbzRE
+2g8uoqCgteOcoxQSFIMjXGCiwnIkAX3x0GgEFUHchmZLkqQFRanA0Imiqen
vIV4ysUnjOlHj5OMPzy5/Pz5dErGE2sv1jE8FS/QLn8kF8lWSRx1WiwKkefN
JAfaAMgJGulUIOBy2VomTJf7nRht8VGdChBmMXa+wOB2o2FMn5nO5Go17oc/
/fyjj0oi2JPvn/08BeX+tLW1Woi0/4MSwpar7Fzebs1q1Saq6y3r6Pr/E+DX
N5+UXGTjmVmaclJAqZSSxCnGR7momWcriwtJxoFg1URQ9rmoIMz6p7JeXErg
lfs1gy2bTJUigr9jRDMZxGsFWtzMyrnpilw14kI66/MGCgXe7xVlurayvP1D
1OzNRPsbdzCi/wPfrXZ6dFazdkQNhewWUnZrLVK5MB0tLVKBAAmgKMbIqDLA
IJLksnaNeb3BsJ8EAolYibeYqfzPm1UfYi8+rG3XzL7JqYUHczxullyWtaVM
/o+PfvjXSc1wzdY3/M7qi5/rVPzR8xTiTyYfJwnKq4MrdWkxyFTZhlAUavNP
Hstmz3zWI4P+ZEwQxOpULcqFJ//48y+yLVbdeNrISI1HqwDIbBBDF+by2c37
aYhOMamGgPhC5YgDRQPc8taj1Ye/zDAEjUMYpLD88Fzvtg/orbbBITVM6Iws
q7Nw2VjSPzCxEbQ3qpvkZcsFOea1OtSYyQyKxYyoMWAQSXhg6/h+1A4/2lz0
G/inU3UMmCrfL81OIQg40OXfqknC6kZezuSMlnBEqflxixFDsez6VpoKyT9k
cC+kM/LpSTnOUwGXU0GUk9WnP7HKaMER9APFXDotxKXtKERPcTOppSwrVpZL
k98/XQ9DDkxOUHEvwxdZMBuU56S6n7VBXL5EOgMwHck8/sMP2394lJDYUCyK
hJ78Qnvys2GNu+tmp0VptHMmgI+V2Tjl6OdmJyE4nwqm4n5A3e5C1O58KSJz
jOQd0yCk2wqg7BAu3bT65ooMtjKPTqxusrN8ssg5SbLf/HA9k/LQz0w7X99/
OrpASH9c8wyXHmXlP7++WniSYiEsKuJSiEhu4U4nCLoiaENp1xhhJiRG/YQY
JEK54mRKguIJrtO7vTzH5ZJuUIgl61kU9xE5PxTOxoRkiQ+CrJATXL5UISog
OCtHorPP/uWnf14nJut5lIhvbCZzP/2UYCHWxwFjQa9TSwfoNC1EyVELiBVr
AovpmlsA911H3j7qLpUpiePCCdzngwwOlollBYfbKkelbG2Ot21te3E+Ec9O
5wSOWXq4ivdLwa2xJmDy4fe1yJOlDI+LAsEXa+Q0yj+jAZuJZD1YsFCfSpF+
r3PcHSRTQqgcjOkUSDpFMxkqE+YDqfLE7e7qP/8s8OshR4/aES4GMBjwYCgU
K4XlTFhmXbHpIoeGc2yQIbmplWBQXvy3f/kpSOUXGbe7Oju1uRjKp/jSVMwM
ZtFxFSmgIT+dYHpH7OEkDQ+c++aYssAMWPKEethZy9ohPhiK4f1G0ksxVqsH
iwsjkBhM5lKxxRVeIGwhWQw1MqrUCOqE35lZCA0Xqo+flacmUzIhS0iATWFM
Vv6xbLD6ccpLhsXq41+yeBQMuCQ3QrhIj58Yu+XLpKUckwihPkKrVA26H/z0
aHJjSnNFrYCZEB6ox8dN9tRkzOsImVTuyGxlLsuXWBhnxPJ//YHJRTf/j3+9
P7NZSTEWEOJSUiAMOx0MbrQnsF4jRhtRVpQNlum1KOeDjS1nXTLuNhq4mHng
lqkcc/rZfDqqgyMQGPQ7OdhIe8l6WWbCi5tTbN5yCyxlhHSOT62UK3E8HMmV
hEJh9fHD5cWVlenHy6hPzGHAdGDuXzxW3BVK+QQUj8yU/UsViUpQVAp1+b0e
jz0UBLuNVkzySoKBcmnGhJ9+XNt6HlKjvphYDlHLz4QeW31GBtEA0goKmUgo
Fy+LjsZcl+YeLdepuac//WFNCMV9Nr3ZgTCcGFO5vADm4ClNgKLMYJQNRR/n
4yx6r10quZEsy0d95GQuw1igWErOT+UzEbJJRfqKyFrE6+K2NtNRE8GX68UK
ZIKpSGqKcaYWZiSM1xkz0fnppMBGxVAqW8wWG1Ukz5vFZOBfIzhm44tOAkWc
wZDP52NTLCvOp4BExEJlorQP0g11YkJ1bnV9rhx0r72YmV9djvq9fjsXoiI/
/t+psfUHGZXHl+IYMRoKUGgqaacjc1Pl6a2fZ1eqle3SPXB2GVdGpqRkOZbW
dyFOqw1yidkYTwcb1XFGYZGImxcu7jViQpBOFLz6UHr64YvHc5MrFfVtRKhN
jPrBVCZRSIUnoNl1n20YF5eelUkKNAbLIu8Pz85zGD06EKOZVDXIUD46WKwV
vJAcYhk7lplbyYuCE0vYQbOLVGkQhMS8oj9NMIVCNJH2C5QpTufSTCIWT8US
uUyk8OPazPzi1uOlGWwAYsXU/Gao/uJJlRDDoT9v6d4IUTaNluoURkFscOvf
HrKRYrAwX60MeOLzvsmZNrvWbXJAIF7AjbG8KrS2hOtiiKJlbKzPFYiE4uWp
DM/Rjaen9HS9JljbWxVilaCi2RxJuLSqe77nWEPhjMXHP1aDMr9UCQoUU6wl
aJc7wBN0ajIYCBWrSzO+cBYh4yLtAFP3N9aTUS8x47CQKEBAFESGCQLieb+8
kptBRryN5Qam1pMCwXIB1B8ThSc/5NyV2uwf/vjTpBtqpGMmJS892nhUDdEI
KZBOwYfGpmsupZnEcY6a+RmCoIUpJrOgHYrVxrDoiBEzGT0+koAxnyyk16aY
ZExyQVw6WyplM2wy7KcYLlIoRDKl+CTPID0DfWo+lkI92rtDA+Ojpi1Jc9vu
DT+cXObhYGJSFjg8X+Uwd6ggCTy7GA/nqpzJ6I+Vg3owwYFQ8cn0yv0pGqli
Rkg37MH8KIwSMCLEnS45WoonudRyRZheeppEijq/WsmE0dzTWWdIspoy3//w
YBoVZS5OCQuzCzM8CuNJSC3Fa4+qftRuxd06pyM2O4Nh8wI5IzvVFa7f4q3a
bHYriIAUPbUkAvEUBoUSciCZpAhRdGesHtjZ36fVYcFKWWQpIVUTm08OB1J0
j35QpaTDZktVQG65AtLyckbZDDPFlBxNFcohkfAHSUq20ywpMARgQek4B7Gx
EB6Zrc+VIxnWifnMdqfegnvdLguLY+GIrlUbS8bqMWa1vvF/Pt544MP1essI
FYCI+U0ayKgOKqlE/fl6WLKPBgJUtLYUbTh3ldSkNrcyHIh6LbDVYiTWHi3i
9SpczkN8F0c7MA2pgdyjNg/BE5GclGiki4OfEr0pcrxv3BHkTXbtSPfAcIct
6uhXupkAzlZDZtjtYgOu/k6Gw5OhKufw14pi6lnOAIT4eIOn5elKIpLgG76B
+BE04Jb8Uj6Ki2XOI4oYg4VXeJanGafF7XRq1HDMh2o5AY/xmmu9Y1YM8cIi
lZxZzMzU69I1DA3phvyVzQIUQq0IkH5YX5wLJhVX7ASWWo4FGHAmh07/eT8r
EGCDkKeR+LFnq/R6wbRF2SUnzoUpb/+AMuAyebgG/YWi4YYCNhP5TDk85lRZ
BzUmRauJgscHFNFA5+0TV/2ZTDCQK/OjA4F80oJ5Fb1aMVloJHcGwTfXyjnZ
L01FQ7laQWLKISHlJYiIHmdZvlCpzs7er1HhsAEWKnkyKXtoCgYMdo/HkMhT
FOgj0rIA91wZRowetwWnaI4XI4UE1gjuIQ8QexxhSrOiRXq0uBaJSLx88aSc
iyymWJquT8t+byTMmFyQByajkDe9vFCvhL1FG2LQGEca3jaod5pijApK/vnj
3ht3OhCTo1rLC63DiO22445G4225fhtJuCj1UNOxLicXS6QLMWwMxbl0xXZn
yDTUiEIG9hF47sWUIzoZidGR2Wo8wafjBpcvFMoSPhIL1pbn16blXNInuoHC
Jh+gQVEC/AFM76+6dFIQNWFkqlyLCj4C1ToAIyJQvBROJvOTxSLvpk3OjVne
P43AD6Ol7//8M2FZGo4uZmoFAotPrmYR0J1CBnQkRpKxgJie23i8luR9NjWg
ar/ZS5IYbRg0c3Y8y2YpBLb2ewAtGem3pEFQ8uqPHyCsp28YE9RgA4u2o2du
UEwgPP2gEojn0hvbc9KEeqg/HvNiTIxiVhYVI5v3GVRYmW+UgOBkEnP7gqAU
cdh5KjM5N5sL0FyI76ZnEdIEFJhhk7fAWeeXLeWoP8QiGcHkD4l+1miEfLNV
Dy0ZLIGGHKwv1HIilt1e5kyq3IY3Uw5tlTkuE6ksVdNJuPLHhZSIY7EgrBiG
IK8P1pqC8/c37/+84oU6M/igd9HuyHr0apUCSkQL07EwhptgiYpP59Orm4vF
iHb/sQw6FChhQ8xkFO/TdfW6gjw+GLq/VrJzy0trM/WofyKXowhfAyiR+1y7
9Ms0DhcXchFeTGenUh4AAEjSISRqJR9fLWGpVAzNZEwAGCq6BlwKIO/CJjVW
go7IRCXsJqweNium6UC2LIWzOMwWVx8trzx4vFXPZteoW1Qh4AuO3ILqk5yP
Wby/mQpBU3E2LDDeQNA8NG43ADKrVvCF7UVy8Ze02haQimraQ4sqq03bMzSb
m65u1/1+ZvPnP/307EVRCFuNCIbAYa8UowOTWTJH2sWxtnZBXGbv4KWNqSyf
SacW71eys0kOdkM8iK6su6BnyzzhL+VCDVPM1EUSQa0GOlmv1DaXsqkpipHl
5JpgsrlIHHDRTo0LNqNqBUjIISxTyJcWFhJEsEASXFAUgrFMPp8rTk3ObNxf
nsKCwDDBqB2N/nX4cmmSHAGXHm6nXMK9Nh9J4iZyrG9obMLgi5Y1EFue8bHV
B9G6KvVAZ9ahUVBt16s9uepyfGW1Lgu55aVwfGH0A317XxDUOAGI4SPPvDs1
s0GfoDONgvGfk634OHk/6MyXQ7XlXLnC4cG0xLn0yRUXMj8lkeFKRgpHWIkU
0l6vB5ravv8oruSKUYN2jAnk1ilO3eF2AV4SdGq8NgjQxacREBMMvePKIKIZ
glC3pXGYkSLJWATUeQJr3/+wwAKOEbULhkk/1qtblOyIO/vgRRzgA1dVhMcL
uyd6RkaUSqcHGHejJCO4nDMPolB1RTdiMvJRbEzrgQNz01m5tpAh7LFqdrUW
P3bH445z4/c0RoaOVE23XEt5Pu1QubDog8KgzePIMOXcdLScXt7K+MTlNQ6C
LeR0kWroqTtYyIQ5jvSRklMKkVPP1qKLj0uw7MdAhQmqrvLxgBGiUNxqUvYo
xl10eL4M2XFmeGxMjTsVnSazrXu8bUwPeUkuKDDRRz/XG5Q3291cEHKbzE6v
FHbbfemtJ/crtEfSd5OocsJEIHfahkcduMEc8BKQS2sic4ncypyue9wak4X2
dkSuzC7EgNDWk6oUqS08KIbge5qkbqD5qs0ZEOMcSMVWlmXOZU1EAjMLMi8l
CXemkipIxUo15acrghewuwB6vtxYTUSYlCUWJvFoBqYyqa0XGTstry7NNVbW
A/nrGwsPtlaWVucKtOBtbrujdRgaeIFIVme2kLbRocudCqN7tGfCH0RpxOOX
y6vznEnrVk4AKFfNoA6rDUC9NFtbKlReyE6EMNCieWzcwLrbh+6ZKICtiajZ
YHAB1kB4eb6/c9Cd4WS3h4QDlRWWT4UfPEnxueWpSDbmguOjOgCNJGAkmXbb
2IXF2URAoJzOeIlngpzBC3qZaqaSyyYlL2z1gzas4BdXKrTXz3F2n88bzsST
K3Uy8OBxnWAYQmus53mZpdLxZDJdX370/MlCJXy7ubPHOW51Orw+EbtrDRsv
3+wHdKo/X7+AwoDSSJJeNjWVF+Fxo9UhbeYFxmO2oyiSncyIwV8yHgKBRdoH
q4f7RcHWMUp4+dm6C/U4fYRB0ccuksOdrjIbj2hTfqgR25EME5pcaKCllC4V
6/kaEa2kilGFSowHqPjqfD4tNjTc5msoKWG4bXRbzcnJyXQlR0CA0w4QvloU
n1wr82QAt6A+tBALM8m6IG3OVmQ84NYO4jE0gHElM7XocvikZGV6Jgt09PTp
7WZAb6Vtdkc+Zjw/FpzQYzCLsL5u2KggZNJkBalwPiv4AlZ+LRSKBzjOOBYI
Byir5/mC3eolBNBgdHlvW2i92cgIYqqIlxEH73GAFnsO69cZiqw8WV+zDaHp
iCiDCBWN+EPz66WGukeMUiLTUGwqwien5zYWooLf7WVsoclCmEEM9+weA56v
Z2JBo8nptnqjXLaSTT16KPkdFoigsEahJyGGq2zEBAlhGBIiuEaDLVR1C5F7
PWNOzOJOFjKM3WgHlB4fhCHIZM6l68cU15w0yiBsazursvjgMDNgoF1YKFus
AGBR2wRmq6Qe9EJOwGWZ3xZvNMu0Z6jXjk5M6LuVFI3Kcpj3OmxkY5C9Upwj
cS5GZ8oPmOZhlZhoyAcJy7EGramZXDIP2V2eUlb+8987Mol4uULDiN4CCrX5
mWlfl9sJA1AkW81QGABZXHaI9LPpQqS8nWK8Spvfh/kpAIQdaHUJp2OUKAkY
zuK3FNUiVRi/2dY/bFJ0WyEpkWBJ98AIjHhTs4LTo7cALX1myZ/eIo90g916
osHuRgke1IA+ORnnfZ6RcKpnBBwnBNYLGIOPSuov7WE/rDS47SNGsxbru+mA
3Q2c2m2Uy2TzBFdXbChpjhTSXZfdAMhKgj8oQVyWpUzG8PM1xqVEAkIwQGiF
TDhSqcRsLhgFsdRUfS7pACHOSSxLQkTymlw2i8XkZRFfLhNIyaysMGIk5mNx
UwAhF9cdzlwxHvRhnnhep02UZbfFEcCZib5R27AWRMMyMNrTBYSmZRtqu6ND
hwcm/MXnS5OMH/RAgMFqGlMN3uyc0LgJXkadXMQ8pOy/NiELrNNOrm0mutqY
MD50d0g3oXO4Te0TTAgnYNto381ug9Wg01rtHur/B1Xae90=
"], {{0, 224}, {224, 0}}, {0, 255},
ColorFunction->GrayLevel],
BoxForm`ImageTag[
     "Byte", ColorSpace -> "Grayscale", Interleaving -> None],
Selectable->False],
DefaultBaseStyle->"ImageGraphics",
ImageSize->{126., Automatic},
ImageSizeRaw->{224, 224},
PlotRange->{{0, 224}, {0, 224}}]\)]
Out[3]=

Performance evaluation

Get a color image:

In[4]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/afdfc50a-fa1c-4f89-abe9-e08fd960865a"]

Compare the colorization performed by the net with the ground truth:

In[5]:=
With[{grayscale = ColorConvert[img, "Grayscale"]},
 <|"Input" -> Image[grayscale, ImageSize -> 224], "Prediction" -> Image[netevaluate[grayscale], ImageSize -> 224], "GroundTruth" -> Image[img, ImageSize -> 224]|>
 ]
Out[5]=

Net information

Inspect the number of parameters of all arrays in the net:

In[6]:=
NetInformation[
 NetModel["Colorful Image Colorization Trained on ImageNet \
Competition Data"], "ArraysElementCounts"]
Out[6]=

Obtain the total number of parameters:

In[7]:=
NetInformation[
 NetModel["Colorful Image Colorization Trained on ImageNet \
Competition Data"], "ArraysTotalElementCount"]
Out[7]=

Obtain the layer type counts:

In[8]:=
NetInformation[
 NetModel["Colorful Image Colorization Trained on ImageNet \
Competition Data"], "LayerTypeCounts"]
Out[8]=

Display the summary graphic:

In[9]:=
NetInformation[
 NetModel["Colorful Image Colorization Trained on ImageNet \
Competition Data"], "SummaryGraphic"]
Out[9]=

Export to MXNet

Export the net into a format that can be opened in MXNet:

In[10]:=
jsonPath = Export[FileNameJoin[{$TemporaryDirectory, "net.json"}], NetModel["Colorful Image Colorization Trained on ImageNet \
Competition Data"], "MXNet"]
Out[10]=

Export also creates a net.params file containing parameters:

In[11]:=
paramPath = FileNameJoin[{DirectoryName[jsonPath], "net.params"}]
Out[11]=

Get the size of the parameter file:

In[12]:=
FileByteCount[paramPath]
Out[12]=

The size is similar to the byte count of the resource object:

In[13]:=
ResourceObject[
  "Colorful Image Colorization Trained on ImageNet Competition \
Data"]["ByteCount"]
Out[13]=

Represent the MXNet net as a graph:

In[14]:=
Import[jsonPath, {"MXNet", "NodeGraphPlot"}]
Out[14]=

Requirements

Wolfram Language 11.3 (March 2018) or above

Resource History

Reference