From 008c6ae399e96ef7b6c3920bad6e661162535249 Mon Sep 17 00:00:00 2001 From: Richard Evans Date: Wed, 6 Dec 2023 00:06:20 -0700 Subject: [PATCH] Updated MLE.md --- data/{maxlikeli => mle}/Econ381totpts.txt | 0 data/{maxlikeli => mle}/clms.txt | 0 docs/book/_toc.yml | 2 +- docs/book/basic_empirics/BasicEmpirMethods.md | 2 +- docs/book/basic_empirics/LogisticReg.md | 2 +- .../struct_est/{MaxLikelihood.md => MLE.md} | 57 +++++++++--------- docs/book/struct_est/SMM.md | 22 +++---- images/mle/Econ381scores_hist.png | Bin 0 -> 91716 bytes 8 files changed, 42 insertions(+), 43 deletions(-) rename data/{maxlikeli => mle}/Econ381totpts.txt (100%) rename data/{maxlikeli => mle}/clms.txt (100%) rename docs/book/struct_est/{MaxLikelihood.md => MLE.md} (82%) create mode 100644 images/mle/Econ381scores_hist.png diff --git a/data/maxlikeli/Econ381totpts.txt b/data/mle/Econ381totpts.txt similarity index 100% rename from data/maxlikeli/Econ381totpts.txt rename to data/mle/Econ381totpts.txt diff --git a/data/maxlikeli/clms.txt b/data/mle/clms.txt similarity index 100% rename from data/maxlikeli/clms.txt rename to data/mle/clms.txt diff --git a/docs/book/_toc.yml b/docs/book/_toc.yml index fc1ea24..ddbf37e 100644 --- a/docs/book/_toc.yml +++ b/docs/book/_toc.yml @@ -38,7 +38,7 @@ parts: numbered: True chapters: - file: struct_est/intro - - file: struct_est/MaxLikelihood + - file: struct_est/MLE - file: struct_est/GMM - file: struct_est/SMM - caption: Appendix diff --git a/docs/book/basic_empirics/BasicEmpirMethods.md b/docs/book/basic_empirics/BasicEmpirMethods.md index 7c1ecc3..fcdabed 100644 --- a/docs/book/basic_empirics/BasicEmpirMethods.md +++ b/docs/book/basic_empirics/BasicEmpirMethods.md @@ -386,7 +386,7 @@ results = reg1.fit() type(results) ``` -We now have the fitted regression model stored in `results` (see [statsmodels.regression.linear_model.RegressionResultsWrapper](http://www.statsmodels.org/dev/generated/statsmodels.regression.linear_model.RegressionResults.html)). The `results` from the `reg1.fit()` command is a regression results object with a lot of information, similar to the results object of the `scipy.optimize.minimize()` function we worked with in the {ref}`Chap_MaxLikeli` and {ref}`Chap_GMM` chapters. +We now have the fitted regression model stored in `results` (see [statsmodels.regression.linear_model.RegressionResultsWrapper](http://www.statsmodels.org/dev/generated/statsmodels.regression.linear_model.RegressionResults.html)). The `results` from the `reg1.fit()` command is a regression results object with a lot of information, similar to the results object of the `scipy.optimize.minimize()` function we worked with in the {ref}`Chap_MLE` and {ref}`Chap_GMM` chapters. To view the OLS regression results, we can call the `.summary()` method. diff --git a/docs/book/basic_empirics/LogisticReg.md b/docs/book/basic_empirics/LogisticReg.md index 2e352f4..2a7f5b9 100644 --- a/docs/book/basic_empirics/LogisticReg.md +++ b/docs/book/basic_empirics/LogisticReg.md @@ -442,4 +442,4 @@ The footnotes from this chapter. [^GMM]: See the {ref}`Chap_GMM` chapter of this book. -[^MaxLikeli]: See the {ref}`Chap_MaxLikeli` chapter of this book. +[^MaxLikeli]: See the {ref}`Chap_MLE` chapter of this book. diff --git a/docs/book/struct_est/MaxLikelihood.md b/docs/book/struct_est/MLE.md similarity index 82% rename from docs/book/struct_est/MaxLikelihood.md rename to docs/book/struct_est/MLE.md index 039a0c5..adfe405 100644 --- a/docs/book/struct_est/MaxLikelihood.md +++ b/docs/book/struct_est/MLE.md @@ -10,19 +10,19 @@ kernelspec: name: python3 --- -(Chap_MaxLikeli)= +(Chap_MLE)= # Maximum Likelihood Estimation -This chapter describes the maximum likelihood estimation (MLE) method. All data and images from this chapter can be found in the data directory ([./data/maxlikeli/](https://github.com/OpenSourceEcon/CompMethods/tree/main/data/maxlikeli/)) and images directory ([./images/maxlikeli/](https://github.com/OpenSourceEcon/CompMethods/tree/main/images/maxlikeli/)) for the GitHub repository for this online book. +This chapter describes the maximum likelihood estimation (MLE) method. All data and images from this chapter can be found in the data directory ([./data/mle/](https://github.com/OpenSourceEcon/CompMethods/tree/main/data/mle/)) and images directory ([./images/mle/](https://github.com/OpenSourceEcon/CompMethods/tree/main/images/mle/)) for the GitHub repository for this online book. -(SecMaxLikeli_GenModel)= +(SecMLE_GenModel)= ## General characterization of a model and data generating process Each of the model estimation approaches that we will discuss in this section on Maximum Likelihood estimation (MLE) and in subsequent sections on generalized method of moments (GMM) and simulated method of moments (SMM) involves choosing values of the parameters of a model to make the model match some number of properties of the data. Define a model or a data generating process (DGP) as, ```{math} - :label: EqMaxLikeli_GenMod + :label: EqMLE_GenMod F(x_t, z_t|\theta) = 0 ``` @@ -31,45 +31,45 @@ where $x_t$ and $z_t$ are variables, $\theta$ is a vector of parameters, and $F( In richer examples, a model could also include inequalities representing constraints. But this is sufficient for our discussion. The goal of maximum likelihood estimation (MLE) is to choose the parameter vector of the model $\theta$ to maximize the likelihood of seeing the data produced by the model $(x_t, z_t)$. -(SecMaxLikeli_GenModel_SimpDist)= +(SecMLE_GenModel_SimpDist)= ### Simple distribution example A simple example of a model is a statistical distribution [e.g., the normal distribution $N(\mu, \sigma)$]. ```{math} - :label: EqMaxLikeli_GenMod_NormDistPDF + :label: EqMLE_GenMod_NormDistPDF Pr(x|\theta) = \frac{1}{\sigma\sqrt{2\pi}}e^{-\frac{(x - \mu)^2}{2\sigma^2}} ``` The probability of drawing value $x_i$ from the distribution $f(x|\theta)$ is $f(x_i|\theta)$. The probability of drawing the following vector of two observations $(x_1,x_2)$ from the distribution $f(x|\theta)$ is $f(x_1|\theta)\times f(x_2|\theta)$. We define the likelihood function of $N$ draws $(x_1,x_2,...x_N)$ from a model or distribution $f(x|\theta)$ as $\mathcal{L}$. ```{math} - :label: EqMaxLikeli_GenMod_NormDistLike + :label: EqMLE_GenMod_NormDistLike \mathcal{L}(x_1,x_2,...x_N|\theta) \equiv \prod_{i=1}^N f(x_i|\theta) ``` Because it can be numerically difficult to maximize a product of percentages (one small value can make dominate the entire product), it is almost always easier to use the log likelihood function $\ln(\mathcal{L})$. ```{math} - :label: EqMaxLikeli_GenMod_NormDistLnLike + :label: EqMLE_GenMod_NormDistLnLike \ln\Bigl(\mathcal{L}(x_1,x_2,...x_N|\theta)\Bigr) \equiv \sum_{i=1}^N \ln\Bigl(f(x_i|\theta)\Bigr) ``` The maximum likelihood estimate $\hat{\theta}_{MLE}$ is the following: ```{math} - :label: EqMaxLikeli_GenMod_NormDistMLE + :label: EqMLE_GenMod_NormDistMLE \hat{\theta}_{MLE} = \theta:\quad \max_\theta \: \ln\mathcal{L} = \sum_{i=1}^N\ln\Bigl(f(x_i|\theta)\Bigr) ``` -(SecMaxLikeli_GenModel_Econ)= +(SecMLE_GenModel_Econ)= ### Economic example An example of an economic model that follows the more general definition of $F(x_t, z_t|\theta) = 0$ is {cite}`BrockMirman:1972`. This model has multiple nonlinear dynamic equations, 7 parameters, 1 exogenous time series of variables, and about 5 endogenous time series of variables. Let's look at a simplified piece of that model--the production function--which is commonly used in total factor productivity estimations. ```{math} - :label: EqMaxLikeli_GenMod_EconProdFunc + :label: EqMLE_GenMod_EconProdFunc Y_t = e^{z_t}(K_t)^\alpha(L_t)^{1-\alpha} \quad\text{where}\quad z_t = \rho z_{t-1} + (1 - \rho)\mu + \varepsilon_t \quad\text{and}\quad \varepsilon_t\sim N(0,\sigma^2) ``` @@ -82,46 +82,40 @@ The likelihood of a given data point is determined by $\varepsilon_t = z_t - \rh The likelihood function of all the data is: ```{math} - :label: EqMaxLikeli_GenMod_EconProdFuncLike + :label: EqMLE_GenMod_EconProdFuncLike \mathcal{L}\left(z_1,z_2,...z_T|\rho,\mu,\sigma\right) = \prod_{t=2}^T f(z_{t+1},z_t|\rho,\mu,\sigma) ``` The log likelihood function of all the data is: ```{math} - :label: EqMaxLikeli_GenMod_EconProdFuncLnLike + :label: EqMLE_GenMod_EconProdFuncLnLike \ln\Bigl(\mathcal{L}\bigl(z_1,z_2,...z_T|\rho,\mu,\sigma\bigr)\Bigr) = \sum_{t=2}^T \ln\Bigl(f(z_{t+1},z_t|\rho,\mu,\sigma)\Bigr) ``` The maximum likelihood estimate of $\rho$, $\mu$, and $\sigma$ is given by the following maximization problem. ```{math} - :label: EqMaxLikeli_GenMod_EconProdFuncMLE + :label: EqMLE_GenMod_EconProdFuncMLE (\hat{\rho}_{MLE},\hat{\mu}_{MLE},\hat{\sigma}_{MLE})=(\rho,\mu,\sigma):\quad \max_{\rho,\mu,\sigma}\ln\mathcal{L} = \sum_{t=2}^T \ln\Bigl(f(z_{t+1},z_t|\rho,\mu,\sigma)\Bigr) ``` -(SecMaxLikeli_DistData)= +(SecMLE_DistData)= ## Comparisons of distributions and data Import some data from the total points earned by all the students in two sections of an intermediate macroeconomics class for undergraduates at an unnamed University in a certain year (two semesters). ```{code-cell} ipython3 -:tags: [] +:tags: ["remove-output"] # Import the necessary libraries import numpy as np -import scipy.stats as sts -import requests -# Download and save the data file Econ381totpts.txt +# Download and save the data file Econ381totpts.txt as NumPy array url = ('https://raw.githubusercontent.com/OpenSourceEcon/CompMethods/' + - 'main/data/maxlikeli/Econ381totpts.txt') -# data_file = requests.get(url, allow_redirects=True) -# open('../../../data/maxlikeli/Econ381totpts.txt', 'wb').write(data_file.content) - -# Load the data as a NumPy array -data = np.loadtxt('../../../data/maxlikeli/Econ381totpts.txt') + 'main/data/mle/Econ381totpts.txt') +data = np.loadtxt(url) ``` Let's create a histogram of the data. @@ -138,15 +132,24 @@ plt.title('Intermediate macro scores: 2011-2012', fontsize=15) plt.xlabel(r'Total points') plt.ylabel(r'Percent of scores') plt.xlim([0, 550]) # This gives the xmin and xmax to be plotted" + +plt.show() +``` +```{figure} ../../../images/mle/Econ381scores_hist.png +--- +height: 500px +name: FigMLE_EconScoreHist +--- +Intermediate macroeconomics midterm scores over two semesters ``` -(SecMaxLikeli_Exerc)= +(SecMLE_Exerc)= ## Exercises -(SecMaxLikeliFootnotes)= +(SecMLEfootnotes)= ## Footnotes The footnotes from this chapter. diff --git a/docs/book/struct_est/SMM.md b/docs/book/struct_est/SMM.md index 119407e..0c3135a 100644 --- a/docs/book/struct_est/SMM.md +++ b/docs/book/struct_est/SMM.md @@ -28,7 +28,7 @@ Let the data be represented, in general, by $x$. This could have many variables, \theta \equiv \left[\theta_1, \theta_2, ...\theta_K\right]^T ``` -In the {ref}`Chap_MaxLikeli` chapter, we used data $x$ and model parameters $\theta$ to maximize the likelihood of drawing that data $x$ from the model given parameters $\theta$, +In the {ref}`Chap_MLE` chapter, we used data $x$ and model parameters $\theta$ to maximize the likelihood of drawing that data $x$ from the model given parameters $\theta$, ```{math} :label: EqSMM_MLestimator @@ -271,7 +271,7 @@ Let the parameter vector $\theta$ have length $K$ such that $K$ parameters are b Recall that each element of $e(\tilde{x},x|\theta)$ is an average moment error across all simulations. $\hat{\Omega}$ from the previous section is the $R\times R$ variance-covariance matrix of the $R$ moment errors used to identify the $K$ parameters $\theta$ to be estimated. The estimated variance-covariance matrix $\hat{\Sigma}$ of the estimated parameter vector is a $K\times K$ matrix. We say the model is *exactly identified* if $K = R$ (number of parameters $K$ equals number of moments $R$). We say the model is *overidentified* if $KR$. -Similar to the inverse Hessian estimator of the variance-covariance matrix of the maximum likelihood estimator from the {ref}`Chap_MaxLikeli` chapter, the SMM variance-covariance matrix is related to the derivative of the criterion function with respect to each parameter. The intuition is that if the second derivative of the criterion function with respect to the parameters is large, there is a lot of curvature around the criterion minimizing estimate. In other words, the parameters of the model are precisely estimated. The inverse of the Hessian matrix will be small. +Similar to the inverse Hessian estimator of the variance-covariance matrix of the maximum likelihood estimator from the {ref}`Chap_MLE` chapter, the SMM variance-covariance matrix is related to the derivative of the criterion function with respect to each parameter. The intuition is that if the second derivative of the criterion function with respect to the parameters is large, there is a lot of curvature around the criterion minimizing estimate. In other words, the parameters of the model are precisely estimated. The inverse of the Hessian matrix will be small. Define $R\times K$ matrix $d(\tilde{x},x|\theta)$ as the Jacobian matrix of derivatives of the $R\times 1$ error vector $e(\tilde{x},x|\theta)$ from {eq}`EqSMM_MomError_vec`. @@ -324,12 +324,12 @@ The following is a centered second-order finite difference numerical approximati (SecSMM_CodeExmp)= ## Code Examples -In this section, we will use SMM to estimate parameters of the models from the {ref}`Chap_MaxLikeli` chapter and from the {ref}`Chap_GMM` chapter. +In this section, we will use SMM to estimate parameters of the models from the {ref}`Chap_MLE` chapter and from the {ref}`Chap_GMM` chapter. (SecSMM_CodeExmp_MacrTest)= ### Fitting a truncated normal to intermediate macroeconomics test scores -Let's revisit the problem from the MLE and GMM notebooks of fitting a truncated normal distribution to intermediate macroeconomics test scores. The data are in the text file [`Econ381totpts.txt`](https://github.com/OpenSourceEcon/CompMethods/blob/main/data/smm/Econ381totpts.txt). Recall that these test scores are between 0 and 450. {numref}`Figure %s ` below shows a histogram of the data, as well as three truncated normal PDF's with different values for $\mu$ and $\sigma$. The black line is the maximum likelihood estimate of $\mu$ and $\sigma$ of the truncated normal pdf from the {ref}`Chap_MaxLikeli` chapter. The red, green, and black lines are just the PDF's of two "arbitrarily" chosen combinations of the truncated normal parameters $\mu$ and $\sigma$.[^TruncNorm] +Let's revisit the problem from the MLE and GMM notebooks of fitting a truncated normal distribution to intermediate macroeconomics test scores. The data are in the text file [`Econ381totpts.txt`](https://github.com/OpenSourceEcon/CompMethods/blob/main/data/smm/Econ381totpts.txt). Recall that these test scores are between 0 and 450. {numref}`Figure %s ` below shows a histogram of the data, as well as three truncated normal PDF's with different values for $\mu$ and $\sigma$. The black line is the maximum likelihood estimate of $\mu$ and $\sigma$ of the truncated normal pdf from the {ref}`Chap_MLE` chapter. The red, green, and black lines are just the PDF's of two "arbitrarily" chosen combinations of the truncated normal parameters $\mu$ and $\sigma$.[^TruncNorm] ```{code-cell} ipython3 :tags: ["hide-input", "remove-output"] @@ -394,20 +394,16 @@ def trunc_norm_pdf(xvals, mu, sigma, cut_lb, cut_ub): return pdf_vals -# Download and save the data file Econ381totpts.txt +# Download and save the data file Econ381totpts.txt as NumPy array url = ('https://raw.githubusercontent.com/OpenSourceEcon/CompMethods/' + 'main/data/smm/Econ381totpts.txt') -data_file = requests.get(url, allow_redirects=True) -open('../../../data/smm/Econ381totpts.txt', 'wb').write(data_file.content) - -# Load the data as a NumPy array -data = np.loadtxt('../../../data/smm/Econ381totpts.txt') +data = np.loadtxt(url) num_bins = 30 count, bins, ignored = plt.hist( data, num_bins, density=True, edgecolor='k', label='data' ) -plt.title('Econ 381 scores: 2011-2012', fontsize=20) +plt.title('Intermediate macro scores: 2011-2012', fontsize=20) plt.xlabel(r'Total points') plt.ylabel(r'Percent of scores') plt.xlim([0, 550]) # This gives the xmin and xmax to be plotted" @@ -975,7 +971,7 @@ name: FigSMM_Econ381_SMM1 SMM-estimated PDF function and data histogram, 2 moments, identity weighting matrix, Econ 381 scores (2011-2012) ``` -That looks just like the maximum likelihood estimate from the {ref}`Chap_MaxLikeli` chapter. {numref}`Figure %s ` below shows what the minimizer is doing. The figure shows the criterion function surface for different of $\mu$ and $\sigma$ in the truncated normal distribution. The minimizer is searching for the parameter values that give the lowest criterion function value. +That looks just like the maximum likelihood estimate from the {ref}`Chap_MLE` chapter. {numref}`Figure %s ` below shows what the minimizer is doing. The figure shows the criterion function surface for different of $\mu$ and $\sigma$ in the truncated normal distribution. The minimizer is searching for the parameter values that give the lowest criterion function value. ```{code-cell} ipython3 :tags: ["remove-output"] @@ -1071,7 +1067,7 @@ In the next section, we see if we can get more accurate estimates (lower criteri (SecSMM_CodeExmp_MacrTest_2m2st)= #### Two moments, two-step optimal weighting matrix -Similar to the maximum likelihood estimation problem in Chapter {ref}`Chap_MaxLikeli`, it looks like the minimum value of the criterion function shown in {numref}`Figure %s ` is roughly equal for a specific portion increase of $\mu$ and $\sigma$ together. That is, the estimation problem with these two moments probably has a correspondence of values of $\mu$ and $\sigma$ that give roughly the same minimum criterion function value. This issue has two possible solutions. +Similar to the maximum likelihood estimation problem in Chapter {ref}`Chap_MLE`, it looks like the minimum value of the criterion function shown in {numref}`Figure %s ` is roughly equal for a specific portion increase of $\mu$ and $\sigma$ together. That is, the estimation problem with these two moments probably has a correspondence of values of $\mu$ and $\sigma$ that give roughly the same minimum criterion function value. This issue has two possible solutions. 1. Maybe we need the two-step variance covariance estimator to calculate a "more" optimal weighting matrix $W$. 2. Maybe our two moments aren't very good moments for fitting the data. diff --git a/images/mle/Econ381scores_hist.png b/images/mle/Econ381scores_hist.png new file mode 100644 index 0000000000000000000000000000000000000000..c7c17e2ae9d9ee302b953175b9c49c412bcf3b7e GIT binary patch literal 91716 zcmeFacT|(xw=RqbicyLRs1yM!MZk&>X@ZD!q=XWxinIif-kSo7Td@!jk*0(e0s%sh z-c%Ho5_%#aD$;wCe%EUd_V(vF_k3gA@tuFRV<0hES??-y%{9yOOz&uGDl@~lV03hJ z%qo{I=+MzIkm%?(qnNgUcP?HN0B>wEwN+HqR#8;s*LHTiZi}*_qZ54S_TYq?f+c$> zpW1GPw92%kiosq^qms>a$Fcd?*2cgV=@na#}1@_NQ$BH5pM*H%f?F3Iqi zu&A&7B2M;7-n8Wh%9+RQ=J0c}gp%8U0~;+AXe|QL;VlSxH8| zg2S!7f;|WC<}sHMBwIcYdDlBjAMh3bg!>*zbTVp>wVTKjEs5#AT<$=~^00!vyVW1q z%zxthoi73(j@@OL*lF?To7}0&1@&8^r)L{F_|;exzbk&R`Sy-UTwC6XC!1g{v9Qh1 zXtjud8tQ=fy99@ad@`u42+ng>LxzN#Zg4djfH>v0x8VCKi+3Fd( z8fvJ+uR5Z|EU!6Uu@dt{J3;->NqfSs+|F_I2$eZ+icO!N0s41Nqe068{nY%CVZr^lAUPLf4H#@gUVtJ@Oru$4A zsdBW>YqYR_u5b+5vZ-jyCBCnuPAX_>;+e}39B%$*`z8i9dAi^KvZi-qK-L-mRipGD zA2F~o(K8@j8UFOOXz+1}0Mf=HDLzJyVR4iDV+d&3f-Y$G|U7P~!VTS8Pr~@`T>p z4o{mu%{Ewc;SXH}2C8ie7LES+M)BYuX3K_Z-1>+0W-x)Vp$btF`?-GK3Gm=Ri@UOa z=!*ZdPN2nmtGtPZKh2f_IsJ#O=r;ZTyhXUqu&(le^3ui8;?~AS%bA%Ogn@xU@!n>h z-$oY%+h)^+mwNfV({EvVo69(?Q&>NT0h#{erDRIrmX7mA*848~V^)LhZi7RuSuyf8C_&(Cim zp*nVL!32w%{m0rpzy}74ii+)nyQXP8lWe^B*WR6+Pi*G&;>{S4UUtm!y@B*x>gU{t z3G-8Z)NKD55oC$y3FJ=JwAZY^Z|tR`+LUn5Pd92hQe|g^VkOW|nfD&&KXy{Zz?o#Q z8)Y@5V#~3sz1SnWbgJ&0gvY2!@%;DL5sZOZ&cXYVRa;p1t5@g?xPBMhP0P#QAFyp^C$id+VQ^s0K^BM>eS^aL?4{-#cdOe;D|^+h=w- z=V*fO$>AsDj@CBw06!+u!rWLw@%-6JM8s3fS$$;UX#$bsa%EK1#i(6ZgYk9+EOW~L zQ~-kE^0q6cloR>}277`<(Mlh_F2!2l1ZA>6&j>PS>JYGl!YAs6Dug{qE% zpe^Qif4}p)7CJs9v3jKad0P?NxCFr@aq_|MLgu=29BMI04qQ5sKo5o@%i$&b&(UsT zU=jO_UeO@9jn95dl>%QNI9I*7`kV6PZxEpk$Lx%_V^&mPL6E-a(b2oVsU>+XTBFnW zkNJ%ZCCr&N$3HRse*ShamV7MVMmmY|4bON6btJV-^>;m<=uF@By_*4v8QLECED1Ei zShfhke(%YK)~K;)%URWfpV1cBbdk0LzbP7b>V= zdqs}&hSf;E1kZCDr;O|VzEG1wKNkv9@Bd2I4O*xoY1x1a{}dbfijcqpAZJl`9bPMX za~l%({&VN)==am+FFl*f<$E2BXG`pD;omekyZq1L;OOGAI5WV+S`=cnmH#=KpfB5K zLYB?SnR^lJtN?eAs=z-d&u&KRX52J)FUPb5Av{S_^f%REr$KAd>YdiK0EVj>a?XHZkm?H9G8IC-xWE6N}fxp8wFV*ACM* zOaHyDrV`Xq_!Vu*-}i!J7j16Z%8BhLs1eFu%l3EScttxz;Z5Q%)l`85iCDQ6Zrmyr zF%>JziyEufTMT@BirrjY^^&)^xw(Ox!veZs%)w2>QOFu{AW(jUrWRTp%-!@|z?tU| zW8FAF4~#9k%lt@wi|*7ym(-Dtyc!CHq7&U=x(T|5vd!h}ZEbUcHg7f7)z$TI{OSV+ zDSw5wpszGOhMa%~|3u=`hLx4*O^?MG54B|%w6-FU#qRhKO8$FxSN$(oeq3Q;$HjYG zR=baC5_cZFNN~pOYKXie&rDm@7ANZ&JvT7E$D;M&>j(USNg`3lzA~86eRX+``dsO5 zRk$cxIrEhol`a2@eQl(W`v5kY(wweqRH(nrruIT&C{4`mUgmjc6cq&ZzKgQnxL+=b zCAJuGpLHMtElqYP9+B|5v7#_!({J6{VB5&)5%7 z)bJHZxqNSb-im2GIl0%K>vS#Gf&|i3QNr57jX0J!TN;Pt={B-I*Cv`Lw7aI`Oqhxb6A^o`LCgHl|YQOIWab`fSeUN+chRad~_0D*?f4Y2dWDu84 zGRnK&@#Mr+R72u=F@EP1sv=FL zwO#tdU@W}g@bz&@dG^#}P7gGR*FKt((s~p<`Lc%rx$mV813i+nsbD0R6c>wH96U_2eRoyD zc33+YuYOGkoaG{#h`XWBuRK$Oj!N`xY4(lP`3}tXn8!N&^h|n2G`&;H8%mv1ub@^8 z8sR-RjI2NpVA_s}zMVQ4C~rWM@+{G}2`eq6 ziC^1__Hbd3y$;={pxeVb|BLBH%PwH*mlW{VN6^um(uA;x#2-+Hg|wU6h5Z}5#P6~x z82>>xI}|ABDIU%rW7UYv_vVb02)opm4T?@zm=+tSivq^NjvW}xY) zZ%bR7!8ciUaC6Ae42MB4>SUcTw1@i1Us^ZL&Q8W~F4PFMpq}K|?QqUxlqlmfYluW5 zmGro|x$)-Z{;g#4mCajOh$$&4;h(}3z+U&!_FBdf*=3{zoyKszPaBRb|8x1qU_4nB zKQq*Zw719MO^aLIM<`aXBH;<02Q8mIp&kATQ^GhQ`Wj<+|Jp)JlE}bms1obLL&0!>=H*SHizSdbW_@~(PnnKshU>3h{K6O zyMxSVrv1$G8(#J^&{-T*xnaG2Zi1fBD{Utc;r3tO+?)OOSc`%#=)5joXKusGy!?TS z3!haM(b8h&vIU`B<3H5Cnq}k@+mb()qnjFctSpK}&xVEaMS z5v}Ib``x{U(BV6AI3jf8@Fjr>h{oaauYct7S(>Op*FBRWrMY4A|B`<5bj!Wm(6Wz+ zwg(^g8S;)SX8Jkyan^XWQtI5^qiSE>aQ}Fwt^GsHsS!do%XSd@{7$-Qvjjm?Ci%|9syJ{ACEZ+YN3B2+RHDV*Q5DT@xmYu-I{46as9(y z;8toMnM+~=3c8f0e8#9xL^Ak+j?@pjjodKp_0A{Svk_feIJsrLNX$Gk9bCvzkI~Qe zb>Gq{RCgn4od(HoW*3$C?L_QpcN?YP#H<;kT<6Ek36j-Y{(5@t-pA_9vUC^Zl`YxJ zi*PyH8*SIAIqvCbJ3sP~2yU~73)jZe*n{Lh(k$#2lRYy=mcSlLJ*l-3(rG{jWu4as zYdbbGfcBXLm&$+RjHLsu`e^4HMbI&LLcgKhOqkeEhwYF2+xFje_$d_s$^%W${<{wUuEYPD zp1f6hWvr`|1mdPm=GPV|dDaml-R0ebgM*n`39^t)Qyt2j2YK?VE6de$-Yz0;$(_X> zVYvmUjw`?nI_;PJ;zczV7gr`Qfs%&L-`vvix$-97vAe93vmE^Ggpa>5)QV41$GnJ+ z*R7Yh_Lfao%KJMol2{{O_zW6N^9WnFWtpZd;L5xQyjVV@NI0H1t&aAdkks{pdK`Ws(B zv%~>Qk4kkcN2^7P%2tIdmGDbsci4y2r`Zzw9dS}?17d~&k4bb1+un#nRV$HYsgb<)OaF(T-u z6pfSyb#a+`fiRvc%DirRp7FL#DIsu?Z|?_t(H2(vj}*!7&<>r37Zn4^k2Q03wPafE zOMbacP{^2$-`t51&;iyfQzE;K&|f-Eq#Iw~5V+La(w8=E-g3)HZSIMbONz#m<2`;N zS*pAf_s(_^tmD1jaEcnUSDnC9gad5zTYtQGLyg>>+8$#azmHxDzop4etdyvuCJ+_h zlT)*Q`wYS~|B9Np%w02B0nr&nPF9fsfo`_QbQA{j8mGs%98KVMEIGmMwLH^C%{0oZ zl2NUXeO)TJdEK_43yO!ai!uot`*~CDm#}VPSfM$>Gi{`y#xV)UE>w2u?9N(Z`tq5j8|+XdCv^8 zWeHg{D2QP3egXAZldN zUX{kLS=km|$6*DVXTJuF;)Bu=U8%Xbo?hIBM$IuNH%z2Sacv@hn$66+0M;+tcD_xZ z*_Gmp4@f;SD?h%@esZY~8CRjsvmeulgP+MvB*TuP`Sgy@H0x@GoJ!4l`31K4R#|q^ zY_BT;X{Uri7|AS8Hwm=s$h*yUSch4tt{RZu{l?3 zLcg@(k$Vt6w)s(AsTh5}#r7HnDaw)b=qf>(KEBl-H3Z6V$yD9;dMT^g$T@nU5bM>& z@>TnKxz!R;BcRKlA^3f}?34Sau9e#44eb6eqDXI*lU#MpO z^Sx}ZczlL)QWBMs^!di=rKK{@La-c~9-nR~w<~y3Y*RxL`!e(Da&^XC6RBCo`N9lW z>SY&EhG@6`q%exUfvJ_SqDYx*Xf#Vk+n^w$3M_tFCPXD+Wjb}_)^1HEC41aAMOGPh z1kzSN3Slc`p-Y#l*fm2Anufg{E$oXYc&m+FzT{rXxaU@jrFTT#&ZheO!#Z)unbK*p zTD2c>LESPE5i`)|-940D?ob`ZTJTs)wnMD!#}~qx%e}tsY9i&=3>I_>d$Y+E-jnZ+ zj(m@GWN|yvF1+W<+=jo|!X{|Urbk@Rd!gl|#Za<65<2W%$lTaIgZAemo&{arI8k$w z^ewqmExAr2Sp~`R-kef)*1GN)7++7bQ0g=Ci$rByDbY>4&9R*z@J!zgI7qjK-oI}f zOz&PE^{_mQVF_!Kc$MwjD`L4lkp*pFF>}{^=>UT+>=RH5?ZS-a*j#p#e%yoL0w+E{b^u=C{ZD{rk0oul=UrhKM&$Jc++Un^a zVhl4NDrqBoS~Jadv#7KQ9}7W-@%jH)%xVr7j(dp3sKfk;&(%b-U7svfnt88^106qC zha)d?92!`djF!cY&wJH}^?9L2ulDT7a`c~ywWL00NoxzwVXJTpHG{{hB{ZaoK897) zn!{cb^>c_V+H$Lt(d9f{*>f2B)0kAzNAMRzrxNDX4bP`k_vDjRw^BvMn^Yy1&2VKY zS_uxhLUDci{`I5&*$e&qVc*!zY<;fItF08Zfh)DD&;r)VM5+z>A@eJ|6exd&A)J$e z125aszx&hfwYbzcU46E;+9r(bwu!BBKY|9>=HH8zeyEe`<2g6_CA(q=M=6&*uY$rtSR!@EMST#2|KHvow z?|c($zz@Zy){B0Wq`Du24kF3^Sp-ca-wf7An7?UphTJn@d=JZU4OUL%QcuDl(gyy* zgN&i*LvMFphtCT6Q1qy1C)6pACA>cwPyo)wQs_{$}7-5C&z% zM}FRy&#bw%&^}@yB4Fw3_+q{P_dE%g)(>ox{Y~56n;UPNS?pewy>%Xs$B$8k4Ohyt z$_+fH^AeViJ=4dceT{8r?r8ab*(`!*&$K;*VbN{ltN^!C``NMy=GZg#2uVe?Uf(gF z?(GSlKPDmqq%*&frA$o?qA>J_d0n#=zs*c>JG>MqF|SXVDj}Y60{8uFs@(FssAERE zRz9`1aw=q!eb{ZlwS53(rANS8kG)Hkj~c8;yn4h?zTvcc?xLMP&p|a1G56XDB6w+xBV0@2Z|BUsmR|qC^?2kV!I(O2$%H7jV4}WfhBMx(f%Gw{;IS zaXJ?EMzNs=nzrS`RTIS0hTfaQVqIBfBI?NG)KmV;L*@lBr>rT!Y^wGvU+f$U$>J_jBIxNF@5oYQ?3T5|Q4$B?-Ahx`8u6th7v%D)!-f?kuM4;V!x>32hRn_my zb7}VL1K94N6-r7P{&8<5VnC+n-qv|OmIyf}7P*bc@>~3x9Z|p=wpEe821G`s9G;zy zXg$+TPEf0CeW<*DKB=l^)grrk;3`fSpw|B7I<3Bo2qgm>XkL zw9&J}9X&h5@D%d-Ng1#7XOr1kZIN4|f~Dhbq=-IzmeZntM{Xm@_==OZ6|Y8FKd}ZK zU;?VJmfN9Z;*B;|y1sk}UVZESfeM2nmQJ)%$z-iaHB5;2WmhNS0^W%n)DL&W4J8e4 z_3r9IDd~MqPE-yf*;5lp2kc%W)YR14>C^GIaR%9>N&M{>$`StE`1b_|>~Kl`W@Q|g zaaH?$hEFtQh@t)ToUr5S3HJ^6qXr1RrrG68Zt+h@GVTTBVZXCuulYdO%uMfI!G`om zt_@jo)L0r$uTQf=2Q3YxKKD$VI>#Czmt`GRW9~DO-*f!O&Cc9iVKqbiKfpOOmBSK- zdM2*c+-KfoG%)VT%M-9TQV^z(gq2PY{AHcs=ibhht~fYDPFjGK0=;MLWcxlPSR8K? z3wIl=%%FZv049Y@)?UYrJg?K*m3pJI(9@G4;}9*EP<{V-2pU!}y)BpmE4V)Ml+Zhx znO6_nTH+xSX)5w$v~CC%<^SV^hS3M;yF_%Lit0r7eYe+9Elg(r=o@qr^&sLCO8Nj5P+Z@(lznDnkx3h{t z1iG0T&3-sl-yY#Veb8s><5>;r0ao@0r7H_VV{t+$)?)*$?NgCe z@Po8_LP%5V#Sx}gYYVH?t^E>P{aG!j@z%E8sD**1kxx@~%n5%Pnnl*(pPsV5FKSwJ z+SfRd#i9Jim%G)fDU)+<%cWhb6`NU~WVLnZ8_&Cf%hA^Y8;!Umvov09`98@%TYgZ) zd|s8Q!*!)Om~nem(j@$4S-|Rw0Dtk9aFME1&Zur(x$?evpQ^?D$wA0)5W%}Lm6&1q zwJKYUtm8P!*d5$V=P}cSrAce4_OVjt%$qGC2^(n+cI8AH5R)6!eHMRFgYEWF+Qz&u z%du&E?v#Q-ToLLgZyJDXD#OcIExP8^#R2in_`o6nicyqw8%$3PZPA23 z)*@;JEHsTV8o=9zlEOd=r~~O(YDKw{hB1fBn{E9#lyBcA)`U>RAJcWhYMp7$GjpnB zOli;*XY^W#ycl9_JdX`@09Lo9QWWVxt$bfZ>Fl8WY?$0i<+m4F8ju#q@_mj6pa}E*2EejRo-Wj1_Wq1GF=Vr(PhckG8vTC~d>Ovbx zQO~j+64E9QilVjjoRT?=C<uMoc~B57dMv_R96^*>3JR z*0ebvlLN483)u<-u84lity&>AE zAsO-&)I_ROdu=$Pg9$J0!-W~_;pfzFg+yT}u+f9?zR0VWjgK%6N||&Y>p}~cFSQh; zXEX1WU*|(P`~VT?oz3Wzj;Rer$U#JH{6apluQDaX8@usb56$bQEuUu^?jua zN(OlnDC&?rQtQUlyq1it>MHVIULYvg0vklYQ|BeN1!0_%+@NCl5Fu*2n21mWo|YLb zn!QeNukPLy{(-9P;yXkNYWb9Kh3ta1$QKB;0l1%5Wi$dU)U4x_BJTI)3&;i4iU*9% z8Eas*Wc4Id9;jQ}+Acn?5{=8pa+tun*$D%53*_)@olOgeE-|9vi)!a%yP{9ig9l+m=IX&HPjdDEn^KDPSX3>F^_i!SoPAf62P!(Oo44S)zx%ruIkljdA& z;ESmxFSDl%z-7W}8(q>({LIA}4UL6Tiqd#Ot_q*UX{*hF=+N|slD}OIYreCPKWlO* z(>3v*`V`~Gm-q_C1k!mN{E3$~@>)D=kp<^=E6+m=&TEYq#a?e#1zD-*EgRk1DXyt$ z?$QUw2LKEb!b>4YBU%8Y2!#U zMF0RgjCZU{VWPyEYmwi$J^e6!2Myg|MO}$R2tc_)g&pjFL-3qh8$n&s_K2H-+kbAc z=$_8-FaW)>IkB-)lr=}R`J#Ao6w@^k+Onb-b&W%ybRq17_OW%jk*`}Dfz*(gGz+~L z!%ys-iL1zdGFa5t|1RnpYBK2GMWs2x|4Qn=$Nzszry8GcxK+Ki1BwR%$BM-3|D#93 zrXf-NnqERdw?8%Bdr+wO>-}@X`32oMJ&&!a6^)PaIaV*voR+Y5PM9T`wvt>Pqs>cS zNr-xVIC78I&cNI+ADDpefi?Wt$nT_=A?$qO8=3~R%e?80&jPT*mYBYcU}t}ve10(g zM%Zi~1V7h2H*%U=$~hTWV2K{A{(&B2?K#!0c%-#>$#^hi6_K2)_Xp!Q8&04Z!fAnU z{pNT>f;aH^7{8{%H!h*AY_~Z9RK<3Y~jHMksWhB4NEDs9tON@*v5?anFblnT|q1_2{rdbl;c@`}f!L7S-FW6MJ z#*dIZs9e;peJHfiN@hp=Jqh3TLLXaXv(!#=|>hK zyNzJyD*&JtLLDcR={IQ&g05Y!UiSLcbz&&P6-MOPsDI5Y$}7+1b?X)CGEMN|NFzkPS{`vqp1;Qv|oUl@@{^jr#tbk)t8kpD%A zx;kpceSFez=$vQdnQ1hsC^SXAd2Vd6T(NV`0cl?>fzN&O-4cKS-|8^Y`_r~`X&V33 zUkBW(DcpB!*JBlD4zY>0Q}VH=+Kn6Mg@W&*hr**XUG79N^U8G^U*tjiEwGXn-jv0p zl@}Mgy#Lx-fNl%zovov08!9%@t-%uIXvTp50zyMB9RSXo?kV;D3X}$M?Mg!s^xGP2 zs*lzYHoZ6!qOU)r(ZRESbuEJ&sK@U`V5*DP#tT zX?y+63~HutfstHEt>f?V(4YI?wYH!9#Kje>+Xf)WR4Xe(>}XThYyfh9LH6iUWX<7- zj^rH0-(jCI&0z{`@`-b6=_tpmS!uxlf2YrQz~Rn%pir9;01B`5Ga%o60*I2mw6-U_ z8RssE0?EQ=9cSEd3tzJFx2v|Q|J(N8b@(Y1|H=bRwg0;g|E|Nodh*}<0{QQ~`LF%( zzh_73?`j9k<$-Rq1v^K_g8TRHhfAxNn#P?uecCWV*00LI&(BY3RChpO8|e|q#T<6| z3g8zsjQQ(+j!y^kA+kwP75h5(Pfz6!R5IboL-+2TQ2ZeBs}{iTq_Y1{kGQ}oh;KG7 z_b<&SktiVKkA@IuiMU7?t9olift^}<(?TSM+dzO#@n@(_ zvO0<{H`%kLb^~np1)LBY# zaY0Aj%zO0vp{e1QFD_DA49wAa*FWZ-xca6&@%+t`g<6>kMMa{f`L9E@;?bLxwMcKG zdg8WsJw4?W>EVIS>5N6R&$iH5cs*%3>*k>L;d6d~dEZgdQ3~o@t607UUW@YWM)fw5 z_h(f?%jQE}!F2R7G!eHV53}poLcvz@{zyQxqV;`EwBby9Msw|p=lJG3y%z}>w}@0} z1~$nDKfU1AAmc}4eNfoBmSaEbI=Et5o*y9(T=AQ4`oD`ZAUhMsMFGv}HZA9UnFiFZhn0mz0mVvPGK#wrkvrQ1nA86~F(DoJ*{3r%e639k-n2M)%c8{haAC+> z)aU;awZ1;WkPqjZ^9is6@>vZ80_DeOJTC9s@GhGW}ixF53TPFhPc!m_e02v#Vxy6_#6gF9LMdKhku2pAE*N2-9GbUXTcOks{aRrc?GjE%)w;F zfXNo#;Gk!kUSlWEp+9iwLj?8=`B$RrZn+)+QM~(crEn)8IPm=s1_QRXIvjuP;|1Db zS#zHqZsPCXXQ2Z13K5M!T8h`dySe!qpwy~gPSkdy1KYB> z!uM^#Rq%MfON!%H8Q9(elHw8-O=#)r+OL2I=)86SX65yl1MpjJxU0>N!S3>a>-SX8 zw}7N8Aw&Gl?sfOOboXtJ2ErAb$B2$M$aAO!n*$~j8M<%_nMv-WpKJk|H?P!Z&KU52 z(G`K4Bs|AI6S&lH`Rq1$!<+-Yc!0Mn(URkn@g@TNXeGcQ=SE5Krrs98>DHm=Ogaw$bky|-&3wcoaK5^n}4HeR<4-{EW2#A}Nw2zjr`p3!z zl%cyyq%zyd5*JDxzki4UQKr)naGI(-uD#srI?<-UNGRJ>8%5e8#GlgiiClzJV?%zYhXmGeQezs`T+ zu6a8yyNs&}x3J*dWc~)^Bk*K=ANCV*RvLxUFrPL}%Kv&;MHtgh>75Z^z7fc`8^uc(>Ij&r-`6%V-haUs zDtm#dxzFp~qHoXvzzS9X)eJiy?VlogJG*2u9Eyd`hKTzLEM>@*L-<mo}Ti;9Yp+((Rc`DfpouaLLw*lV#mDZW}2j_;c5F){1;jX2D5;HNTeJs{a8{P*>jOyOfOmFze!VyTE_A`Q8p?RzGFV;iUPUuAGrPa$F!+hn zCJB%V3c?!Hg4!WyF3aOpN?RGHf4uE>T|<3Hj57Fdtk;p?4}qv%C58?_wNb)wzhZnq z4Md3|uSZBvUXt8l*3$|Ko}fQh1f^&xHAhuL!1z7oKVCaY`ntHd7*@^z8>r*1c-h{p zeLx{BSIaHkjPo8$FLhMPdK(Ar<2m{L>m@gSu&wQ5vH;6CD08y8VjK-rnF&y3=L#xY zK^>J50Pj@$FYij_Pgi!N}UpbrrwKF849vmr7aWj-7APD@3{8`)O z{<+bYAGnPe!C)a^=JkcGfn@-Lwp-z>j2dLMw!XEj4h=c3+2=RlQ|gs|Q0Rkh2a<%! zi96@XP_s+;d=2_lAj5l)3!Le&zqf>bf@a`QXeKf|3 zfx3f;g_Q?zi#+`+zJLHslQqGjz?IT8Ry%Te0nz2kKN;=?p>MvcGl3iG#4eg`NxSf& zQ>WvUtOZA*(acCTjtt`Ig7%jOYdQe3Z(kXEZGXN)Cmi|GDG6w~H$>4J09+%m1Ro|; zdv}M+ynN1HB6O+g=;=XUahW`YqxfS@Q*r(a0}?mBTxc3R;g@|j8(dCdfer&?tR)8} zoI^ytTel*)!lpUxvSrKD;Nq_laySs7?99G^4nP_^M&c*%-WvbMv`W!&ry`}GBG#@<(30QGDqc948(TIuP={U&ijD6aQq;FA8 zBS#0b`)LV;E9^sKbq$lpgczE#7xL({4vfsCx8itwOoaHsa7@xjlKZvim!iTSCD_QD;hMkU-VI#!TxFCy_-3_)uCx)ag=;nB)_0p_9uKDLBedj}U71MJ zdwM&p6L^D9pvQDe*$qVv3H&@+*^2{#O8|FUef`4|O9&3C6ivErmmkH_6Wv=QJDa0@ zKlo@VF<7mB{-=!;1Yy-YHSr_>GKwzY2p?F=U0ujtg}ni2tLJe}Rog1faX_*bvd!)G zbSk5D;fuJ`73{3oGM9tdtQcc*|~`*09>iTeJ2Iyi<{Q7_md_Q`&J zuVpillxcG#@Y?AIaBHPd=ROfE6+IIWszNAo#-Zh7L#IeY=Yi4UF;cX-ztk;1GCNP&=?ExDiNR@BSpn znh7}x7(MxsE0fKfZ@2RK9$#VvXeZRdbgCRL{ZS156k)%2zd+MbB|H5nX?B88ph%n# zwn2MShs@lk-3Z448Q43qm!E?Pa3F2^1_MS!#>*nFzGXSHo2AzkmD3Bk`7dX*IK%eL z;QU)3M^ljQScuai%$)hNG)&q2cWtpsL?K)GdU-DLHg?f=VSF>W++(bKb;ap)*a|N` z!14OLju-Z;>NFn(?{%mW!5Rp>$-q}gwzNl}j+X`jle8NqrHl~5RlW)Kdt3I`8egX5H7lk6K_%kaJT6B++KktDZhn* z6pSXt*>FWC@$}!gAy)vW?FQ4g+U~ZQIpYzmXh2b!UcG%*;Sm$_l`a55>=^T(D|UKN zyulDrdtO&UHvbInc5wy%v~iX5k-?&PqHuuF@uik*e~!9Ia>m&Wib&JEnXwLDTP-LK zaVsuhW!|zl=^4i*+o9Caq|bPwu^O)5dx{(&qw6;lu_-?X@f=gGx@w^v&@d%vaqu5R z70a*a4v{~4S6L?w>!gIMA#yl;96SSvY^bJnCW^xg4t@ARwV(&|2W5(Z zmODOwvvhjJd}T5^i3N6d+CQ!l3n2gSJwtzBqGS@U9Aj{bONy@hPCk0t?tL;nVKQ3b z1X{oOD6jCbruy=wUh%NGA-~C0Fr@t~_u5)D{RTdW8};+S0C@Wur6`TPW zt(0e3kYN(M+Tvk+E8}4)f#U#8zQfwut$;&E&cHBrBy@o+!|`B1{dlQsR6BG!8d%`< z>LR)9&)zA19~}1{Ab0@~sY>di4P$HMjAqgu#NK6~ikL-0YHx}%io(B{2;8fy&`F!bgPMY+fk#HJ zVrD5UL@oDpNCg!1`uaHm<24mff(i$L#bcf z8Q0b)00CSs53#ZY3>V&62nZHzk+xP*vW%Ch(j2D;OW|s2)UEm9s&1`zSpc~ILZG5)YG_%; zD~BWJsA|c&JC^oLzK!1oSgIV`Y_2eJnGWS$>pj9Pdn1&2?-`-Yj10%x?}{m1o%tVU zQD(MKJPj?f$92o~Wsxtlqgi!d;SzYl2yPFxl|$UfFQF`JY&VWi&*7E!N&43u(3+xW z?i`R?XpE>jIx>I~#KXyU2Y4wf%VPm@D3xQZB%;0swT->u6R2e+@JWc(hAbP!^0Zk( zl2HX-VuHiCa$Q_wl7Yq+usml3P-Kd zcsir0V>lj06Zj6Zq^QxP#LM-`Etbx0v*%k3wvz2*6XTJ{GZ9Ox2^EGb!=20V)nAKs zd!TA*KSPJ|(zL1_c-59vn;_lH1vb=cZtACPL0zCM>I`C-Zp=U)i&227mW_vq-x@)~ zYgDXl-1eYsUq>l$F&>9`6R)UQAGgI_;mppej9M{)k=VO->^u3*uV8=+CB!!PuJvtJ zWh=!0M9A|Bwv%BeV;bMtt}!IWFsVuRJJ_gST5ea&=#=V zMwnE0YeuO0Ej(8^=~wieblC203QJ5|>T$d6ScQur&I$7$tgEueZc0_~xoXTCv=w01 z_&p=(zwvwCLFGm|UIWs$J1{1$E&mK2Yd;MTanu%jVG!Tf?G|ULpuFQrSC^s)$@O9+ zLIupP&X*i6ir+CP)=ox9@SeNV(6ya5Ra0IDm}L6J@BDZl_4)Lx!6m`A*l zTOQErVOyBQIKcMiV=o6i4_0Cf0XQ!=rgydP8}r_p&giG5ZXQ}4NLYDCTt0C(8hEry zaYE}9=G#zIM7?-z?+y<+uGL6^{1n)w+Im3DeXg$nuSqYqu0!i*TKdL`V)0 z+kB z*u8X5e@Z9+ADx`cjGF=1&UHB*c|sBel~^Z! zhN4uC#<@218NGC}Y&T9Dax*OM%Jv=vJkG)S-zY{L`jq`?7UQi*N9Oa1vJWp;eg@3k z#URCg4tBfk&%3e#s?Q+GwzsW(nOYv=Qsv54Qad2lZd~ZxyBI8*zo!!6b$7z&my^KX z4TWzhzs}gN=?CNgic|}GYR~w%?k*W2kKd(pDS6}u>?tt=a&~+8=K(h+)Z${F+Q2g* zUb8I45BYv#xaiSzYazyq6o<3R(*y3KUVWij7mt-kAnt#RIdv2j!E@ru1w*v^#Opl) zHVr8wk*w0=@`hdz{eIk^_ffZ7Zwvzai-Baml71H+Uig&;Him*%s{Yk&)FHkEJe5()fw|g%w zt+h}VB!luQ(eHHKG+LUM5;e?qzP6uh93(-z}3xwYsiT{%X8d9 zo{7ZnJKzJ;6Dt|7Vk;5UCVS}`mASr&t|I~(qj3LIkV5-uwJYg^&~Wvsem+K+=V-~4 zrmiG*b6fslE#xLvz*VA_c)6s7Ap(V5$#bgp>G&YnC=sib6c0`{U|*@GuVAYw`Ok+c zxAN@}p#YAO!tGSWPPIz?HvEHMtRxZ2G`a8^MIsGj1!9QT+wIbu$mRs`MG6n@f=D@nF-;>p$fUXIvE1u@2}<=~!C zG)!i@)E1SEBF`)+hXkL%CzDBKl{dGBeb%%R%A^7h($Esyw-+QcB+m_j@H^}I$19lz zwy$I3r_PqoYk2PGUCHWe&i1zDFd0y*{0suiyA)0WHu;VjopXYhnp#!S5B1X0<}!E~ z-C!gr^lNOJBCmz31-^Aeh_l|=e5L;?=F~`^YB(bR(q`Sipp1pvN$hq)Z^?GgA#xLg zAzacl>7joLhawt{?lCx%Mye7rx%>bds8%k#)HqwBT0-o-ob-^iFfdO%tc`oYPKPNc zhKvJx*>+qs@wuv{4iv5ON>QB1|Dw9;*#OI%JcZNqXwgD$&gZx+lEmIG^nu)AgpCe9 z8Yg~)8~+!XQyp*eIK&KlKMj<4AQ3}!QjB*Zgq7oARYa^Vj_7(lCir|v2o~cYtZW&Z z)uMU|^8(`no7TI(Gy_UAv}&~|Z*3R3rTf6ZJD%)vbU9y-4}m(&@eTey(I zsRpg9fbLz)TG<-p&EBn?xUd!~VekJ0nVIe67%fT{;vL4YbOdgEKILYc8!0mQXb2J( z$et$j?XqFAOAgUbTjUvt1z-a&_52llBE%B*$o0q*#96QePt|ZQh%i-LmTO#D>Swje zI0aEg>jehojqpz?#c>3bdi6zhg5o5kk{ z8%&jS-0l>7LND?>$A#euV!oG*xnRue z`d+jE(A~l?L*p~jpwt0W6JV9DA-b+rr$4#k5-yJDZP!k)6ll;H*EQaRJ$?z74RXFS zF;#~{6tJIh%_=PqiJdxD_HL0`O>@^?aOeQK6FOjlj8FPq#Blodr`^{;tv!Mx}b5?8j2k5_gRlxYfzPqiqm5o%sbnhYDi8Gba4eG@(d% z)bM)TQmoJK)ZAJaQr_<0gM~?dzFN$)VwsJm8SLifyb+Wk+=2HkPl@x zHG9;I0!wW%r5vbvUk)+=z2?ujewn*?C>3geS7HSv;fM+ z?5x9(f&Jga=++ulcN2)nR7#Q0Jszo(zT^b|-d=N%_L~&fWL_g2CALDFp6{6nN~-`Y zz~4_UGszf<903~Rp3KyTQz_j;{BC~acm+|vfyY}}IZ;pwvrcZ+w<)`1X*M}U=vuBqO}z9PU?t`NV!>>w7F+YI0$TAHT1mRW4$x4afr|Hj zAdR@A{Edz0=w%})bDql;d4C-)AP%g?52ydQu#hnZIUyXtSCL)-qRj7a}+ASbx)`uu>Z ziZ8vLO#bDtRIEZfIJ8Y^`?D)8VHP4r0l&~c#XCrFIJ~P3h`I6D;*kKsNU-<%g1*AP zGP7p{z@#%{*7VEYZ(%<|zQTi7`Ce2OD)T@o^7}B4b#FcwIOD3xI)d6Za*@8rEMw}~IMZ)>p+yCyV zZ~p12LAsM_F$6F=`C!;;N6um6ao@@0U1sStDvY;e_ z#Soyo61VL>^2)|z^s*&%2&#D4H!w3$x!}Kuf7DOF!unYm8>+7ZQmuJV))8W~!tfw9 z$IbGW!A+RS)(`fvU!Y^C7Ikb}c^l$YIVe_C0P=*d|C`7C`OMWzPyMyruPm1f%qc1` zyzTks-zpCS#ytTX`xyo;%G zQyS(#r$>pK#&iD46#^-zY`sFep%LW$HEv9L5NZWVo;GR8p`kSR5B66gSuYPTAeHxf zB{T0knM_4Aryeewc*C7C1d2X%jhBH%l{IM_Vf1anK zY-U_2tSR;Xu=ib2QKj3~il{(~A_xiwKtUuFC<;nO1r>^%p$LK&O3olz5CaOL#}SAW?VIl>Zadky>%!T}o4Kie)`Bpgq)G0NF*b+h(i z{yNH;B`{rd@x;DVS;FcCZW|F?P9m?bS;*e@=bBJVAELi+Z*k;y!iu$Wu{igz1HFMe;3i zh~Ap+V!*W&P#Nd|(X-=q9sCOo*OPI$&!C_u#Ta<-r7NggYzJ*45 zJg;qcd+5jm@zfabgA2djxZs_xpc~}dNup;|oVVVjY?6yy(Q8R>iuXl`$AZ=_!sY>2 zVzRNej0oGX605?v6=MS(Y-Rxi`#l@G z^iN(s-TjT4+l{1^BZxnVQd@u1T5(}foB^2$Mvw}Px41OPd&WT1D2HGPNj_HBrnR@& zQ5r)bz|W72)9V@`gXzZ5_2f0^X9~My@9TX^h~MKj*{27~$jPp~P{i*(dC#-+`HTjF zS5Ch|Hc@P~uJ8E6kokdR1pAJ?-oF^+UxwO?;Wsk^9*8ckm6Ro%qV7AJi84pP$J7m= zW^iowH{%DZ4W$`WD2}ih$`z+oXd6-z zM~!g+1?Z<%*`Z-!h{ui9sXi1`A1%-N4pQ5FDvO^V5aCVljU5sGoDhLQTjZSV-aWKC z^YGzTNp#$Jv?!@DU!9*x`)L__R%2>U@q)`^?G@$JLyoR?*qP!rL{xe?;AiD_Rz^a!lK= zdxw3AL`iBd9&97@b^c7DcAR~6d=KHsOAY0M%o)#KLAg!=gmLHA?V*` zt$*Jyk_NCz`>^m*{^pn7hzb>(aMu&AML&9OA83ErabO=+4dt)2ON^o#8u5mHoUCLD zRCI&VO@T*A-;WiRoI4l)oNbPXS(=SeL#Z>1&44ZRjc*2%F(|u~KfFM3-1x)3qce0} zt|CpPiv6JrE9TSP7YC*To~82**M-#AXmX1T)x7@v;wbv+RkyoX>|)1D2Tc=`Ec^Yb3~3&3T6o!@9wCJ=1xDKKo)0P%KQVm7n2#&R?=jxK{vmtW?*5i! zdUb{Oap|;0(EP6QWkxfJ0IvEdxuAgm+q|~pB#%`$oMeqXFI^|eNRhBNW=<((~2oB z)-p6Stcjhy>2r!&C^s!l4xyWc0HM&^HqPtVUvOG_H6ObOgPrmq;ZJsB4D zX`!48en4tkt9X!#lJf4{&q{r&<;F$@TEPctF*sbvFJl9P*WmoEy2H#$SgEP08#_7- zBA@&QO5Im4+Ka|n0@!l357}z+8Q0Hu&#HSQjP2<+wBz0JO-j zkohp5IrC{EhzDDBWd3e>mbUE!bR!__vf1;L?#3=k7MbvK(2w_pjU~ZdG**_D|NVnqifL?+RMng8 zEyf7{{=O&QX*t7QQcq7WdSudb1+C*S{`QGh!DcJ)Yp?)nD`;vOm!(D)=NVrT#+J(< zCn!MLOAy0<=8c@7Y9A2`2EZD!*1ju$?LJPFQd@fcxG^le3HczYdZ(qewF*cZ#wsQT zc{o`~VJC&e4a$l;~8eD5I~-UxDd z7qh8}msJjaR7u;YP$y}pFgMva*?D`suA=c3dNaEC!EAJMON)fFbFq3YE-~GD8bzLS zsfG@`s`>bj2+NJN-fXu)A3>h&dd1D{v}hW263+RraATy1Qga>PAT}n+hDuD;+E`7}!0B)CvR&h&v`+eS*ZQg^9K#8bA^%zGA%y zt6ogPOkfLtgO(iG(C9y2iyqZ+*g+8Vj$h!n7s0qWfy-hkAU z6yC)Ag`9G*;_6R_L^z-(@q=n41Ve3dNf2AEg1l1t;b(-hM51y)ZWcfA_;;uGfmb5TY7bDML{DW{pM$W`|b3=A#KHU3ZE z^^f)w5D-vV%~fiwVr(GCbFc11rXJc1)y%?FvefH1w=$j@)WitUMQ9SEL4xH^e{?jb8QS2_^L`(Mn~2*|G)22E!(iN@pOBEk0u2G2+?^*TVxUyFJMSLf^m zA~}e7R0T!FGvGgYa?cM94ptV}jQ@1oe=6
    gy7yW94uc=0{>yTvkwIPF^|d#{!R zTnFnewV)y5CDsgceahe7nP>5iGzY5ERf9B$0L+L8;!n8cz7Y{7yq-GQh5=r5Jd!Ee z#cXuXieUkRZ%7C$@G&nBciewa^(^_J(H{03x*`FB!8;=cA6v3lCI?WY|M^wC?fmB7 z9~OZG+>lB+TzpN0C8~1I;7zv8xXL#cM%Xsd)NPmt#dBu-S^&GMFqJV-cHM4Q-$>1s zdt$Vwnut`KVEXci=~o!AlguTb_&2xN3Sz)NVb6vys7F^8+0Wk{YmPtL-bmiRXz&c2 z=3GFkllY}(Hcrk!fBz%!RRC+(Mh*Y``4x2X?)@6d_%`Ez7=xUkG0y`#eE!L>f}Eh# zPOos#Bu!acn-c-eYqd2dCntRa1J8nLw;xc}D`Ea927(*-oE|0N%a@m$qQ+zAK!^Fl zs)f+%L}teAk3|2qO`^0yaI!)i?CgF%!Yj8-yYqC*l4D|+Mn^_oLEKX>cF4{@DX(fY zdSkA@VhxD<(laPWnL<@GP>{{(o@k zCyhJqpRY29JkD+K37aZGfMMt zw-;ZnQKWHz<2%s3!|{PMh`07t+9l-86%3Qj;}OY+53$I(e&2bw;UPw-{VnPdPv@9& z-i$C_s73>4wa;|tHJ$nh0-!H0@$!b|z?*#C+^i&V>(;B36i#_5XJ!@_G~^x2LYuU~ z$X|5pkZi_p?$Jd4V5`H>Ck+r&JNEe|i{8&=Ffg(h8~i z+z*uq3i?0N=Dht|X*1vow_qorAb=Fyt{YC3H8q#swgTrs-vyM%eX1drbB&IULKdy} z?j-A(yU4k`nzi7=$CMA7u(5b0O3eAx?d`)BE~F9>V>p5&T?40(-AS^@pi3n%=B3iI zAI$1iR#*FFWN>$Y`_IbGu3ReJbXL!}>$%YFdrSg7KIt`4OYP}@{0n%CZn;4=LdvxX z>r>XhM2nQpZ>Ovq3*0ZoKcpZP06fb6L#OB6;(`a0WMuZNr?$UV}|{_K9w<| zhn@irp2}FfKo5QSZ9Sy2j#eF$ME*2?As(0Gzyn!=#|=d6*?MQpf;Zh5pk4}$HY7|T z9*C8L!ygpXt8nO`pCEetu0zbFB|%w1K|v4FGi9~cMb@$@;2Se9t1^ADlFCCa!jIOU zV+15%6O3xb%7dq=BTOC7`&lw9G#E04LIMUu41NUJ_NlEQYzC@a%eZJiuckEk@Ty|)q z`n~5KF+5e>2g|pATLh9EkCXl5RWUR%p0N&7 zvT}0rvge;Jrj|D|YF;lu_CLPsVLelUSluzJnP2$t5N$>>nb+Gl=6~eMFCEibn$Dd( zd9ty$mlFW8A=`4GQCGe^DNs|7a1le-Ga8$kYW)s6)-4yM($dm$8@HeTmRi((6tc7Y zzk5fw-#|-auaU(+B!X1Nqb42^5fRb5gHDh2VxT7D>EqJKx>eyrZx(6(KKIOxxyAtq z6t7RwS2|ists&<#>lcl$V?`N)k#zR*Kcurj2EJSe?2?%1=rRv2hvWUwO_%zPODupb zn>F=|Cwin@01O)n+Nb`yH=(^yu;frKk_B~#Y7#UYXAn-yrU_-GtkZTK;oJN90!jJJ zKP5h@#}OU%>OEJ_6>VLccgrx!_RV_`RzufBjZz(f(H&+kE?Fd}3<(Jt8eaRt1$QG* zXQdf0dAZLWetMcp&Og;XiNiVNka~nj4=$2=u9S=Q+}7S~Y`Kk&a(SPa1>2iBR(_V3 zm0kNLJ;3vQ%>H?F%1HlU3H*%|CFGj&G?K>O#@-oBD8JP{pM8y6F^aH3B2DRT1;Dq4+h+s}DX z#C&*)b_6*DOK>Rs0n4m$Nl8-3tP!8xR5)q6wFs)XCygboaXrVhuf!xW@;$jG_!k%} z@$bmIR}J%jHNBeuJuI>i|IBr@tM4bQA5nk|7|apL%)WX5v=KqAYwLjhMjx*ouS$WM zKX{r&Dwx4)xQ=Ic8|=3?`LFSVUe`_M<;7H26x=xd$0`yMo=Ixy*Cf&CTK!pvu(u@( zwx6M9o6SBUhRw+9;e}zkoZ9LH!ngdBJg z_IXqe)rR46a#Xeoyvt^hq={r?WMXDFn-4KNkLrv$2t-3k_9HN5#sTIfYR=Ayto=BI z{ZaNf%7n4RN&)edOx#EDThrdsG;9vH#6213J+Z>T+~da^%H@1t5u^_C|LO-p{YRR4 zo!8Q+cc*RHj>CS;z>~PjJ3d1CL(a>~%R{=i56|3mdS;6C3k=k!k-mHHUKk&?oW+WP zXphwP6*G8GG&#BaXKORKyEC|k2L4> zu3t!~jCz^V)o44P&|R?hwpq34j6H2c#)#Ygn%0Y;2zP-|iFoZWee{bn)&j@1&RqaJ ziS~jXF?;Um)`1UhXAwN8`C}`9g0+eiQ4jh_spP0>XnJiV4C@u{ke35%$@C?Va`t=- zWI59S=SV8hf{!>9BnwH+D%k$-}Lf z)b!0#QR6sQb?wycafiXe)2KEkvk|$b z%IaAe|825ODjiwKusoR?ioD%T z%T6ip!LttMyfNMOc(+wWms%Rv@7mCJz_g9JjFRT7nl%aTb8(`+3Ky9eNBcH)oKjU= zI$D_~bH@u+zcouLu8T`l9+L!9PsnC?KU#39Eh;E1NT$+z{)^PEBQI*#lUw4JZ`e6= zeodZQ(G+o2f*pkcNo$q9{8tLr3lasATHfSL>SS9MP9;f2amyg%$C*C!kjjXwn#{u7 zKK=sZmVTAO@T}!T`-S+F6q@3Cji^`h0P>k$l#-U_nS+67efpHld-b8~z@6Duj74(0 z48s;StF?P!&!#nxAXnpl4bOH}DSBnELr!;lp)Qw?8Ow1H5D(MKf7L#S*$E9}5eMyUGv+@xH)q3>y0R3hL?xkyUd*PYv<@(FA#h z!Z*nNt@^TsGk^@RTPr&GxHhCBDk>?Otl-K6VD(gJi}s$xzgE;S5NV^5MU%I4LxxjC z#gCE?aW$iR{c*m^a(XJt%m>FxP#4HjWkcz3xVMgbr6+GTKKSuy>H|l_ZFVA6cz{Qzk+*V<{bpR4z^+qn>E2KnIhuXG9^&{H<(GL+!30E*(y0X6B*8=|Z_2y&bpKdWDdrz_Cp8 z!Zz~aq5Q@!ZSmXZ@e8Ek4*SUq+T=M*ZN~3!E}?ubbJ>WHPk$;!SGIQt6O)(ZcNo(5 zRi(G|m8~p)v|*$W+ipBWO8>Bt-A$ggU|XJ$z&8Kjj@y>D#n?<2CI1{vt-0-wNM+xJk||0?h9*38s+;TYqXxIAABw};-|5EMSP z?$0dE4C+lf^`+lK`ddW~W*yZVMdj}!dI@np0#wG?xw$W<-r?A@7nU4WA(rNp1H>bN zMqgwM+K3Etg&)bWsqv6*n~o-0MC|m}edE1wAj?RqN+wGnF=Hul*$z%JNneOI^A1f3iK45}o?#UZjoL?%WkplDb zIWhv{YsYsD_3S@HP1k^Y6=}{sYdf2`x&Qm3|N7{lq0p0Oah+2$j{(3V1vRyEkM76s7|Ev$pa)>)G;o~TYDh{#RyN4r-(R>ZX1{hlLSABG zVX21w@inlr_QRnF$`OM;kn5<5l@6=P20oECGXIDenvuuI1yzMJ6XUd2W3vtQG(lGo(UwO1-W-oMt0%7OhV#ncyJjnSy zBcZFRZ`BP0S5#KoQ7kjpx^Ze85?A`JA}RV!x6HN zl5NIY*XwU5W@jreaHRshq7p<0n0R=UUDgKZs~d6X)eesnr*0r6`!VwTv)dQLT#;~s zTR0b73{#f^^A?f?6q3MP!b3jK_bILLTaUb2gH2DD*b(TN{Vy7hS2Z;?!m7rj8ycVr zrJ$q31p*m<2q(ukEQ}r5a>k)uls6sNANTLgREB1vlD(I5-1qf$8zTTW9Ny!V>mDzz zzllQb)^#dl&?Tq@A1_uqmKYs<4K|I5c)GicqICdGq0A&hFV&5$H{M2X_9w*0zXo*$;(r7nR!GHf+b6{PScS{g(Re#a%{U9QpUE!y`{Ohm}`*`A4dFJ=xy|;_oW+) zsb~Eaegh00(xM8LDmQ3E(efDjcQ9y1d9|h1eXOHm)Z12f|rpPbh z@adCCJmtGp!;FuoxAx*)RR;o3eNl1!YxK+hpvR&NTQ3gg-m0p82B>NuUQgiaA<_PX zB2S*1pRcW$_4M@gef4T4&BB*9pG4i(lrw!w?U5IUwH< z@8RANJls~{jYNvy9UplZDU`OZf_pM(vs6N(SR76iFa-FodmyDeP!S@O35}P1YQ;J$ z7qxH!uU|7l)z9PB8YAH+#Sj<}(6VY?VuwJTgA5mfytchMBlQli9z##BNkineW&D7A zls*7xuXf)Yn=1g(%diN|_*c6F0>V}Kvm$Ui2wF_2f5iO=oGQirXoYHBY!6DPBvmM% zD4T=jvohj|RXND;4WHDVp19@d!2sHsZEIPa;wu<#g$dH5RK~DYQuC~Aq*Gh%fMZnk z(gxDrTQk~i=z)(&lskUAcGMyN7or^77q?L;hf}A7qrkps3SM-Yb+$@9S+d%tQ`5BC z$nAN+oX?HFJ<>aFPl|Jjc`=C-7~fWR{dv;yFS*&-o*Ovp)n6tC(AWlc72y8oU0bcB zbjY_FHZOr|Rbt(}8uGu|>lWObvw~3*5y|5ucHd5Pac=`R3T!7souIh{H857i9Pz3^ zII2ZGUm3P@q&|AII;eCF`wtim0|(*$ZGjz#ZB`HOZF3X7gg*po4b@cb0BB+J*-iwL z=A`J9nyQ$x&P62VtQh9OZL0#0DONhS11?J=`AyT%PDbZsm5E>-FCjn&*8i2V$L;?1 z&$gNE7w-VYrV~OAN>$a>wM~=&KymUM^XGSe&;0O>jO1ixWsU06hMMs;XwP{9 z2l(GH6y3u+EQ2HJsOEm6Z}h%aHE?er_NH%Wq?h?Z{GbNmlx-jL>)_7<`BE!WverSUi{+=sX`W%th9{ong8 z9^yMpSa-STJ3qxUJV9iIsT=@k$?Ml2alp-jQO+~K?A80Ma>*h7(KNKTEnfwrw=*WU zS7bY0exuJJ&cED5w=U{&;|O7A4ba>nk55S8KK?~EQcxYD)LG~?2YGmShY+vqie@lH$a8s69T=tD8S-cwg+1p;5-+?-`T+=h-zE+DkRty@MsFIK*p?N%f# zenCV;CnFxNI>&S$R;zz2f% z9o~y3r=C1zznkM$K9dTs4LMP4@2}NjGhB*R#@TavlQ2>| z+DGwB{x%cS!~?gu>|>4kx>CkXGz%IQ{dsy_BA~(v!1!pH8a(wm^mF%v`gK zU^xltg) zQ@oCiay~9&V@C7+*H~g|CxZUKa$nzwD;8x11qDfO-+l}Y4eizegl9ZkI5AtxY4)=% zOjZPIcl5KE&;67zGBV19k)g6v55Did{AP8!v=0WU^k4z94e#O2bHHU`*tMHPbQclD zgUfEnPJ&*idFdsA5zz5TC+|9>Rvmuaq*>i6$zJRQ5APw@rK@{|`Rpeio;2$*d}^1H z`v?=2dsX`Sm}NxW%R8+WS=p=nsp{iKrRg2RUYU+l`Cryk!i1d*fT~#_I^E}d%dp$2 zR@Sg>7Hc-0BP#3$#fjdF(&N{ta{&Gr+nDXmF~Kkz1;ZTL>X%tqM2oT(-2Ygdf%gm9 zi&MpJyZ-!E27RJ<(n}>CPnz>^XRj>N`EWVDSUz>VkMR{V1!nDJQku=eh&=Rkqeip8 z@oDQ-!>nPt8dQ&Ag4;ooWF)PXs-^w-L6H*S)08rS{o z#>%o!RM588*6q+>PA)F)DQ@#E2V?Gp*lP^fcr34bs_LZbMz2x4-r}-7y|7#R33jLG z1qvdHhsaNLQzMBv+whx`PGKN0^!raE!IBA7;4105hv+Gy^v``C|H4ovDCwX#&riY@ zhF{Zt)4l9eTZDB9zq9Jlc(JjyEtudeJ_vG9>!b6Ms1!lJn_=dp$%7<4wTVi32oYrwTO~kaHQ{V04X}) zZ(0c2fim3+fkDr|pno^AW7dfEtdf3vUL@MGOE}AGjITmP2hjI#d53W^p35Hposlay zb#!455tiZ^YU7|vyYI6_;KX!6fSph+T?wdeSCCGY)NkE;RP#0pG)pfYyxRjt|vN*u-&Uz}qkMN(a(V zetoZvZ<7z4Oo1nuVcjeqCnGOs7prJdzXDm+%XpN$cUn-AYPQ|}h z(eF~EBX$U(Z#44Q#@91G?N286Ct>TndB>9YS_Pv&(}h(p_Y9&NX0|N2Z7s(3q|Ptf zSeLq43KCCI<5|13vR0_bCN?+vHjTJS*D|V`LxBg0(#&h{JcpDV5w9|>M-?zx_g_;f z`1jy)D9aGsEDUy{QI#P37z^5;28v0FVL;Tjn$0imY&tsHk2n(eBW)@9Lv*kvH4vE~ zn93}%78{V4PA5Okn|A>r6JygPL=`?dp+1MhRzy~-$^I8yw%HYi5jy=%fSqg6>o|-u zFTK0f|AyC7c34W}3ul;Xh(J)eD7JL4BTI1M^HH5F=u7cWaXYmLBJzvHASRIsQ0y?D z^b{}L;VHLo_$#)XWZ@d#{BxU$S#$=#vlQoD(7DdQwrz`)(j`D@;*%em$gH<@ubf`* zoz8c_=qy*MY7V{jKzV9aQ7Gv98X8M9JFLxq(Hd$%YR95SpDC6FQ4bsx5Y`e4%drP^qNHvZv`}2b9AXB4+SEHpAb#1EP5BXu$ zU!mnFt#<{jYRm%(de1=Z;t5DsRv`;lD8KmYDi^bUG90qMXwr+DKfz9FS+!45$H;^~ zmADAnp#{v*{JF%!CZO$Rb!`CTm@cr4wD6aV+zXf$L^iO)TFJboZr9CqSTyh3^ecB{ z@yfyN`?;IMgK#JWpYXPuYh*uh6~iA7xri*r++5MwtVTxDuG6o)0!gS3oCPDNS>|rF z%eBg9r!lABw8=ph)T|=vBDqO2WnTbW*QqEfu{wx)t;E|g){+>J_Ri}ZmW}nggzSC) zk$2Mp-(r{@G`N>xWo|dJC_E@z`EX<}pYD+N%QlfRMB@vgGMkr02GX&Oqub(q z)-->ShNX*Z5 zcMKF~cY7rJ4v7=B=xi?P6t2ictu)7li!y%{4*{K`xu8D9ttkx9=vj}T$|(U@yud5P zx8SGDyQH+i>Yy->%bo%nbu79P=o$a3xB@i|;rG+cM_W<~V69L0oFH|E>XYVCUaS&p zwd6oh8nc_Ex&dW{etqPmp`JuqBjpE>xp1{H`ceuX}&fks;Y=`d`$Wadzh4aL`t$1J4U)J*aL;Nm$;;O z!PO(#(6Z?1cCg$1xlkvY5$Q%N@t*9T z3SU&s9Q1zHyKY!PIyBsEDURB+);%mdT#=}usW&@m=GxU#>ihldRdkukFV@!yJ~+L# zst=9ax16~c<+DE~^9(@k>iB-(rYwvn+!@HgKwAm!vq zohH4=b}CgvzNIWM@G&DuN#)=rDDEN6w6 zCah9KO6~hEr|fHsTJMO;$G*@|IpV7x0%G?vY2b5hQ#`wnxJELxLX)UQj|bGo#m zhO`CSP_#)Ua<;5*0y&^@Tx=SzV^cWch!Sy{`MAGF@o4rXxq;tbZwmJZ3aaEP^R}(1 zwiRx0I68$?Yxp&sxxprvT9c6qg1Dykv5^N@cc--~e65<3MmaY}rLek5O=c@c5-r1! zOQGIu!ZdR%1|~t1gRkFi>brqt6X(yVN2{SFU^>+=*V!T=aLmqiY<4< z=T!=Xi$ch&j!X-A8GNS{ZD%u>Vm3X8zICmh%VqYjB@Yo#45et+3Hh*$3?(luo;al! zXgxneHCDe{w32MCDJ*1x>B_{XSOaxdJKMS!r&h|favoIbE&aTXdYaVBRI?5+sC9@O zOcKFA190vwlNr+ie2d;@PM@%%2BWx&*^h@@X0$?7=n}b;vM$NeDmt+Jh}sy93iUYE z($HuRJAS$HSx&JvN$(v|9%m0zkFDXt+Qp{?J^wnsIu^Ye5GQI#8trjiC_o;~#+)bd zwXwa5B8P2VE;h1UN@zLVL2+MDM!`ai1a(qS+Kmbo9!7%0c*W+{OiZV9?Yd#N0#&$^ z_m`kzoWrKB<{rOYWhxT$oQx8mYeAZI+E8Ltd39yKWOR_nK~I&cRt0%0QclXCENj{9 zsG+)u@koaE)79-vkrGO;U&Y|etIjE)*hnR+Zdbvg+vsqQ(^#4KmjRtlfUoE7{fj0% zM<4<93&Ac(F%JB1JKL%R6G>dhd<+dnN4<1@mmM{Rjjk_Wq^zpCE~=*dYI!QWz_?T{ z)>5i4nLZ|+oxX5a3^l7*%jvus^Z2dbshH*N8#>+I2#D7IspkSZ4R&lN;+ zmslUfqFFKMz1@q>v}H8t!_{dDb-Vu*st-k<6Dv^NcOoT^w?Ipq&+(kn`4!k-{PY63 zCJ_mSx!zg)dnce}R5@7(q^FrZ?)#B(R#IqN0Jj3ie8N;AxO-i?XIw zV9V1sVvP=zGYp9uGbvV;|5dfjp6aarJP#P^L05kvfe*N0gMp{n?j$;}Mn9iaTxhihgHPJF-e-AxNI68 zwx^|6p|DCpj`Vi1$q}DvZ;C#A&^j+iimw%i&?J(R{3R16Z+fGS*@Yq@D^eB3*BnKb zoA@#?aMxJzb5A2hmB8mX^nLjVR+NPQE8m+^qiEH>%m-`z$GbDWK-B$pMJu^~7O8y_ zrCZ+V>c^e>JfKWcq9CYEbbB!nbt;t(o6v$=(Wd?ZlGvxLJ}(E<;+TW?o$SWKXz@F((isQU??M>=bz~Nte(n4I*f>%{4p5p*?{Qx0N!iyv%DDl0 z(D}H>{Myc{wP`?<9GLn&Ho$aNh!A-=nGy&>%bkl-ZV?+*U6<-u%H#9opAXV7ifP^jXrhYx|%T8EKVxHHkJ}S z$EL~ULirW8jvvewO9MG#ZCZ6hE(%@viqiK2cXH#u7e+d>(ei>3L_I^6exQ!z{6xAlU7`!o5kJo2C5Qj4S>d)z5O(nKlR;^_NkF5Snu=A^2DV#PR2d`R&m`+G9KMXxv7 z_`k++T)@52OaI6;K$vGo+{s+Z0)|Q)dIxKM$#4y+W)I z3~ej_+?AQ_f8dxz-`VEQ&M~qq_SKuGean3{!pcvha7`8(`x9&t&HcWB zQKj+r5+{%l=T%M9Dr1hs zyrEI{p4ofYRzzLC3#L6HZpASBY%a)d{uILjdY?bRf;BU+kIs1KuZxR2Hg4iBCQIg-vZM$QlmCaD{sup_s%8>&N!^c*f8i#o@t}U-i?oPXgRALAQG_& zu$9f}5MqAn{`$=Ls6!%aX{?`QzDQh2ycw%${4(I_Nt3@H^c3Mi@o}|!pgPM9dSri$ z-hXmokju3Ld&PO^Vw8pWH|B7qzzHn*wGg?gy9%*QXM6o3@@$VwMr$y0e} zHBVQvL|LPW0#4$_Rn0W6y&rr~bz9*^KzYp2kUq;UM6d>{IZBUdA#eH(H&j|(L4)Yr z4@valfWx>>UCPV^BV~yPjh1MXn`Y62^!VPO_H$@S%x4N_FNq1iSFs0BSP$j8cO0Cg zx^NdFsalzc*uv22y1)gU5-eWdubX)Jn$w}}sJn}}&P&vg$eZgeBadl3sZ2@5Ky`0j zO*J1Kk?!I5BcvMSUlb45-!^43CVi6;D!g1NTYz)Y$cd9m(g@r5(db8H6K;r!k2rNR zK90EcPLtm*A&NOK(-FJm0kFpUfvdue;msKmToQ`sV`bX|fVxqAgBMB4CATqR zbT@*M<229)Isu`a^o4~<%$ zio&tclZ8{`aRj$pJ1k<)Y;SpR({`T$7pk0q!u3xeDiT^ZR6yc~y&=UxAs|$lD1~1k zfe+_CRyCpl<0kB#DSAUI;s2=xL3l6!vjxFw94}fRn!u1BnG;uoBGrU-5f`XEgGO*f zBW%qn?t|3wXjz7$g+_D_x0sOUE7VD(#lioRmJ|vt4(iYtiOeS z=ObnWr-zU9P`uxdd$R%S@PV8nRoQ-?-A^^PGDLvcOtD#{moTNwe%W&9SsoMq7(6NeJ!8Q{=h8pn5#0{b%Fvkk-kE`YfpPy` zvQ%78stFY9Ectgg_$#JuHLRoXjqOv1LMbJSIDY405zu8*3|$W3^RB2^2JHpyZ|8-c8h?T^PtH6ZaE zG}1nEmLlwqxL0TjzIviKMyTMONUz3gK?T3pw(E@m#g#MLuE(e~DSdOjbTcx+<%!GgwJq>CDVHVEnJ!oz=-4To<3pT9#vMtFEUeDT<_ z^bI}5xkuYwpdX)}ngGu%w(WG-^J>{J-dgGqE=$2L9{-`SiobFH+E`_N^~HxC*~iE+ zpt*{yHH3L45c{lZiM*jp1-_vyR)P=@o_X$Y(|QGQ4XPvyH?|XDqJq=rCxslEv1Tv{ z8v>0qWqbbUa(3s~3CKq%vtf+KD^1xg_=6FoTDvTt5XuV%!flvmCdMIMFPg8D?xvz| zj%*zpv{e(4C1S`nE$B(LBRioG-xl)2`*QDn5Gc2ll~gK(77M=xx|#w8U2elY!LsVH z>i2B2c-E$`L>&_%_#kS655hY*^q|Y9D+~wXNKCW;95xuh<=&!O^0-wJg zk@PDBp{WVz+9TS#>JT1~l~bfTRo@kylVP^M+Y=M4|KImIk@<4iNY(**^?M!rJz2tU z^XS(cQ?(n1$4i^96R6)9a9#76s(cFTpJdt{Rz~BrRN7#&ucwK^EPs$DOvurecD{;l zB?Q7GdwF4UG$?;@=F=ezP;8uG z-qZPJJO4M^vTWy}j(ZabR_+Z-B*L)IpbK_qN=E9*w9z9zb>nt95B{5P`Ry>z#-i2p z32xgN(KsS&m+HDco?|zy?YOzxX9oT3WQ1Mu3mA;8B@ptvU>Z~NMr+yQzqBe;U>ujK zSVbti>ePDOyOjIj3znn0pkXE=XL2Fl0!n0wkN3{wBg6tb;mYFu)#A9cJCHf=ZrhcH z>2H|e|12~(LlyPggX%rRQ5g15U^O;~kA7a2s+rzcviLRE7?e4U0vB>q>oT^1sx0AZ{hj?93?7|~w>nkq)Io$@!LHdGbB-a}dO z?cL2Tn9G%|P@z>h`GF6Pao4MFd+`V|f4V|LLGAA{jN)eeI3YxH_+#aXFpX6nhUE@oAa*F{4UD*#av(BYXmkej4>MtK2*?2PA?kl3&cU{r`1ncyy7U!erP2`;I*Aji#L?3je&qP9zh0!o zf5pn*^+Acpo8}a#Qdd@mKQ1!Lo!!bCAEP|gKtD1tF#_j>DmF7QvFQ)w&So-1db_Oav%6IU@IZFR3SsOnrH&{~zXK*)8x ztB))!24WUD$lfeyr>-f?0P`lE{|mDt5WS z4kEB}4;1y&wz)D0FI^zeIBsY-*`41cv{$}Q;sZkMaIE?P7s>#d$SM*YdNgi0^whB!O6A(kO`#g|+1o}VeH(s7FII3m+8a3)d z*AI)h{%8$Ov~v&pSnZWW()c51dff3h_Zs2m;$8fUnbvDTj5Cp2F8**1!Z^c#0!wwP z2IzWFW&6LyF+b68f6x5f;dDRD&}qH?4gHdKKxgj`N4ah|%IwpLxsGJ=PX*5qvdmIK zE=9;POQ)$c>4~tENQ!);UExWq$mYBV!V%K+Mq%%PR{j@Q{q_^i^d)jzCi_bJ_yq*C zDq2C&><*Y;eqz4z5R+xf3<<)xIvjHo@4caD-od!))fu@=Y#pV?SBs)gqYow)Ol{aK|Ad* zyRhk1qP9n(LR;zinJvMAUzc6yFLRz5|3tr$MwP^53I$fqyPG}*begil%Q1-De~S5< z>oV8oXK)GSO%&q>-`v6>RsSz4uG=2TAXw9h5Tq3kFUz{FG$9(WVcgi~?8mqYh#dRN zNipeqkEuQ^#FUGyyc4l3YXu5fx8LEukiET3TeX)d(J>0Vt*;Y!34hmKwvWD-QM})? zZR}XRb)U^yk?;q|TFpAsydV3HgP;|nDEj4+MEgo(P|018y7?8g(d24L)DG8HbDyfu zlL;myul-+25PYKjFC~a%`^xm@dZ~c!xa`JH+3+-Fb}Dp;7G(}-6qthCQYuvr_yuUg z@cwL&IU087afQzYA^GMaL~?xc{n_#9+hK%F2{-nQNAl*czKyW>TWj!wv54v_IlXXq zOMODc8uYq45!8=6#}t-GE-8s&ZOKpaXBC7Xew#Y7t%>@-j!NgDd2fAA*9a0(nPz^_ z-Z7$i+?RgmDUx^c^Ya^nS{<^zHYm9TrrGao=+}iCAU%{xSdBP6%O_w#iDVElQmObz zMDluP22o_!z(R6ltCICf%ZAxV)gv3w&vRjEGDJ9$rDM{c2SH2O1lR;DXW7o6qg8Su z#8PFQ`l)2r5`-g{{?nJ>{q?nf{Sw$=y80`()FicJB>_w=Jpo9s>&mh>_RauJQKoAO zPBI6UZ+=>+q7iYcDiXo-qjvZF;YU&C9^eT}Kf3wsZ|*0>Z&yMUf$t$!tdkM3ZeQWf z0-HpMtahNIvi>>`6D=$b;eaRM!hQS}xR99UYG7f5zn=<(m`3QL`%m$C?cGY{NRZ$R ztjk{^+STn!f-e2#o+;0qXom+*tfRq9$8$e|R5XckLp!+zAuN*u~NkMfb8cPrRW0`^RB- zmK5q6b@mJtM1Ou9MbZ=D1temO*@#}k=X$${_z2D%e_v5?0!;^q;PE zFF$+X0g_#A89RP(ykGPx@G12}QFX#6@MfbDSZZ~O#6uKv>f^fX#EEYY6(je^;Uoq5 z-yf6N5d^D8>n=SKJ%Yr*!#UCC+4*V1?p(Nok1?WeI0&D>SDx(EeSC+AuWcW;drQMp zZ3KEMKKN`O;bROe1b>82=LW6sg_V*zqOVH7i$?P_WpUZ=W3qIR1B%r9K*YNigPb^ zIFqV5yShDY6n_r;@ok=kx2i}+l)5MSZ(90Si5=0FuO~#G;+gxOH+w{1oaK%chyLGv zSO1-$9(Gw^*iDq=&tYFnqg!cUM{aT3?r(&q9m&s~WSH(apS*54tply6?ZM z3%yf&Mk6WR{sC*-EOUc-gkDoXb}oLh2^pgT&2O&S-sg)i5hD4Yz;}M~nKl+F8AGNr z|H2~&A{6({%yK_HJ=6u?2Pz}Akd2QuzJMnFagfwU5`IK|3}*T{7y0A=!dyIiFuTjF zv?rxx#Sp0*Ve3rB5to}UPoxVu{%Qx}JtB~wW1@4sOyyukJn4NB4By#^vlL(d7-W?c zL!mB`f|SMFO5ILD_B;}0Ry3OKEe0t1Mpv<8E&zNrOx7)T{9YlVlSmNIZ>R!{gnB*F z0`_pgFS>QfO|<_{sHMF8Rp|+oOt`WLi|wpSk4dhwS)}8^SevdY62$S=s zCb_j$CG&^h;T+3{;_X$tJ$m)qsSi#e_EFO|ZHm)8{lwJm4}) zirGK;V4PPVBNNl!Sx`|R6Pfnfk1K=mPE|xy%CD_SJpq9vM^?jeTccCLr*zq$UOjC* z_)v_Bgh}R4x?}N33i|+(F}7~n{o5Gq4I*{M`|HOYjk|01^6E7%w_Y;#A~!sxP?c7| zFVkK*@N{CNx7PM{btdmpQKgN2r-rTI@f`YFl5a+3-X>PBg$wo;|D64PP$0)Fv$%Lb z+318IH>+ut`2ed=aUu8T0mpk{f05ipFpV?D*n6+m8IbJu>a`ff?eb{%H3oL#;+*6b z4SiI`YeoISq=>7J3WJyf(`>TsL=0xzF}8P?o=lw{VK=VxqVY&Rp_xg0mAnTnHhw{b z#ph&3PHw@)PX)un4LR98?nCE2-Vn`G6+V2D7I3H0fXMB@pX*ygoN*j(@(FgS=L=+i z3bXU)N$x&QcEDa)CJT1wl5_Xm zJGx!f?e-ytOuU3hUuQrTA(2F<1;qT-v?nYanyPr3LTWv z!i{uVXaQwo|5-pNDF3s7xE24rf&c$|13}FLh$UsdwBH#5b{JW*$+=Puw~2}Ki;KTP zl=%r@d?i^I3JArw@9A(^mO%jWvw}9T&V%^sQMRUhKo+A2s(F7nS@W^`Lkj>10wy}v zS~A;PmC2iQ7ti4`4#PIyo9s!f2F~Eud>`XiW-EcMiZx0GkH~~T@ zyg+~)!bAWO*8C42KCm2HSy@$p_TmWSeFPkVIh@xg6$<-wng)XdL|1l4t>Jlv^&`LF zNK8hWf14vd45n*KAA0_(Nckcbxp3Oe^8>3+`mTbwnmQ;K#=M4b1;hn2pCYmU#=LnDg{ES zvF9V=wV=DcQu1l6Y0$e{F0p(C9k-LjBxYkRiTQd@BiSEDDPCLxX&CQ-!?*m>PsBW<%ejm!rNG^~7iz}h(@NW@Y5sI| zKTxUnfkn1WR8(BFb~Leo!7p0pI}U4vh(1Q7Wm3-CEg26E3|ncDS`NBQ#a$zH;%FLN zkZJS?!il=55JH<{D~iay|M|Zi>;Ej9e1K>Cx7j44PMF+8M{1G7mr%G>DK&46FAQd$ zzchn+oBJo#ZMIBJE3K8b{ORZBPL9v^@V*YN4Dz}2osTs!LFkjSowRERErter+Bz

    zy1KBxF};dJzfNK!rnwWJBC8}ZF;ulvmJB@p5I~H!Oo(kBk1gq-wK<~Ys)oIB5^=g+O8^J((9D0E4(T>q@E%T`rkv2Ed@Zi9;ydZRS0=Er& z_-kF_N@I+P<*G~M(&mejWoHfx!AUWU#S_YBH?8QxoLF7>DnEs;j9OT)IS&w{hi$Ww z8KrD&w~WI`KIS+*E?3{5dQ+qvB#-Fu zUfkhI2E7Q!ZuX(dSJ@`X4l1dq*y%oZ_y=8pBSQrM1&o5`1%B+EyAsU`#RI@B2Lamh zXDoa%RdWZG_x*pe@ZErMoYwtk<;q_N?K4viV!Io%Q*@5~qDuGuR(e+f#^W72(KZ&ouBcP1S-5tQa`dAtwu6Gz`!n%DTY>S4%oCpE>| zG&$HwKMtX-ucWjX@ap|NL?j$nD^za58;)U+cm$-3!Xq!$!bh|HTu~)@@zE2T>gAHo z<=ni263{3gH3C-dhP@iYu{JbcpHM}hF$c)w^(gI$4C{`(@HBcNf&9iERbQC3revb2 ze02lhsBGkV<#C?{7<7!}jz_`Rw`$`#*1owoYV9ey-)UDi5n5Q_6$J3270~#q0IZWG zQh6zlr!OCp)tUYD!fMcVUC1bqb8RjcDBP|(NDv!WZ53@*!FBxukj=&KonNF}c$H2E zB@?idvAU{y7FPD^>U~Gu-f|z$gS6XWf}3Bt-d9NTd@~$8*THkTvkOW1LW&xdMy`0# znR2PCC7Ds>8Eb%?n-SPyc(C6lXr-!Y>)Scf!Lrk-t=+cNhJfI)g8Lt_H37u_zvJAy zD#4qJ$xqh(Du8$oU4ewIqE7h&3FMTbX^bc<%^YTik54_~3Q)l7DWhjxAwS(9mqCpe zh}F%a@l$WQYikbh0;``T@W2zx%({K-ce+lV)R;jNP>1QVO2vHwaNM&*crw2t`bV6z zTgp9fp<+kz0=E>2O5g78A&)GgOaHnT9XuBY9%J^t%Be1Mx99{+xl?^;j2P)(OFCqi zxj${{2M8stm0YOxH@;WDHw>(C1Dh^IFgkr}J9~Sa^!(Hy&tJew0CTANDRhlr2`9{qw?<7v`8w<1ukK zzK8tA7a#_KEj*SD$n87V%6tm{op-Mdgj8gmBM^!}X<@14jJ%HkPMj+xFk0E9=b&l< z&(p~m(dr0Pz;y_C0@1Ft24e*J@ERd}5=+$#o|D;c6#UK(g_syL4on;Ml582VS(?l`Pxrug|tfi^xgGL5EKa;QI3UANa)_yXHebUGeTL)zCC3aDmomDaHB+p&j1;!Oa zCVB$F^afU^a{LK+YJXjAdxFnywq6V7d>C{s%h#=J@KJOeCzt9KEmY1f;n~d-#r>6_ zBxEX%rxg{F(RBxtvBq(=1h`IrWv&9078=)qM7CTdYN<{Xnv?AI8n!|u>5~_UBd4DP zk6w4SM^I$!=6=nX&g102zfon9YJdE*8^6Jpz4Bza_oRziLrII#D3}|iKx13sH%2{P zOIf-+YJ4-B$EhEn<`;N4)a0qpzCG1)F06T=2OiC!AyN5JBu{8RTY*I zexR+N93!wOHxwn5x$D9~5TdF>!v$3uGZPGm=K(%7ANZyM%BFn^P&ul)${t_{Wfk!{ z18o5fDNk}T(mG>~51nI~)V@vAhYqx@ll{?~cR#P8 zw!-`|SSs+&deylow$uuXVJNY-i&um&857Z_8Cee)K1s$a^huXwrXP5((E1dnuHaho?mrpttga(dAf7tV zSu8C0m=$K-THIfhQIn~DbMQf3y(8+N>1MqyK?>h?GU{=9HDvsK<-#jZc5Rnio*%h)ss@Q8;6R10c) zQt{F+kJ?@$P4t$);ZY7k$MG7y1^sJj_9}*^7$1&-@`yAQX25WN|0qknDF$mm3ZVdF z_Be(3?R8ih(OH}ficzTy**@TQNJl5PMN&!)$3P_$Ha)pkbq29C>FoQ>7_u7!H)_Yto4p*a9v(Z4wr~(X%={z%zG!mF5$z z#p^T-kpzZVYsZ-zIth(l?hL4|pbPGM)YSCT?Q(jL$aE$Ss#Tp9oiy;Nuqy<@yX4f80IgrJ=;sJPx={M4Nmcm#rFi~#@+znkIOUXuH)i4{QVFBzC2fD|1 zv{zn&V?H5tC_=DYr~1W04O`vzqh=s5@zPd~y59{`A@vYBpVXddK|4^VAn(F_i-f}> z4*!fnUNaLR<*KPTAqX;tf@1&;M(*X3sxUX2A{nk1&IWeejohdGYWelOSb@&shgJj< ziJX^)ktK6?%UXmwB)nTEI6qe?xc-$OCEa*n^tTC88R#9U_DOcG-`Q6(lcJ}V8kWo@ z$RJnj@H*&ms-JGof%rE{XdqiFbduVE>{WPOx3fDhME*#hb;}LPfGI?2$%#k#KLo;; zu?rc&j7E=415(lM8of9MnKZ}1nwf>gi(86LKrEWKDOvzJ8t2eTCXQE~GHKp)94Vl< z5F`v`GBNQFJdHQPE9en8v{2N6+&*fg^Cc6(m8niqQEGk#5Z**a#03xJRTa&b5yi4U z6x;&L7C)S9YrizI02T3D26H68?wUG(o-rE)arVjjDHotE3{u12YyLZ5Ljg5>4y!`S z*%j1S#$AXfAqXz;gv~Va>!|+b)M29_?}RCdS8m)D_Z^Md$OF=3!<71WcDRh3Qh!9- z%Y&?!C7>%=({&fH`X7L^1~@v{-cnH)xoewCl>`!P)R>B~8mF2rP`6Qu4@5}$DB`*# zXW23D2sg#V5K4MM1aU}c#B1nTnZ)xxe0=48ZA+T>5{8P8RUI;81cowIG2J`H85GPr zlha#Yf=WKI=p^;8+hWn5^A+ZzH~w~(4VGNg*-O?h?f6#@fPCf2uu862MlciRV{lmC z8z@qccQn$bPn49pxbv_Hln|EBag{3O?bj$_{=QKGyo+;j!Dkwu0dDQ#q&IAGU)frm+8iTWWHzZCx++Zz{twM zA#1P@FacBUSkaG4iiAC)3lLK;$(nWTcoTt>E;AL;d<)Y6@6(+-9y8SBw(%MRTJjvR zDel=uAsyHIY!9THVq*>@`Xg*I1qb#NHg?q$*}Ar8M0e!HCTQX$(jqD|)HU$l->H}I zQ9wD6Y*n4rjHcRl?JkvQROB!2P!Pj(|6i>UK^b+xGj+3Ua@oJ2cXw_u{6X)|1-G;S zDfMbq+Th$>Elw7LU{?N4Yk^S>h1rEKKYyWZR zowZZ{qERUH<Y{EKMwU;jhswH^GM*Y+=3cW5$yj`jQkDDYPo!2kLmf=L?x zUpz_gN0={w^Ycl_HyywTOXhLNWudlw{=5j7a|5EJ-OJaPuQBZK0K}w)m&^CEuZ}>W zXA}^~EP*gKhwJtah1QG>a7cUto-L-p_%z#cWTT?ItLp3HWcIcM@v>(ITk8r1(~S4# zd0H~BoV{Fo$yh6{6NQXJ-w9B6&bzeAa^{oB3U7^E{4&8lJgk9a=#BZno+Sh{o&S}}pn>-^ryLOMMu9&g#A^b6HO6?=i^US{!~*V>GVyFJ#@z`< zAWdOlx9?0W8u08N!Cot+uF)gntGt779mWh*visDkuj|zDS_kInP#6Mp8_c7(>ZmU=Gmqj-!_FVQ~CPbt5-v zEy_tN@ckTzzgy`oHO^&qkL+r6?r(6E@Z!sSOlwtZ&2;%fLT?Z3CI&3_4WVJ-%GwF# zenqolPn+3qzX_L>+e_{UU-D=u(DYNq=>OFM=ZAUQ$36xv8V1YBB#g5#`|3{Ga{lq* z-YDQ>E<-#McGgd@Z=J(%c&6#+(FCA%7l^aDc9xp_5d0gYi{$)YAYB}L`S%dO#MtqW?d>wq#3qf^>^_;@f z2yZJ&hdh9FQoy$I7&p4iCfFqhUZwigp}JlQW(u3R^J;L9dk^<(bbzLaLBFGU-j+Ie zE2M&OwbaNf?@~9U&gya;297CCtWr#ROJ-}H-=>P;%k8Vz###nd5lOe0B``GrEQWLG zoK7L>t3MX#k^`h4| zRxaq}#cA0q{Vf{j&SUU55I$7T8)eXnE-q7u@dABB;~^4)zs39o;DW!w@eYIdpK1KF z8W6+dKLIBHY^wiT-VH3=AB&4@!}TYC|G5GV8{eM3QDNY)H;6z$m8Vr}Ky0fcZQ8>j zT1PvOLR8y~Jg))$d`nQ-f#`I%`CVVo+rRto6SvoC@>i*$Dj7l;;pkPQY~jhp@R!Iz zUyT3E*3Je6X>Ppu@$~k6eV-6w_6{JCfgGiD$M!~%OKb3f-NN@9S%FVjL4g{ECJ}Do$BF?m(mAJq?yp?4O}FYh&11Xpcv~v@2a1J zL%t@AG%h+@gt8YI`+bd(`Jbk0RQ7pfY1}3XXtkIcE8?)@bphUzC3f?%N~HQ5r{Tkb znjV(gb&=b=j`{c*ofjB4j(~Pd1(4c=D8S`p)*=AGJE+>&b>8~5$wQ%O6SwJv5==0- zr~CiK-~dg~4M@0ya)$%x&voK16`S6A&sG7t-6rrc$HeYb!PG98tJhqT1n&CZCu^S+ z-j^N0gJt$^j=4ou4UCMKak#Uc;Z-g`c$M=E5)BSz6&o$r^ztsd3k1fTBUj2YIf{75 zfLvAnuwI!F{!_=nljB3nmOz4C;7mG_a8tWPF+>zr(;mWp)vV;r8z|@D88FJKmQ8SQ zd?sp=sO8XN#%PSqL<{8P~YQBG) zXVn6vm#vG8x^q0-v>*OR%`0c21wVth++aS>$)C$u>N@aov=Kim{q($(Hmw=}vsX%h zuKB1aHoxH#hgB(VBx}*f4_VKq!@m!Hc_w=;UvW%09~H3+cN+c5UpQK;_P_#|$5a5T z^+YJ+9AR*?NWwd$0N z1%lBz4-enuMBU3__P`QAtgrs2i89c%E69^6FuOU!HZ z72W5^c6mY{yS&JyM;At3^1+RB)SJ0?1i3{!3emcyVwj47lR)flS-ApFrvwEM4$4op zLU=X?;q)h$sKp+B{5(A-MTWgvWN^qaI?AhE-ujWcE?7L?gwLbcX?hnb%Z<;!SGiI- z@c!GO353E~^yo3g#Xf(W&KBlEdk?|m{1VwR*&8Rsq3iiuPNsI_uW~Z6oU@z^2-n)o z-T*NDK{=1J^eke@R*=hOIlR1XAoEiIi_6j`fz#-0C_0`A>XU8GfiBnnh(wKR zx%yk5{yJa!0Xml`4+I+z9x^|CBZQ4we-tHDfxe66wdoAW6&J1CX?*L3ZdE0}^E{52 z(^GXPPTP@+?i?#z_3>KNDiA)oXrxVK;(@V`755WGCH3cY{>9f`tb#Jk>jr!+M;D<> z>@tM~@1&D*@z3wse+g{f21QV>cye8j1VUobG;}ZpNs7^nZmXn~vyTTpP$FF?+v_gE zk=d>fr9*BvwDqXI0rr`LeV3D;61#bl?P z;*}!+Vm|e{4dt&Gqu2Mf{{CR_pqd0PHTtvEOy5?VAJ7!=+IJgMek}n8oQ%tEt&y2C zm+YG2bn99!Szo8@%=x8(inG%D+cAQ~Jgw`>bUya^sxCfwndG30vp$=ZbIB7RuwO!r z1TRt_OZ1}x1`ff$&TqX1aL+bWzFeP(0me2o1x!OX-QHhI;G=;}Qsp!mb6Yv=O^)rf zlXBRC5uM%niSqU~QfR>woG4G{d~FB)&R2COUa*Jf9>)G=Uq7lKfR7789k0ggO%)eC zRS6QwL$Uq}3oev;zM%%QOyAu{bA_gWR>EI2|7!`FVCbR0_UNbn%8i=bw(Syp?(>(l z{hC$U3qkOb-rZ{AS`s~9)q^f_6H+2HfUh_6CI{Gea;P$;i>^0{(vTO}js4fxk2JaS z4Ykx)_Js(QA51|Jb(joc|7!`p4_*}iwMX^Lko#H#2Cl;azlbR_mpPwSlY^ro%0(|} zhw^+?J?6qnuU`f4*u(P1u#FpA*ii4I=ylPG*MxG*oZpAPMB#s-JKs<{rs~hHdXs@E z(0f~8FaNOwHcQ^rzxL=ocB`X;#rt*3kA6`+>|Wo#I|c>Qj@o-Y2RP@eDzMe2hJN|a zF!2-ppD^)HnE3x9OuXb(QB{r2!9G}8UM>TTm4QnhK<fg-nQ z?sRT2>!7L*ihZtNW6_d+3^u$T;;o!TG;2o{(`#_7fcm;7m~|}!?7Ir(SNM?9OVyna zRG1>=mx&E09d0U%ooJ$!Lti2A8-eUDWP%k+#%~6sMFziqmA`rYKa+0F*d-;|gvCmR z=^)S`ZcJLlzPDT*H~R@@4H!D~D8>_W8Q+Ml2=F}Id^IMVmDNANMBZB$0=4=bC*=v_ z=)oMEA8Y$Hg(SEY2a>_N$0BJBBzX}Zi~f&7DAt9g|15;&O3kT&DYqqUw{d%1>dHC3 z@Tnk3jS(66=bYo}KK>EKbB+rb04+v=@{7)fle{N}hS+D@BvsVo*=NyL-#`*H(_w;E zz?tKZep1Ih723Xf#&>%tmaO;V1zw%KY1rT_CZSb8!i zG*9~Ru?q`WXS1_)R=EP#hn}{#OYu+dv$M1Dy8DQ*8Hm?v*y0)BnNF>}KLVwph-t1u zw1duKB~?(!)v9p+Y3BxfTP(&}EjjczFL5|;-!La(LlP5r=6E~gmX7hry;FZtcuhfmGtx!)HE!_-J!2bLeO3%mvoA{Zn?r)%+ku(owvW?7m*7)%?p6yZkpz%%s>ES_$KD{#GVNGgi_ zcQElvirZ>2jZ_R`d!f&A`^VP|%`CeJr8ebTI?IK_KZika@we{43%aYs**9AA4@@d&r=u*RXH|aXeri%3zu;4oF*Fg@Z(0`TFIM3#Gqr#m2q0 zpZcYp@;L3{&?a!Nb8}+!fjK?+9(e(CJJ#SQmZtluXxw5Y>ccS+EqzZcy$6*aVMN~g z;y+6`QMdHBpg!0IM=#R-vWpW9^FiG#;=&Z<(15mfR>xpC731ALa;SH{=!DKP+H4$S z>*Y>keR*-kA>b#y54aB!aa3OF32CfFpq;;l(97}Xi1trM-0+_><}#k1-ovJf#A4`3 zr;{yuG7x@)oSlDT6wE;D{HX zWMd+Vl%oUc>4j0%lApN{5N(bW;5{5mW5{{=sz@RY^AoKVO^*!zR?&4lPTSXB%R-Jg zeb?_;_zovlFvy^)0f%YRCkQN>td;k8?TgJFd5z&ij$-T`RAkb}_?K0Y&CJ0<2$kYx z=o?OO{|Da?$Zi|pLEU9u^PboU7d!I=^EIQ@6yakEi@ayMnE@`(PW=-rg@qg`c>PV$ zi`%yJ>hfTZC`i5Y=Oz}+RLi4$V7b!UGMLa<{`@A%FE3u_rk`LoWrkPZvzStuLO#5Y zy+mP)&xz4~-4Ol$3dOFoK)zFXG)DSA59ObS^3P8B=k@s~ru=ULt$zIK?#d`B$L9%v zL_-3Kf$A60K0%$A$K`IvJ40iyu?2n@dmM+1X6-$N?|Ij0Mb_=wh9s0RwQn$tYK6(d z#sCIk4A`e?0ZyhSNI$RGku>0Qxu^i04x1wayU&3dg*_`sInPaePQ!K_C*YQLGf9K; zmNLAsRvUZMF2zh=J9%L6D7kW=E4vzRN`*8!p#dMBa7$U0${NWKme=9o{!}YVcQa*LN@oXO^7?GKR&lT$z%N(3O(iadPALj3yLiaf;Q z1%4rrW-bCfz6C;7?PUPXoQxB?X(!C20R%B2qg6l+J< zFRb=Zw{oJpMl=b)Bj16L`ceX?wRXF)va>lix@Zj0OI5+pxHMK55G-G5IW`GA>Aw%M zBurYAZp^%F-982JR;ISgNn?nof+DkC?L#m6eM(()%Z&SHxaNAX*4r2 zXsKZg@pqZ@7*Ry?GNjI)FX~h6QDuuOt-R(xH)pK?MsgJ-)V6E&r*97nf5Pgh)lz*&*|%-*A>6D-X=~i9 z--30h$hMTK+@!QTM}NLRzBERRPu8|-)^aidz9hWb(|0Wo*-Bu~$cL6yHDEtcx|K=9 zy{9j8^dhO~@O5RtZ8Ssl3`=yE+vGl92UT1sp&TtL)d%*}ZiI!wJ6|@xR%YKl3+&W- zkCuZwg@d3WfCNO}#a7@iED+D#yE3rxbf|3Sgtq_T^Zqglzw!FnQ{QS!ZAIM0O8t8C zZzUdYeP9Cw_;W5+S}i^<14yT_U2D!76;iiqW6QwdltRtdk3*JliP!B=5c-!F;CjhADIu)n0UVtVBJ9`o7H(mWg_OkJhP76r*R@t!Hy)stIjjR&moGP7GBX8M?R= zc7Tpra&Z_1O@C~@>A5*0ZuVKWTb?)D^9`@UvE++7CjMErD;#*1HwRia6<3;ikqkx_ zSGEY~e0rWf!4RX>Hbt1n57O{93G2?!5mPbgZ#}h;-%2H8Zouz^Y%jS?6D|qFrW8pA z3UNWX^gH&LA_7U-Q8AkEfdB&YS2qqH#|=3WjNUb{?Hm$CkfYFCn8Nn@I>a}&OuA3@W?EJ1zjdbY|L zbomifII_^UKM$d?@Qvtml*cr5&~HhgXsO4YE7xk_=y(Qp@Rv>vH~rK?m*S;o$QLPZ z)TGCI6wAdOFQ5#2FH>ZaHdpuh?&MbBd5_?>o1JLYsIM|H>zUJCfkoegUbScV=?Kmy zD3Z~V1SMXc(&N7sq_Mok9{l(Kra$Ze20en`IsPJPyCR8+-b#XaA;c}C>JKe|p%!NT zJVOavsc%mjQaG)OSM3)t^U^VV@WF2)RX6zUd7BNTXt3SxMrq!1&YMa*fWp~yCfM!y?! zuW*H-0<&!@)7$x(Tsf2irXYw{XDEc`z@^JV=64))`Q~ahu{`w>x*@@<#fA06K;b2Dl=~y z*D|xhr|f2kjuTGWeqDb=7<|*SLbLzjcBLTuXS@Etyc2@X&zG1}vTXd8U<}U9070+h63kFh}CFFyYZSyYWKa#hV zYRqoQtsFpAdH);)=~erHOjBjFz{h9D#p^p3mm)NP^~m9NpWIlPX&(|)P{)8o8tD}F zdwhKrPzlyLJvkoJ^CO+R&bKZE%?%QzI|OW@F~B27f{86broQ{vH?HEEpC0cy)w%6C zdK_%q`jLB9E&eE&F-ur~OMBYY|*MY+Tf zZ0XH7^_$-M(2+(uLWZePe#Z~JAG*H)jr>)J0I~-hOyaX!oPY;TS>1Q#a?@;T*W$|MM)Tl*0adJH!k=xAns1eplbkny`N zPub31v+YUZw(5wYUyYO#xNm#45H|A)uCiw}HuhGWbm8(Od6jU~m6TeXB?-yQ zOg-wqWk~qOdsDPl|ILx}q`bmzR+7_vkcrb_QE3mb^?3N+<3wK<*w@`kXfh9)xC!)E z-UR~kTXxNl_d&?}Z6pTT8Z}pTw$M%7F3+wtoYuPmH{7n#z^9_M^<3A-)t%+OyR_J1 zI6)*g7e7+Fc1m8p0-TX`Dp3+Q;PyQRkoz*4qPiann3fGbzPdDbYjB}acC~h^_W^DJ z;1|;-$!8)~W{FH=eoHeLLzo9Lq|Y-Vg|o^2?=zx-2qX}{IlCD^{5JeljCs*#i$jH} zE>GXy3|bfV$r;jlIcBE{^dr9|zxXCm01(wNy=HQL%^$E>^T3jP@?l2H^}dLnaBnbA zm9;a#S9b?Eh*txn*^hR{-*Fv*y-+nk5?29#xY1R>iq;?3b~cy4&RW`ZeMzK(P;RDwrSe6Z7FCceR`+H+FA4M)N#lw`@&45+&CrQ z%4*Tpd@rZ)GC*lRw9Y%;9qk2TPb0lG1kMJs_cLu=u8Uq5l&9X3gHA~Fgb(++)W z{23}u|4P=sAr!Panx_cOFJw;psELB?GC$_eG&rm$b( z%W=THOL&&Bue>c%zGCOhUk66M3QXf(ls3d!PRX6(AMo-eV{l8eU`PU<+x<1><{X1C z{=sBGU%XzJ$ItBShRf5w3$ZrRS0Nh$%T(klkaw{G*!3hgVE)OZ`SxBjM1&K;b9q4h zY5-XNSYxi@B5}%aDr(nf_ry?-#K~NmYilrG&+K)w)O5YQ&GgvKgP%)fv-jni;BkO2 z9LWqv>g${WqxZ5S{m)6d_3rNk;wf+oKHEaPUrvJ)tfBgOiXpSpPnwZ2D_S3DyRX<~ zTWWAt!EUQmI|P|jN2kg;^7E8C#A@0FxGeeBx~FAIZcw2%DQ+;hgO(jKzfg7ly%xgL z?qeoaRrA#DN-Mk7bjMGtHuQtmOi*+Ebv_`HP^y!z6M>b zz-6H8G9q`=qd;}u0|}&CthNZgeOxDWWu|({g2vF>a4W2}-IlaNyh%mUPC9&}mFf|W zp3&DPF{>Y6-d$&9YKF-s^Q@H6@#VI^729+|qmHPZcJb9M%ktfX{HWiC|zZZV){{DVPfJpP8Gw7UCmK6(#h z*6TYQtr}rA$|CZ>S!ZpISuHWh(e~pGmDf#4{O*D+a2}0v!0Z!;nDUGKSjR?u`?Ov^ z{`FWT{_DrOTEuc#CB<6TKgZZb0IqAL12G$z&U!&E_qMFxH{(<|+iC?7GCS5Z^EJCKv|WYwvmG<}hph3LC1_w0?K`E`Noh50iI+&99*TOcB@$Z^Ch zLt*6dG1L)-v)~c?XAN=BCru*+swTVHS4vG)Qf=jeOYdwF5${k6^QRaAS)M%VvbS*3 zi0QoQr%yb^^QpJr3Wf#Q@c852&9%UIQz%J~qFnA-};b!gh` zj~nW)TNaBCBf7FP6%s|_1mKufCsWE+g`s|x#xUOM$1)n#ZwOwAQpT)|R#aMQa8!0r zFxf7}b~%!cRNS-0efO4Am1>;8jxs6Q-3lYoKTS&2x=XA^H|jbNm-`D<;biljS`8H>Z{aF(X+1RkmY$wKrN?OGM3Ecb?g5>`kV4wd z%Y(0iRUbt!YY3U{p^{TI>Se2ai^H+* zLlwZ+aJtlRl1Iu3!o;BmKgPR9kEXiAK3`h9U0V(ODGlCvAH9)cwGM6;?Up-;7`>rS znvqaHU+K(ci70DfHyg~QLMx6pe@e-rbd-hTXS-UnLU4Lop)XW5Y&lxn!*2eqy;+il z7{jToIQ$Be>CKdDh;MVSQM<1!L@pZH%R8zVp2d|OaF4{p0+D5;HtzdsODW%Wjqd>} zS8MugQDIb@XLnkMWT><;R88g&)#N~w5RSWsP>F6uXHMiSziD@0M~`o!(zXRD9E1^t z>D{qqP1WunCt}@rSfp0k^oWH33sj$??zo@^S9*v0Fp`mp&AfkaPy>9&W-a|{hjHFv z^wMuWlb@?zDtI85%}*KuXFmCk{qmmP73({ES_WuV@^NUeN$)EKDA;I9^v%vpZsL&TNtF zJQc9ouP8SLEAmCZ#a&ffG)vIm(L_iY2oA&-n(L_Hglm}yYqc>Eq72&?x0^M9AMpS7 z1C<7y-o&Sms5*FF{Lt0&_YJ+9d-7wUA-X6!LfPS+e{K17Z2Bg¥eBsP86oSD3qo z06nh+a<8nP_J%Fkb#uZTj?2&Bz}1IW1UCo!ySc!Px119+qPQ-fwCy%h6nv;4!L=A4 zPp5bOEzb)Jue_s9dU%3u#8tIFvmxL@+C)fZb`kwL)8`q7&(-EyXbkw-Nx^R?3~H~w zv*X{k#=Uz7W!Qnz)J#VWAB=nKB9A8}W_!vN3B|*X1g~XoZR~q=CZ6?Rdvuzv^v&$| zvu>CvY`wi>G0@d6`Q_aJZ5X-cTfX`xW>%TaeMn3x=BlThAJgMaXS}w!ZD?B1xE9TM%b0w%sfUoYq=`?VR@kf zB}ENObjnkHEMcKtO^m=1@f`46V1 zek*YBgZc`yp&Q!m4xJZY#X?siucWA~)gFh|eYKfN510etW?fWKIi1}B z>;`zUQpKuDS=D&qT0_xk6|6L1e=96r8^|a?GwgMs$zG8P01qI_^}~xrD!n{H5ARB` zRR&g~ogZI6C4V_Z<6O?y+cj>UDQMS#EHpWR&rgsjjqRb!eDb{O&SEr^ic#_=#Kbo^ z1dw=*pNxgIC4pHttsl7qgD>Wpa0fK{(?PL*rE!)g3NCm<@aJDaBkyu z9)gRZ+3d!XX(u~B%IucsYb?>jp9Ei^j9w%;R&oSGG#WC>wZTFyS-KIflx~61p3<3_ zr}Xs(_=KJ!=gBa%L^7&kuxya*PSLt>bHmn~MfNGZM}Q!WIhk_pCLU)Ucq*-YX%MLr zKP!}Xl}j9NA6(9cMF>?3G#4O#J(;|}tj~SC!VZz=6h8}`d8zaQ0C%O(#r3;zi#hv7 ztB62d^OwqX$PWP4X3g)Qb z{^Z3ipuCh+a6Nuo~Kd`Cze1 zS5ZhC!olo(({ntGWQ2B;9I&J2HfSscN^~HW2IeTxFnf9GfuD}C0sQ7tmk`Atd`YJq zTC6dY?l_1)73p96mJx)h?(6|MkNVJv5Mi1wt8W;K#SG-B)&QZyXBd;;fVy}yp3!CJQ+ z(bARTFsK$2wA4-=R#x#ewYmZhoKKfCdxBVW9Uf%BIQomkE4N_N;Byq_dsBcruHL{7~ZIDN(}rW zH2$~}s5M@Z-%e5R*TfJ5hiI9L2&H;4bBc<;1d~Nr+s!5-?&|uZO)03T()rE-B*8qj zO)<V?DXVN17ilzf|y@?Y4D6ZwrR2+Df{emzqSH{K1t*IJ74j*32bdm3_Vb9 zdY~7=NjKFUE#0a;v`e@M!#rXC0QoxLNRRBlCO>K<`V?&=eU{)G_RfA&%rybCDEMs_ zPj>RIfn~FNo@0>%LUybxwvnPBriM2%7U2E@hH=$#QgGnwcSUrCUx+YKu@ZiN?JW8o z*7B;dIq?{$73UL9L=X2o$re&l<&rhaxcif5)Z=;7A5cTt$jJV#{F%tXPSf|i2?&2Z>!BE%rrJ&aPy)P&+jMwU|VL$FY$cLy5S@G4Lb zNpOQe6{BGP6K`PT)sOx~T(w!CmS_En5Oc~Sn?#@RosYi~2D`r$I*Co=_3xdk=wW%l z5_zN^)5(yx{!=FwE_xwD#OE#X?5S0eliD_dwoe*YfU zL+~#a*-0-pyE|@z}J-Q|N8UnAvAD2FRiLhi~ZHFf0X8q`2@i* zQLotjH=k?cgD6l~qZ9Hs6WpYsfF3}B_JBLEzpw%}9IJ`e^~DW9F1ZRgeb#_TR@oYI zyQXe;q_f<5VtfVEAPU5S2zRyWP}5rpqZ@%x17K<)e}EWZ;#jHo=`tMT7j1h$IrBl{ zD0s!Sz<+fU0!ajQOAkaFz%E#jLvEgk9q`#aF^9zuV5~d^V$&=*)5%K>cE5t~ueB?x zc7o`I1O_zM9lpDDfU;Uwg^A#D*w*V~b6k{yM8pJO^mo50613oQVQyJ1N_1Qq8K!eo z8@|pjF&waQ)it26+}P?2{LL@k~D#BFC3B@ zX;_wWheD50LUf5nN*9pOR)OrL10=}_Tv?nT$gVKcO<}8LkhnSku>>dRKEu1-cyXmB z!ZMga#Sty;A+u)X7Y~pq8vB8okP|Y!PUuUHBArtQeO9IC@j;c028c=l;!yh6%c7h;B6~o-p)&q~Nx~SDMn%_Ue+0#A$y2Wj zynH~iPzw}~yJHdopyN{LaupVy2X6d@Y19JmU`!y2F$>wxykg$%YH|qM5T5Tz9+MX* zHJX0SqE)@jrWZjfzWR`qCC;nC^X2|`Kd3}j;fEJYK;&f8Abxc{Xn7>^i^(7w(Ox~s z1+Lf3x^_F%RQvkj1nU*g2eDMQliD`p@&d4E zi8u86_;#8dE8pmoYbyTa16L?V1jUf@a`;-^wa+O_sVh;yfbZnMd=J#p;`j|lHN%Qc z){1ABg7+w?n1|-@Ww}05T2Z>n6JkHsish&Gh@nEMHQTWJ?%fQDBST{71{W3sWw3$8 zFQm#$i`+~u=zaP{fdS39H)VwuxXasD0_b=G{QJq_nl-joq{DX8AgDY~dv1RqkGI^a zS>~3ISYpQoF#@0RIE5aC^{&nRnKwNYQFKbuNL=j7;Rk865`?Qc&ctck_IfiCFg|6r zm3kye5+?p@$&(^ZG}`4WHKMvyXu5Jug!)V=Y|>3cQFb5ZOrxCOt&SUyXCw%zF6uY6 zzonMRmU;7vGMjRpssK#s5yc$UjQ&!8BwbaWD$LOhK0Zo1w)WiVqyH66a2IsMrM2t3 zZ}tR+q5A;1pzDk!aobDhPkCR6^h}A+ck<6nb>9TXBPxAlCS(d!t5qa=dQW&)lX00r z^A0Lv3ETkI4}(w0ZolA3(hLH z)|FpQ7<)4>mx zC77$qaYiuYk7lOjq-!adHUk+8@3CHL>N@f%EZ{5iaKWR9A`D~REY;(6Cf@XFyrO}i zoyx<$Md7mqx!CgpX=|DGmeyw=bs0W&$z#_pw4d)OFS4n{#EhpA48l%s>De%wtv>uF z#j2LgoN~drQukD|(?P2zQF$hdpU+~lzSkq!oQ;a*ou+0MZbgwjnHere$~NNL4!0mj zxL*t?&J|y(_*^*!>i6&;K8A<2s$5fq26zH$hP@J!>V*N|XidOa*ptSb5?TGmcVk%7h=AK^`h^ySOTYm6DPZzhgQG~!D_P6n|7ORp-zRpPU8e zyU35!j=?KM`k7`Y`y4(Y2)L`pk%hr!!{JU|Onm+$ed*2z<2}KB9 z0U>}?DR<}mkDl|+ym#)G`{mADhjB8QfjrO3e)eALSJvwA?2Ks2BF;c$$*v%VU7tSr zJb_yX;7bj2y+cn&u57Gzv^O3vyeCm!apQtarh&cE{m4!nNMEdoKf_N!D(;}9dqHg( zupm)t1^E_SVCp2xgV5*|Yb2ImiHuMXcr&>OWMbf+K(tqXG52a#G$MRFLgbl=% zR@|Wlntsem?Qy8q752XUC7sn%Rjq~bP`tA}f7>bggNdLgKeNV<-sH+ta~(_!%)PSW zn=_nyd7gmjx5B6|7a5^R!`2!5Ojx zP6Y1M+VdmyTjM5y!hz>w&%%pVlQli^S6807c4Wb}i(KA(zq~hyf`4%?jWZ;TpNEVm zeo&C;Ee*{hl4?$pXANe*aao1zo;8oSP?<+ z9#NPoP|W1#^ac?_CaRow);tV5u4Jsi0E|lmU8quH4mtU&f1@_({LEv;NM~Cnnr8PO zK|;OE*&7gayZU08pO&lR{VbCYXg6uijX_j)iUtQy3Lm)?!H*y5^qJ3VBs-CLpgf%R z?30!gWY~c$%n zG6u@(Hg{FW)-UVwaPbrtwmL4{N_LUq-q?GN$b~5*5X{bOvGqruOgr4u-=9GjAA-2# z=F2|?e=ACh@v}e*)}#C1c7KyQqOOuiwhwMDy=#6nV;>!`&(tzio6an zcNVQb)qUx$bydaKBC%~gsLMO^CbB+5b@Rh!5^NOk0sbflLbVMC$S#FHJGl=mCKNx? z;fE!2P$~JW=U<~U_#FY=T+^jp_(jkm1GdZusurJEuE|`C7R?~;m8z$hh<4wZ__0@3 zx?fLH_cZ@l+mA{D4e)YDcC+-LcnN2lv`=>icVvOX_%N3TmnPR_TscE~AC}M%kX|X^ z-FChCd;bGtu3sa~h8fC4MIrmft67B#rhm+qS1V=&V1WUwYunie!D?ZpP6KN z&D#v&Svl?HlY#;GXnCqo@-JNvGKNLh&*_Kl@NX_`ZFg>Y*zTMgx)gO;zE3!WTtQMF za-=Bh34I|{8&sz6XvM$YwlUa6Vm_F-*md#Js2BjxLnR*bd`7T%-ZX3-sET0w({3!wr+q&ate(^YY-}K4p(Xb~fpd zh+WK`%|m>-o>NbTw;a^?teB!~sk@K=1=+ZSvol^lI7^_eq+_U^(V)h)j^dVIRwdXT z%=1)Q$6#1h2^9^&Hy}kA5AQUC@Sqhn9Uj+SbO&i>QzQoY`qa&D8$SXgOPw_^KRxLT z2LG?NT{}t&TOXV<>-rLpIZu#IuVQ^Pm(vL}ZQsfYw(91=mOHQw6|RZ*jQ#?pd(X@5 zIUI?B`tXEx2@KL z!NUe{cI!a_?spUG=*| zOdTPGHw~mmoidkt@6N>WJ#0%?Vv*Dv>ww0gWJc4=HK;{9pn-C<^LzRfJ^kq|VnO=& z))PvWQ52b?p;IyM@j^E*{zm#wV5}vf=S&oQ)%f=yAnBGV8NWV*H0!Fd6{wrIHNHS! z5IbgveF3p4zBh`=n5cB5rBaX$TW-;`XbXh;C~iAy4u)qNU3_d4PZ!8^PV;P|>X7kj z(v*Ivx4V(xwV96XdgtyW&?e0fttmwnAm}kkY~r=@vx84`PJck=NX}fl=BF~~JHmf` zvhqsLk;*bz4+`IAC>~k{El(7Bv+dDDQ02mz_3equ z_HXNtK%```O0_Ch!J76<(2N(73*ZqB+@;X0n#Xl_6ER4;q-u;BK4~ec@kP+sGTHJp zl81+?MYe(BIR(K8*(8-7$FFHzk(bzJ5~UT*ZV+*DbxSdBEFsvo243UzY5GwOyfcn9 z>@R~`5}d|0xiK#-+Q%=PLC}*7M5yy?(~$aMU|(}KZp3E%ok&LVQ>AeYx8v4H*S4CWc%If^J^nL*8X=j z(!UAOg=%bC2~g`bFBc9(89j#eI+)7P9^0w?(y zq4S}V;q8{nT}vS|gALEe)bsb){!dr0tE!drT4#kVd;{pm zr28OOhI)FSh%FoNsW;;VGgL92@3)8CAiXW~?P;XGP3Ma<^553>RBmIbDa1?JS1lp= z=3~T})}-7Z!LVvO+E>c0?U=8gwoP0tPV>o=%@=BKc?EJ;54NbPiKbm>QsypF6*}ui zg%H|}INe;0(Pij3&r(3wy^Ij=JJvcDG$2xqjj$T+9#l7bJ~sOz8uK|DV~QCuhv@ zq&~n5Ik`OA{k`fr29RFQC(Blj`BEE$<{ITG<+6hh-ziW5re~zI*vpUr^P6`X$!bCI z!(!NBPussD&80u|eVtw;(JhrgsA9}6KWqECyA9)|>+hd@uImV2$n#gHeqHXa-Xw(i z5Io)Yb@E!XnDzaNSEY6GTaD;47uhD_2kUD?TMqUKLewzA%?%AvVn<)Wwm@r!Ck7*s zMbeu#;ar2>Yv1uk$1lAMn?F+;d4R$Jp8XsDAKuxcp1)Dl+|GP16=i= zu4_Xwc6Ob&t9djOC(Igj!E+1sJ_|%O)Jb?*TJgT(GLFwn= z(MsrGRkkz1mLjzm%jL0|cc3lgP((Ift(WDPMbNewi8}&CQ}JZ`^-CN3&taRH zLV21VFzH;<^VOMa6u8-YzTDW>>DL0|M)T?Qpg(ypXsv#x({^&RwkY14oh52xAg#1P z+=T2-KbP}d%}^q7vMZzAM;uaIr<@zn&WMn!YG6RyBpB+2q==xijj>Wvqbn`*h(M>7 z@v2*9ZQrwkQJ@-pFmb`<+nFA$hTr8hX(u0+CZB91JGE=w z+Zh|`@(A`-b@?kWV685kEp&=ghGm4(oFw0(1MdF3a1?hKKr;ePANt?DzaImt&jrX=2IIs*#v9Pj(9_jr&=h)~8|#7$Z$N1^4-oG-qEqUG|Qce+@ji1ZF^vJu1$w0C3BuicdLgVF+J*i)hx2PK&gc<=pG|tJ(d9;9kcyKimCx)(FRv1F(0q=zpc0 zz?&`U02+Hfx%Y(A|K|6N={NlUD_;cOyragUj9LSZJ^1gGkum3^H~;^S{&xiauLt}N z0aCj>Dpd~D4dtNwBu5fd4bt#x(9yF3h!G^F@0c;?z#Csd75gnB6@0~s4hhA^3OERba_8uAmHeZISL5BFVAlByXYVCv-)C(= z%=vm>;}_86R{_joF#t3%pJIBY1aI0bcn;i6kW1A?%bFBhm2U=ZhCt`BxH{J+wRlLQ zk|A^9tv0V%8IY7ra|8$=GGl-|(iiyv=LVfSPY6gJ=lh#LeZyf~E2ms_j{Lc3IE3R6 zz61b%eM3`mY(JdD7Tpfq026>iEC?A&sXOy|Cx>=pUI7k@E<8CVh;9d;ypPxCvfK@ZBXz0T*>`8fUIRMU2}{l4E^ zZnQCAcX?j>p>~#9NIcWw0_DgeUzv}?x^>G*{CngNB%b4p&;Uy4`y301RnF!$s)Pq@`U%j8++^ZvwTNnT z#LOMlYma$ZZHB+zVs(wym%PeD=T3D#vX;{g`*|AHFF4M1R^j*RNX-WONXO&jHh)4@ zD~xtN2DEE^jcovTc~PZxF>KIt)+Gi^uu2x z-Z2fO-GF)xJ4j0jJsh^RPbp~G_ef9UY@RHk;;1r2IeDU|7KF}WErWXe+CYk~RC@@) zpIQ8>`_1BH=a{DH@%sh}q(@0=#mrB5#FO~NGWnEkO-k|Ga>+NODOYCaeKuC#E3X}J zvrV@7I5X6wpFVZrZK2p*GC)Kh-b-9Tu_~K<(Aq_b2}yHnMS-Fk99rLMf93lc)k~r;Rh40A(mcU`0=H<{c{EzG(YA zD>Muh_BKoVHLaS%&Y~qq@SDu!5A?p8@SJi9v=AjyAN8wI75WNVq)Al6sco1~6>`Wk z;x!$=FBO_~tL+?B|EN`5{Rr=Nh#W*@_A){|!KVvw#Lf+|&kE^wBe#x`ouJkov6$nn znVPzqJ0a-gHoOGTsGZEW-?eysZ2hynTP2V_w-Snzp(KmOD+tOP?_f}E>G!e7Q4on6 z*?m{ZDrUK?v*r~$(A(ytJ^`bXUuY$qxE;HvofBqbcZ!U>P%le zOrL|cNke2A>Sbmf^%)89CXNN6%q+^ojqG~O96ls;S72s{GnK2?g{%%$l1|aGoE{)r z43qZZw&RdQaa!yCOH=-yTl=tT!D3|*r)516gTg$bFe;=pm^y`9+dTy{mmgCa>2;n z8!t2ZElwJN$x{<P?lGC-O?i<7N8`$xr1nLM;oqA(ax>>VQJIkZ6V%GTh>qfz z#g2|k4?zhY`JUbe-v>{Jhrtgq`LInKKM-J=r?0SM<+2tLmfsM^;TPamjma{8*mnS+ zKi%Bi#m58f!;w`8h=CuaK!YIS0A|Ss@N7qmV-Yd@RFjT)FLcy^B@r372#Q0fjT%X~ zj49OQX;NV-INNsw>iU8sY}|zqh)$~G@Z%6{*}x^7JVUvq1i4VGNy}p&F=d;o;gIy9 zqy^Ti152_^7V`d_v8hUXUh!$}9k+0qNxYXLuIAt_dyeeP@ng*k$vXU$pi60rEdhQc zAoy+@6xLAM3bUPzUq?SnTUTht_%G>B?l2FB>({_7@g>}tk4qis-@B-d0SMFQ3m?n> zgb8gU&p~H3-_qts@l&^-OlL7;kTL9-rRdYxZI~Q3YoMPXxJ6Q{UItjclpGKO@r$29 z>Fn>>!O>RUO(yF>oyaIMp1^~8%0iG%lUVearI^!Z1Q~1(kkbVWTp)W#Y}}JudqZQu zt=SN-9oLFU?!?psUr!66w?-0f1K8iE3q0H`=yzPyy|;jqG8K939t|LQzu{sm+eDh~ zb+zo+rS4c3uMtC&z}}0c^ZV+`UhpLraI_WU(aJLPZS%&)_fuFD$$22eWGeHZ)-C~9 zz7nWQ=~>6xz*$(yX}1B{%h1M}hU%kXlZbBUJ)*v=E8*o)x9-;U(?U%#u=#o^!9Wad zuUz&S{ZEm9QmPOX(!)k0;;})cE+%ikZ%yO{nE6iGigpNBL8fu^a5T zDzF1sIiN8q7!NQK3)1_ii18Hitk&jzBBWAJE zuI>a9it&nS=y~_+ddgljA=G-m>Ya^0Th=7jnEpjW@2Al3^O753VaLB96I+fb31?x zM#$JbCd`8xP`s;}Df`tgzG#>Uj62Rq9P1I|wtIZmYg-i$${LdOL6(Tb z?SM$rs1DGa`|oH+XMqf-+}W{isX_CZbe{VB8sD$IKzB zjEdoyRCA5J!taN88)u67TI7wHt9U zt{DTmxT^y}M^OUv08rB+TiTQ{58j+7Ib1uyL`z5pn%#H10%%LAl;QPE4aYOfrEnHC zV%D#d_%%X*A$HYQ#%vn?% z^m4QJ_x4CAKEpM5h$`vhDFTvtsY6a4u#|nd5{bB3!j~D#yhuzB+kGgb*PY2FeFS)y zCgk-~*X92lCnS(0@3i&tureQ&iOL>9d|_)b4xaije?m^ zb{FK!9~SnYlrNWvf?hXPDubNtLiQx#DQX?%Kk#bQ6j4N*J?OVMg+%}lW$n(pVs_jlm#xi6OOt9JwV)vK830fGi73trY}%;^-*zzxS5A_Cua$*F7qkFGaWaj z?V)E4LszFGH^L9ePWNx-&-n_-x_#l`0tPOZ0Nb)=UqCZn z-aJes)N^cl6U1D5s(t_RmylJPh>)mypn5CbUyTvyxt_~+Jf?AFH{%efmR5y!#oNgdQUD4%r( zHfhQd(U>rJ88Gb_-PDU&JX&{UO>M{O-FWHooq57C`>4=-({tOw)QgcWED`myP!dGQ zhHN#Qi2Y{5M8}Jwx9~{Sp}JASNaKMg2ihaLM|t)7b|xr`pk?|?g)n{a&Gl!H;j>i*G*Nu*ek>@Rp(cKpK4t7GWJw0C0yX_%o=b!iMVkN|-)R@@dobU@T&}T;de`Jha-&<5y2W^*h{lq6&5%sFa7U^!YuNJ8F+dzbx`G~4prgXPQEi@w5w zM7{V++eey9fsG~%8mZ+}8&RparmT>*!0H+^-M;Jss-FM(@{Labu;l5kkGlVQ>i?og zWlf12Y4BSRHemwk#J(-D6Yj@0#4aIww^~1&`dXdCG_COwZ64ACo3`K}E5ksGW_p`sm zY!Q5HB^pTNdcaO%*3EhIf$je$|iz-?ZAJr*L4WnPUC_k!YI_<*6p z;sPt6(}zB$;`HtxX7$()Qk<3z*FI)mP{khEu?dFNK3wIYukZ|gFB|fLJQ#cp*8YHF@xgGbVsPEDcM5`lWZ}1I?)#R-BLPrhBmHd>~3M1RV=TUy>>s z2m5E^A2R}JBSlpw=j|%8H$K?)36nF$n(9^gX9Vsx!S<8)LBj|>C6U7SFMV(KHb1g6 zOhXqgZeY5zj(8$MN{Rd=j}tL?XY ziN6sMc(e-u9=or{uPgpOCXqZ(_>tYLf>?)WW=!VRgm(yhIt&!WAF!b~b>cR03AaZ4n)ZqV;ot|yAwDTJkIv*7KkPXk=>;?wILT?Es~w3 zMYOjJ#FK{LXiJ0Tq_dCZdLB9MgyEq&KDn36CY;hl601-z2A(vQ`0D>M7yViJkftl{ zy>zdT@Z_kbCS=Cotfw5*k@)MVJr=&!v#Ev!3CHbs_+-TP#E+*v+)&|*T#?ky@M;%d zcsCy5jSu=VImhUw`i5_C5`xEEe9AL$oa&E^89_7p76o*&twB*Gzxo zbsRXq5CX1FAy(O5)YZ-DUt32i$c@F)A1?eWVrX}d%w`)LKYo|M?}`HSZ~@|PwddVM zzn?<@(;efPWj=i*Q*?^*_oKNP{4H-Tmw3_NM>ISd1B@lt$NEn6{QZ%4UVzPfXTiG7 z*?(QbB>!9Bm9qjcXCxFxYIFSq`|Isd@n|MSAe0_-#bz!m#X z+3@GS{(3<^04{1z$NtoRV)CaZ