From 9a18aac080f9d519016ec288c61cd7d759fc7491 Mon Sep 17 00:00:00 2001 From: Simon Meister Date: Fri, 10 Nov 2017 16:22:37 +0100 Subject: [PATCH] WIP --- approach.tex | 48 +++++---- background.tex | 248 ++++++++++++++++++++++++++++++++++------------- conclusion.tex | 7 +- experiments.tex | 2 +- figures/rpn.png | Bin 0 -> 170510 bytes introduction.tex | 5 +- 6 files changed, 218 insertions(+), 92 deletions(-) create mode 100644 figures/rpn.png diff --git a/approach.tex b/approach.tex index a5f0852..a63a54a 100644 --- a/approach.tex +++ b/approach.tex @@ -15,7 +15,7 @@ region proposal. Table \ref{table:motionrcnn_resnet} shows the modified network. \centering \begin{tabular}{llr} \toprule -\textbf{Layer ID} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ +\textbf{Output} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ \midrule\midrule & input image & H $\times$ W $\times$ C \\ \midrule @@ -33,7 +33,8 @@ M$_1$ & $\begin{bmatrix}\textrm{fully connected}, 1024\end{bmatrix}$ $\times$ 2 $R_t^{cam}$& From M$_1$: fully connected, 3 & 1 $\times$ 3 \\ $t_t^{cam}$& From M$_1$: fully connected, 3 & 1 $\times$ 3 \\ -$o_t^{cam}$& From M$_1$: fully connected, 2 & 1 $\times$ 2 \\ +& From M$_1$: fully connected, 2 & 1 $\times$ 2 \\ +$o_t^{cam}$& softmax, 2 & 1 $\times$ 2 \\ \midrule \multicolumn{3}{c}{\textbf{RoI Head \& RoI Head: Masks} (Table \ref{table:maskrcnn_resnet})}\\ \midrule @@ -43,7 +44,8 @@ M$_2$ & From ave: $\begin{bmatrix}\textrm{fully connected}, 1024\end{bmatrix}$ $ $\forall k: R_t^k$ & From M$_2$: fully connected, 3 & N$_{RPN}$ $\times$ 3 \\ $\forall k: t_t^k$ & From M$_2$: fully connected, 3 & N$_{RPN}$ $\times$ 3 \\ $\forall k: p_t^k$ & From M$_2$: fully connected, 3 & N$_{RPN}$ $\times$ 3 \\ -$\forall k: o_t^k$ & From M$_2$: fully connected, 2 & N$_{RPN}$ $\times$ 2 \\ +& From M$_2$: fully connected, 2 & N$_{RPN}$ $\times$ 2 \\ +$\forall k: o_t^k$ & softmax, 2 & N$_{RPN}$ $\times$ 2 \\ \bottomrule \end{tabular} @@ -61,7 +63,7 @@ ResNet-50 architecture (Table \ref{table:maskrcnn_resnet}). \centering \begin{tabular}{llr} \toprule -\textbf{Layer ID} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ +\textbf{Output} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ \midrule\midrule & input image & H $\times$ W $\times$ C \\ \midrule @@ -77,7 +79,8 @@ C$_5$ & ResNet-50 (Table \ref{table:resnet}) & $\tfrac{1}{32}$ H $\times$ $\tfra M$_1$ & $\begin{bmatrix}\textrm{fully connected}, 1024\end{bmatrix}$ $\times$ 2 & 1 $\times$ 1024 \\ $R_t^{cam}$& From M$_1$: fully connected, 3 & 1 $\times$ 3 \\ $t_t^{cam}$& From M$_1$: fully connected, 3 & 1 $\times$ 3 \\ -$o_t^{cam}$& From M$_1$: fully connected, 2 & 1 $\times$ 2 \\ +& From M$_1$: fully connected, 2 & 1 $\times$ 2 \\ +$o_t^{cam}$& softmax, 2 & 1 $\times$ 2 \\ \midrule \multicolumn{3}{c}{\textbf{RoI Head \& RoI Head: Masks} (Table \ref{table:maskrcnn_resnet_fpn})} \\ \midrule @@ -87,7 +90,8 @@ M$_2$ & From F$_1$: $\begin{bmatrix}\textrm{fully connected}, 1024\end{bmatrix}$ $\forall k: R_t^k$ & From M$_2$: fully connected, 3 & N$_{RPN}$ $\times$ 3 \\ $\forall k: t_t^k$ & From M$_2$: fully connected, 3 & N$_{RPN}$ $\times$ 3 \\ $\forall k: p_t^k$ & From M$_2$: fully connected, 3 & N$_{RPN}$ $\times$ 3 \\ -$\forall k: o_t^k$ & From M$_2$: fully connected, 2 & N$_{RPN}$ $\times$ 2 \\ +& From M$_2$: fully connected, 2 & N$_{RPN}$ $\times$ 2 \\ +$\forall k: o_t^k$ & softmax, 2 & N$_{RPN}$ $\times$ 2 \\ \bottomrule \end{tabular} @@ -108,8 +112,8 @@ Like Faster R-CNN and Mask R-CNN, we use a ResNet \cite{ResNet} variant as backb Inspired by FlowNetS \cite{FlowNet}, we make one modification to the ResNet backbone to enable image matching, laying the foundation for our motion estimation. Instead of taking a single image as input to the backbone, we depth-concatenate two temporally consecutive frames $I_t$ and $I_{t+1}$, yielding a input image map with six channels. -Alternatively, we also experiment with concatenating the camera space XYZ coordinates for each frame -into the input as well. +Alternatively, we also experiment with concatenating the camera space XYZ coordinates for each frame, +XYZ$_t$ and XYZ$_{t+1}$, into the input as well. We do not introduce a separate network for computing region proposals and use our modified backbone network as both first stage RPN and second stage feature extractor for region cropping. Technically, our feature encoder network will have to learn a motion representation similar to @@ -172,7 +176,9 @@ and that objects rotate at most 90 degrees in either direction along any axis, which is in general a safe assumption for image sequences from videos. All predictions are made in camera space, and translation and pivot predictions are in meters. We additionally predict softmax scores $o_t^k$ for classifying the objects into -still and moving objects. +still and moving objects. As a postprocessing, for any object instance $k$ with predicted moving flag $o_t^k = 0$, +we set $\sin(\alpha) = \sin(\beta) = \sin(\gamma) = 0$ and $t_t^k = (0,0,0)^T$, +and thus predict an identity motion. \paragraph{Camera motion prediction} In addition to the object transformations, we optionally predict the camera motion $\{R_t^{cam}, t_t^{cam}\}\in \mathbf{SE}(3)$ @@ -180,7 +186,7 @@ between the two frames $I_t$ and $I_{t+1}$. For this, we flatten the bottleneck output of the backbone and pass it through a fully connected layer. We again represent $R_t^{cam}$ using a Euler angle representation and predict $\sin(\alpha)$, $\sin(\beta)$, $\sin(\gamma)$ and $t_t^{cam}$ in the same way as for the individual objects. -Again, we predict a softmax score $o_t^{cam}$ for classifying differentiating between +Again, we predict a softmax score $o_t^{cam}$ for differentiating between a still and moving camera. \subsection{Supervision} @@ -190,29 +196,28 @@ a still and moving camera. The most straightforward way to supervise the object motions is by using ground truth motions computed from ground truth object poses, which is in general only practical when training on synthetic datasets. -Given the $k$-th positive RoI, let $i_k$ be the index of the matched ground truth example with class $c_k$, -let $R^{k,c_k}, t^{k,c_k}, p^{k,c_k}$ be the predicted motion for class $c_k$ -and $R^{gt,i_k}, t^{gt,i_k}, p^{gt,i_k}$ the ground truth motion for the example $i_k$. +Given the $k$-th foreground RoI, let $i_k$ be the index of the matched ground truth example with class $c_k$, +let $R^{k,c_k}, t^{k,c_k}, p^{k,c_k}, o^{k,c_k}$ be the predicted motion for class $c_k$ +and $R^{gt,i_k}, t^{gt,i_k}, p^{gt,i_k}, o^{gt,i_k}$ the ground truth motion for the example $i_k$. Note that we dropped the subscript $t$ to increase readability. Similar to the camera pose regression loss in \cite{PoseNet2}, we use a variant of the $\ell_1$-loss to penalize the differences between ground truth and predicted rotation, translation (and pivot, in our case). We found that the smooth $\ell_1$-loss performs better in our case than the standard $\ell_1$-loss. -For each RoI, we compute the total motion loss $L_{motion}^k$ from -the individual loss terms as, +We then compute the RoI motion loss as \begin{equation} -L_{motion}^k = l_{p}^k + (l_{R}^k + l_{t}^k) \cdot o^{gt,i_k} + l_o^k, +L_{motion} = \frac{1}{N_{RoI}^{fg}} \sum_k^{N_{RoI}} l_{p}^k + (l_{R}^k + l_{t}^k) \cdot o^{gt,i_k} + l_o^k, \end{equation} where \begin{equation} -l_{R}^k = \ell_1^* (R^{gt,i_k} - R^{k,c_k}), +l_{R}^k = \ell_{reg} (R^{gt,i_k} - R^{k,c_k}), \end{equation} \begin{equation} -l_{t}^k = \ell_1^* (t^{gt,i_k} - t^{k,c_k}), +l_{t}^k = \ell_{reg} (t^{gt,i_k} - t^{k,c_k}), \end{equation} \begin{equation} -l_{p}^k = \ell_1^* (p^{gt,i_k} - p^{k,c_k}). +l_{p}^k = \ell_{reg} (p^{gt,i_k} - p^{k,c_k}). \end{equation} are the smooth $\ell_1$-loss terms for the predicted rotation, translation and pivot, respectively and @@ -228,6 +233,11 @@ numerically more difficult to optimize than performing classification between moving and non-moving objects and discarding the regression for the non-moving ones. +Now, our modified RoI loss is +\begin{equation} +L_{RoI} = L_{cls} + L_{box} + L_{mask} + L_{motion}. +\end{equation} + \paragraph{Camera motion supervision} We supervise the camera motion with ground truth analogously to the object motions, with the only difference being that we only have diff --git a/background.tex b/background.tex index 087d3f1..db54de5 100644 --- a/background.tex +++ b/background.tex @@ -1,22 +1,6 @@ In this section, we will give a more detailed description of previous works we directly build on and other prerequisites. -\subsection{Basic definitions} -For regression, we define the smooth $\ell_1$-loss as -\begin{equation} -\ell_1^*(x) = -\begin{cases} -0.5x^2 &\text{if |x| < 1} \\ -|x| - 0.5 &\text{otherwise,} -\end{cases} -\end{equation} -which provides a certain robustness to outliers and will be used -frequently in the following chapters. -For classification we define the cross-entropy loss as -\begin{equation} -\ell_{cls} = -\end{equation} - \subsection{Optical flow and scene flow} Let $I_1,I_2 : P \to \mathbb{R}^3$ be two temporally consecutive frames in a sequence of images. @@ -48,7 +32,7 @@ performing upsampling of the compressed features and resulting in a encoder-deco The most popular deep networks of this kind for end-to-end optical flow prediction are variants of the FlowNet family \cite{FlowNet, FlowNet2}, which was recently extended to scene flow estimation \cite{SceneFlowDataset}. -Table \ref{} shows the classical FlowNetS architecture for optical fow prediction. +Table \ref{} shows the classical FlowNetS architecture for optical flow prediction. Note that the network itself is a rather generic autoencoder and is specialized for optical flow only through being trained with supervision from dense optical flow ground truth. Potentially, the same network could also be used for semantic segmentation if @@ -80,7 +64,7 @@ shows the fundamental building block of ResNet-50. %\centering \begin{longtable}{llr} \toprule -\textbf{Layer ID} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ +\textbf{Output} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ \midrule\midrule & input image & H $\times$ W $\times$ C \\ \midrule @@ -144,6 +128,7 @@ is only applied to the first block, but not to repeated blocks. \caption{ ResNet \cite{ResNet} \enquote{bottleneck} convolutional block introduced to reduce computational complexity in deeper network variants, shown here with 256 input and output channels. +Figure from \cite{ResNet}. } \label{figure:bottleneck} \end{figure} @@ -164,14 +149,16 @@ The original R-CNN involves computing one forward pass of the CNN for each of th which is costly, as there is generally a large number of proposals. Fast R-CNN \cite{FastRCNN} significantly reduces computation by performing only a single forward pass with the whole image as input to the CNN (compared to the sequential input of crops in the case of R-CNN). -Then, fixed size crops are taken from the compressed feature map of the image, +Then, fixed size (H $\times$ W) feature maps are extracted from the compressed feature map of the image, each corresponding to one of the proposal bounding boxes. -The crops are collected into a batch and passed into a small Fast R-CNN +The extracted per-RoI feature maps are collected into a batch and passed into a small Fast R-CNN \emph{head} network, which performs classification and prediction of refined boxes for all regions in one forward pass. -This technique is called \emph{RoI pooling}. % TODO explain how RoI pooling converts full image box coords to crop ranges -\todo{more details and figure} +The extraction technique is called \emph{RoI pooling}. In RoI pooling, the RoI bounding box window over the full image features +is divided into a H $\times$ W grid of cells. For each cell, the values of the underlying +full image feature map are max-pooled to yield the output value at this cell. Thus, given region proposals, the per-region computation is reduced to a single pass through the complete network, -speeding up the system by orders of magnitude. % TODO verify that +speeding up the system by two orders of magnitude at inference time and one order of magnitude +at training time. \paragraph{Faster R-CNN} After streamlining the CNN components, Fast R-CNN is limited by the speed of the region proposal @@ -185,17 +172,21 @@ In the \emph{first stage}, one forward pass is performed on the \emph{backbone} which is a deep feature encoder CNN with the original image as input. Next, the \emph{backbone} output features are passed into a small, fully convolutional \emph{Region Proposal Network (RPN)} head, which predicts objectness scores and regresses bounding boxes at each of its output positions. -At any position, bounding boxes are predicted as offsets relative to a fixed set of \emph{anchors} with different -aspect ratios. -\todo{more details and figure} -% TODO more about striding & computing the anchors? -For each anchor at a given position, the objectness score tells us how likely this anchors is to correspond to a detection. -The region proposals can then be obtained as the N highest scoring anchor boxes. +At any of the $h \times w$ output positions of the RPN head, +$N_a$ bounding boxes with their objectness scores are predicted as offsets relative to a fixed set of $N_a$ \emph{anchors} with different +aspect ratios and scales. Thus, there are $N_a \times h \times w$ reference anchors in total. +In Faster R-CNN, $N_a = 9$, with 3 scales corresponding +to anchor boxes of areas of $\{128^2, 256^2, 512^2\}$ pixels and 3 aspect ratios, +$\{1:2, 1:1, 2:1\}$. For the ResNet Faster R-CNN backbone, we generally have a stride of 16 +with respect to the input image at the RPN output (Table \ref{table:maskrcnn_resnet}). -The \emph{second stage} corresponds to the original Fast R-CNN head network, performing classification -and bounding box refinement for each region proposal. % TODO verify that it isn't modified -As in Fast R-CNN, RoI pooling is used to crop one fixed size feature map for each of the region proposals. +For each RPN prediction at a given position, the objectness score tells us how likely it is to correspond to a detection. +The region proposals can then be obtained as the N highest scoring RPN predictions. +Then, the \emph{second stage} corresponds to the original Fast R-CNN head network, performing classification +and bounding box refinement for each of the region proposals, which are now obtained +from the RPN instead of being pre-computed by some external algorithm. +As in Fast R-CNN, RoI pooling is used to extract one fixed size feature map for each of the region proposals. \paragraph{Mask R-CNN} Faster R-CNN and the earlier systems detect and classify objects at bounding box granularity. @@ -205,16 +196,23 @@ to that object. This problem is called \emph{instance segmentation}. Mask R-CNN \cite{MaskRCNN} extends the Faster R-CNN system to instance segmentation by predicting fixed resolution instance masks within the bounding boxes of each detected object. This is done by simply extending the Faster R-CNN head with multiple convolutions, which -compute a pixel-precise mask for each instance. +compute a pixel-precise binary mask for each instance. The basic Mask R-CNN ResNet-50 architecture is shown in Table \ref{table:maskrcnn_resnet}. -\todo{RoI Align} +Note that the per-class masks logits are put through a sigmoid layer, and thus there is no +comptetition between classes for the mask prediction branch. + +One important technical aspect of Mask R-CNN is the replacement of RoI pooling with +bilinear sampling for extracting the RoI features, which is much more precise. +%In RoI pooling, at the borders, the bins for max-pooling are not aligned with the actual pixel +%boundary of the bounding box. + { %\begin{table}[t] %\centering \begin{longtable}{llr} \toprule -\textbf{Layer ID} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ +\textbf{Output} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ \midrule\midrule & input image & H $\times$ W $\times$ C \\ \midrule @@ -226,27 +224,30 @@ C$_4$ & ResNet-50 \{up to C$_4$\} (Table \ref{table:resnet}) & $\tfrac{1}{16}$ & 1 $\times$ 1 conv, 6 & $\tfrac{1}{16}$ H $\times$ $\tfrac{1}{16}$ W $\times$ 6 \\ & flatten & A $\times$ 6 \\ & decode bounding boxes (Eq. \ref{eq:pred_bounding_box}) & A $\times$ 6 \\ -ROI$_{\mathrm{RPN}}$ & sample bounding boxes \& scores (Listing \ref{}) & N$_{RPN}$ $\times$ 6 \\ +ROI$_{\mathrm{RPN}}$ & sample bounding boxes \& scores & N$_{RoI}$ $\times$ 6 \\ \midrule \multicolumn{3}{c}{\textbf{RoI Head}}\\ \midrule -& From C$_4$ with ROI$_{\mathrm{RPN}}$: RoI pooling (\ref{}) & N$_{RPN}$ $\times$ 7 $\times$ 7 $\times$ 1024 \\ -R$_1$& ResNet-50 \{C$_5$ without stride\} (Table \ref{table:resnet}) & N$_{RPN}$ $\times$ 7 $\times$ 7 $\times$ 2048 \\ +& From C$_4$ with ROI$_{\mathrm{RPN}}$: RoI extraction & N$_{RoI}$ $\times$ 7 $\times$ 7 $\times$ 1024 \\ +R$_1$& ResNet-50 \{C$_5$ without stride\} (Table \ref{table:resnet}) & N$_{RoI}$ $\times$ 7 $\times$ 7 $\times$ 2048 \\ ave & average pool & N$_{RPN}$ $\times$ 2048 \\ boxes& From ave: fully connected, 4 & N$_{RPN}$ $\times$ 4 \\ -logits& From ave: fully connected, N$_{cls}$ & N$_{RPN}$ $\times$ N$_{cls}$ \\ +& From ave: fully connected, N$_{cls}$ & N$_{RoI}$ $\times$ N$_{cls}$ \\ +classes& softmax, N$_{cls}$ & N$_{RoI}$ $\times$ N$_{cls}$ \\ \midrule \multicolumn{3}{c}{\textbf{RoI Head: Masks}}\\ \midrule -& From R$_1$: 2 $\times$ 2 deconv, 256, stride 2 & N$_{RPN}$ $\times$ 14 $\times$ 14 $\times$ 256 \\ -masks & 1 $\times$ 1 conv, N$_{cls}$ & N$_{RPN}$ $\times$ 14 $\times$ 14 $\times$ N$_{cls}$ \\ +& From R$_1$: 2 $\times$ 2 deconv, 256, stride 2 & N$_{RoI}$ $\times$ 14 $\times$ 14 $\times$ 256 \\ +& 1 $\times$ 1 conv, N$_{cls}$ & N$_{RoI}$ $\times$ 14 $\times$ 14 $\times$ N$_{cls}$ \\ +masks & sigmoid, N$_{cls}$ & N$_{RoI}$ $\times$ 28 $\times$ 28 $\times$ N$_{cls}$ \\ \bottomrule \caption { Mask R-CNN \cite{MaskRCNN} ResNet-50 \cite{ResNet} architecture. Note that this is equivalent to the Faster R-CNN architecture if the mask -head is left out. +head is left out. In Mask R-CNN, bilinear sampling is used for RoI extraction, +whereas Faster R-CNN used RoI pooling. } \label{table:maskrcnn_resnet} \end{longtable} @@ -264,17 +265,35 @@ information for this object. As a solution to this, the Feature Pyramid Network (FPN) \cite{FPN} enable features of an appropriate scale to be used, depending of the size of the bounding box. For this, a pyramid of feature maps is created on top of the ResNet \cite{ResNet} -encoder. \todo{figure and more details} -Now, during RoI pooling, -\todo{show formula}. +encoder by combining bilinear upsampled feature maps coming from the bottleneck +with lateral skip connections from the encoder. The Mask R-CNN ResNet-50-FPN variant is shown in Table \ref{table:maskrcnn_resnet_fpn}. +Instead of a single RPN head with anchors at 3 scales and 3 aspect ratios, +the FPN variant has one RPN head after each of the pyramid levels P$_2$ ... P$_6$. +At each output position of the resulting RPN pyramid, bounding boxes are predicted +with respect to 3 anchor aspect ratios $\{1:2, 1:1, 2:1\}$ and a single scale. +For P$_2$, P$_3$, P$_4$, P$_5$, P$_6$, +the scale corresponds to anchor bounding boxes of areas $32^2, 64^2, 128^2, 256^2, 512^2$, +respectively. +Note that there is no need for multiple anchor scales per anchor position anymore, +as the RPN heads themselves correspond to multiple scales. +Now, in the RPN, higher resolution feature maps can be used for regressing smaller +bounding boxes. For example, boxes of area close to $32^2$ are predicted using P$_2$, +which has a stride of $4$ with respect to the input image. +Most importantly, the RoI features can now be extracted at the pyramid level $P_j$ appropriate for a +RoI bounding box with size $h \times w$, +\begin{equation} +j = \log_2(\sqrt{w \cdot h} / 224). %TODO complete +\label{eq:level_assignment} +\end{equation} + { %\begin{table}[t] %\centering \begin{longtable}{llr} \toprule -\textbf{Layer ID} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ +\textbf{Output} & \textbf{Layer Operations} & \textbf{Output Dimensions} \\ \midrule\midrule & input image & H $\times$ W $\times$ C \\ \midrule @@ -297,21 +316,23 @@ RPN$_i$& flatten & A$_i$ $\times$ 6 \\ \midrule & From \{RPN$_2$ ... RPN$_6$\}: concatenate & A $\times$ 6 \\ & decode bounding boxes (Eq. \ref{eq:pred_bounding_box}) & A $\times$ 6 \\ -ROI$_{\mathrm{RPN}}$ & sample bounding boxes \& scores (Listing \ref{}) & N$_{RPN}$ $\times$ 6 \\ +ROI$_{\mathrm{RPN}}$ & sample bounding boxes \& scores & N$_{RoI}$ $\times$ 6 \\ \midrule \multicolumn{3}{c}{\textbf{RoI Head}}\\ \midrule -R$_2$ & From \{P$_2$ ... P$_6$\} with ROI$_{\mathrm{RPN}}$: FPN RoI crop & N$_{RPN}$ $\times$ 14 $\times$ 14 $\times$ 256 \\ -& 2 $\times$ 2 max pool & N$_{RPN}$ $\times$ 7 $\times$ 7 $\times$ 256 \\ -F$_1$ & $\begin{bmatrix}\textrm{fully connected}, 1024\end{bmatrix}$ $\times$ 2 & N$_{RPN}$ $\times$ 1024 \\ -boxes& From F$_1$: fully connected, 4 & N$_{RPN}$ $\times$ 4 \\ -logits& From F$_1$: fully connected, N$_{cls}$ & N$_{RPN}$ $\times$ N$_{cls}$ \\ +R$_2$ & From \{P$_2$ ... P$_6$\} with ROI$_{\mathrm{RPN}}$: RoI extraction (Eq. \ref{eq:level_assignment}) & N$_{RoI}$ $\times$ 14 $\times$ 14 $\times$ 256 \\ +& 2 $\times$ 2 max pool & N$_{RoI}$ $\times$ 7 $\times$ 7 $\times$ 256 \\ +F$_1$ & $\begin{bmatrix}\textrm{fully connected}, 1024\end{bmatrix}$ $\times$ 2 & N$_{RoI}$ $\times$ 1024 \\ +boxes& From F$_1$: fully connected, 4 & N$_{RoI}$ $\times$ 4 \\ +& From F$_1$: fully connected, N$_{cls}$ & N$_{RoI}$ $\times$ N$_{cls}$ \\ +classes& softmax, N$_{cls}$ & N$_{RoI}$ $\times$ N$_{cls}$ \\ \midrule \multicolumn{3}{c}{\textbf{RoI Head: Masks}}\\ \midrule -& From R$_2$: $\begin{bmatrix}\textrm{3 $\times$ 3 conv} \end{bmatrix}$ $\times$ 4, 256 & N$_{RPN}$ $\times$ 14 $\times$ 14 $\times$ 256 \\ -& 2 $\times$ 2 deconv, 256, stride 2 & N$_{RPN}$ $\times$ 28 $\times$ 28 $\times$ 256 \\ -masks & 1 $\times$ 1 conv, N$_{cls}$ & N$_{RPN}$ $\times$ 28 $\times$ 28 $\times$ N$_{cls}$ \\ +& From R$_2$: $\begin{bmatrix}\textrm{3 $\times$ 3 conv} \end{bmatrix}$ $\times$ 4, 256 & N$_{RoI}$ $\times$ 14 $\times$ 14 $\times$ 256 \\ +& 2 $\times$ 2 deconv, 256, stride 2 & N$_{RoI}$ $\times$ 28 $\times$ 28 $\times$ 256 \\ +& 1 $\times$ 1 conv, N$_{cls}$ & N$_{RoI}$ $\times$ 28 $\times$ 28 $\times$ N$_{cls}$ \\ +masks & sigmoid, N$_{cls}$ & N$_{RoI}$ $\times$ 28 $\times$ 28 $\times$ N$_{cls}$ \\ \bottomrule \caption { @@ -331,11 +352,26 @@ block (see Figure \ref{figure:fpn_block}). FPN block from \cite{FPN}. Lower resolution features coming from the bottleneck are bilinearly upsampled and added with higher resolution skip connections from the encoder. +Figure from \cite{FPN}. } \label{figure:fpn_block} \end{figure} \subsection{Training Mask R-CNN} +\paragraph{Loss definitions} +For regression, we define the smooth $\ell_1$ regression loss as +\begin{equation} +\ell_{reg}(x) = +\begin{cases} +0.5x^2 &\text{if |x| < 1} \\ +|x| - 0.5 &\text{otherwise,} +\end{cases} +\end{equation} +which provides a certain robustness to outliers and will be used +frequently in the following chapters. For vector or tuple arguments, the sum of the componentwise scalar +losses is computed. +For classification we define $\ell_{cls}$ as the cross-entropy classification loss. + \label{ssec:rcnn_techn} \paragraph{Bounding box regression} All bounding boxes predicted by the RoI head or RPN are estimated as offsets @@ -345,7 +381,7 @@ predicted relative to the RPN output bounding boxes. Let $(x, y, w, h)$ be the top left coordinates, height and width of the bounding box to be predicted. Likewise, let $(x^*, y^*, w^*, h^*)$ be the ground truth bounding box and let $(x_r, y_r, w_r, h_r)$ be the reference bounding box. -We then define the ground truth \emph{box encoding} $b_e^*$ as +The ground truth \emph{box encoding} $b_e^*$ is then defined as \begin{equation} b_e^* = (b_x^*, b_y^*, b_w^*, b_h^*), \end{equation} @@ -354,18 +390,18 @@ where b_x^* = \frac{x^* - x_r}{w_r}, \end{equation*} \begin{equation*} -b_y^* = \frac{y^* - y_r}{h_r} +b_y^* = \frac{y^* - y_r}{h_r}, \end{equation*} \begin{equation*} -b_w^* = \log \left( \frac{w^*}{w_r} \right) +b_w^* = \log \left( \frac{w^*}{w_r} \right), \end{equation*} \begin{equation*} b_h^* = \log \left( \frac{h^*}{h_r} \right), \end{equation*} -which represents the regression target for the bounding box refinement +which represents the regression target for the bounding box outputs of the network. -In the same way, we define the predicted box encoding $b_e$ as +Thus, for each bounding box prediction, the network predicts the box encoding $b_e$, \begin{equation} b_e = (b_x, b_y, b_w, b_h), \end{equation} @@ -374,17 +410,17 @@ where b_x = \frac{x - x_r}{w_r}, \end{equation*} \begin{equation*} -b_y = \frac{y - y_r}{h_r} +b_y = \frac{y - y_r}{h_r}, \end{equation*} \begin{equation*} -b_w = \log \left( \frac{w}{w_r} \right) +b_w = \log \left( \frac{w}{w_r} \right), \end{equation*} \begin{equation*} b_h = \log \left( \frac{h}{h_r} \right). \end{equation*} At test time, to get from a predicted box encoding $b_e$ to the predicted bounding box $b$, -we invert the definitions above, +the definitions above can be inverted, \begin{equation} b = (x, y, w, h), \label{eq:pred_bounding_box} @@ -402,12 +438,88 @@ w = \exp(b_w) \cdot w_r, \begin{equation*} h = \exp(b_h) \cdot h_r, \end{equation*} -and thus obtain the bounding box as the reference bounding box adjusted by +and thus the bounding box is obtained as the reference bounding box adjusted by the predicted relative offsets and scales. - \paragraph{Supervision of the RPN} -\todo{TODO} +A positive RPN proposal is defined as one with a IoU of at least $0.7$ with +a ground truth bounding box. For training the RPN, $N_{RPN} = 256$ positive and negative +examples are randomly sampled from the set of all RPN proposals, +with at most $50\%$ positive examples (if there are less positive examples, +more negative examples are used instead). +For examples selected in this way, a regression loss is computed between +predicted and ground truth bounding box encoding, and a classification loss +is computed for the predicted objectness. +Specifically, let $s_i^* = 1$ if proposal $i$ is positive and $s_i^* = 0$ if +it is negative, let $s_i$ be the predicted objectness score and $b_i$, $b_i^*$ the +predicted and ground truth bounding box encodings. +Then, the RPN loss is computed as +\begin{equation} +L_{RPN} = L_{obj} + L_{box}^{RPN}, +\end{equation} +where +\begin{equation} +L_{obj} = \frac{1}{N_{RPN}} \sum_{i=1}^{N_{RPN}} \ell_{cls}(s_i, s_i^*), +\end{equation} +\begin{equation} +L_{box}^{RPN} = \frac{1}{N_{RPN}^{pos}} \sum_{i=1}^{N_{RPN}} s_i^* \cdot \ell_{reg}(b_i^* - b_i), +\end{equation} +and +\begin{equation} +N_{RPN}^{pos} = \sum_{i=1}^{N_{pos}} s_i^* +\end{equation} +is the number of positive examples. Note that the bounding box loss is only +active for positive examples, and that the classification loss is computed +between the classes $\{\textrm{object},\textrm{non-object}\}$. -\paragraph{Supervision of the RoI head} -\todo{TODO} +\paragraph{Supervision of the Mask R-CNN RoI head} +For selecting RoIs to train the RoI head network, a foreground example +is defined as one with a IoU of at least $0.5$ with +a ground truth bounding box, and a background example is defined as +one with a maximum IoU in $[0.1, 0.5)$. +A total of 64 (without FPN) or 512 (with FPN) RoIs are sampled, with +at most $25\%$ foreground examples. +Now, let $c_i^*$ be the ground truth object class, where $c_i = 0$ +for background examples and $c_i \in \{1, ..., N_{cls}\}$ for foreground examples, +and let $c_i$ be the class prediction. +Now, for any foreground RoI, let $b_i^*$ be the ground truth bounding box encoding and $b_i$ +the predicted refined box encoding for class $c_i^*$. +Additionally, for any foreground RoI, let $m_i$ be the predicted $m \times m$ mask for class $c_i^*$ +and $m_i^*$ the $m \times m$ mask target with values in $\{0,1\}$, where the mask target is cropped and resized from +the binary ground truth mask using the RPN proposal bounding box. +Then, the ROI loss is computed as +\begin{equation} +L_{RoI} = L_{cls} + L_{box} + L_{mask} +\end{equation} +where +\begin{equation} +L_{cls} = \frac{1}{N_{RoI}} \sum_{i=1}^{N_{RoI}} \ell_{cls}(c_i, c_i^*), +\end{equation} +is the average cross-entropy classification loss, +\begin{equation} +L_{box} = \frac{1}{N_{RoI}^{fg}} \sum_{i=1}^{N_{RoI}} [c_i^* \geq 1] \cdot \ell_{reg}(b_i^* - b_i) +\end{equation} +is the average smooth-$\ell_1$ bounding box regression loss, +\begin{equation} +L_{mask} = \frac{1}{N_{RoI}^{fg}} \sum_{i=1}^{N_{RoI}} [c_i^* \geq 1] \cdot \ell_{cls}(m_i,m_i^*) +\end{equation} +is the average binary cross-entropy mask loss, +\begin{equation} +N_{RoI}^{fg} = \sum_{i=1}^{N_{RoI}} [c_i^* \geq 1] +\end{equation} +is the number of foreground examples, and +\begin{equation} +[c_i^* \geq 1] = +\begin{cases} +1 &\text{$c_i^* \geq 1$} \\ +0 &\text{otherwise} +\end{cases} +\end{equation} +is the Iverson bracket indicator function. Thus, the bounding box and mask +losses are only enabled for the foreground RoIs. Note that the bounding box and mask predictions +for all classes other than $c_i^*$ are not penalized. + +\paragraph{Test-time operation} +During inference, the 300 (without FPN) or 1000 (with FPN) highest scoring region proposals +are selected and passed through the RoI head. After this, non-maximum supression +is applied to predicted foreground RoIs. diff --git a/conclusion.tex b/conclusion.tex index 4087cb2..b2cc6e4 100644 --- a/conclusion.tex +++ b/conclusion.tex @@ -31,7 +31,10 @@ Although single-frame monocular depth prediction with deep networks was already to some level of success, our two-frame input should allow the network to make use of epipolar geometry for making a more reliable depth estimate, at least when the camera -is moving. +is moving. We could also extend our method to stereo input data easily by concatenating +all of the frames into the input image, which +would however require using a different dataset for training, as Virtual KITTI does not +provide stereo images. { \begin{table}[h] @@ -48,7 +51,7 @@ C$_5$ & ResNet-50 (Table \ref{table:resnet}) & $\tfrac{1}{32}$ H $\times$ $\tfra \midrule \multicolumn{3}{c}{\textbf{Depth Network}}\\ \midrule -& From P$_2$: 3 $\times$ 3 conv, 1024 & $\tfrac{1}{4}$ H $\times$ $\tfrac{1}{4}$ W $\times$ 256 \\ +& From P$_2$: 3 $\times$ 3 conv, 1024 & $\tfrac{1}{4}$ H $\times$ $\tfrac{1}{4}$ W $\times$ 1024 \\ & 1 $\times$ 1 conv, 1 & $\tfrac{1}{4}$ H $\times$ $\tfrac{1}{4}$ W $\times$ 1 \\ & $\times$ 2 bilinear upsample & H $\times$ W $\times$ 1 \\ \midrule diff --git a/experiments.tex b/experiments.tex index a79a979..5e2557e 100644 --- a/experiments.tex +++ b/experiments.tex @@ -26,7 +26,7 @@ Each sequence is rendered with varying lighting and weather conditions and from different viewing angles, resulting in a total of 10 variants per sequence. In addition to the RGB frames, a variety of ground truth is supplied. For each frame, we are given a dense depth and optical flow map and the camera -extrinsics matrix. +extrinsics matrix. There are two annotated object classes, cars, and vans. For all cars and vans in the each frame, we are given 2D and 3D object bounding boxes, instance masks, 3D poses, and various other labels. diff --git a/figures/rpn.png b/figures/rpn.png new file mode 100644 index 0000000000000000000000000000000000000000..10e5dd4ae5c16684f75b28f5580cd8dad290936f GIT binary patch literal 170510 zcmcG$cRbdA`#!F+ql73jLlh}9d&|fsBPk*?l~IvVgjBN1mOVqZ(2`O1-Xvv{hLM@y zaowNa_wl_yfBpXYU61?mxV?3GUDxaRJfG)z9LIT_uV7v6v(%JKlte^C)aq)=`b0#e zZ-|KYY?1HAZ{kk|Yvcd+xGJg}l9Q7UkL!HJe;#&IF>y1ve%0-^rHc)bt@HJ3HUh3! zTx@KdUG1*B{oM0jmWYU*NL^XM@J_-)qK7&4#jUmJFRO{15eufY9K-$$2Qn7)i+^&> zM>=KbN;h>W@uh#ZFr{r`qD)cN;IPS}%^1z70wsu%^pee$i4r%7EI?Z zKte<$f6>a)+1X(zt03cl?qH2h&6tUSF zr`eH8*Ncz;=MR1b5&gCP|GxXK(a6*PHXPZ5YS&p(@%Ek*i(@(r!OHFeD%%;?FBzz) z)K^6zK)zN)#%-+izq<54Uyo(!$8g1yCsgXT2M-=3+jnH`*ra{3p`oF)*IHQ0$7BWn ziOHAw+4eW2egAFE_rHp0E59~O_@A#nt2}M54Ax6h_xgqq=Jo`MqqBL@(a{wZf7$5C!pL7A+w*^y;RA7rxB1S$?VT7dBBHI+^7iEa z_uq)1r zQz#po^qsFie&na88}@%AqP5>D`14me2(yz<4a4GjzqQZtM3b8zgE*&7weA08Rm z-Aj1=7${21&;Rl8z)|NfZ%q3Ydw0eC=ap|(HaGU(_m#z4+>voV?XGXTH2*u4O=`w( zEd2Pd=Sr8*>^lq*|M+KjzIIR0Y+Je{jvN_}_tFo}Q7B67)5mUzn1T(p^CDEzt|5wfPwfz73KFOzV2usyFeM$|#YwWeIXV_m% z6vA-M{>+W9AF3jVap@4Ri+lg?H|u#=XRIUE!1{XSxR$38QTE|bRvE?pJ5RcFdFiJn zC*S&PZ+!n0ir7_5(KR(S)zWJHOn9kTyi<*%4p&=V{V{CY{=$1>Ik<{?@ci&sm<-pzt5jP*U-=y8X8hnRZWz-9TgcFXl>-S=;9Zf-(C zLQ6}_t4r_ytb3`bsO;I}&nn|NGe6&-v^*x~v!$k^|L48ms@KJ4(etZC@Cs3t#D4v$>B3H(bZ+=>#{1j+R7^DbFR$y&xAS~TT)&ge_mWd!bnGo zlDN~rWqbQZ{)yV4{Yr|8R#sNnIaRrrSTjq@@o(R(u3Rw?Xf-r58=C8ViG^>SUOHWa zMLb_%E^)QxP)>-~+Aj`i>8;P7xqM?}DqJC-Lc0?(&t?2Oe&R%i#cmP)jpfs zqGe?K^lNx_e*VhJ?H@xb+antr8-s&`?wn$}6|YZT)OPnSVifExJ45vAy>v@+>1ef| ziAdPL+v3yq7d6R5v|q@2KiZdg{P=OVgEd%k`hy2E<~SF}-nV>5!MXwpf zkBFHlOOo@wn!WyGSUg;>7K|_4n_n+5t zIJP)^GT{9A^Ap^Qi;MavUYV6q(*0Q+ttKTU)fapg8M(B)O#kv(b9;M|h&?T{o|N0n zuVs5tQu0C(rvY+ezlDW`t*xy;fBqm6ut)Ch?njRvU7c#0oSdAUouxWh=X?a`Xe)_Z zRMd!zR$X%VjmOfhA4A&;H}8#ojuyOh{*{H+_6VDtk9qa2AC<1tbLWra$ojFrc;h?k z{`+gda(Wd5@tPbNLq{Rv{+Zq(6?^-f>gvt3v!MkSs{;=whf^_$d>$EjgDgfRe*EM~ zB%8L@)_UjZnVIaYtg|%AdU|^2&b740i|3RmC@8qNxFGTc6=;~4nEw6yhq~O@*m!$! zq=u=zsR?y#CtW@Ac6Y_OKr z50ur@r%&5$;=@ta-`xG{jBw=|ecU{GoR4oEueCL`g9{6TC1dZrt6e53FRE;=%o|Hd zNiBSu`|@`4lJ9me_AW(t?ayR0GOjsr_ULGkl?f7uDiRr@tli6 zXg!=Y!PksBdGXOuq?mn|_3}(7iViY1mZ-I@ji(vOB}(7|0*sNF`DJRVV($*_9Uu%L z^jQhD0ASk0#Kg|gacO;PVqyY0A~$?w#9T_Mx~1iumX_9(?dDE+?s;O(TF=j@)%#eb z?XFz$kEMycQE6^qU?3>?YCoH-+BFSc`hQ#N!h(W7XFAiC0}7&|82sBvc>o!o?J_sSz%HW!`;c?TRXl1Bx8V7px7M>6wSZLvVzZ?Su7W+o;& z`qPwZ_3fVx1*0Xlp}M?vxruVVtOf_Uxw$RfIgcJK%2ZTRLUmH}DK9KM<2Iw~>A4Xv zeqBgNNc)Af{k3aa9j|ZyWF5K+WQ!&0pj=&Dt*@^i$k$0YbH7a3s*b#L_f`As{CpNR zHXR@5?#z()?|p&#oSmI--CFH^X*}JYq+{F}q3Xda&AQ*U6`qDSM|VtnW@aW? zR7OUo$N@qA=_!j{Vy0F+Wx&VeWcEKDot>v;Wo0=yEGL_uyVa^jojenhF^;UvAZT&U zXR-ek&dI;{XV0Diy2i&dGcr~<^d8^f8tLz6bQBg5q0eVOcC6ysuo!D-kkQ@kjk3f< z7PFFz)wf7UNbC|nepI}B_pV<Dv8&Lu$L91yYA18(8fOMChJB zf6LW%bA9PpbJW-y5C0Vf)YF~(F%qIV;bf6IyF88Bh=_>3{0mpMr?D#o=65%i?H4`* zMr=FZzdw|U*M$n_nf4$)>RHtQpj&rt0o@=VyX<4u-fgbZukRn#x=5~}^tHl|#9PXB z>PdPW;0QpDVXJncFk^xDm$S-<2u!H zpgIf~MIuwqVW`yJ)by~K?O?HuU4X?~Pk?z`N>5odwv7lt zRU=k#8ZM{Kue{pIRLT@od*CHsnm-@4`ty@PwS(2SUk$L$AUUR=W)FDtMmjgSrKw4g z`c$R4v$b`4dbP2Uk%XjVYX0Ex@D+FB-8G;)4|l%(_#u#^W!Nog`SAe(03_m&T%Phl zqP4o9si|2@UwrfC&54|tm_xa`dV0TcimWbtZf`%k!-jYiRCl2o>>ecYWlhn2?Kayr z+nvRD@StA6+}zxiPftf*-%NY;>Q#LFg};lVIR>~hsap$&gsjPxUeG-QDYKH$`5=Cn z=$>kOmHWJ`*Bbk9*>Z^8j2JTwcHuFDfK|4Lh3`fk_Cdc*6yF~7fK^=2vN z3p?(JU;mPqmp6jS1grl*{rg)zDr6~eYUu+klFI(GAxzv#skVOXSQ1s5IzZiv z7cbWTlSi!pU67pfXq;XGe#=(6e*Gt~!(2yi??D5lw9n1WK>-0}*M_5xSKj%}In~=X z>9U_y`r+H!nwlDLIL2X~$yjGAE0^UNy`tb3+gn>cTWbm|l2THdk2D8f8Vfr0Re7#B zbiHI+d2(7<*b(S_fWgh(J&5YCB1>_5I+aFkpt7Bv-Q~;dW{&mFb_QndT zEo$lGM~@UYqf1LmffyfY(LPYfP3nyJx4qdP^$T@>C9JNl4(r(HCDCU~ORSfFV^luT zzir>F+v11^kX}!UpPyfJbpKDrIq`@WFPyItUfxJjRZ$Qk&QCD-B+!j_w+_z^)Sggm zJm9|TYkOD}0C14wXjR#O^dY3W)rG-yY8Hfk(Ur3|ZWI<5yOb>k*z26q6n} zX8qZDR66zy6VcPUj*c4uxm3IFczRkrZAZD`ks~G}OAC%*VPQEvM)K^-+q-ES%BWfl z0%oU1jJ35-a&al=Jq@nw>tnc(m3E)S+uOUdZ0E^g;ZwEjlG@tZ*;^n`G41>cG%=)0 z8!PiTn`7@-x@S_v&RxBFb$oohr6Nk1ATJbLJRPb21W3v9~eNM%DS_pPM^}*gou#datXn=GT0`;QsylORZ-* zQgEna0%K~&&&7#QY)%tU7U*Wimv4ZsNHi*%1XV)*Xy%(Y2}cA$6Fw$vXwjm4D=8@j zhL|U_ayIhb;{-8ji4~sj%U@bq(bCj>85OK7GqJJr;Yj%rqI(-3hEv=3KD{>o?Z+XX zod|g@p6A3wTOOxu5=?|^@ex{ylFxf2`S|!i6}Pwlj#-7&#~haNT*mq5y_Q^1P>>#H zev+~KIXyl7W%rrguV24DrL)J)&8_hBJ$3aRxs{H_$YY%@{G6tKo>D!u@kD(#G=c7kdpNXQI>+=Tou}D7XO_7to za^=dxIt_73W@fo%Z6K;yMtb_lJMZ|n63i^OZ{NZBSwIk9kIgJ)_UaeaO+hyidU(rLg00{gWEG5_fvQvQ!rCuhfnhJr|(mz zvr{)6IHvQ$$;KuiYT?UUX78i*D0zOY<@Viyt?wD2*6pzYUxr%4N&d06c8``x18d&! zgOI3(AW%^dj&ksPpKVXKr8oCoz!<9-0N-c5_)nU+bdB9etF5VKTvNQiyqD;!&V8Hj z87HVgccJd+Buc8#JV3;Gd2Q7N)7ex<<7z0h_*VkdJ9sc)h_u+j+UvvcBn18>4& z?hvvy`R-N=^Pe3MGDn5uk zWfx*3A_|ECKiJ&dq$BQ3Q6e)oHqN0tS|ndVA;M0%cW==wAM#y@Ah$~M(cA9J38g>V z29N79?q1n{HDi49@6Xnj7PU`C(9CdE)Q`SjPf_7~w^#Z>P`F~7!!GmKFUJHfUp;)} z5#Omjs;r@BY3knI{oA34)U;>Mo~c{AckkZECc7{<*VORlu~r^)k>3Fk8N(!5@0ffh zKF{Tu`?)sjUOrYJcrB1nPmf_~nD-icUS|4e9}q&C zpj_xDR~Hw~kWxexNFTZ1Om|j0k|y_Pc5dz?!*#vFo3=I5t#}Xd?p<*-Q$TS~pFXwQ zM@4@!`h-6ZBaj*?KW|D;2457>i)7Qfj+zFdhb<&T@>en-TP==MTI;@mHU$Q35GkmW$}u^kBvHBqxS4u*hAMn(wO<2fJ%rzrA>neaUOK=rc0BfuZ5@#){({#IZ^+ zs2)qmB#=Wti}3vzDnkH%JomxpM6Xdmsbjx(?mWO$jK>wIOA0g|9v*fnhQ{CKJ}#sk zKYywzi^es}MljybZ}+WBRkkNdH|F~#@qMnY$6I9@f_-|g|KWZsy82gTKer4q2}wZY zF&zhcdp&LtT_Em^j0^}aG?WC~7#6x0Y|NefqB?i!@ znumv{v$Ipt@<&}6QHG`et_zPfws*2lOhJcg(Y?e=UreEggV#Ym08Wh0A@@V2Vv@SW zFF}%b5|r$$QOKi5k9=KRsCp zv62wElb4%`h>K_K{&4Y~*I-M_6Z7j#%*Mb)@^%H`l)734HN_HCO1R`+%Lcdg1gXLqla%wp6<>Zz?l0d7AqQ zgAh4nA|%2Zz_TD`52LV3IBJpYKY(&u_V0{V-!`F8>d~gGV)z5y4lJaOwI64M<;W2z zJwPJ7!$M>M&vTCJvK5w=CJ5VVo0uH(;1x~hm6_S&n$Qn=SRsg@O+=~ZuRCvg zdU~LREZuqYKl+2W_cp%rd2!LZ;QHs!_dRA5{CqtH;0?cp%2*2OTji+88XBFLm>3&t z3Z`jfl#!gA{OZ+M+S&c6S~(m`8PNBiW7nQPUs;;gL6G30&_ypxkv=;F1f!{=gOq}{ z0~7?5Vqq~c=f1qdo5K)b`9wumaCD_vLz^dab8;Y;9%g3l-+D|WUpooS?8}!g#0tP* zKL<-7VymrK9@u*|5o>aN%1)|m@8IlPul3^BuVdKBZEPk{`B%VP0qA;vEQ6#Ibeh=M z^~q+&;%?wB8}Oehw^_f2sDktU)`D?KNjhSXJ8Yet_+9-0bFdLX_V)JR-ADGFH8#H8 zk*okxU-#}zQoAqEBqHja*E%N+Iocpko;)GlcjRFKU2c41(Nu)3t!-9TmM_%gBY)z= zt~pp*)^zR40yRLM6Bqy694$DZ`4Vurpa4Lz3yT)`R7Oy+9jqA}DWv&S_MB|etMqhh zz(6M_^L&MJaBFVvN#F%6%a#hT;1vglCZFT`i9(+Il=EG9=W+aN!)({3-tt?~8ov@t zirOO_-VTzU4Z6Cq6!awDk&lTYrt3!C!vhmRU%y^P5^c(}M}aJ!jE>ZR6p3`cmy!~x z5wAPa7VA5@qr{g)O{pz!*7o%F+Bj?Qi-uDhFwM=(6uxJ8nGhe}tQ$MoqjGgXQ``7S z*i?a#h{`K|dPD!`E`$-9G}PkSDfXX{pax;SKAW!jHDO- za+9Zm%PA;~L)QUliTW1 z(Z_FEY>nd7@-n=utLy96uUOA3HFkWk9Q}mX zug?@Q)bAnN2Z3iC=jPBi18W-_sJ=6}Z7cDIfbayF%CeSpsxXGAgZBVs0C2_8yPpr9 zI3JcH;Gkm9+b6jRLE6H?qUFc(!r&5CmjAlh`ST34v@Qz+MOivYkMcVpE8qOykZ)e8 z-?bH(3K$Av1-uIax$J|J0MN(G98a>BVPI%Th>SLs>CK(rR(5tXpw?WX5Q(a9jDHM2 zdbW$%?e^{cAl8?-1qIPEyQ{&$eDGkI^QdHx%%gNrCh!~UGcgGXM|^#E5d2Y5QC7W( zeID(E*EY>jIBW;LsWt#MgXd~)plA{P08q*_HeEdiUkZo1@wXx`y{fYE+S*zN_c8KR z)B`jN8JTqx*dGE>K79DFu&@v{vuJ1BXnKkc}u&^t#3U-l^+3=lG&_3``i(+f_?o6$EN>v(Vb#-+dB{#QaNEx%= z(Jx2{2&jix-tgq`lJBQqbVM3W&Hvtdc=oSEav$vK7X92%6lCmU&`iBov-V4}NS;7p zY&&{(4;05C@!wh)xF1DELK1@PQSN-4Hh2*XX8dCSQYz3 zHqPc1r@y|#)hlEjzH3$L_wsb;R4BK@Dh;{c+UgHxrETP{W8B;}SFh@Gqv!#sM0*}R zcZ{7j7`Hpv9BiRRqYR))OG^utc*djlz4zwS&!6ud`wP?~d9DqWs*_&#)MXOmI&-8A z^}&?%NXGk-7Cx7O2TMBI+L`m^r%@C#9H=1GG4h-I!mg?xROZw}bHvK(8^jZcU>xeB z9w2DzOVix~clBg_whrM%OIvJJ{|Hg8u3sv1=htPV94NL_T1F^WX9+KFU1^^G;{^yk zEIb9>Gkg06HhW-U3M?4e-MK40SdGi9+{yIv<>Y}VkZ&hT%kcR46Ex%riHVKRbRyJ1 z=CrlW|Kdt^p>T;k~g~xoKK+b4pPwSya%;L^1@#6c3#~Rj_XS-3ZR-q9%4HgqRN^!?B zva)pI#i)ma>+wl%5cC0i8Igpa9+CL?#&iXLfQDD9D^C%10|<0o9Tl7G7D_-&xM=7n zy5VqIV$=@xhE@u4aucBB0ZDW$%8QCT(1+5{X!=(Fn48b9xw#oa@h#9}WS7IjR|z!! z;F{MvuG8ccgNL|#ajG_h_Onq@Q30O;V}cHGi^fMpfaeq2+D;c1o^pyxh>1CKZAfG) z{ONzrM0R#II!ycbheBXO5AnYIhg0KIzwa!cd10cC5{GyMMm@BMZsq*!Y$Xm9E>bke ziqlYS-55b=75Qsi`R}6OVBU;vL=cw1tH=uYc$eprrsRa8BHM z0LcXzS9dAoYHJ+O4^R$ni=L5D&U2af^y%=Se$LxfCofhh&ss(cToALWdx(>poz3gC zvNq93r>WU|^z2i$u0s*eqM|(S+#yK+)G~a>j)_C-fh^G6*eG*%bpe~z^Jo#xKeQkL zNtJy0?#m;lU*lSFBEe6d(9errv9mjaoQ6dzti;bVTkA`m zv{+so^5SkN&+a#FoU*Csp&`G8=5$h$@4~=o&#c=MpTkj6l7(}TgX7}k0mm==PcH}b z5v`yUo-rtX3!Q17kXHow`Rke(SXtlQ7*||A9br)cj&ofZ-C0$~Jjhd_C+JSi%w548@^fUt7p0g5;+5MTl&xd4ckcO8~+;kwlX zp$Yr~0v>j!kGvG^pHsB1mL%2X>C3N|&e?mX!ceY*{RLM{V z!V>%#hSDpFZV5E09-ejRv=G)nBa1?Kozh>NviyjTY212(<^vjBl+@JR+w&BD51%~g z>F&PLr?*?)KhW6pfZK_opL0d8(=-7_|} zgYPL?nTRA?Xn>L(=!zFeTPmF4X+9#949=^&tJas)`YFBDwRiUbAZv( z`BBJz4Xw*NhO^pKqTwMvmvTZyjvn1(VBqQD0V!TRPVDhpJ=`do=($H{cY>DVptE;M zz%d>ZJ3EpN4$*o;_DkMZSLzdGf^M3wj743{E-1_FY`ZwFMHP$uvEVQeK$r zhk!XU;?Cvu{mDU5%sHm;gsedV4%SgTZ16L1fVO$6F8%eNXppuv?$I42M$0P zXh?XEmiWxKZ;C81J*cXvSoTStJb4nN=SgU2mIh>C)N*vxIvvi2vVxOIxPP8>f#OgF zCdc1lWNDdFT`iUIj-Qve5UCnM$?S@gqobx+HL@JgeOfNDQ0M@*i^A{JsZ#@0?k~f` zJCUHhyuDAW{J=71zcD6OfNay+;wPf zc9xrqD>?f1Tu`Qc?s}Ny2dbr%!_Bufrs{Zi}8CV`F4wJSXA)t&ZaC z*-xjhG$i+;$q%F%(q@Q8CD8NZABXUkyB;2TZMldvLOop^ReF0#y$wA`AnQ^EG{GUY zwOB784FX{GEbQQiTeofjv1d(5qN&L$edm+{4eVRYaFReH%69DT>Qd%BFsnVBu1*Kj z#r-{GKAWr8aACt}H9vc*O7lhRCCojnJ>4SWazi+Fi?b>*}a6+Z%1b!h#=Q9<&2xTObKOBUG=AK>8N01*)O3}j1XRn^qcMHB;Q z1Id9UKul>&O86Y`D7^p3!OK`%csx+y(lo@JN6tWS$7-es-$!_*H4^vdUpV{sVqsM>qoKbYbSA>Z?j12X8$(nuU>13c;!Q1W93Q3iJE zc<<}0asAMdBMY;$ruhR(L>=tBukr?x?)*zSabP`zmO6G<@(X_n6W#AW!#I?n75@SX z2Q1gKg|0U_Kacgi5PN<4$Bz(uDsb>0^Ya(jI7_YweP)=P5>Ls_e&~6_xH&eHht#ud zA8KYh+fr$EqY^v0pJ{#g^XH=2IzM)>jcgK<%8+dol0aqpf$tzZGehY-hnSg>8&66} zVR=V!aPI@_@TjO;|6|o) zW@18h1GCEq9zZKTI5hNAo?qwAGz10}RaL74WM-vy2bh_MvGTbSAPUb0d(b|(wK$@l zI8^2sJ#ddi*tSjC*gK;5^B(!{-?6t7#1;5sLXhdGKcf3YG%9%^V&fIa2cSSt3Q|Z? z65HWt_c9s^Q&X*wYi48}AXLkkg*|$t0huIovsUsgh$F6%!qN}?Snf2a`1dry+k<9p zwS6~FZhBrG4)HF7LLd06z#r`VaeDr)u4YQbqN62p>QqN}cjR$C0f9tm4vB?#86^mNcIIj^-FjS<|q^}fD7bgPzmsp%n3pj=W>Q%kVqH1F&363-% zbL56_*J{8?wW@^LasG={^ut_NuA;91!g*E`Gyx3~^cwHnSwpuCnhfkE9q2LQ>wxgR zkaUNkV;{2&>{0UG4Hm*U`{%BxHAx_#M82#7#G)Gn^1x2af37+uWt^SK@W?qiIXMws zRp1Wb$&jZk#0tE8eA`Rya(0)M!__8`jYDj}NP9G;Y?|nCv0#O(+?+XidDd;cr*8$>`jEpkhoqr%Wx99udV%QfVspu*F`9D^l&9${; zZsZ~#AD@&On z;4f|45^8e8qlD-Q!nnJ~*{4SoDwrS=9iKhh+1twv6l4H@O-%83gzQ8+)=Hqs!xSuO z^V>{?e8nzb0|#NE?CZz`x;$4P5$3LoJk*j`iok8^lU@RJM9MUO>*-#0^)1vpJ2k05 zc^Vj=OoXWOx`*F+OT{VP-dg)@l0ZhKqodPP0p7xhj+tG6@4t02(E!&QH$L35Iz@(4ZEzmOb|PL>_8OsZ$2fWRV#8t@s(cg+BY-FB=VK|;}Vp}OhjChfCzhU41z2eso{ zEB%+K4$fR=xr_TyQ3-%A7tvU^I)_YNr#}gs8rO*vkGG!uxE$PD9UfLaLhtk@gzyoZf#?v>>xK}Ihx@VRiamk(i_nVYx z|M}U(B#n3Q=o0A~oXSXf`&K44{<#~;bJ#}wvX4mL36qE<-E;r`{p|7#w2;z=&&%x6 zEHr)fJu zjwXvZCBQ)tvH`-m0Q_T~giZ5fhol4&q2jYRQ1l&9YI@a%0~uRE>9a6D6(|_*{xd)V zJ!9{Ri)l9{vqvdedQOI}{d8FbT1Mcc?1MurHXeLfTiY3ZCcEIEAhPoN_Xfq@;iREV z0>x&$lFs?b4GU@%RD?K=UU!g%PHu4p8tdaaJvHV^O8y*CZ!$8tS3KaNipbq1f5mG3 z&*VglJ1BznscNQ=#lzCTzYw%Ow_^j(ojXTAU*AHM)c#4ouyJG-_P8rAWV1O^OwJtH z=RmG%p?GZo{`A9p-}@O|?M{Zb#iZguso>=AJ5W*^8XA_p6M@e31kxrUA4NnQh=dRH zq1H$ZJ1O(b3btD(_Q7Kte3KZLU+8>IkR3aAjNpF=ugU8g7&s);(%5JzH{37)y;r?) zht#~Jq$G0LlaA0(43waTQXZ_FO6c+4i?*!0-e15{#vd!~QHL-P zz2l#cbi-r5pXf|?%PT3f)4(;B%#GX=WBCa@xkH6I^wUvBKB0Rg`_Uv78Qo8n*4M3Z z>R!%WxN-kYjKk9({T5n`F7d$sqD(w*rk}Ib9Jn8yG=s z0G>@03y|Z|f>Wres5%Cc+&WsLcI_ce9^Yz9kno?|1X>4F~@U+$K`z0!XGy zgndm&swFki0>?vdPbvPb)RtgTk&)+M*}L~K!c9L2C=E`D2uo~f`?5Cw_rSLl8eNcYIuMKgD z{X9%0!cMGM(>d|l`HQw|=r!-9d?r#^BHi}Jwl>ueIS~<&nVVsdkc$TRfLv<=_e5Pj z@CXP*U?pPAGJ~7K>7coV#Xo>oxm&ld7UrKwMLcS&b5h1;xt&r2P4nllm-DRX-t!Lg`Ul&%i2 z{y|?&5+)yb4qE~gD)5v7*c&M@^C6HL5?B zm))W~hf7OE4Arxc$9~xTv&1(!JzB)|nf?nsLK|sENL>8v-og3rg~k{2j5T1pKRgxR zJjp6`iz?(0+A6@Tx(0+Hgw|N$pn|_&T`EnUusF4~zW^G56a^fpp~e2?)g?JPI=Z>A zN%I3B>ml%%>sQ^9k$DSoDra3?-@u?V`ZZ(`lk^ry=z~K;5%-JW8^m2J?T>9Ig_2=7 z=?jVksBvC=hv;hGF;V5W!NHB-$ygJ+Yu7@C?BHg+a_grRbwHh8?>RNK%LUMaeMxOC zMU;3IXrR7FI}9l57{G7sf9-E&Wu@LeXQE%ev7(NBVLoxlU6FAV(cA-GyZ73Co?g0W zZq6Ip1Ct0Gz@MGIn2c(-4-A~) zMIT<%Qvabs;n&gMZ>#n@aGMjS!Ya+@n@i=wU7ta$j+5Bj>7N&SToMR+N1aj#qF4d!*9UUK}djMi2Z!J7*avlFj!p6q7dLV!(c^3ygWg6R<$nV(2 zBU3A!JBjdeB=7PIDTz?^a&&a$&-{YZoLfEe{l=MrF7@1cQLNktg@=X7lL#7Pkjr(7N-UId?^PjxXFmJ|6lX}r z;ofgwzMO{|X7cWW8+tv3U$;>w;dGB7YC)+am7$MkkFk+5S)f?`(r41jC}?qNKdsvd zs%P*SFNd}MiC_&8J$B3%&AA!QN;EVF3N4|X3Ej=!C4clQ3Ou;eF>*x<3t177XEoI& zW~bcZVF3Uai*E&GVo>xJ+CsgfU=m^6l5zPnWJWjv{@X#qzlyhWqK z_KuEeLC`B~vstnK|)+oj%Ea%oQ%roC#o~Ai;h*vavU(}CeKve5BAE2IP0gSm_`TmHR zmew!Vt+lmvR;mxZ?}-8r_^%3`$mj_+7HH)iwBq98{Ks*+a5IKZ&Jm#Ang=7Sfng!x z9VO?)=n~vp)*Q_K8g3FFTt32I*iZ{358*6(=L;x&lmuJR>3hWy+o-7E93Yd81j{2} z{zFk^+jb1g6it2t0oeHa^ZK=GEExF+i%0-un!x7SQTUHj2oF#isH+Eivsx}>f{Iz6 zZ#gZf(V3o2Ts^Y7d@TuUg-S}lL=IaA{EO%*T|`f)%_)*``2Aas&g<(j)Ev4El2o@L!uu+#i{(f~z{ zKDsW;MYWN7a&hsIE@&iXOS(c;Ij#m!H`2a1QsG*-NNfL(r%0-rWQ|_v_{8 zNG6ON2`{c}+OED16V;LXNWa%^0zrP+Ih5+w22}nAQW71V_Bqv-59Nv9zJ22gVd4z` zJUW`_oD?v)Ku4@+ZvKR|iJ{K_QNo#<-;`TMT%eEWck{~K`Sqc%PmGU`qP6@!1y707 zGaa(^7xQgAV_{kmAm6yVQy{{ z8*Nu|MRxXQXq!ki)*g^-`(9nT2@DKr;9s0rH}@#(we<}uXXSBWCf zQ2fi3m}Z?utDdtOJo)tf$7nU|G#yw$Sy_wRs|-X;)Cm0asy8T2tgNhMy$b|UT*U(J z4Yl~i<>h6dTiv_f2!Dv3qHh>4o;JRa_yVd%54i&UvNq(-{f@feq28&MEq?dH`X0kp zsQqTpniR3N&zSUg%|@r>BD4h_h?;+#Jr^B;S9B z;qlM;R>OJc^7DG0hmn-ev$SzS|HyLPbaAmNWUusC;)#=eT06eBxp`&p4mO)#WEFhT zQSbWNZ1=I2&>*UJY%DA?1bL@lHCV=H>yBZ39@HU%5DzsWwYmpQreUlQ*RDI#Um(17 zj`)L8)SHLW(bDR&>}7aCu4HNohZL1SV{0o0Tk5R>9LS!_XC? ze(_>r^JHq|9FBe0!9;qTF-T}Gi%>)A6JB$3a1XI`u+s!91C?fq(SR=3e<@hO3e6NK z!mLqRRtAHsj6+X0{G0ld09a+u?Qg@)jDm6Ed|vY3EHJAR#swplZtaeMQ2{To5zo7KrOuq` zVXI{zT4O=-(~Yb|jGCP*WbbSF)%5%X-fuVc4qG%aHPzPg7I6 zVa85Fm~?11;-sgXZi#h8dmD{vG0t$zsX;4ZlkqgfN1&mD>O$xq3?ebUGro4<|9An= zMuemK3#fiwT{;j1AQ@mEMpq$srRvY1FLke&7)HB4?|1$^9<{;E652v^=L;?21A``- z9_Iut4LykDBasA0zxC%Z)E-H29ytZ{op0Ru1vb#pA`GqI(??zampt=IS|cVJ;@{9u z@CC`p$eQ0IY3UgnieTc(P@)5g5s`_ykLhMy00SnyH8uA=onerD7#jNCZI-rA8@;o* zJR|gn(fd>iq)NFO%q-^gxwUoljfe1*RZK)gv2`QuvdquE{9$yBa+G9~-eajzc$XDM zD_zNigH_gljqhRL!I&n}G$vXX;MsXLlvq2C%3uYxJA1Ldzh7Kh+NdYLeXwmoI;n+u8Cz`Ig|Og6ah^G#B;l>_=#0cT5e6zw zT{84z$7BF{Zt@%MWpkE1JsQE2$)&n3x@j*TTXPwM|GM|J~QWbCtNu-)M$R9X%x3d?&X0EZf>E0RYm_1gepicDJi4O$d9 zzxsN5>#!ENxCoWYNnZYN4-5y>&RlSE!jpvYQ24KwV+XKVa2G5gc=ZZ86xudJ5RT#Lfkv2YpNwo!inkQW#&Kaj)p|gAUl1KV zy|uHmw$|>^(NTXtzl6lZ;ZpmwygYM#eLP)&(D7g9fQ%7Qp5R0<2p^M_jH6`k!P$jj z!y$j9@N{x%DVilI>7=N(i-UuMJwj(>WVQi$`K1hU2!1;t0#x@Q6pyIXhYv=o;p035_pCV zH6fkz@{;fMCnpNog99qG_7ut)bd^HlI=8}7M`8t(mdnh!<|l7|{cyj|w^|GA6w?^@ z?h*YdMAL9)LXe-o1?|ewrM#161GAQMF%D%)uUf`$ICMSkI-&XFm~Z4wH}9u;pT7Sl z;Hr#le0&8&(y}slz@^UyO3oT{1|Jqa~uh8mrEy}Lh(^`e;!Dz5EB~1WW~ohJT)%C0xx7f)dyVOiv{`ImR%Z z!Mqa|b{WxyN&lY&nj&M{c0|MGytz3ljZIO1L^<f3q}D`}Ar>itu2Dkm9wH^74YH zu2ZtI{~$A2>4X=p!sQkj6Em~0&>|fP3#y#k49!%R_Gz3=Xf!Ct=q*KZqoM)FS@o7d zJ^}&%{p%OI=y8baRvkOE#0uqCoSe?_QNvFQ^Y-T(6bdvjD@|;EpP$#%jy0=r_MUEI z0lR|OrzUzFk_&pLW@QcsZA1u9J^-pgBlhuztp`pp#?tC6zTnURn3cGlM@m*dq1SlL`(cXj}3t#3WvVz%BOur z+s5YRBM3ND?awnPXtZP8)yw$*%{;d$H%hq#!!c}n92XpU;d4QOe|bKi5*96T z^5l7C_#e?Pf<`*s8gINb3_OBrxZm(0DQl>$qvHvsqsu9V;J0tyn6uL)SGMEn0_q_t z$h}DLuwkPCj;T`HKTdFi>*|J8R8&yy4(plLeszmEPklk|WOVcR4c!&CyOs|9g#`@Y zW`y>pXdv<}79$?D>$>QtTA%YsVnSUvsmE`eK_ZF+@q!FUuw>dRs$<{+B?ioTPp9y` zjo@Oxh?g0)BMNCswC&rjlXeO@&yp#LI?mwmBN#;;lJCtuKXl{PEx7n#K*W%+1w6>u zg2v9+$;rB%?7I*hX&{D6UZ*AQ1l5iewymnJ#>BRM02B&^pcQ-`GNtR8BVs=9Py-HJ zaK3z5{_6LGdn=dAOG{hY+GMd)C@C2Am|0CXx&T4D7Vd@+piZIH#&kE0*@t*79gPsEg9e4NMbA_^G7wq{z z21~L^N|sT6T#_)JQg~q9hUqBwzxN2JGtLck70#Dxdko(jZX>>zu$TlQZvq; z9Q<{_dL2tWAx9+spzLsPaNwy5z_rVgjpIgKVCZ?tV4;~z1LckW;b#fZo-+`6m*)w{6Jww7iLyUy_Y=@14 zB8(2|>F5d+1<8h+_k(oVJ{$t7Xnd2cp2q zYd2S)atU>(s_gAi&$<=FwJ!*EPrM>uyjYx`z71gQ)Q48tIwmJFG^6{zY3)kBrKe|T zA>q{I$MJo2RJLS$t}``>w)^ic`S0?Wdx0mg_~Og_{I}rmFo@F-bI_1ufCV8xV>y2q z6C~?^#$rwbaQSXR3f|sYhb)r=$byWVsv}{JeY|+lhR%rgKek(K{g){AlK**uU1BLv zV1D0`Mm@$3L+s^`jmLZf{BzVK^8-akn^iG~3wHrt07@v?7H>U`EG$j}5f&7D$-f|u zM<1y(G95nLYpzX-269~{Q*?78en5ah$H6$h8rub+Rl=h!_}0k%PKk+q0e%9vg4m4` zg`rmB-S=V305={gzjoqQZC6v%BMyfY6CAzskwKYRS<|z#>X|3dcR9(+`*mp@H;!0^ zK5VnE6Od{V&dsT*hUq2DPGL3-2@#nJ$rozR7mSX@-cCecO?Mr55vIM*B1{++2b!T$ z&*S=y$tRK3H>s(FAvp*D>Th+xO9(gZkPfq&j_GRONqh|T-o3+xmbLggIQSg~Kn(b0 zK4IXy{#avEQ-s!dfGGUYcubdIl!e~2!v_M=-}2=k+mt%=&LNo@22?=jp0m8WR@xFC zdU(ehgyPP;RXEQ1+NsoelnK4dn+G^8Z+PZjC@h?TfU~)W30-MuQfpsp_RNL3*yUct z18Y2Bh(u()uCBhaeiCQ~k;L(uL$_VeK;KlxYKU{`_x+00?N=;2Z6}`lP!ny9scLIK zbrz^$MHC6$u#~}|PASx?zzZqwDfR`WWn~4$wu!J~w4_l@K@{3l2BKB;`gBEVa&iZT zVFnA(y#F~=_LAyAJJ+s#?9>rp_-Ix*nAv_xk4djz1WS>7UNFAkw`Er{C}LxUg|2T{N7EB(0+# zyT`Y+wV6Z_&Iv?}QfmO$WiK9nxEMa(%CcEpo?+HpkaErIhJ2iVp`xIk6tQ?(>GPal zmDT}b1=Z4L3K5=uYFr;b{ymp)MsLoj?3+UfwR*O&PkuMv4;?|5LqQ!4X7~+PH=QYo z${q$54@t(yddS*}54LyfM|Gwi!L0Xxp2^^J=b>^c9{Ca0SPs`39*8se;KN=D3OpQ6 zm_gA`OLrKea>{|H4-`3+XxYRdP{0%Ps1jw;mEaeQ(rvg2IY6P=#zGE{zo8voz-hJe zxOVH7m?v+1L)6Ef9(oz&5*{PQ6x};;{Y}}X78I}^J{){BAb!*#1> zgi(A`9plH4>(L1;%Cuf2Oc-Y(!NKd4V{oh#^m@8jbO6=scvOkg<=m628cM7C$8}JR z0#u(KGocbe>M2VVsS8X;g3)*+UKfa#0tOC*4Zq$vgW~n+TjrJfFNu<0q-CVO>vVR0 z(QewVS0eh%Fyj(oHHfImFbg4#Zj>5#{8+_xHs<*~e(ayal49mh#P?nYBtI3MjUH!_ zME_tRNF!W=0v2yf99ju06l7?ULGXl50x3sai>({Kk(Zy}N!!Em_;6mPlRX2kY6w#R zE7Fi=>k@YqCWrt~P0juv-rhT$>%I*i*WDh~%kkrB}%BgxFn zitL>cQX#TOL{@fLWzYDY@4M%De*gcD@A3T{&v745WxU7h^}4R>yw3Byw6A16IgU7) zq!{fJ`o-$YNfNVkq6zkmEvI)Bf&LVS2onA9Dd!wT9@zJco^oZr`1Hbo9H#qQWTLNS z2(j^pmDD0p2HKx^jfHB@&wy(aTjvXhfru-os43JC>k74T=EOG_{TP8G3 zGv|4WFXQOM4K!OAXqInlutCK`ZSoT!%b%J@3QyyjOGCwm@d}rdIIGLsWv#=LJ zb=RMvy%z;7b?mcGNHGj5CTc!^=DM0qQu2KAu`$WSJaR-7+V5wz@$<$#vlV$@nI|X7 zbWkiN1_fxSax_Dk`02MT5GKly*4Cq%%n8H;kSROK%d0YeUF}43f{k1+18hY0*=?7n zW=jz#B_*}wkSkz<*xkE+*E|x=l#hnYVs=s1B*{%bl5gIk8w(t>Tl_Vk8azV%pH;I! zM<{5}23O>g0$l(Oe+{!A0ds`DctN?@PHJpq6svUxH+t$4zYi$+hBiwm;yEfL_AaOd zTJ$_P><#{9&{0AIc!1&QOYO3|f|2&||2()3uqW9xAI2>fq`+jDTYpA|ql%d?0aKi& zj+=*USGHW3W(xsSYGz8wJ$&@2VO+Rs z3DGmzu$JRtGNac7`guI@W*EQeGpKE~a}g!Gi~zl$$tWcUAF|Kf(TC%d&p=X%m9D~K zvgu#8w*RJq`R0yyagLRq+qY0t?X7-Dx}BQipk?O(<07|D*%ALT`pNAaK4p8o7&!K5 zP*v_Y=0!#Q{ABcnns@G#bJNz*7bqCXzPq0PYA34cz*i!?`=ya>Q@v1|PAs+*vmHPy$zslF=9xYRPZC<_Ena$LF z@)J-X<_)TdBL9s_Qj)~VS9`tZ`2OWcohu*D5?6g2@5)qoK!CNM0DOzuK!KJ zEA)G$?!^_w)97?hwkGZk@(cJQdy{woY;Z(=WucG|7)yy*Gc&o+j8pi1sy`PxTPd zTi2OIba(ieFdz1(!*_odfVWE4}R6WARkug2F0_1vpp^1i+cbOk!7^Ido0jCe6c(MefXfAEdu}KyH=!?0JyPnA25N z3xg`L@G2=obWz|*^gJLyyfVdGhjQqlLDIX$A1#kM>#ISE9DK%$ZVG-gBGp;oyz0-a zUBR9_JiWA&@e=8siRr^7Y!*Z9#buKr)U_c6)G%GRh8Mue8>hLu$JW?)MP*sZr+?29yaoc z4?9}cQ|O2|X~236$)n%eXj7z+lIt9ag-FY$rLzxyg2UeUB(FFJJZ(M-ux%^r^2eN5wS`IGo^|83^quacn{t zBj3cpAU-;}7iTCcK#1^~?)h{shj|fFoQTURA4f(sYlV`3mjVRQK!afjo3sjoGxBkDHlD zNYsQxy!V<)Nc2rFSNn;IOgRT0s`Zp5J9@MqiQKGy^|x>57#Z(7TjCD=a*5|&6D57d zeM2P2#q-&-MxdjKP14kd4?_|zw-}>~MTxBkMD{|wHu??dZgs_^!15D(TB({@bV>op z(EMRBxpk`-{0%6jV0a0M3lJy-y4ynY2!c6zc|$K}VO9vDlNDQYW1|&%FFoP=1r&Df z*zvZg=(gkz^g2*dpP)s7aUR_((0{Dr0Qm#W07I{;Oql8o2w!M1Sb3u^K_?LKKuh)w zNffmx)qel}jb#gA=6$_@UX&8JC=NC@%g*cpw2Dlc=ucgLk37)(@lA-E(25|CfU$$1 z=GXQ1>bX{IT1R|7Pn3nFsC8wGuAFtN!|EjseC51x?BsurL?}+Lo(wq0kudCiG*GadJMT zi&fLm052;Yt>6a_#KD%))r|sK4M7S{A-GSl8dtGju`mkU0G*=<{4mWCjW-`FzID3X z4s<`%9XeWCIo9k7Z2}7Xb^A-p$^eQ>Kh}kg4)i8d?5%_ZX0lzt@ZhFMINbm%NqYEK zA_r|Lj+csFFMt={0d!PS*!sld%SOSqTR+R@!^h9Rci%pEqLI&C`&_lYeQ(#-AKy>e z{CNIhaAGSJ$2k_aw(=9A`aRt7EMr02wSM0EHV`nu%gs zUD)*W5=4?K;Ha4&sIa(-%7z3gYPTQ&Wv==`PoL519bG)MqG-7k0RP{Mi*`B!cB}<=xa(nb&rlZdaiGC@JCk7R?a2i7plx#+b=%%KoiqGe<`w*6G*hPW@ z0>BagK1kOe82|QW0uL5BUw8%18Z^-1lN437OB2mWAU+xR#KvYLEVLwVQD_5bq+9M* zCZI!k3dsB{qCFg^l&M03f(&omm_eqvs0C3A);Uap#O36Co5s*#>jiy#O#+FX>(D1s zgJTrgWCF*(gvAos2*5pdsy$fvsDqTYy)Ht=L$n_f68Lcz0v4FAH(+K?fP>qN5QvX> z*T)?8bik+;m?aWfEQPvgNiImSi9gA`k;w!Ig3h>EFxTnRp!&Ql3dZ_@EDRlP_}+of z08)43y^fxqb;RiU*M86iBTijHH_KC}6NWY@X8yWWAj^*sJ}s%D1|$Vbv{sMbWwbsV z&jX?wLeuifUtChM0w@c3)^c!>#dD4CqQ-H~3l@txMK8Kx=(VUpVD{;gCmmfPk^6x4 zG&V94lJB6Sr&k6+3brO(@v!4?YF~#|3mQ=&P39Niv!mSuysfP4UJgHp^dnCy>6Fw| z01=lbUkjSI4x<3Xx1!Sm+^J_3y>>j7aR4oI4cPke1n2180q9z(_jmxCK}LC0%=Hu> zpTW!7nHd`ghtQ&9k+7e3z;Q&*eDTy1^eLEjB))b0B?gEdF#2wfkC2)7fEyh--3;)q zsVQq8(}L{%Sf(0fqt-&dLea*uw;Y_RjjC*1+h7b|yH&hmTwFuE0xJ(E!4)cMGO}iH zgbH-^_0Kagi5mn~f?0{kPfh1Ys8a({%daAvLhl({;GVengUVWP;_>^+!)UH4D-+sj z(64-z;-poA0Z0M6A8m}*X}yV-4YW{Is2x~Xi$QL=3SALiVplq{2U1#rAL#rIDh=O- zk)llaHUKR6Mvc~wMg+EoXl3)_K9q&>FGaKlkBRxjgFcJm_#F@g z-?}vgaEJ41cI+#oH$eYzo^%)6y{=R;L|lW$*RrQbW&LFV{som6BHoM20=!ZLF@&23 zorU*s+$4TW{@T{o_W83+eox0_*nhbIBFB##)ravSIWTCX=QFB76ES=y6=Wb@^X1xD zkc9y1N5S4JctJB$S8Fo#Ow|t3#1?>|I4`o9uOdSBoPD(q8_dMO4UHYNEpQZ8>7HU` z^(3Xm%c7wHu}r|cARZ13YuLBN9F<5q5eoryIe7QCK(PiW9XK8J`8%4=$ON{9?^SxL z(spEq0V7*+m_W7F1qc_SPn#HZWo0uo`O$`kHzWK( zI0k!w4a2Rv9F?muEHFWMmJ+J|YmGU9^XG&@#PQXWBpb|FmYUy`)DVVHrP(%7#56r< zFS0R2L+Sqg`#3?N&mnwmR{YVJD9tCA^ob{|MnRzg(2G+WU>NZbIwzswPr#7_95wQc zP5kmQ;N4mr%3#3ZoEjM#nnMhnZltD?{y9F50|&{>x0aTGPOew4_F4gj#}1#|?p=Tr zstyehEF##6RJrXyPpm-Aou*yckwUN; zvNkm|JS=6@_v(ywaBl7_Fcdf%yP{W&y+<)a9PHR}=79KQg$on|D<Pa2w}IJ^ChvEqd3Tdf>f2239Ccc!NXHYZ{tXq z`7Y}5-ldv5@2boCwyRZEMRu4cXQ}GgmPMo)28HfC!5QMn0sDFm;@Jb(J=yf@8sN#6 zpSH$;G_Rvi3ZtkXIYH~bsD`6Fm#|M+NAJOSUn$G-gDy$(8DMOcxjlSCGft=-da7QH z6bGk^oj(k?2TK9FB_*Bv>S+;%tdjzgtVm#wD@@L!B`vKG$E_vdQF#j~ zXIJC)K?aOb5VU}6DF%N`3%np97_%JFRylC>y z&heeDgX{H)%vmc;N_fA?X!dvb1*5xOWI-{%ds9%=lePTdlX_=RTzg0~|DC21o0`pM zzI1!w!6o1Q+L6J$D6dHF%m4R#mEX);wY2oN|9Ma+R+p66cmdO87%6uVBRlX>JB$0t z(i$_B^WPA*)-+`fzr}#x$;d*Sw8STs4XxttPkgsmyd(2Ae>T7JD2>GW-emgUPcX-! z{o|;{2M#5@nBw+Q6+364og^e1lsLWr{RzoCKsc;fEbGq{yJ!CGlLqP2;a(u&$pioU zg3~}_KWWde{r*~P7xeW_a=%nx!xobJw0p(>PSm(>aZyFw#^A++xSLHbimPc_y%%S# z_G1#th5ucIoUD(mV(*S8Sta&8Hi;@zJN|wF+yB16u+~Rb)n%XWRri(OmBu^XjDIdU zOG2V7`M<|BmBPl}l+SRuc%XK}CX@8ec+J0GFvb44;t$`2{(D*Y!+Y!hemnkf zzyH72gFldzi~r{h6Mt}f6id9=zn}GsGJT}~{ljz7^54PD%bZC0jb!w+Y$D#omFmps ztBl7E7;chCZ?9L{c0VF3J6pL=A-Q#(K{Qk6^|Ml)EhNeV%j>SrLb$uyAB#_3`DP7` z^zM0{s7+hDi&i7z%1%;I5k(CtVG@$qyTvvYa(xPaf8l+fhi4!C_jlvQhX3=P@xm1K zwBjCQM&O4V2ZZok{>=T{kz>R6_)gN0;9vnkL5?_>KdOi32IiJVGP(n1TRul}fAb8r zu`4}RqGoTz&*<=P;(ZzSbI;H$8?L-jnzgEO4;YS(D|YT4s(z*>p^9rc2|>@tuYX_3 zd-rX&|5?bmp#OhQa>sub8F5F;q!0hU*(v|GU)le&Uqq~pST^hjMYo0Vkgz?-(Z6(b zWIhfh6u(GNUzs8Vf_lOccbK?o!A~IC5N>jJ3y?PhbW--_XC!ARub7V=gNKhU zvSis3o?z90?90S5ygy4pwR^aCTQV8Mh;_BK@JMb$#eDL_z&_$92kYa?Q<2k|nof+#c!}(_|+f72B-2gUF}wsX?RwjM>>?Xh0croEOc~} zUF2~V!Vj!HR{s3_`tMwwqcJB;l4?u1lyjiQ`vXj)26{|<7_~z7(n7vxcL1>*gd?u3 zPheKmPINGqE$e>{Un!2EFkJhk_nSjV?%#8YCem&qEwHQGy-^`qYN4*5|YVDA1UIys(BJ5pD<_#YCh+wUc@U1 z2Dd1fKQhZx#6a#pCs^{aRNZFSvOIV3ie^b z!`}<{NKBvk8@y@)dNei;av;q8+!qj zD$v6B?_*KCpu!8*tmqtti5J$^X*`{kt(e;a1|7x@pgo=;qKSs~;@q6CpC2}}I_zzD zd4HmW1b4)!-Bva+y1S zGEY>URG|$fDku}o!P!9-7ZRclD5BE+1=1%EPfsM9t0>Zt$f8C7xd)77^d1t+Twn%; z^d0yZ9R3q9V+-6fTx>4oZoHHm0MzOKbf26>GDr+c0zWdbOw$|{Khix=l);h&1`9?z zl32`A0{QaJbdMI2v3NR028A{xvLGnI#t3CBieQj@c!h)rKa;bC3+N==`~6)UfICK6 zz(f~2yaa7)h~m>ik(`iF$t{Xkz#|pa1M+X+8{T3%uNCQ$52GqTCZ{$})Lk|QSTYro z%vVuAz#9XdF!TedG6?!P^eG^z=c3+CbS58}Mcn{LNGE4!&@Nx7y@q8lK@}BsDTs*p z2CP4nQ|s7FF4ALMuhb0ySwVq;l~IDQbppa=D+l}zLkgUMqZ#z`;zn`cLD5W%(dwB6 zS@p#6<6a#{1qJXl=+9El5if-)W z+}b}R7rG|h?VUZoGVxMdTx*7Zc@{`#ETY0NFbg`dNh6VHz6xRmU@Ht?80pBts1}z)Hr2D^#IC)X49Jr=zrk>`4(`8$hSPbwXSSaO%Qi)O~C3trHj!z+{uM zKxlQK1>usBm&Xvg-*^rXs!C751y3RD)D6^+Sa_B<;AfZvr@c|JNBoz;5AKh(uebrCnPp#u;YFjYim!YD1r@wR#0hasUms;7GKz= zRRt6%#U*D-Vu2u|^c;iw!(d$M-%k{%uXsi%++R%J368$N7zTs```s+!pguo@tL^Oz zXd2-Za24Cc4xr$2~HYn$;tGJKr66b z_(!+IG#!O~D1wbo*(4Ukj(vv!AUxrdcP3L3h-133Im8i>YN2zfK49wJ7LwR~A}@P& zZ#x$*pA+QCAdvD(SFOCE2v@OwFh$`{RCVnGt_d)4kA%DXCQAQ2{mOXAjh})0ySZ@> z%!Qc;+3uTvz*hT=l9Ds!BLXQf2*BgO5Xh3TOz1=z7dtYB4Znf1u6OB)AiijkZmk|0EKY^|bE4ZPli6_Mk<6A)8QH(m1 zgDx!kq7DrTGD3G}2Di^Kvalo)2*>^kn4w3D-8+Xy1G-JF|Bi;Nl{XQufEORWH+Soh z2t!jxhXbw}ry-~S;D8ank{p=ncjnwVIFR0c{SrV4d6~LFE}e}q>`?*Jjw*xdhbkV} z$Z0-4ie1~#VZi${T2ce$%;ENJGxUc*MKqeX^&J}=B<4F<$-s5vwS0Ihn)8B!Z_w+) zTXk`Ag4Be!38!59r%$73I^ew$2QzO+g9WBNl$Di1`TRo_iwpoKf%s@Rs8Y4#q(jSw z0RCCW%r%C1jf?>uVY9E0A-o1O5FHP4xq7O(6Ob|Ow zfrXxjxdSd^15R2M6+>v1PO9Rdg~^T^F#_X7qWN7sINYtJ46rIfH-?J^(2gUL;Bblx z!9P+*NBH=0LM}uljZ=!y6=LVZ#hzNAUxAGZbCM|OpF+_K03|Xa0-O<36O+}oHMCM9 z{JLPs0+3_6yRfOb`8VJef}62e3rZ83a1&orC{h}?Zso*Bkk>iXU=rO^AZ7BU3~$KG^C_pdcTEyuFwS* zyb^5!{?(Y+a3zOaGPJPJ5q_U2?JK_!$rGEf_Ce?YaEvxxQw82$SLdrE&PIj9`gzep zFuQ8e9j~|rU{9g;C6ImNapnqwJdf^y_c??Z;ATVDnp<0Q^RsTEQ{BS>2uxCbq_R@V z;MTK@j9V}SQi1WN$1YDuqj;5^@80#@1ilH14(QTVeSCNd+{AFUx>l1 z!Fk5>(;35gA(dN&aJ1V)!YUe!lw;xpNZJ$PEW$-g$gUw`yItUV5~buN;Nk<QK54`vXCc>rhVChX#bDLu zt-Wn{*D!6yt_9BRl^h-tBHbncNiaqe6uKh(0zWGAP-e5geVefnbp%&|DBi@N20}31 zOc`$G(ktH^$8?`!{{k%vSyVU#r^ls-5caX9i~^|hrr`tZiKrj14dE^}V{tp6^^c|0 zJhj#ntt9KFTCT+rkl@h zZMbp%8eM-3 zBou`4zE;2kVH;E&R}YNwAm$Ox`P8zWITJSel;WnjIbN}anHgC-R>RQGpPxy2P@;r> zz7Lf0`0CydfZhiiVTNgqxG_l#cqu6<0clEyf}XxX;T?WATp;ods`eNpR3LlmDRX(z zeT!g&>}q`G(>f38^}5XO zCdSOveCTa{Vd$U19q-Yq_Dn&po{P;U(e#cG0zWh-c5(<@0fB)ZUHu`O+_r5S%ydg} z!lOZsqUlB&phftYHVgb2i*bhuYW4_@7O;V73Rz*j<4M8pV?aq}=j`M(+4sjitqQHy zw=i@x*hWYXtExP7KKi*JN61Hsbft8f?-<2bocq|vm+VyokV?^?o7ai2N0WZ zks`lEpO0(-csw`n?gW6j_|h&(_+AhRRnzR!5@rnWyClOf8@)(C_O~M;pCLLXcSdl! zoftiZ<6?bfET$(AaSVyHG8Kx}bo>)|T!nD0MZXrNYA3#Fkw*hBBq6!z{Kyjr1r`}3 z8OCxO_*R3tZ^>zC@CFXm0k+y88q62Rq|k;~YI*=YN^o~IEgR6kvU!pV0&%5fkFFb_ z+nx%fJi%Z5ZnmdVFYSD!%!b)`+;dLoNlO`9O)Bl1W?V$<*D0|4e9r-IuxHp4;xI}- zMZWrmhOzn;u0Q>d(Rf`?Ln6CGm=S{6jNw*bjtqYiLXKZt*92j%S8>e9i;5;+IS$q# zCm9p{PMg!epILU}HU_s#c|H}%jgP&|SbS1ig-TlW%UEUSo?C&gHf!fW=7SMCP-g{2 zMTExd2Q&GUP)}f@WpKt>STLdikNnS8)8wj-PM)yI?Kc+fK5XO-n#-W3PYYDwE!_eK zx^o8(>qwWYtq=QsElj)E z$}ru@YJ{k2^5va?ANOP=qQihuo-$vU&S}C)$7yvw1>PiBi^aW~a4WX@-U?3&dk;a} z+5Plj4lh=!J!pE&<9Z5=-rwKS!*0C|hvf7C<_vXKqr(&CbhWkPGx~fZRzn;*>(}1`YnQ7JZDb~07681gz2+2L79YpnVg)IZ2wMWvyMuz1O zufbzDJ-IAbA7!yhYXI*qjK15&2B-5iiNohKD+6!4ySq=f{zRVZ^^-)EY6O!}L}A)J zy1^eSbxyOsityTetN<6oE)QYsj-`c#kd=dW#I8uU-A)P?#T*DKxlfRuH8d#5U(dw$ znH<7lgX1_r6pvvY8QY9i%kKRdzbP(!0fii0q%MGQyQ2VpgaKsF;0FK)8spC<#wu?htK$}LO2V+86sFU#6=15$}vghZ}ckkSAQ`ikvUzj=qZx6kU9~yvd z&R}%~Ru(ja;nn!Up4Uk2y101~I%8P28JrP4dD3TX?64Pt4iWj>S56F217~f0{#6z+>U>*TG9T-;nF6XtjwRt~&1ojHk zyL2s(Nh%!<$A_Tp`C0UM2OAmEqoOCk1O|0D?683$G6M*Mw%0*b1^Ka5B=4w7N5wea zChQtOJ|q82?rVnhAE0L>GE(Bu1%}tu(vy>mbGouij*SCs8=P2zDf^%xku0B7JN(EK zg&_{^@1dYSmx)X$CH9{y7Jsxu%K8o(m9hMYR2` zZi~JA=b8B5)N3vWu9qo~`>CeVwD8yyOiGQzc@Y0SfBEuXMx;rOAc6}>8J`q1z#9Yh zv?YFMW^qbq`i6}6tHT&iU+^Rhep<^NwZ6-ZA!e&IGoKH4HMWCcgaV}Ep3l`ki1H2G zu?HXh&;=^zeUu_|{W~YIRN|hBgS)G5}l)19yI%r zKZX{Qs{8zKT1Is%n+oHhuBb;?15*B|_nx`b49!%IZ8h-9BptI7V+Mgieo)XjGy5`G z$hxF$S)lRo*uKs&3ffI|TXn#ViUt=g`bFmqu|3HCA=m2oc_Ed_Ac={EqMky! zZ5|t;hy~G|9>Kt@0Z|JLbpz?2q^__yyjhmP|;V;Yby(!*~5XNek6uwE4r}|LdrKoxvZ#*YxBh=UoK@ zmaMRAZNOBEb28xOt7h9lBLP|q-Y%G3F9QN*a6AA)DBbk0sg_VyRwlqB%&6!J1jmR| zry#wv_X*nG^&%KVi7Jh4hO?y@Q0B2ucYUd`uY$d_SQ1iFWG0u1u7&bSErkFSs5shB z7fPWiZSX>KU(1`%gOYCOOM{q@{pJk|-$PZZ1(^BZAjJ@;C=ur?5qFBH2XGmL^Dq*M z>({Seym+|nEbq_&po)4XkaR)f!_g~!Gq=tZ@)#_7Z&UB#5sA3xB?E8fE3;w=RQhnH zmZGtu+2zai@)ISdeP}cLjSB>S-oAxoBa&}6cjvU@3R}X@Zb4g~;8;f61H>c~84+L7 z>p4X!k5K@lP{r!Y0Bm`C>2!*I`34SH;9Sa+f(r8Tot8CJ($IkAo3%*fYa=w_SnYy5 zRdV`<@t)ngF_Lo-VH5?>?5u3rQK5>iE@HUJVeDLiogn7DgETRy6>k?3 z9?m}37@~U@O;!?;$GMzJ`{B&*c@Z!|My&FpsJhp>*x*-Ievtd%hYvQ5@?_qJ4}0U5}MdE`7{f(qw$%&9Jrf=>pa zn7Q-}RCm_JwQj&s0MaY(Dzq2s&qG#AlZqKB^?#GYo%Vsu{~Fm5V$dp;m|)2R9RWko)- z=2(I;A~J!qSL!oC;X0Gh0?SMfm8hk~MlCX}WL2!d+7PfX6pItZ$M+*#`Ikxc-)$quUJ;>m}dXeP( zf-$}nbMNuGz?_~xRvDI$R0|R3)m_nY%DiTsYkx~i?yha5b$QmdSFW^zHB`iMWd|v{ z&)|#`PGFQKXR$eP@L?!}*B(UDA7GaOv#RS6VaF0fui{zBHNqPm<@8CV-18BQnOL%ZJqUcnm`LhMFzfO3#Q zz)TsB?v-mbb`zoeMG-gpylD)1z)e^L!d4L83P?4fTC@CQL`v$1gyLW%U5*5yu;^s@ z1mOiD3rbJpcP7Bu0C5eCja46M6}X4@51-s@YO@FvVNGm^s&_F`W3sBMd7tbviRD;S z+e+i_kBI6+swv~GNAeM*(m<0h(U=KCn7Bh`1H_3^?}RXn6Q?K3_vC_yZdFN01cA?T zZ7wnbgdbu6e0aNm9lli#doJYEnX(1Y|LJu0ZI4!FTVLl_v@N@G_isZ~YejHa?;nz6 zvxmn%Ncq1hl+>?TuXXlVx3&79+YUw|Gw)2H=J5sz(iX+9Kms9qzQBmt&KTIVf8Ra~ zGh0W#!^z1hOKqt4#scn!bS|>u;vcXG;k!_X`RLdUx#Sgiemi#V)O4%F6Xdy?eP4&q zcDM#LPsj;_8bn%b?VgcO2t>Ou@rS{8I;1RO4h*F&v<4WNEAI)LX#jD%Wz9&KWF<3|#N z0_K5U+whDNU$LG!d2|{<9kcAZvd(_RArq0nr0Hb?7qD6*hTC6RU*vl}qWGzcnvRQ| zT_!?@=w7Pv;7vn)`-JUt6VF;c=g&u;n30Fz+EZ{Va(Qn@nx-{Oo_Yd*{P6kHPJZy< zCs6@@erPtR-S?D?Upj_V8y{rb@yT#B!m8l+VBw01W2=4t{(e{uU?n+rYrPQld7jTD zHy#W<%Oa=ReR7_ggruEHp-nDxN_c(vqDNn%Nt3`Nm<PmnuThBmSk=Nps1yPa6uKABell*x5jH6dB9;Jwt2d-y_+}9Yo7k(Q^_m8Mb$V>Ofw<0@@jb*fwL8l|}Yz}^jRQWMJzT7B3 zq|psHp4olC{B@lf^q8k>R!?em$TW*!?YqGn#ue68frJ!$2|BUO*K8I*rKB{>2JHu% zs`bAAo75QwvzW1+>#S3BrUZqOV)p`4iR};jS?xjuitfR1BI(5u1_o}M6&Yku#ZL%y zEXmpUqP-EzCpIMQ=;YYW@2C+VDa?RPuLy^4=g#xhF9Mlp}$ig!Ot+VGNeyq&_XOU|% zKg)3G;XA=Gu>IVa7lx^k7pqsc2#5*;FnetHr3`0%=MZ(JhhWniGDUI6uY zQS?P;7CbQ-dEM=BZe7{f*9{Puw<&(dl z814eKt@_4Nk-#60csmxGQaeOHDevX4#yFtRFJ1f%PB7%!^?KmWLg>Y`;Q+;E3+6%{ z!0dV?WK0`Wp6HS|IG79-q^C9)6Ld>PECvj+li!w#Y?sI>MuYKxP*?!P7es4Fhx^+L zu9Tokg>8-LP1F^besQlJfL_b>pX84|#b+49r^4SQ3}pvY`GJ`~F;oz9Cbu)asu^JYWd&)(* zVg|lL;jQjZ+g^ca)ewr#Z%?u6po1UU`>6dV2au4Wo@MzXXNX0M(enN}gzCsZbCr-^ zrn!LX5f~(S@r_vo;m#K;{s;~$urVuojYy~E`U~;+VPw&7lz{bHd3Dxopqe3N5DC+P z?kg8%4(Kih6Xnne0}#cjn820_xVdt64CNc;Le=`&lM*AZKlM8-)hEen`R*i9Hr8{x zXZ@{zZy{|^ zi9;$RMFY5wm`MjR7|# zolv4-oKvTO#{|nbB#>dz5#G+hpkYL(f%b!Lqf8rdJl^-I;Fz0zf#m@ z)U%!aDe48H`Y7#eyqMYE-Km!L1pKUV!UT$`#<5|gj^&r@Cusp(5a%~Mer9pZ0ziG zN}YJ;!jkZ_l`!$r?+4M1#7HotZi*41!Tk@aHAnT2X)>ul)Jve@92XY16$Rb4dQI>C zpMVf)dt=k$=r+4AA9l-etR(Sif;JEE%Fh?okBq&}Szf@7z)!FHFt)aVHXK+?06a9k8zA~L!+Rh@mK*j|YT2;r) z!U7~oyXLM!(%F?eg02KBci&&>5$KY#Z4~I>e-7lhL8D*I?|)0X<1GK1N3PvljqSL9 zSI^$uq*F{wj9a(S<5?*fD^Hjd2cQei%(F?jm}U@Jh`Mh?(R)Dqk^%2epe^%RW;m|!MnMa?1i!Q& zYT|HXyom3n6SOe0v@A8Q+Z@$}D=U~!wvXl|D0wf2KYj?_nMCxcl5G?iX5Y0iR{UM-n}26)HeCh_6ZxlUh%Kz@G}Yo1B{RW_Lm4 z!vCf~tstf<-8FD^Ee!}DcVx52&@F5=|9Kv?CoIu?sEkc_!iwkvR^aAFOz42f1+CzU zsa_j)4h}@!78po^PN<}04dO~`KhlyzyJ$Q1?V_!fD}D%21^JFWJU;5*eKL4TFT98{ zqEdU`CzL2)9gHxQnpNnyAl`e=fI=<&@-uGG zQ(B07w(8ZsaygjbTA~=n4L&lzh4dPSabmfOiAfTkV9GaE(MUZzhD)B8Gc~tRIf*pM z+@vr0Ry#6HLK4Em$XDsG8DA>#Za9@X_JQKmKHv%r z=1^3@&Ev$o2;jOUCB%ek>6qH4ChYBXkeCp|YO{Y}+G!9trHP`x@yS^HuoIWl!uXbP z;EnZQEf_&6wx}G8VjZGaOl`6&Li*9d?m{IEUc%c~ECaao5)7c7o|RJZuj@0>9SIjtYHz zdubxwIFtdA##SD6G}K704((W6yjnFg01MqNmmV&X9NSJ8Zw z7+h9SGceJ*MGEB)4V>vM=GLaq}j@ z0&A({{7Z!del7o75!^To-N56>Ehs>{N^X6Vf`S6nGjyuP$Hq>){aTBo3XY7&Swm!H zcI-cL)pExt0~=vE+>r}X03!B8ogpV9drxl&&)g44jj%&I5Pl1qAv2DokF1(%C4)Q+ z8c>iE(REPuHU$<#n?0~@>;^R10GTHK^as-h5bp5sF!~t) zF!2G(8QghB=^CV3KtcjB)MjV~_ya&IV+tu@jDss_>&F=nx`IaIUdYz|mDOR+;Vak? zK)Hs0%FPf-6)-=0_)chzb9_wh!}zb0^pCm^=J^~UiT&iX#`V!;sYl=0wy$ad=zeeH zi*wKB^q+@^qcH#cuM7_&0rUrD$1RZwVr=Qaq_z+Eve@DooeM`f#bSOfGqwt?zC2T)Xh^rDn@vGJhsNn zVcpQ%Rh>@Nt^W^+8Z!h3tH1OZm%Nw=@qrZLBrg20 zORcL?{=C3ual3ol>PqOK7E!yA;Y1jeuyKrLsw6)%WD9(`-$F%h(SRpX_~mm6(saBoLyY< zy#Il11nd~u`k`+C48sl?6djTgqoW8WYv)TE6UW!@*8kg-6g+cUw+{-r=ZQu6mG4|e zPZ&Yi!L&&kO2+kzRm~9jF;Gb31F)u2{>DL;cCM1gBJ-MA#HdwpeQ){>;-|B1BTe%K zc#MRP-@pCKm)Ziy8x3TkWX!)<*JGZpM?F~66Wj0#bc6!Vh?$r9P5;Z49CPnktD4P} zW0du2DQE%Z`U6tc_S6$AqAO9M!lS-)SRl;;TtMfQS(Z5vCObPc8V?5%yDOCzzivyM zauHC{IOUoniZ4ZqS^ef-Xtal$*IaJ+JXvlnZqzD7=v87%ujKKsPDLLLU)Sp+T~V;q zHrRh&$ySt2@yZp7wX2t2e~DCX6ZrFcKyon~h$1rCT=Vc!_``tm6)dU?5}EvBIss!P z%fS6-2P!68@RCP!z&=h(OT!DhbauV8C{fA8eqM$_($yEh-0 z`>d|>YxHF#12EgLt}B6r0GmE!+#a)R!G>RKuBwtY;R@n-aN!gbyBKXYnvHmj3Ov9% zhKp`ng+J_}r;DZgH>%#;U?^pcU>&#h@} zqIZ2ScxF1UlgnoI+JQ_~MR3)T*Uyi!o;Zp@q9{!-A?O{E>kOd^J$`4Ge5{Ps%~VP+ffq zq9h4zEpPa*fQd#(-E&w#x3Rai4K)0F|D3r0BhU|1Q1BtbgDqs^D1xcgh?0TGNsn~1 zI;x*FLoxHWRnTh3M_dYtIrXkRmyYU?BO>ktv%D`EHn1#OI z4Pvk~_}((xx6eO8-|XCOrJ7LP-4_l9AJ=>!YXj3UcnZL!8sCP+MbQLt5Au%}H`*#_ zlg5=!uLQ3D>6s{j95aB}Bd=^v%o|^qPcGGA5~n`On>q;QCM>bafzysYPx^^u09YD2 z_LP)CC>GN&{hf+xcx>zvNLItcWUrTgwvC}t0S!k;k7al#o+h*$KjIasgD8rCuF!CO?dW)kp=&tGEkJHr$w;;M zNRg)N>87z{GOAkA?f2$KTH<2NOVT|g?RD7>9p|T0jbOBYFbmu}bYfzn zq^Jll39c83vpPtCl~1or4AlLY7s?sd-g4jP(T1Zs><*HF7ob>%xj3=`Sky#@hcEnS zF|2vcgcvhxr5B=n^|+Vp2GcK%=ksqCqTZ6eY`vGPwbjtMa63sXr&@^mMaN10w}vsl ze#*S$v33zDUlC-v*5|kCcj4+h%TJdy&rh>fmjzAGEFT-Afz8m7BS-MWj~rP?FCXJM z#bWd!FvkUCWb|S;Aql|*Hjp-vgeB~lB#nPHzu8&x>wQ73%jVFfx?Intr-jj#S_MOs zTNEXB?bb5NlVg8TPuG%KJn5?Vw(G(Z#k>7PwLnmX-rc>Qnj{I zPDh+^bod}87^F|~`0>-CVg35jxiS9f&fBH)){*Dkte3uOJW-fVYCI{ds%Yjk0_*`F zc#Uv4GP~u1=3A_TWz}?b(zJj7i9;(t_PR-QK8RIuU}fxJJs|t}?ajd`;ZF^+elpKX zXh?pz$R^s#Zq%@^FmJlfF{GG1Je^dO!9+r`%aL-m+ObP-tLnZHztD^m zIxcg8^~;x2N+isf1*DpvJJL{PXxnvn9(GiSvQyt2;49cd;zpUkzfolL!C&s+WR}>* zYN)z3 z&Peig1N*o@ZAoUMk21&b^xgsAe}77IcALqc`R^Sb%t4r7+@H1Q;qSibEL{BK*GmrW zEY@2Ox@bz<-Lf)>D7}U6o#%QjxBj&wHA%6=)aZ8VOvH!X0kk1c^|n|&_T07qbwuOU zNxj*{Y>Q2kv@(t>2_c1~B)`iYNA{C$7iS~hB6RW+^^K|-FDskL;@QSKwZaCP+2 zcd}iD@2!9FWwGOwr6Kul{iHGK+#!-es(p|32}S%pj>cGw>s!bJ*NFCQ;1f%N;Xd(&6jLdIQN=6s2ea$w~3A=UUd2P63fJ@tghZ|@={Kk|=J zf4%#!ZX>%+<}b9xbX!O^1O9k=TH)6PmSE%GuhZ0Mb9>trNSnab_;iNBzs20ZzO-73 z(ZHGSp`<^heDI#C%F>T36iQnKw1mGheB(CXO|pxvdYofP5nxmdNd|99L{_{T=zQtFXtnOt|EZ1ZPTz^6=j?n4{rxuI{b@tK zl71t8CQX+giu0GSNyjk?NZn>!zCiXnU7=B_F-pj_wRC`m(spl1?OvWtePt1?*5>T0 zo{f zsY_}M)Z60H3tY#)w*)RCk-9yS?l2bct+AmxQj~HZ<-Q0TXPXuF9auQz5|Z1tysle# zva(c9ca86i;Opuj8Y-oYl}qjc`7u~c3TC?)k1XHlAFg74eQlE6Lqa7gTt6y&b2+L+ zN$Bf_OFwIZ#)&)U@r*UCT%0E|_XMr++Auw&iDP>Bem-s}=>YxLx}cqjCI@61#ME?{ z^v)lh*4v!TIIo%~ari=V(A`_MfxO^i|L%(lr;fS%E4V7~G%mPV^P+L}mx#Af@n`rr zJd4Oa`1S^oKRigK6fnQ45PfUn*>q_BHjW4-P$CtBj zZQ>Yr?Z{aD#A{yoRN-i_#c^J#-^T=`c4_0ej#K=SEstg+shRjAI`N7-XL#@?k>LIl z(>~p!()ZT#=E6=F3%z(YN8t-can$XijTSx5!7bfNadf)(yZu=ft}P_9ca;W}E>Y}j z37WektBt-#`FaH+T3mCxJ;IQe_!sv)s$B( zBo{8Kb1n1-AAO!*xjqmin67|qUc8FMr+WNcf3MbDQ-=D^)b40*l{5{-qoMgu_7Oc9 zTkiSe2g!_g)9z_)`mlxM@QQK&@Xc4;xbEgBQE2`ZjH|DWr)sB92lPn&p{L~YOq5Ou zmgkq+g}XL8dH9q_3mLPycSz!dF82%K61cCpb5p$bf9vr3kc=F$K{BRBG8U#zFBvXv zt5{*a;unw}b)km%<}2*#cD|XR`BJxZ=8`B~h}+LiHow*BY7p~^Wt2K3`!p;7w!Ii_ z%=3;jeHd5zhvW*LTFq<)L+!7rP}da(lI8xLi4%!^>(;z$tXoO$U+2Gbd+?4Vr8S%4 zjfYGhx5Y6@blj2rMEmBPj9hw8o+4GEn<**D2J6DN;UAa(2y_tNLt^6iw$|cX;funs zTZ&Dw+!E@~8G6ruW;8Use;$Xmk+enr*8}uC=H3>n6+$PL~R}O$lfbE*?VuEbJh3vJdc0A-|(0xt=!A4lDo~qiNeaoI_JSMXv;hd zpKQ~6PmwkM#kfeNQ2meP84M0AyoV>|;^=6;+}ve^@>jeYxGo!7?$jEVR0 zmwRT`E{m>jAGaS&|5+%lQ9hGGdiC=1`?J2M;j5hz@by=+;yQ=Cbp=%SR#z{xx0%tv zf(7nH7|`{o=OB)ZD78*BkBh}xa>rjtovyXHEl@t9;77+^<6%!ddIm3>eus=vof9sQqjn+^w1zWSqOQREaPG2FTh|FB@XN2NhADdq&414)D#spG z9B0n;G|v^!RHadJcXc=2-SDh7KRn!dOEU~YM7??O+D(-?ExO`OnYoMPx~V9hryY9M7PCjB{h@2Y^e-H+k#;cOfV~U^mJVz zla!XB;ARvv`uo~39QE@&qX(&T5*BIu?fOm{=ABD(r+vxLo2Mwbh$sejKZ-KPP|kee zUy+Th8CiExd)y$gh1yAvI=D`$uQgp3z@9N8)x;7_VucKrH(P8@E&@U@d7i zeXo|lWQVkM*P4W68gg{Bq%9%?`EX)5`LN~+3+?LK$WiBJEy)E3hF zIf~%a247_kzF5JBmjL5GWq&{S3=#U%^$NO<%v!I zcuS0Pd0gxK_%or^E9ue_6=&}t8&v56(+Uia)`Mv!|E_wNANBs^t(kLO$)0RnF(vq~ zNt?n6bYX9{XFPa2vQ||v*B|JgaU@C78G(57KgrclTO7S>5}5u%td>mw@0Jh_0VT)z zmcb@1?y}fEqFc{sXZYkFnyQ=^+p-HxdOJvj(@7m6?i?7zrp;qq(EX_Y(kVd&b>*V0 zE<@OF`hOR_f+py9ZdvOytxUDW%uK7|Xp2^+M$$Hc&>#3Kc<|D0R4pVzjUN#boRjBN zIX8JQ{?}0Qt-m@?#$z78O3XJ;ePRHg^yb2lg38U!Gs-;j6#PB*Vx5-}YrijB*>HN^ z75sR2@J3S6=I@E4`k36bnqyIElPH&5?kae&d8&bh{j@I^+$-+S%ns}`Ngclie*|af zm&o8Cwi#XIHB!kOJb=?ReF0?igH?n?Wlmz`;bdmrEM54oFD6obCt!GYLU69I;aruu zI{HCxY2N*BghY9)kRRQl5#?N~+3#OWYow2v|CEh`MAlDz{R=v8(_Fm%<}&ye^!1Y( zUZs(53`=04dtQQ%p>t=rd42NM216?`oyl26{E<;`dI(N8QGip_mpFK+@0IZ4$AnE~ zdnYzWcX>pUm0mTzy7A_!2$b-yR8kS79()zv)d4$)QBfKd78;MF$cK2Oq(JLN!a}B2 zWrg8Hu3@R$iFVW@+z}{`Ih^`)^r823dM5HMC6DI@g;Y8FZpP6=lU&Ka%O><=$=-r% zxpQ}JdJq5ztsH28X$Pp>Z^l(tpth#z$em942feD_waeI?7D-wBxgJ7P={wS^T@l_Ne-tx*|3G1B zRx++`BH|X@{RsKU5zCm?-IY*ph}R-FL;H6>D2OWFwYu+BlD3sFs!~xkA7V80@evty z&n*m5{zd1e0DUwQfccpe+4}8X=>z7v-*3#HgAPZ@ zjBUv!)#BwWmckD=mfmdiOSov^v^|R+PatB>L%c-wt-^^ho zr8%`*@wVa3;UXKxO$~>I7A^nV;=a$ziWu7_{(OE(CZ5*9C`HALD~t!*Cr&66Yxz?> zJ#*&?RtcGjr?8Cp`jU(k2f=5*K#(yD`W81LFv}t-=WP1HxQ-qcz^Hv#WSW6!n0R&IwQVf zV<{GgR}RnSr+^6`?suGf{K)LZr`y-^!5etL?lLVZ*0@2Ml9bR_^fP>o(7=2)uP|yc z=9RQt<~0ce*5h#lVPgy?u&)2@iJHC}gVZJwPEKAAjWL3Zp6W5p7cZVbFt=5)c1?lr zJsw#k(fuG}8%-eIm_IAUFlg!NP&}lARQpagcvE7&`~1G66shiJ>=u)vJV%dM^DACS zsmwPO(j58NZT=T>A6u9H7`~d?$MI+>%bf6`P>RY0xd)BIR}d0>QXC&j4C(}QzE9cz zML>^`F?Sw@OL3Oz;RQc_6a3vWbaYm1RlBl1d&#momm)ROi6d^6aIe=C*A5P>0^SoxNceUYbY+AyamGP0^9D^26|l__&2i zim@bo^t7rJiEpt?NFEg%ZYu;$59ivgWe18J&5Tq&3Be{BlLd)i*g>}*xT$N#|J^~K zNCz#Nic|wf>->`K^c|=Oi1?Yk+Hw11qh^ABnEP9Sw zibJ9qO|Bvn=Qu)+^qIp z4MaOJ@z8cX{Prk8GagwKjyFeou^<8NlU0nc2!i;CV0&^+^@ah~kH#8rxzKB-1@@Q; z-(kkJ_S>;RV)H*b=XKxa=PN19Z@TGl6|p9Ks5nPPdP$~{PG1jg9c{~Oo&pfm}c@A}J{9*_l% z1gxp{ov*sR;$EJ(kZX5)0bQvk4*nI)#piZc&ecM!_$eDjB*888@YpLgzy6f}DEKYo zcncSy6oeE?!dZVGO5(x{2;XwZK77(A(?o`Evf)W6+LH_UpQMX7J*lI4FqXGni z{oM9b-PNp<{~SE(u<$vX+0$CagF4bHyoADb~*W zWS^7AaflwP|EoTef@^_BIN3w_uiEjw?N#4rhnK+tLIH}Pp7kj>iMf|*D#$*SlnCR~ z;9a-bFPSF_EfEWvKj}UAqHiQNrhU1^A|acvVEA?rU)9*|+$&$Gg`y_LOSak)qIrL6E`SU6StN=^0RwqJ^+(2+#l z=VjH*evkOx5*8W5nYyN%rJ8Y_aXLbeo|SR8m$}P|j3*gVs}Xt^5ynT5?rKZ*TJD&) z9Cg@RAshXoTe-fT+FHa=web6mOe8E9IiU6mykWu%$quBYdnfQSU}K(t$to&O&iT8E znulrsTrT&-+c(N^*{7Y~2+Xe_f{y3>^zFA^cMOQc>$o4!ozxoXFLQ-sA9>gzQ8nXx zw$#sK5KPcaO1j8pSQ0Q}VkuGY5){y#aqMF6Iue5O{*_MNeEcYV^`R{#0Kl>D04yp) z+7%#Vx>G(^l*?*(G3IlRs6+=bq8*4SH1u5kvs)WZ#-kazymJtmmFX16Z<|h}!S6nK1Ppsw$dqnangutWH9E( zT=J*I3&@^u67yE2nPvI%2Qs|gSa-Ymw5B32)u*NllV*`2`lv%bNX&1_ttjs?x4ZU9 z97Wj^XX{?#k__VDlcVb-9VSa2&S&rTP-8nSx8kh#V~)G`mec*Vk35gOJZ+5Y$+XnG z6v{^CzB<@>N~sJzIt)JOg$~~FhVjczY6X@|Rt{FL%yYRaL`0C4ldud)=rX7q|Jd zwEl^m8;MlbiYQ1pNUm}Y`w9>Bnkbod#ACA@n-lMplsE_H(ammtZY`B{g!V3Z^)L9A zpOi*>3zsC69Wr}ZlT+B;&DP)#>Da3;}T?#<7;Nsd2Q(jV-SqIR7#$~W1EI=zlkheXKFe(Tc+3xg9PAP z&OA+`^Hl6r=yTjE`hKV57lB5{?j5m+x0AAnLCxr)AgVpCwR8=kq@1C?X?24~WC_|9 zFa&au{gw_t<{m7G_zBx|MQ!9<;^2n&j;D~3ceW_!01`1J#>L!~yxk?^#^H3BXGXu=JWn7$eWU&V{;Ndx@La#{ zi0jv~mOJq`8iKkSei0z{kR$y2uV>>Zh^A=fXeaCI$!;1fo=3_={)zW`tC0S}zPHM; zDQlhbJA1n2kI>nHv#u^!9B*<008{Wo8S&r@*-+q_z?bg#mF{uyCvS9vf{j!@yn-#j2{ddU889N^rC&|zw0sv$ZOMK zzC*_3=K?YMu>6onLT~sj#bmZ%ImZ&h!*%5g6YIY(hujrAjeip@5y4Lh+wx+@?2K(0 z1tAd*M$-4+(rM$QN!!!tW+r~aR6W4nyX63dm&MYJ{3=z#q?2^|1F5D zF^`(rAi)ID7mldnkXldCs`~mzm5FykXGwcXO*p{UL58%UpA3KxwM2{x;aQ(%_ zNB2~P@Gq$p55{F{_}#JPIE5Vj_ZGf!#h>+eoG$fixYMCYqJ@JTQ~!61yUS<&ElUIU zm`ig_nBjT9pZ{->0DnPVxY$X>jfNO~vm0SgUr~9l0MBtTfW(ASp~^mxR}5|q z8N)kyW`h>K%tm5Y%ll3752O&a*T9Kvc8&wkBKR0k1p2_fC93%zH65vh7Y&!I53Y!a zX@AnbcS6wnzjg4vi|yKM34G6&Cd;@w^w`k-k8k;o{bdoPgcGU^La~z2i2hF(ptz)8 zJwo#~l(xJpyZifZIh{wm+jDdBEY$oO>?E`?NQJ2J%dxHXgkvWlZJ}RmFQjR*Y;lVR zeO!F*o75_%DZ==^p1HqPU2I5fhN|%qSHFKR5_zF|x7hOthX3F0@&4G?X~LAn zb|ZTcK=-`=Z779JE@2QKvVSj%Qavz9y~AL@?w5T=!*7xy+j8_`+4%qc>rVgmY^Wcl zL|9}QV%wDC*weahXS_S7L3+}=UGU$##IVHayz(;7F5s$NPiPDYz8+2{t&c&}V5QNO z!#B9j*g?yz!f;h&k0dx>|8gjSt-~oK5m46OfBum~-BTOgQPY`x^8r0X{NC;VofQm= zSk_*~2^+_>sLNT437^#H)I%;H9wGnz9TX{7C3FH>sKg#-ti8RkBov{<%+4Vnq-m`*S<*0ts?} znxu9k=n@uHEbq^EtBjNTP9AQ%Q%WN&^}7-bEN zT#aOA z;j=()J>Ki90@^K&G^CIyoultMI_XD|CdaYRCF;q1f*3(?2M!Bu( zPunl^B51?~meDFl)ou$=UJ~5b8i2*H?c8D%?*cSr!j6O&yvY7-8fFZo4itUkQuj75Iip}@OZm^N68B}w2 z8{U}<;TM=b8_U^+Q&p!!h4@mZb?vr`;K3}rdRk_fOA-Up#XzKlwv38;fx`_{o}g;2 zk%3RQ6o!nk?B!}^@x|ML&IkOSNvEtkoVf1Ck7l6Di3{v?SQHP#Sg_gQnzy&5Q8Gi* z%%_~sDjB24N82py@C-Q7HMif#vmdU!4S7avjux2aJ-F%QC5gPOZA2Fzx5kbah zP!0rWWL1a*uY0RgnrU)++D#0Gr|-aHdcZ#{i3HTFCLHf3xK==oaet(iao)CcZhoFA zIe}cH24yvG0# zIKUseRzC21lo|OvF*KY40vC)hDsNNF>PMvOyxzirEzRpL z(98VFzn)r)eOoq_%H4*?5qJYj+F%}))exHl0G_vIi5z%=f zyo)di+;e|$)b-M)>mg_}j2@nnY> zOmnFQ{|4nn;Uj| zrIlwnBrI(C$OrU-z}~>#*!0*QMw{Ptfe(XF#u3L&)peNcXAZ6@U-;pN;NL~5Byj=c zuBfhPyp9J9(7_5+?cb9v*Qt}K0a@UCngugH@?6(nRUIELk5DfaHV?b4RG*>n+<9_5 zME28w1v=T9O$$NI6!b>pywV=B3!gwJhf1{dp9R;>UY^(>u<^I!@RtkiNUgio&`*qv zDS;!v@BoVr45P>1Y#LB`wY9XM3uWMN*AY8xbc>sN&&Nnt%(y=k$e6$*bq)jh(TsYv zbd?MSrr`=`2dOw%Ytr-Fedgj)y6vp0+6(mVuP^3NQ_gjCK9sv7o#z^NGnfju&YfgYM zg+t`xVt%q1AQ+(CAnh+eLBN#)G~dgCGfu?YWhY-VvJgL%jBfb-uN++$Zuhje-v>fJ z(5IH<^EI>Zcvl2LWKdb2uNI6GA{^6ooBYU9JIXjDs$C%4>+OgDhx7?V0(FY6Kv-2> zu8Fa#+|@U7H6*3EtuuSi4Ty!%khaKo)5>_po;&ic#_w=q($G3GtMXhJ18kGX!3|M2qBgS1Vn{GgR-uHfl`>C z(v~QdQa165htWB;X)+@xhX|L7hGwdf1A)z>U;%J#v*ma-3er;clgOEYMrbN_<8--d_f#{9$fj<`0GR0NCJ>mvk z!;H>`%F5wgvsd&>hq)Sxwi z3XmPY_k)NZW)i_c=mBpjO9OQKm{0S-H2=v!uXI==sR1!QmGDEjaU!~&jE%GQ{F~j% ze@y`B*&pcBx-I+I!U;#|*~8dvXyyVLCFntd#@fw2XnzDjD(E%=?J>pV4twQ2xCuWs zGc&UfWr=n+oXg+CPNT!n&xdxGGNa{)Wl$nuHF0Si3 zn_|h%MZ#tK2WBlDf!ElwfS5v142T#`j|yHj1+R~cL(Rri<+7*kS>!bVz(MV9c=fY< z6KMN9oOvRi0YXxs4Hp7K^s!p`f#n9?@*U9(Yw#};@u zAYJ~PzLP(IDq})xNG14P5GD3HoYw<$HoU0jXfon5ePg)>zGURPclO6`6Au2+$_pIY zgWL@TYE%pqM@I^eB$OWoKf?D-y+k&6^Q2(ZmdD=j?KnX{06b!0%sQVI zE!ODRAtyI?bv7+w2|8Hd89l2B?_LYuP83db-Wd;o&(2C-0F8gp#86gI`4j(qMp09f z8-xxLdVZ*VpS5?*lPgBIa(zPx*lTE(^yq=Y(QChfDQ&IZmEK-cCqTLwn3#j>eco;C z`59;)9tvAuTZ2;^x)jdRCQ}0q#K6#mPRrD^wEEgw7@(Kg160itPf&9ZI4XgupzCiz z=k9nX?3j*$!EGwq$enS?gZiNmzNn%*oc~io5-^DZE{$&KwtJUa4A(*BGTv>8tL>8n zlr!=Z4)5N)`8!SEa5m!Ld4b$vO7L9Ejjoz{B3>osSQ<`RovI9Ux|g)reYf++QrNudhe_KL81$5F*s& zMYT);yL0i4IC<{iV*Dz+*dloU0=o_J1|yt?~bfEfMPXvsjTTxJyD^2vsVkWah6vf!5MXe%>zA3WznzU zw$R^r1UzJzMC2MKuru}ox==@`O$(O;xWN*4X697Rg%i6P7yc?9_2gk@?qJm~9z_3y zd~iw8!J%;biIh}O_2ZeA^U#i-Q_c8uoZ-4B!8ueNBubs6HOR zkGKca>TBc~NNQ=t?5{M?Cd3MOd5*r1;&sVuD;owtp5PJfv84+0qOjnL%c{&l%?9A> zJl`6i+LC;U+LV>jFY?Aw(B6m5QG(aKaA2lrr^|i1$4ku^RFY%+Z2)-jY z(Xs21O%Bs>PD=y%^&XIi-ac3`5(rOfzgIlO0@DeX-{A4@e6u?bTL>&-dBWxZ(WE|& zeEb6I8u0n`EW?Z*u(P{T%xzW--&agj*pc}R5$=i4J?2EUc0x}~LmcPEel+j=G!eOM zMV1gxk`tcGfp~|Y_l*TB<&L79`t#TBI*%?cQ8R*R8;$k1K@-kk#xNUnvqsH zM+<@?*v%oUgYoE&VDXfzhN99N0r&_d-QQ>lyv~%B*S^;F1iZYxtl|-+bd}BsAi=+H zlQ1z!pU#yx25$(Xhr{~1XYFC;{0cc2gk_WL&hO>b(^Yi8GhG@nK^Ij!$+@gy?V(0Z z<~+m5fa23aEWOs<8#=Gz^?+wcyAXc?fz$R6Noi($TNqNngqQkR?9P{i@GNbAfUD zF(jxGy|fiw3zQnfJrM6@Gzv}p^2HoF#g}{;7#QZ^)T{@Ne6l_w{{ewueEqZ)pZh|8 z?uf5He^Wr5T$=TWdA0QKe7*b@!}&uqrdc>GlUAWEtXRXI z+dD9GUheJW3$E>KI@AJiL!rVJ%@<~O*S9sn2zg)!)@IP3GV zYZk&j!UWmS6g$t4*T~4`UB0N$)5^up94-J{cC*t+(4sJW_Dsj}h0(6BKghl;z%bk(fZ7W!nGhnXYiQhwEgt;|op>N!nFD4Jj9yygMTn_sM*>Jg4S{?= zLJfEqvJo5Cx#L;^57~RQK$`>BiGOuQ_VO>j7uC@7!@2wh;>e9DKXjj8eGkDw7feessS({?V&E`4{xA*4rqh#@ev-atd`=V!>R1&(Irz4k_r4@WE9$ImmBk9p#g zkj(NOjlIBQXJ15+pVMzmpqaw}^?2RhVn+G@ak)MML)0@dT*_jv8`{{HmzbQm#n z633xEG}gI{1XxuaIzsE%Hq!ufwC)c4jxhfXk!qRRu<7A+@7!=&+bKWOtKsK>`T#^w z!Ttd->n$9k3nR~Vredr-mrAG0)lTR^j?8tnmIzHzl;Kqvk^WlT0u+U%$N{w^NHXy0 z&VkDW3>QxjDz@JLG%U_*3L2mxHNqPJv4Kvh?PM*ZS^j`7ApPYxdWUmr6}z3T_5hd2 zE}UK#CR*sNF)ic=KjK5Vqf)Gyr8WYOxH2C7xZkwM?KH58tJB3uus8@VPvOu-#9I&M z$+SYo!JPfP&&M89a3TqiE?Cn0ZCeV7rQR+Wd@ajy}LJ|mwT(t>u5dr zDkSz;Pfe>m3ZS>!{R|urGQ3aTOoWHh$9VaE4b`YkX8f>WMf}t zJX{(^_DHHWHt7m_VtnbLW-@*oAnbu`5R-_#eL>q#umd2pQ|P|ChVFFuKgjO*FsaZ- zKwDYmB7B$1UBl(!hkws!+lL495I4I`gmMh=JlKz7dDaN24|GHazjnmJQ?SC&RqnM| z*mtyieAH%i4cPGzrh%PAlME1cERr|0v@A~PX=P10K|;F1HmTBXzTy4fl(F7*Q*F+? z1k?@+Q&{yQzLEd36cS0nnJO8OO*-5fv??B4M3X0A;2|y2V_G<1P8v#c@16xTgMrEg zcihricEbDjTlE4*tIoSYH#Si>=fT?x8gds|JVlyP$CsXgrTI_v`t z(5)pHJ}?7GGZ4v+hUsxiN^fIh7d+NrtPFGt9=2bFJuoLKGTlpPG$G6pC)U%l3$)Lx zu%0t+q@k&RsbL^oP%y)lAb3$A@2e0Lyj>Xpic9bVmv8I`yHC~Z=9m}*1e!QP9&TEo zO+(!k8j&DeKyaG&sAaM8G3DZ?(@3RH*O5`XoH+w5UC&AMUj>o$2M#W@{)hSn%BOFC zsuOMO2lnxWf0=hdBv66j<^!qB;b;FhU5=$j1UdgzQ@oMORxfv|A4Na9*Bgr)r6A*rZY8iRw6E64sJ(jsfAB|87yg0Fwc<^TRb zAj{CrBxUxoYyed8aI5hXKXBiA16f*pV>mkK7qm^HBhTTW_+r9j#fB9J(pCpo?n4sA zA*fB^d;jkM2F*bqEdxM9J_Y0gV`UuKH-E#J(mHhiD(;{_mFU+uUW>-C_~-7jPp>WZVCiLS zIee`DE32qAPF~)ARA|A~(K%`DaysH1)oo3Nr)r3B!_k~$cr6uNoBH3itslxg_|QjT zVY*b{CYE#)tMCkfRYu{$wC#(j~Dt~zJu6zm~D50Y1E+wj=m#m%mS82otn>Ss`c zbSJ~rb0lOZ@yowB8a)h8Deo|68HB!v%^{+~`_$aDidD8wE#hwTe-`nd`25>&j9r1J z^Lfxp5!^=HpzfB~!=J-fyvL}6^mSsgk{Yx}jI$p&1dhInvRr&ObIQ>nE zh%eQ@!fT1l>gkq{CzA9jH+h7J(iIzqFL{qCcNmQ5_2~BK=l1K2Ost-kl}qC1YTT~; z{M1(e*uLxQh&C7XtHob75C6T>k7}CS7Lc_(LqoP#8}GqDZuRmd$NfV5Zw z_5!L5S2fO-#o~HcCNXL{jg$Cktd@Ooj-wzGMGGLX9Ny$FlS=jrOx9PgrAZHY{IXw< zD(EIoU4g=LOwFSaW6QtJCn?0(ua8e&+~A>^%DG6o)LJpaAG0b_mH|O%+X1?)!%!^t z@cf!s*FDJ}qHt$`kVMc5>Tv-1lO1lr7Ln^4AiE01)e0j5(tz1Fbp-Z+8mFa_BgvG*wl#*W{6Hwi+KQdMvCqsp2% z8RB~>vZ-fX5}~YMrI{ZbG#i3&!M;|5nd>a~^5E+e-)gUCFRt)I#^E##86<;eZQdLZ z3*kfk)_Q*;zpY^yt>X*7ZdkZ){ODx(Mc8IG3t<<=ZpJcpOj*ML75dKmHP5u5rqKTqNSCWCrg&C25 z8~Xb@x*%GzmXx$X`j~rAjZC{Jt8^#&BJNm{s3Pw``)unui&y#ED&zcVm`DHpgiEI@y15KVr| zEa^M_yUj1RxDq*8r}hb1+>7{52w(kcQoOBg@~Rg+I6o!WG=B7zxVU`$d}Y!;@X@{Z z%AxO-Wv8Z~eY3o@%hS`CIIZtB@lAJ#w&8VPh=AWh_KN+y%za!a8c}}XQt&Fs?%Fpz zMXs?rG=60qC}r?5!4kU0m#PdAj%k^cd_iaZ1*3S>hjkrOCk|FWeSZ1HdD!>J=E&+S zhVI{T3d?2=cqSz(5=55RS3@I~|xyi>mc?Pg{VvwjA}EAhXZ8&gy#jlx^TSKOf-lOG5S8 zo=dA`OEtfu*4#5}4~b3Bn>qm}GWWf>h48n;0bU;p=t^h$8$dIzkcmOP2`~M_#iCg9+jYhk&)N3+o&1L zHobQ*D?58>^#HBhTox&Y0a5(CA~jG{gcu6M;U@qdLSSQCd@{CcU~(tv%c?>0fSUp| zNRqs}&vs8LI~$SsR5J&B=)MBYxFpU3Jr%j65kB&C#bJK_iu4-tWVQT5<}o8UfZiW% zU_@!c8_3u+bYO5=q3VeiNQLkr#{#kZ3S{UVpq_lp1T&c=dbA&t0nCTb42Pl8)u0v? z-f+uG+FL8KW-=V5^rT&sWD{Gvb{eIS{r_B2-U-KmnX}O5EfLg7^U!3{7$4*Gm$CG* zc*E5rj2dG{A2A1Iy;xQwx$QKP3#StZ;^b&Gl!YeB=Ck)}nLGi4eA8IdbyXwd_;SK1d z?N^0D@C2G~pxF1_EsG7O5yn1Z1)hLTn2pWofO$v4F%#(4mmU+DnwhO@b?DBmCR|!c zdt%Z7Adfm<9|*zKZw)BlsV}rnY#I)_?8p3=>7KBCIvpwzk-~uRqi8|sDQK)gw)L^7 zs4pc3MC>4?zXNhe2cvbc=pQV)&0_YZ_#QbHje z(xgOG(HDdL1R9B1ATgMlDh6zxTeo%+8NvjQ^d!>%5I)<{f)-Kvlv&Y=0ltZueL~io z56pwk`C_0oug|dS^VNpYiaUv7Da|i~LpVhL(9E$+IFMeg^6(Z!jUa}_wSu34K{|WR zFy95jHrq$yqaZZ^fukoQeb8@l3Xeb2NrH$ir%r$w&sMCZ*};~s;W@w8;UB}g%>Gzh z_!eQusRJ$zzBROqW0kMqj_>(=qB?AqH4F<*6NE2J0 zFU*mpnXbC#bQj-!C+h-IJzh0(G_fvC{!)SUSXk~-z(jglJW#1{nH8^~wgo3KV}H%r~l>LtXov?2#m@PBWDDO5_t2dos7uiyD} zvXtkZH-BV;zIA}VfuG<{OuIn(31sAGIyv%`kn>tT$5Q2-zjcgx3Fv@Y_S(LG_s*I9 z+g03g(5M3qej;8w2t+|EGF}9dF`)biT2eRoTtRIgYMI2Lbzo=UtT1U7&_NYsdJ^(* zRXepzN085O!}Ekp3M4c5jjmn2dXa?X&%tRBMud@LrJ(N%IqWCn&8?fzrq^Q|5i|igaj(2zhdRm=z>7;g#x4vK^09U!!G9R zt=qR{Aq`?@%pNo9I-do_c&HVAE-ZvH9|$qQNEVn+&8U$ln$wq{SDc@}zF0h}3(_k< zB#Qa)0axfgX()^sg{qkxIo~N*z**joj;OON;u$z(5XjyDl_zP-rpco8A5hw(X zR4_SigSZ`}dut&1{qFty-kzRPZN?YXpc`Wi`2vtJZO=~#96!!?8eLU!q&zPGxZ=UNkyVboh zvVW_Kb@Aw^#JRLIT*m)(gDg$LcCf%yJeSkdA8R8PmnoN4Mf%}33vf12X~4Sufm9EC zT~|-XprlsFnnAt=Rr!7Yo}G zFmGpj)T>HmZnkL)1o@$Qo!9mY1dW{HV64WOval9IH>e$yUssosI=4Osgpu1k?llA? zB)bDf2{xWVcmfCad3npR%8-@>ZAcwPwaDW@;CBJZW99mzZqq_k(D+5XPM-Q*sOQDH zy$nrMNO_;bmN~*fwx!x2`$@)X7!n}<@f0rZ7xmnv z!ompt4k*IuV7C$j$k*<>UB~0wz@Hw_($tLi+%st-dIF_*P}C~hXIhDOyIu~*V!at^ z8AJ;&%LAfa=F-#bxo>#D_>&}#{9IsEw6b~vL5eT_k)~$j1`{KrZk|El^vk)I1k{zs zM{ZmB2m7PPOW7#hV3uDs(*e~}a=;$#%@6+imAzW^^eY&1@gPlpPVLp6Xl`uspc)5q z!Qh6Yk>ZQh0O`=>8+Y|8ihU1rzK(p`p4pR}Y7(gBJwAE3oUi=A9nVN1=+H~xzTFPA zURMU39QnTOh9D#}^P2&<0V)qwoO4ecWMpAW@`5d3 zC(92uE*5Bl1lbH3nw1`PTjpaDS5j3q%vHxwEm!U@@of)vf%cqE+uq(D45`R}L4($m zQNk7I0}tPU#Bpvt`TnAnm$c%Q>1?PB$w%EEhYAW-E^YmOxxc@ETD6po4Ug~h!YPHnfg*!WU7u@31pY(CFutvRs7AIlJdJ3Y1o2bNtoL!$ zCiw$Lz!H`LExOcHe|#t{z)PNmO*&e$-~wW}=^!2mTcLOTp}Mm2`brHR%*^<<&bxgt zD2W^hOpwWp6Yz>327)4}C=MGLf*>mZ!>r6qYTc@W;p4vSxnmpHV!$H>rTF~erP;W3 z7=#2i)&q)7H!g{XwFw(hDqV4ageCf$VFw-<#{s8^X*podYi^L|$SG-0MMDc)tJrJK zu71Nk41;9B=T7X|`^~ z2at!5#sXt#37D(!7R#n{fS2Y3_|`{-{zk+1KA-kP%`8I!rW^U7_`eN44MPRW zY#>k#*7VbKgH5S6R0N@@^z1DM2<*<1l)#Y~o~MH%;sAI(LSab%ZbK~49tvc2duDV% zjseS(lmXBz#%@Uf^UD<^cHs&#(!#>u&rrB4fP5#62Ljdu)C3~pHcP#ZkHCuC373Jh z273#Vvs1h}Qc{g^Pz!`FDj&__N74*=WSCi=$H*#)dH~x}z_=*e70MX$;XMODSgnFj zL=Lk6GZ*SIz&8fR4BL5Y%dzz-jJz8G#+f}P!o$V-vK)-p8DdI=dEu!wz{Oz$@)(Tl zNxl8xfdep)0a?J3K_2~Zg5vmRcI^mM=b#9s&mL28Jeyq$kZ%M0@K-Gwx!4j=ldI6x zQ&tvfYX(_O7#A9K2}+f)u#^uKfd>R&9C%c7EWGGUU7$mC=;s~)hMgQLgxCW+93L0A z7(dAYAqj{+^2D_QyKGI`9UKSi4InqbsMsaIBrxj>-U$k>7MvoE0mVQRmv-C-=k3m3 zivdWp#+o|qo!SBZs3$OOeeu&(xZ#2TBb>Md$PL@VxOBQZe)qP?^%X=h)|-k%yrryd z*VH=s4X@yX`I+-t+Cx?+C4K21E4?-B_+gxmyNZi=)vEdYx%+PceMG7Lk*$XAafqcsEATSY;2gZ`PGlrVKg&sE8VqVKu-h1oYh#6 z`F0^*1m)y`t8ZXf46N=_PcjUsnh-5Hf=ZM1UVAkU5L*1ZWEv9;!{m+OFs!HEg{?w&c_b?%BZ&1F>Mqk8Bs`9Hu{Cf=%G{KKl3-*4 zC*DqR)nSX_L?xKHb*_|3`$nT>?ME(afE5S$U^bj38|g4MA~MpB(bkdq2*`b+L)O`C zi~!9XKQnT2mLw)VFWrWqcp+H%<{KFO&~tdo83q)}$*e#{hbgTMY+nBj(Bv~J9t1WY zh6;8|&gah`v4ApS^Iz+Wii(B__#LO4<$Kx}yw&-(GmuXDqPBIr1R2YS;E3rxSQ9>4 zKFZUXo@45r3GGeXu@!S!D->~AqxAbog?#(AdGscSi=c)In@831_`dZunst0O`%>*# za~5uijfMO+$Ks_Jx7bx6c|Ueq;Wvh($2mV*ftmn^sdvPgJ#zq?e zmaI8}+}$5m3aDv`+&BLk>?ykGXOp(2 z{ZBLkMe|^Ufv|+DD>tpg-bR;vK7}b1{dJEhkE*-JcbJQ9E;4ABEY)`=G!15Zq{|{J zh3xr@`k_Y(PuBhB9JkDhDu+Mlq-OSN#ER_ug?t<@6lu49EzHe~d_#Jk6j$smp3xb3 zfUCy->D1}doy>QwFt|95inpkKn6raq<(?G~hMY1ZT3|pTv>#l=PNEFPVFf5D*1mFt z=m?F!haFIM9SoL<7OBuZ1|MR(_tooKP||xQJvj`_GQL6kd>#@LPc5ZyUEXqR@Wsy` z_|jp_ZaD-e7kj+{LW$Psg5F+L;hzB%oCdSC)JGURM5x#URY7#V_Ow%qh& zVJ@J}8a#pK%06SZ&OxQ^m+1yX!LZ?2hSplk6#aJci2ALg@F;={ECFhZ_cg?pC*2b51BFIk0bmA zSyFwqukZ~dvdw?4l%FyUi#tYZ{+_-N$LHb!Aqo`w%e}--(^js3JBjzz$#)DtP!t>- z1$GkH2Z?Iz2G;%->@ln?ET|KA`3&Oq!V-@kkNKQB4+`XOKYVDMT>*s3fA{Ve?3 zE40m@0{whQW)p*HL`}ZFqWbntx!1h%GF`JpV;t#^UCy>)OU-WGB zE%Ro56fg+CnBymv@`2Up)igg`Yw}jDb&=nu80-Xq^#k{SHK82V)`T+Lk~y?i#*U$i zKD6!xvgJL!=+D+>pPm!REDE=>O>gm4GR1dH+IF0c2_dJo zhEn~42zK}gc*Yh_=IMKx&BtATGE zHM9y+($Q&g+0|ZL0Fm&ogjcZz&a3*{br5A17Z%RimjN@vMylg5B)vxQmjS&63sf5)(F2F_G~$eh+M1$&OGr; z)#H+sf$pWmlX2!dJ0aM83klCLFd6gArFZY&zt7JfA9u1R_+uOB#rK;?QBhIx`Qr_3 z^TQx%5&_Th0AVI;4k)$T z+q1a|Re%Jn+>|B=N@GI`cz~6tnN@WPGk{Ms^V}1{^0)w6mk5x*$p>RhL%> zbp55J&w~I{dMFUa71LGn5*TS|jewt3;k=cJfl!+h$nDF40Yo3Y0N3X= z4YsP0#WDurul>BZs^2jwSgJrL(RVu4466&wzP*Inscx+s1Q`g;%bfRt;oq;QVp$bv zRGWSX+2(4}C^prama5+r79&8csUWR*e~gW6@Mv@5Jvp`_cRD#B^&SR4d8bJWC3&XC zNTujgY~7r(TNxAD8V61n-#f8RNWS2lJUMVeipDQPS}DDN$VCRIdh^Wxhout9*igFd6hp)prF;2u`9wWr+ zz_WB(tcKQ}J8PgSIDwOXay-QrYaJSidy~h8*JdcQ0~%NT@Ld9ebzHx-w`|APn|a3X zkKIRT{|#W@B$3nkK(5CA_cemjh%MG((kDgOWH@W={E9$M<%N>>*Jc$lR!S=7wnt9P z>yOZbWzj5H5ePg{m&0ElC#{ZaR$GK-2K++K+K!U__a_wVDku|OZ5tU*R+ z8c#g2j^=O`L(;|cM`!#YGSwiQ@E(EQ924xCcF;+ZIhlC0@;+q-w)%# z4wKSexV$463<=$dS1xnsIQQ(+DkS24P1IqDCr5}ti0QR|{i>)ny8Nk6TFs~{m1Tq;i4`M%6z#}gBBoZ3QWRqjx ztGkn=g832GH4)Hklez=X9nSLIm+^)cxztQbEVsLZt$bRgBp=Q=?_D{Ln!XYdX4e!=nHU?r}v39K67&kH}T3KzFfW#CZ_21<>eP7rKBv;t$tbIy5HO!MwL5x)=N*x8aVSHE)tTkKQQ&pmCBpGEoKxWuh?xvw(&Hk>m;j)Zxz zNf*&xvz)_nf2RO5d2L?{9qWXOeyr2S`V}iD$K*@K z6xaWuUO`c@<7&Sg?AoEgdfi}7ZAXt^A`;u&^iM)^>{gg+Nmt?#WG+pCD&(t;kea`}8&Rfo@?3Q_yrBIFgN;rUr6Yd(5Ul7T^njjYv(CgQ3@K`+nX|8oJ1 z)+PXB0LOW*!G}nW!(_B8Cn@?VDdJL5d-iAgYn=Ufo~8#;@ILCjFl0mgxY51<79T4r z*8O7Z&F=Xg()90q`z2t$QPkI`aIwi?oLgDtswVD#7fq)$rOy3UU6pgfqU?&}DNq)n z3AI3!K&BI+MbsY+M^Sre`)SsL*-50i)@ep1mPf4RRLy^#bGscGqBGTende^>H^r9t z4;S$Vg&qh^nBmfgM$fs6XNDpJg(;hr{ukjG`^w8YIdNTAF9=Q^46^g`o~#PG19TM; z2nQQlwy8Jg@{L@* z`PCrqr2l)?m9NE)gf|<+%*YrOxX9irvf0`FlHejhNOHKeyxP(n#JTz4MUee7XmG4` z=!%N?gDDH%+}|Uy=UYa55sm@y(h%H&OzmMWKnUN3Njm!m2kQutqVTp^>2YU1xUE0W z6y7zT8)s;+YPRn|=y(wjJF;DC{C?r@Y_|F?r1FoWxU)Kn7Y?AO*CMy%_={ENpF#OU z8yl9IGT*?VedmJgDX1xmkB5h}f^8C#l9D-rqUaXEk_BvKd27D)&8eDvZy!Z8#mD-3 z+J2}~awJEIR!00F)r~4NwrkOIY}OkYi+S^kL_)+|l3%;G52sK_pvzQ?9Dyl-boG6e zv}fskIz5gO!fbj&kQn7*FSwLDJGPyey>VcAY6!vnLq^B{R1~_8+t9-R6BE^Z26%?Hn?c?#ju_=te$f|B22 zUy=&u&Aw8L&{zGP_*?svK;L<+a>c)N&&Z9l*dts`Oz`&MRXtH34oa}{CFowd>n3WS zi(2ST-yw{!k+&el5&yHcrI11ic&KfZAY1I!=Q!Q~@mdt6@;UdLR=uUC-!Qkd!<=5A1T(_<-&Z|*c)6*w`c`Pd2*pVeezxkW&Ew63lBZ!cJ z5`yMo1kI{|JrUJg19_y@{=jvc0gzfUzv3(p1d9c6gxOr(HGSL6ob%p~mCXzOW$l{% ztf+|F23V6W1Qd+{J~gMFekR(5r9<|0;#FMBM9hQK5ePn_&D%h?%DtsInA^29=^=f> zur#@_<$muvv7?kHOQ*CQsLhMqNW8X#mls(3F1ZzPgH``U$RK7OUDNKXgMFJ6Uqc4A zr11cJCOnh)1IdE(XBltcNzsIsCs|*0k)oCfTG4AYA0KQ84#v@)7pr@&11*6Gv++lB zrei}&5Y!^y+2M$`f#T#<;3Zroc$;4Y6@bYG6y4jWK&kK{7{s>6fTFH-z5D$Ii5B~`&8UILZ3bJG1E#;#nqM1?s^Bw!tvb)^!XIf zoDCuO(Ec?GIiv(tMFq>_HaT~ud$#M4%HFlvC<^#Gx?{m~C4V&-&s@l$lPOLPQy zq;@Oak`l($+8;S>?KU&Kr9$lvMxvDn7cP;5GMBp-M~|+2YV((zoSz=m?>>MGl1~oM z&s9~6TTkF#n@g~_O{VI)x52QF$JOrN4+IQiksGN&0-Rn_B5BOlq{W@Z%yVdoJH4Rf z^Ji=CxO8b;JJQXyF&QW)!mF7yvW)}^?>fV)2b#viN``b~Iu1^fJLFHHkjU@f0j;M$ zyZ%YGJ%H*DU>X+8J_37S)oqX#IA2Yl+;P8h{WPMq6ZurdlBD7ZqU{=!Mwj)qRjE55 zo`=!^(3;Hp<{U}W_CL||No101H5?rS%N)=( z@ST4=C2Sa~hs?F>l>|t3Ho-EMklI{2?0Sgh{ZR^_MroAAgJcxbNCG^+d%qo1%4F>*7K+A zkrE;Acg0iopX#5GUduv4stfQ1Z*_C?H~wUoASkw?2aW@eNn-`LORFU_dq@p6w7&+F z1}ez7$2x%8!sLSSD@j>t=`X{ioZz0Bah>gbfL%Q7n^tb}m|lXhbw>>iPlU!tNFA}DVDhi7 zjdld^EIc%$Jx0>FB@Jx87xW%qwG=#^nJTNA5YcNiXAdGYP;sUwWl!BmmX=wsTh75a zA2{I^>GUZ%dw$39W$LcOiumr`F_F6GL3C)r`SKffagE^XQ5;b`I?FwF7WQvbS$=?KLq013q*Bs63rGjmoJ4f^e?IR$kINg?nP27OO#0T5*tsr0RDkelNNg*bqX z_*BH?{{8#VDV}r-V(%=d@Zuk2VQXs(oB^myx8wvFR3PFUGAu(fvgPTDNcsqHM&N%U zl`^TEhOKv+i{8BhT&Wy5Lh}v{Oi3zW*!!)cdade~JjGVQkuwKE0yOv=f7sL012O1p zNq=t*jTbtlo>iw6kXA}a_$q|Fg+Qoqb#N#H@aM)CN|KoHv?+#L z{`p5jeEit(FqjT^v7Cm6hI{w!9sa9;&*7nytsqs0x%A^<0N6~&!$1fQw_Sz#73?JR zIul5!1}+QIkleeMN&AF6MqvW>6e@o}bsJ~~3XuQt8`)AVHsC_}jc_-CU#MOLa%45_ zS$J!Pu~{+E(LcG5a6>OtgYt&5vVG|3=!tRyuUI>XzhH9S)fq%*ptE`cT~5ImySIGW z^Mtmu?q{>evPJ;LOWeD+wC4$gB&ZuOo-3WpV*`Db;4Oq^Y7%xO5Kw{!4Bi@osKVCP zE1)$VGJOUgj$YimVsK2=UbA?{wvGxf5*$4!3YAQ{{dU!R<>`df^&zCq<{Z3aWP(+g z3)mdMWdOS3?Y1mEj*jg#6L?)K3>={gD|EcxE4PN4KM;qVf&BoqtLOM1q!&Oe_~K3G zaby#83)PH0mWJ32+>`Mv`uBNx8^G3ewb1Z@=8cfo=147r(*lw;AZGmWyM*M+7(6Ak z1W`jaXTNQs?n9NS&#D)+D7CK12*Q6R$dS8z;Q{R>Rj`s`Ba)NDlm|-{KR!YIhLY8L zKHh_K5t{GxT`Am7(d}kPL=W4+#0|Wz*{pw#o7Mk%sZ$Jjk0aKH(#M(p!=b|)jVh+0kUnVgK<{t0T`FNd#!()|70RR zGy-1#bL^Ad`iP*$Xp~@GNJIpg2Ol+dl7tuI6bxvwPAU9NV5`owI#Q!(P-XrHrcz}z+yydciSi5*a1-NsbuBve+>+zU8_^sWq6{lAbP_^M+M%U zLHS;QY9o}P5j$=HxCF`!`LfNhhC{_HK0!-?%r^mNs+#ZK-heOr1Swk^o44tInxI+- zbTH%*sLf?`L3op!;v$lzsalJuxx(-<9wnfYEt&d=G#*WMP1r2hv{auf4TM zV-(=|UI8Zx2?YPtP*J%==3llx68|iDtfw~%QgkFKz~f@wQe91Lxy^@{bXQeHrFkG} z28e;c688ir&8z9Vl9tva;F=GV@vs@0lA@Cz4DtgU1OBit{RKcg$bx3DWYBs|;-`kX z`U)_90o!j+pfryI3LCt}yg*fCJG}b-R;dsqvopXa-_0bB>xX~oTDHp+dNK{bU8cO4 z>iqQ#TodpOA;Eg`?%lhKO9l8nzreg@k}b$8Gm*SiGNlWUiw@Y^TBND_K*W#!t+e%G zr2mF{44Rq4p4pqV7lONitRi>U z#XNl}iA)ix8-;P62p;&NReQ z!1t4MBIxC}9i{wv33mQ8F)bBZxTmGPkRUR7lmVgCl@U+SOWSk2#5)EH`ih^ogXF#>exdgXN8fOfOJ`fRk6L;JzI? zF*deM#_)A#XL>|;*rt~5W`D8@GYrtG3Fui!bufsmq2)NtKp>NgpWhA4Ab?j};Q~-H zt#|2Pcr_2*>mi$E*kF)N!YQ{A$vX0GGk*UvA%Pww%J=WC3IQaw)r940a5R zYjqjH3Pl7YdVAiqgPOBPej# zv(M%zV%T7e#aL)P=f;?0pjYBphl7s#UM%nCNjIMn!wlTU8A!8(5zXt;ub(aIs{Ocg z*qT;H_-kx962WIf6+iIpo9nat4~ldiT3FCiQhxlIclp(|6_P&G__XdZyOBSL!Z{FE z*)xiB^$$m~#5u_DOxv;^EQw_qnTVHAzw)GHi^@ zq>x8n<~BJF>!J6;Q`@$leak9sHVT{_f`pR!hp0uL`}kT6xD|g%Bk&$$fokSQ?#y<} z%7l4;Bi#bQ5G=V0ltqKL_T@d(8teYVKQkS*bMEy<%IgFvu}Xs&@h`L)_ptMPmbowX z4IER)b>7!am(PyT(n)a{_j`VPfiwDodY=;l!bxP9u`A6R_($`fH2W)j6idEm5TU^- zS6C+?B(wl`zU#}TwtI$Ld74`P2kn}&6s#iG63hoE&&G~D$>*~cL*zu-0poA|Z(XIA za&!MY`n&h{xjAx8^Xi~!GxFIRMHb~^PA7~RyK$(QKDCc!;Jv)4z(ukq!oj5S@+1iH zY){MT;Fm*Sy3doj=@cwQ>4X~y%W4_qfy-_GasI6H4K;GsM&P!?^Ms^CYK*``X;@*@ z^Cj`>zGB?3s86Iz{Bva|-Ujm7HXg!~!L`No`SLHG zu17Mpj0ij#$jV@bF#?HM&Lb0lh{jV;gZ?1&S$o`Rc=I z8y+V* zCL!@@bHiCpMqNYWHw36avs;{y-+z#eE^iFahDLyvyC}s&v|TDv&I+fR<`T{$4188r zdlh*QI_KT|?;Amq`giy)%_SWMBl0U{ts!F%tT~HT&r%K6FaBIoHq~N{whN6WJcAl6 zOtJZ2Z@=@fAtrMU&l(H+-ya$=-j)d`f_p}f+`;aqHSyZQ_J%hn+C-tq%i(KK zVZ;bk{H@(wU8@bs?}G~s6)B^mH~dZivDN=_1QTA-~~&w;*F1a9>xCO~a13O+_>E|#O&!YY8;Ioe1Wqe7I&Yx! zRqNMhJVBKPK)aDzdRIVCB~vA1Of~A_Z)~a9k>w`G0TCZ9J2Vtz?8* zbQdpTN7hev2OsFl``#ky?lxSsbgylcBhXfxZWtDNYH7saEsTQLxbfebo-QB9Q~2Ll zFWjXz5P{Q1oYk9Z;wWnT$>EyilDI5ILlavwPRqK96!Cg$-(F-1wa8ji$knEztl#E# zxB`YMF(*)au>P|#{zdnNPw1&8S-Rm=A-+Y(M{wVC!29tu92bR>=cfLOGq+1)if~p| zyh!WmLb5eeSf!IKNdaE7qbb<~G;~BT4+CWWCx8C^J%-sfGW3&8LKc=YnAhm2Ql?ae zo&13co?J6gx3x!_Q08zJH~D*ttls{Xg`F)3mAivOkw%}sO-r|(#h({uLj{%&QAw<* zon5^F82T@2yl#YNULH$V=fV9=J%JhnmZLK$hEB=w-_RdD|9xe4Q}O&$f(vr&yTawG ze)-F;mB#dyIL(0?%Q}wo9!`C^g9AHJr;YYf}AbdAnTenb{vikGWVn}<*VywP-%xmCHX^uNJK!xXHNHm_)2iO~uG;MCJN#Ruszqv-Mi!AP`|BYn?E{(Vyv>x44Ek z;ke&`3xtpyt`elDU%KSeNa_(k8kOKR#6?KqrUPqXpAc0 zp&%?(kfe9`MAlzjysXnd!Z=rAH7-btxx;9UC9f3L`RHeaQY;NOFsq1)wjzH_2GJ9D zcYEb4RegA~VOTZdLrHn*`W_9wpqog4)^fDczRKmq(wLd4=PmPCGkUyHbpNttiKy<> z_+Ni>CX&90kEHVCs#fj|rqh4?5b{$|gJseZ24k>pSJ?j>JRceP-PhhZE;dbg2?qDo z@dtK3Zd7rJ6ff)0U}C-o2UWo@vj_E&^e>fi#@(`Ax5= zjVfEbRcer&Lj-G3Tpab*5zUYKW^k3Ob;JvTsPf7&>Y}7m5rNL{Dd8*#tT4?4Dj{@CIrms(eK?Fwj|Gy}*caOuRhD?MKewp{zZG1+D z<%*gftasgcp83WW%Qlrx4Wg!I&aS9@%-LLXRN~nI38=wDa;(gl7(*>W8GnPP#P>{VPb@(|MrG+e(>@C=K_@WGLqnk zW~G(be1;dHw6|0{S{Nrri3i(2MF=NGhvUO5FkIc@$QJ_cPoAJL?IE;}w9<05DE-!I zGd7z-#NnMMvonkW)W~X;t2XZk# z^{Yj2D~VeaqEd3b$;|oj5}1q+0y|K5VaXx$A+KM+2cu@r?*EzCXI`lWcJ>;t|&{uqH8q80ASh-WVn zbx<5?t68YR7LA+462b9Vew$IvX&Ej{80P;>QMdF>MPL!imYWTlP0UWn|ivwVd~FIl^0DVppt zu1pB2|5rnAdus)L(e+zSTh|9ZFD;c_@2SU%2xVvqA7n)F-SE^tKcbxzQ#bP;tlsVR ze9ExOLxFQlMX$dypPv#dgk#zbr>hzNIM2$`GS7(1!Rd1v%o7g>P;BJ{rzUY@FEM&a zQN$16lYNs6mkmp{RKZP&Wo3DyAV+q)7x4idr`$8GwUNL!Dw&}r1Dl#e4joliqxwZL zKb(&rMLo1a%;5mmHSI6oBDk@RhTP!ba^UhxnNKh0HSu%vQ!i!sdO3uXwB>YnPvxD6 zn$CV8AcLR9!w8eIXT!%*d7wTvHuBY<0e%wU>P0|NzdPzJ7p=nf@de&3nkoWAE0>3q zM&Mpg^bZ&$um0i554NAv*&>JHZ(YOINbhls`>dgK~!CSlHdw4Q^oL!`sJ<9MXvh6<4*K9 z>4n%;ZxeQlBk&aNM-q7VB@&PgV2(VJx8bMGL_aA4E86v&(Ow0F5gO7#XZ4_lA!L^m z5S;!+`-)yX+2du00muXiq8?pVbJ0a@Rk6j^dM$cBeflGRWSa>UrSfU6yJ+_PvGjCR zuNJoAk?k2r@piUZgo{{6~rh3!A-eYj!Q#b|D zRy<{%mrp%3BO|}bWM}M5Qxsn{KZrR3(A?TmPNhMznM;jsIQkHa@%i>!!}n@pEG*4S z_A?(@uwawA6F&n%FHf*aH*I;j^<#8&Vxncxjh#W5vgcF7G*uA!J2*MSViA7z(fiB6 zDfs6yM)#$~ng)DD!*cYuYQ7-l?W$ShGz7JhBCEv#HA{4EDvp;&89GGK9aRc!Y4gO!)aO?AK|rBcFK2AEt>vO zPv)&IwWblE#F6^0sHoQD3^%sACN<%sO>O7BnFx_KMx>r%< zVreiE+Q~wM64ZLm&D)Z21=Oz$}g8&pIsQJ zPd3#G%q-QJ>6#3h3zJ_Z$}oH!KKwCnW@56i9J>_RX}IR}_w1{fw*o7RsoVN&GtIPo zk#*PAH|(&`xPpwFJW91vQ5Bx~vZCZ<;kY}B(qgwa=aYyZYI9(uEx;Rh8POH!tR9`9 zTUr^wpOJ16z066k?MW~-eTKugDV49sajz~X%c7gh6Dv@YC%xz$bDhcA8Ll?@@$F+y zx;)HvA2Rorm4+B5eV+YT=Y0vzPnPl68x(O3Z$ZrhyV<;$k;edklks%-5lj(< z(LKi1+hFn45Px>g14KY<47sy}j#qO@gY3F2R?V^*cTO16d7r%f92ab|UPwp`_UR}i z$7;Y}hf^Anj;=}ZOv9-#votFi&Jv27oQmS()HTZuH>rZJ08i zc@QPsKxP(JHfaA!)L_^Nl+cWnjP56+{o&V$KU(sTseyUV7vJAB1`@%Jb`9c$6!ifC?Pb+HdJAiAoj6N9-GZ50!*r~k z6|TSPn^|T}Qv+k;`7uY7;)_!DDI=;7Gf9EVi^gNV*`Ce~trV(Q!07-~uiXa*o$lG_! zLyrY^h*F(93k}LiK7SF7XY>A^#Lyrnb*v_nT~MO#he4+3CyG&tVV9|=7jaEy*N^tB z(arA~3#PtSYR~T?W!@p(wdBwh&=8I^OW5PFI(|~bc?+gw`s-^+W@{JvnyXj2!%f*` zlBToIDijuMcI#d9Wf)Yj4G9tq7!8oCa;GnVusceDN!!47AhYi4u6*(mtZ>q+?7owd z-}>%4*18^bOHNkCQ2xPBk;QI_YnPOLNcH|0H^s>I()_`>{@mQD+oxvkAE|Hq!Ux{5 zCJ$-QQ@9hjYw%o$8NQnR<;LdG!SB7c$MM!w3xR1`jz2I4~5qb z$q7w)40(85a@{OW5hp7$jv~ty-gk=m@0s4?78YeS0wd5Wft$DB$!hW>xpdXq?yjcB z^jjr1ySGtmcZDx{_D<{SKD_EpUCf|CAo8AH{&O*{G0IL&8oxE|6(R9*unVl`h|z%x zNo##|*=soEO|vOBnR>&w_C22pH+aGSI4NSjGqi2Ub8_Jw=ye}I;mfJZkZKH0OZGh{ zN5`tt_YhhC>~UBZ+x3i9;NXBWh< ze@?<;2X_X`Jvh$86GiI&0F86~Hmda7Ced-ddKYQkpBEvqv~~v*`UflA1XqBFwUG{u zBj3pniC8cKfoWZqS0gJW`IVE!q6aFkMpEk{GF&aN@bP0}U)T@;cP&Vk5f;Utv8RWB z%8ENe4A&=I0i^i_-lASC1>% z%ltGClVwvLOS~tMf}Y7jiaW?DfGD95wJXJJ%Jy)#-qf~QPC!7(>)>OwvU9YB0E+=z zcWmZ=>!F2=5twc_Smo!BTDhr~V!+eDND(I}ICC$)B>S8=M&VUv(ttBN7Z=cP0^$cx z+|pE8!pSI7p}6cfVY+{;Br&yOU72JQ6klLxkjHSURHdc5xi8ky-=3}HZY?Pu3scYi zhgOi6(;2A|6!_60^j?RR7?YE^(3lUax8&ck>S_H{bS>DdEoY0%Fq~tH?8+N?e&LSjr*P@ndcC+3fa=CQo|N*Q6rI2_spPv}6EWTo|?X*qNj1 zY@56s?}>$5^dMz_|D5)SV(pVHBYq2mjr&LZN0p~Wi>dnJz|Vr%7kM+5qrvs(Om700n2Tcv%td_N zXDEci$uY2f#7VqqWnAIAQ4D1|f<~V%dv$Kwl)hZ$Nsfm6Y={6er5rCUkHdSXj#`6e z6&bmdS6i#NSr61OKFtdjY zUT3B!7vDa0c!>oPx6bs`q##qD;m!8;+g4_Qfp0~agoMJ^wSU!&z7R57NgZjc?368! zw{&Sn4`h>+d|#wL$V@puKj+v0jc1zM1DS41zSmzE7!@%ip(Ca{{>4RE4QOpiPm1oG z`Fc=GE0dr=;yPoGr2=4aorUd7Bn%WU)pzUPU!0dR*7OKOCF8tO2joPJm4u-Id2+td zxMxPzUp5;x828_zuoTzKO;Rh`%Neo7X8eT_0Sk&zs`q-{5Bw|KMPid(6P;2aN$!Z% zSGg9^*gumHG0cJ;x~jhBLHXa)iSGU2hNZkaD)Kc(XY~fKJlhskl<^c%1=v-Ssq?q| z?fL8h7yK=DyH02GRA=>6CueM`6!DObVK`PN{)PTv)bN};KNHgV{A(4BiI0KWQ$4D& zgBHIEv5bOQewBi=yW^;Gt5E_>9t_2$QZVG&CnQ^g|2{kBw;^OS=DDuJ>*tV)LTQ|+l&owmy9f)ut zRB&ZZ7{nDc$kXE@Q2Jq*J$_@Moxy|Ydyq!tgI;-!x>u~W2<~^^XW2WWSW8!0`(9Is^f3mJg zNlX2H_tw9g;99(K6=eOC4(~w5m@RCHtD!`70pfYEfDtWi&O!z*E?%>Lz^vq#LOWXq zQ!L3zlUsmu+{wwfiX-MP1K_nek+bFl|EX+*qY4A-Y) zmpOBZ{$~-aiq$L2THn2=Es6+yz(8d4*z5HXxXVEIfk4FME~)c>L5gmeh4#b8e3p%Y zES#hmmuRn-c7F3W?S=KGg7Lki!^LAIxT(KpEu}TtzGpvoLz&}lZpl)T?!+-4E06t9 z%A)@O2pItR^29&SV`OB$$F9xORm+9j_~i#}gxni&QJ;*C$v??|z(yXs^;W@qghiu) zx8;Gav$+)OCuP_#>T>4n0=D+#%a^7+Un+frQGbW@>=gv@xU6{A@5{x9Ful!N`#3~~ z=DOfqll$IbJxRCgZ}C~x`0Q7`{gu)5yGnPBYA#_d5T20QgEy~Npo)yW(00D~^A~Fn z3bbLsK}aT|2Yy}#)Vb69UtmIK_aXj2Ul~rQ3zWB`9!q`Y8a8F5{jmRkVGCuj8ufj8 z`YJKKS%>Q)j)zPiqQrJ(pJsV|Nh-@dGZ}t8M$kkgcp(Bt7(o!w30as8%ll3mLLSjz zf|Vf{85xFMI2?fW4(s{20D<2IREzt%8NX#0a#^mXFQ)c0a4~ILKJBpca(iD(~tNK_WhD$%hHBdKbys~M{$s=Hl6s(shQ|0x;?Cy`Dh|%HF@V@gy)hJ^=c%@8;Q8!QlS>XC&Vw)E1P%c zAPZeC=b3uakX0tCs2wgh&I1L=&^rIl>yY)Y%vfoMUo|*S9e8#Ixi~JdWkukwv1Exs zacog?oSwO368AXN05FY4`=$w2xs|G4EN+i7_6Sy>hhTZN2-Cr?IW~w znHzrB&yq{^9CtKojEv(IIs!YUl~@e_taa*fm>Fl?n_=r-Pf~FpOfRopxczgM?cE@{%k_qE z$k`S?>R+o+`ylkF*GrYt2j^KA=GdWp_z~HwR(wv)%b^T1V@`%t_>mW5&Y=t%Q6ESK zZo9x~_$EPU*yewH=!4diH7>tWc^d_(XY7L%^PMjCr6I!hiM~UAJsI6pMK1$S}&h*Fd5Nl z!Sg5=fB7Xs9rxEnNBc9cB2n5ZjmnhW_aaFs!3@z#rS@J-PA7VoTi6S3RY`grZ}A9_ z$MO0WGyYuf>(0$5t*ojNXP~nj2<^GJ?eQJg zwX{t+U+V?N-QT_w<6{f^UZtOtng+5|{c-Cm^^&6G`o(@NqDla{cRS-ax7pTyNzR;hcu= zuWzFIsi*C)x29elcJcmW$o{kDxik0c(>qd4-}u2RJgpk!It9NBd%H;if~=3YFD zq#7FBMji+GFtjWsiY@nFMiPwr5@#Y3LL~-O7DuK$^l|d!b~k@T-O%UIcg!5JB^`PT zo3*kLTou^T@kP`{etBo=36x3we+vH(?tV@)%XC!!Z(07<^ijVKVY7LNT_0<%RM9ke2LSJ(A8w z4{xPVc=zTA*9q2X^)D%kzD&SWUQP(N+o6n1;-K?$eDT;tnuK7YmWvaH-S~Jg%otyc z?h~OVQI{vX7&M@JZ}_9}P^t&Y@MpU|Vf?q0ZrJH3NY3kgNqaQOIVkZh z42E)px)&P*$C<60r>BU(@`M^6o#*9)ZxRU=_|lSopN;${xF*>t>RJZB#0<>)l)B!! z`8re5d#gHJdnmQv`l5ODJ$__7$M_@VO&y=OF$b@f>=*@)FvgME&83#xQ3G$|)helJ z361@GsnYN##%bwjQbjmg96TkA;)RM0PbH+uadIVsG?q&)3aNV!X2&KT(IxNaTg7Q& z&`%KXm=oo8Q71HP;|QbAyRh7Fvgy(<3^|4jm}@s*c4E1}5=Yj1z9pRG_qE&9chf^# za4plji>KSB?Pp#R;#{K{kWF3IeW;cwmR*kY1S) zmvCC0$9IwPq7If@yD3WKUmFhYt`J7{3R2X?Fyz(vL~V-F;k`8Nax3>@77}!9%t(bR z9>x$Y)BXJWyxxHdo6sXE7YBS(RNC2HUstcQX9piW?RP>v>^2;HlQouMBM3p)qYxN;hkC~Qp;9}EQ*&Jv?M(K|3@wusx>n=|EpP*e5YMDtqX{mGk) z%;j`A$~N!Dar&uE3R+0UQXQOPc4CDL%iIdsI={SKd1U-RR11%gN}hqcIc6Cui$3b< z?mr!g&rg(?co1c)c}tqZ%k<2q&Ulr_PNGJNHPmV;bE!pM%XksP8%VsK%&*4yiMd9` zM@Q)*!N|Lb-NN?g$Qc((^NP&O%%X2A%(}CS6}R;1i<|PJG3wt?4f<{u5#9@@e;lsQ z^S<;*Iy|^3LdIs_>{l&B5>b9n>{ZqF-M?Lgcf;WXxwcDWWf^gNI%?y5*edyC7dDS# zOzowJgKN3wlR1v}Vjs%%|GJ7?pC9m95FcWyP#?>b&k+z@!dpi(;>2?^;m#-L^m^@~ zx#;JjStCseG0}HkX|i!2zDy#1E|TsNAEUyo%LB@oF!fdL9^&Ux78|0KH81)<^RQu` zQ2+a;#hT^lB@wOgNHeO}DKbagtVVG#OmBEEnEvFQdpwQB=z46L$=(M>@n0L&{Td&3 zj>8HD&L-jq&wSp-a2ppaRO|_t!+_IxA(c{{Fl*E)Q{xxQt!4TpPAy)Sc(L%dL2kTL zcJJwjc$KO3oDf$ynQQH)SU5}0 zaQ4BoT_CGMO4P=?YA${0ye#*5pGn>#cT*yaf3c1&l2_|+`LAr_@V5@u^RV8u3l9Nu z+k5KkM-ODQZ5B(PN*DC;mmfDA%RKY!yPC5Q_2RWN_WDM2vz_%@&-NS2A6G#x>(ggy zk22MYbxZ5fFt5FpUHft;wwD?>G~a3)?|Nq3hYRG_6D)19Xe4;xYQ$yxr0`YE)W*jM zbyA6myJ_$5sprv%cu5$UKlBni1nt9e$rcXG2D?|QTr;yu;7?n*M?HsP z!*~_FHv4?+%NV-k<&wRCv`R?xA*NH+@nx#N85>9akTVS^;w7WZt;+>44Wd;EO9SAzffNL z#Pjvt<{h&8yj(o0WG_ekKJk=44or|VD<2!{-<;R| zyx$);HxXEKu9+kDIEDbVacNo|&?R#Xu5UKxq%b>VXg_Et4h-cK9`$_7fng5%P9C=2 zDZhqPR_b+R-o$G#Yaq{mSl6_LJ_#fx|CW_@0vsCn5(DzE)Xu^6K4 zAMe>uw&IlXI`kNK$s@!|*{23eML$aC5nyO&`DKE;IakMTjRa;h)tGndb^oP}lsWUL zYN|WU9&x2OJ*fCTRGEppuzqmnBKT1Chnr9qnZH7@zCwK+_N~B=iKlKo_(ul;n_Iz;sr8>K0F0P%q1+7lO8CP8U zR}AA=Msr(ydEeD^KcXk1q0`ZH_2%UaV=mSXJLjc3WNoAx@vR#hLN!J<0zV!Jj{GU} zms8z`n1Ve<(2R`5Zm#;s>{Fovhk`$(AuN#n8qhGcG`8z9>SgH0HCV^{9P=)0?ZuPg zfrFt$Q;VLRImqAX{3`?N&@a5OHKDV9b8Gvb?u`PDg)7(&d&}tM&yS?bc_S`S?vX!$ z+X<@R)bDp0bGR9`TyY)Ih*IKCw*f;?)lZt7JJvXQ7LBOS)+Cs-1zri997PBS1MAM6^q_CdqP4`a$dlUx&1}BZ_L_}PED(A z)9Jp)^Zop>Agllk{n4F3l?W1DpOMi7`*@y=yW8M_Oq3GmbuYrNWKa&Xn$r?7OkXI~ zeyNIt?#aQOaiNl=8httCDoZl0HPsm&XLCB(Nub7OF|{S%goUo>{)+qDTl^us=P0b} zU8E5iE@(N7S&*-f!bImRAL8q}a*XmiN~Z+RRDv_8xNwm*@pW3~LCXtHLwc~=Y3Ua^ z2TNAnMr(X_*Xbx?b@_z?90F3*-Vm$uf{ivvn`lkSmaH-E6ZONh$6sapTty`;vW!HF zN&>S(N1M8OC-5!onqDxt=JpV!iI|Ura6myL`j5WrnNzVxVXJGfSGxsk{u#}H2%9es z8&E;vP1b`ySOXA)9Y~6B$W>`&&g^;ZRXFo8x#T)_8ZxJH+2MQDHVTx; z-vtNfcCbX|5F<=voEE(6ffR+H+!b8A`n0Anv36l1Gq5p~-Fq`g?K*en^=h z8)?gbExY@rN1kE<6Ffc_f_pcUT2Yn2HnJTEQ zhZd7l%#(-x^Y_|c2=xfv0UvI)<;(b~7dq@)P z>iYUCUYzFDr3*%ZBHv1J+sF>mm}zld52=;nck<*Y_$Fd~s`1o*m3Luj?kOtmKfdJI zHl_a?y?(K;Qrgxjr93Fsom~BpGG3^p?iip{#2OO-CNa|U}*AT3sswJ?CJa`H2YE{cmZ+6zycKv*6I%e+fb@I?W zpY0omQB}s#t*w*LZx!|{EON{AikTuddS=Gr3wQ_SY`V=MXbvMQ(h=pb`*jG z97708U;n1~C$++t!8`3_FVC|4h5YAJ($hacaLDD4)%iV=X56x&`@LDIuswMa-85(6 zaAQPa)vIj*?m4c8?qx^LzK)Lg3BEhl&(ONwzF-d`McrOB@HDdXx|(x297jl+mtEtQ z#N*?if+SKMh2X~sP{pN;33GUOj|-&mNTaN;cdV2C&XHrfr`K1d#iFER<11 zkRED&^2}%BPS$k$wt^HA4m6y2IICp12r;RySFH200&)F_6A|r6<^5u?m}N`sIX>p# z#V~jE+GlcrQ$+CMLC2Jqdwjj*QsKp-ZZ3nH^z~y+?c8bh%q@|;5*Hn)Oz4^h-wbguD>1Ax1Dj<9~!`2 z9_&R*;Db&Hg6tDr$Ud1$P1dQN)19=e13|h%ceQ`BW;!Nu?sL&preeEK;BaHrhJcIH zus&1A)G-{U2wEf@{c&0p{1F~9Ua|UmNiYH9k>ShVO3iwLB10O<2LXTFlbZa8LELw5 z(gJ{wKuna$BUK;C;w8=$RUqcVe;ZoEeup^hqxkFo)1j4T?6BP_mxoP#mT&bDqml4F zo=|+$&B|H6F-Qp%E~#SkxkG{^CLFttTsp0Fp4^l@7ZYoV8)6yN28UqXNio^qmi4nn65s7NTY0{w-u(sk?vr+YAbz@1kc1Z z#?d`rggg_OvpSjlac2rK-A`}$pjNg2ndHuhmnc63!wao}M5Rg3P*!|@uUD$Cf_!_g zuqRogZbi^%SQ%Kt8U)p8nVT+&D5~unny+1An5HpuIo2Sr>Y8qUm2g&|9Tw!zvF?yvw z2D7&J1N)T*$0XUFV!iZhn;!%f%w4G(&eBa)6NARP|rOzDax!J9ynA&Peb{|2Xrg&=uwcKM7LGl?mU zV~z7s{jjN9tE3BSmf=XC{->Il+T(0=ny3YP%y1O^mHFe`AYEH&Z$BZDf z`hv+Rtg{@Lh(!Ew{+^9p#x;L6ho?1BK~2294bA{0)qy4=pJi}x@llnL7^H`sR)8n` znj}Eu6zE(S_ruLuPs!t56E^BCb;Ma4>*XU3Th|Y2sa4k;9#3w`Y0t3WgMun9vKM64 zDVl8P;B~V+`cTt-{^IgEzNFIGJ&xGAmn4(wzLqpf+dlCfEa+fa17gT*UH-xOEDDU5 zvwV%k$2}byW?+pE)MvtoyuLavnWve<#y`!OzzNfC{RSHF{~B@-jvOe2CWb1Y5-sUd zzy8&dMO&tWezB!bHkjRpuXo{D`_2Vw8%Qso8wC%Ui~D~)Q&G2uv!jgzPp{(dgQ6TC zi3JPol_S%MPgWhLHdfyC@982LK7_H(c%h5Djh5p*94&kk2my)z5srPu6g_t!lgSg8 zBz-X<3DuUkz|tv}ZDlbC`4}SLpA6~5JEDdZ1{g7Ur%iU)r$P4g}TJw%X$&G^80cPmS6%kKvK3g7G8fip%6R zL%0rKs5p1dWiH84&N)@E3rwE^RXz%-R73OB%CT3+1$9jt{2N zMe^WXB$7#zS&yQRro4h0MP;TX@P|PEtK;gyw2(Zt_LDN|8$kk3qEzlOp$r+W1p&Cb z2_RWn5On81KryO&J(3)hV(fVxG39yVd)st&CCbUkItUTBkj)hV=RdPhS2=y=Q#u7e z7qQ2V|I(Zz&+pG(hhJ}=MT6euu5Von>xB=byRF|I{Xp$A6VbB5f4HqE+Ouqp?!^di6a?qIClzaI1YFpBm6wS z`nGk;PfFi_6n_WP!$xxgA`@t`m~kMLJipMZ7DS;@UkvUAQ=TLfgarTMEbl|uqS)at4QVX zd}xO3-`@zyfo+oxyZB%y0rx+E@#6ZfiU?9usvJr-1e#QBbOis}GSJ8#$_J?r$`;+> zuGHqiiNs7io%X^)5esEf<@QT%lrYE@8(`m0zDby=%&-xyt-MOdKo*fCJIiI_E?BhP zU@@IRo!2Acg-SOdHk;K;<}$Ry$3h&z=brDHFY+HWuW(r`RplAu5}W{Ahz^ifN8h^_ z;2(yEiE#bfT9(%B8e{tu*%K^iKZU8QYJxZQ>wjfZy0{$RCjlZ1*+0fr1rR7ZNhR56 z?%0uslUp}QM&Wf>pu-yB&6Di*y?MbCf&(=yRIgIxRzlUKevta`IA)bL!ywXntEPy(w7;94ilW4z&+9M`)XQ38#>H>2HS^m^PNPUrQ3lX~ zKv}~7+jLbm>h=cn|F5cnEaiEv4GiHu@alQRmT|i(lc6-~{|9N0I#exbyBJUFv$s{xS)2YE(gslS z(<*q`i#Py=;Q@FH^h;!NorYL$2iPVf|3>%efoQQq$bT(yt{c*3@14#~P_CIY;z5W} zICX&btKhjoIlhY8Xp3^|u zhL>62Q@1zbBSNXq4F^DAI?n`sw-GS`YR1#W-m(fib~4&xXZcj1n$x{J-#hy12Bbl3YaMCL5eel3f z8!!8|jn!N;uPZ&zQ0(oMo5cy9KXKVQCs}7BV%ZFA%jc^&!5=CJFzVl&qGR=vE=+_g z#Am#Eza#YMQO7wsbiVXO<3$m~bA7p+&D>~Pe9Rbo*mGe_D|~*ito_0U58up_aPfQ6 znC_8fPDq^9nODB)J>85dw7?I|xROcE6t|8%S(x!aD4(ZK3VJPkiBwbdbgkS!i)$%S z#FqXpR51GyG)&Svo%T&2xLW6^n-(s0 zLp*`9?D5v!%Et4tSLeBcRj+SsI|}|RVBjpv5t8eOf~sg5P;3MprT@Jz=XWRz+1%}$ zTK5RUSU}CBY7bx6(gE5{cjT_qs=j@}+^~^N&f`V%E5H3d^2m5U0dDh#?2pCk;H<(xM zm*diI6BS0h<@UcAuGcv5q~kmTLT~7+tHO3)sR|#n{#4PLO37vJ##U&Ker@Hl1GY#E zUsd7F?|#4Ak{s(NbFqwa{&g5*-eOivGSx(xsm{ZU+{~BA1Y;h>W|K-i?CW zG}EMq3qJDEs`mgx|ChH<-)`r@%J}_Aw8}&w1%mzci2g5VAa`j0rE&xY`9vD6X1(vMBCXT)L`B?agO*Jbx&(A5^U_bXQeyp2eb{jpm+0V^H=?#;MWVT)u`|z z8tc?2(J+ds(Oe5r;65BR|0U}O4>fyGJNUkqOq>mN?gy*K1CvRytc~HD8fQsn00(%Y z#0D)HRjFx7i|LmvY10x$cWsOZk*Vz3j~K`v0IPV*R_o*9ItjmEtUUZjnHW=Aj7TeQ z>Ci6mF5}}jLUKRh>Iz1DO8FMRbFT7rZ<{O7N}*DVz^+T{`vZx!`#n!Q$C*u=b&z`V!-mLZ+^wc zsY$1&6Vx3XE`ugeGPv8KZT_7R&#ZVb?Zzh7`d}$@fFG>A<%7Wkil~2Y^f~1|$y&x3 zhqJ2WHc(gbwYvr=C4Y;0WS{P>=Tbr z=^RFL1Wn(t%vo)PM?n5|>ic4KLF4T)7cx?Dz(L{b{$O21q0b{tl+0yO1HYf8V#L22 zLtYFTij%5*grD-UZbfvmJf@HaR<0PK5HY+MQ!NPVE$p5clfQYhs^y^HH>@RJC)a|} zd3Dn<3_hCAx}@%d=C9#6rl5SZ-ZM-xzt8VQ;}Og zQt&H(isM^IQZSh!fM%`etg+ed6XWIM57&fe69~8rR3KUy$Pt?t8;kSSoUvphX}pno zwj&cv)hpg-YjxiUUPG;>f2ndk38W!SySD<5A8ks@PLL#kE{xuF7LYId(fp@`g%Ivf zDQUQ3z#AOBz>W&M4m09xD^;=H-#uZ>w?zhWfLH7&m!~jL4sldrTevbzf8>NL%_EcO zXVQ{qGok5&KJpr0g(&PWm-0_3ybB}ZKi+q%2Yi$m_~gjJt&S%|si=W?|35YgYC~yO zPu|kJG;xou*9E|{h>tnOz_a7>pdo1@^Nk=x^3sLpOqeT?Mw+4lHjr46&NN+3(%)dW z!)xG-&wmn{5VKGIsoleiNr(jDd-2TaY<|g8;|b?pJQY2Z&tRN|W=sDE(^~MMLWDIj z+qX4C{y=On!Dk+7%#KVwR)G169UOt?50MzbRyY9IVgk>}%S-6|F%f7yBDc=_3p?~U zcCSE+rfhpTYuv)CE6MgFicb#jf(}uKAo`BIuyA(v5O++A)QGo%| z)nB2db{rN~=%y>h(uS>1I;f%t9O=Lu6BNq+dEQp(&4B*j@4p=HUw?T-D#-c~P1{9+ zpwEO;1Iw&0INUG_Ei#|)3;#zefp!Y;YmXf{rtaP70_bJ0jR9WFMXI1zTJoXQj`yzB zyZo}QtknA6I#Y3ByGF=KK5zN7@r#Hqkw@=C=|t)T-#Y-tK&88{`rnq0o3eHY0V4h) z(g;j$(pDoly^_F-`=g8471Me*5nA_{Eof;Xr4$nfD=5hUbjlclnqdPT>$Bg+i#t+N zqC=<5A6xI^I3-B~kDqmuobuG5kWzklXa*W2gM#l*cr!mmk}L>-I4VtPqblY}IWkGs z@!k}jZH23Z_GC8b(sWR0Jr>6$ts3f%TzweG@ zQl4iR48-%@4kCoa(4u8D-*y=vB;HMhbZjUK%^5pWDfN$$2A68J&T(OAt5(h)aiX5= z69}PRqdj+fr@CLGIdRIM3lcqS?swYVTIS?1N z@Vb@igs9^U2WbvbK-HS?lrY9oqdO)xuEX9iT4J<&WV3OkbYO=g9MoZm{;hCD%kNz4 z9_oYyB$b44*RZp>>9k`Yrqb`aPhkM^z7PfX!j(f}fwcYLcR3Le1s7?D?3f<693YQFkqB_C|W29Q@4 zCk@6hlKJ2}{@_6U=)gV$Lurf<8rPfkh7( z`Gjzn){hv*7e%eGK=${_ClypLviX>d=a<_x0-1N8 z=`Sl}j&IY~|E;^|eb2e|)>eBG;9Tg(HdhMsu{b!!rwSv)MX5gFNKDqrpK{t|c%v>X zr`a#dAW6Y@)4#tdUJwV0PGR_xHYg|!Jx&s(!r*stKt*>VnS(VxB}Ey*@IgCiNCc^d zn5iTz;ae`$xD0a5>qm4&*das03zh;GgO za90n|?%>U)Rxij~bltK-$Ya14fwydY@4dG(zVax|#?$y@ww0_v<9${4M07WLEL@`# zlIdcWVN0rRzJ>S}woSIvV(JebqYaCRP~ODdeRH_P(2XsCR|MC7_C!1Ku6ssdbpm*H z&X7%p?S?puN`G>dA{J2D--IGuFULg@pk_(qD<{7Pl~#B_Iw}b!yj1u1R8FHUYYkB% z5>0&Nj#!wO0)OD9a{am9msWKQQGgg=bIM1LkbYQime10d$`r+j046s6fs3r@+ zHfOmjMLTxc9#4DDc`Avv5DRi9DfHL1{g)tZ0^sL~L`mkj?_BsG02=uI`%15~I?X;v zUslXT;N;&1K?(id3ZJr!XuIy`yL4~%1!Wdxb z`F6MR5Eu`k>7J&~+s9 zZeMbjBlYvy)JR+H*#}Ku^=jdU?q(9x#vV;Q34ov8ZWWb~f)-XeTrFq5DQzpgpu&Sn zE-@_bCwZ>jW+-WtPmyb|&BsvtQGP<7SD#|DvI5l$z%ok}4J248Eq}{JbRyMoM9typ z@Y{9XEebmfmN?@nMVcW0jU?}e`fBvO29m$U30`x4p>{N7a!|>yx0~mkH3}45`>tUP zU|xKFC&hM`&EO)%o|W6ls1BL8usHXfWp3VfJ655qgn{iWF{Z^e4xdv%lMT^6NNG35 z3~i|ov_Y=`{=L$4GbC^CjN`@7io9|o@!5ob?7{cuQ1Gie3PVUZPQN6E@>0+o1u~^% zDdq`xBpWx!%6wcjambQ2x6`JN6A!wu-Djbds5q=~C9jR1y8o!_)6us+uQG-aUBgQ3 zJ)*xw(o7SfpZncnG0tqmKApah8` za%X}5X~GTIUW(XnQ5JmF1&T#*&8cMHAcwlD4?D6;W5gG+E3?B|g}Uyy1ky2@C^`sg zoBU4$Q6?XH$~e>RnHFu!xq8SYu68Kj%_vJug!cPLO)MX2`(`v74|%@yQDia2iC0;? zzQq(VtI3vjFE89Q4Qgdfd$~Tk95Vf^Vzi(D!x;pp^>;)2)gc1)CIj5buNS@NxzKIPQU-+f*X%mUCP6FDoG6mEd3P=)Dr;FElp*%Vb zyhm;`KW$g0-eL!t)yxa)dq$hI$R3{6XBgK?*y8ut@g!@lSgKMd@LF>`@;LwKUeE~d zqNx*FNTaCKbTl5m`f0P0l26uyq)(bQWtusHlsS;^w9Bn55TW;5yDYrxT?y%(N85%S zQNGTvQ2i(O2^=RYcc8llJ%L{IvoFa2o*#Rw0{qK3^bc}5%vQJ0URE`jeKVl#k>6S1 zz(D-CUKJZv^M)({hDNp+vBzy_pM?AK_E^d(?g&~E5l#h=DWYEOx~~#qAL{*Jg>{D~ zLs@9kcGz3vUR4BFaQxRb1VI=4!%P+PnR`0E_dpi7drdk>^f0W z&sC|@64n(t7U`7DnCxte`9niwRRG&&j-$p5i!>@w)K-hUbbo*Z_@I}8p6~oFiQkNB z`RQKHyKWQGoL?qo4y6?_#N+p0oVZg87)wau*p6!!Us~GfKySF5n);P>UbT`IjCH5n zh=OPBu=s^1pQ#=vm?!_cMmSAv!eY8Ug8KC!JMTozJ`v&2vV#SW<0 zw9bnct|ywb*)vwj626BN?{OD*XOe7xAEP-yKVlp6X4sy0GKJ%CT{ zqLH=qHAccr&YOjq`lIS+u*)souyFHg9QkD{R5ixfIBj-6JD&09Z~ijVa1hAinhT#rO83gin2Ii2Bpwq#iZm>!3|?zv?vYT+$o*_x$s zt^-6G>Pyd7(cp?!gf&aiW?ni4kf;CChT%oIK@1ekkSqFG;9$^q!P8NW+9(B-$lVUt zijNItp@*nL%;qt_$E{iL0};=LN6q`p;|i6FBFTH~{D@UcK0u;U*4U4}ZFL!9^O_{_ z9A^CT4E!Zy9(U^JxShcJX>tR)$##s-n-i}e8{bV?s}D`l({AGF&@f&)dJQCZAL1H0 zUpQxrDilri$B9FpV4%`z;rUr45%hzpL!LWMZsaNk3ltyW|NNa{=7obQAApsk9oNiM z)ajSc5C&Pr#V*5Ek1wr=S;(pKYS)@iYsr|eSVV${$I^Gl1ESnS1+r2SaCIb@TwkzF z845<}vtp;0miyZ_OJ~e%X?vLUuplMS6a8DuTQXqBF*p-0GOksnR--en!6nRgejS!c zkwtvwm`Jhf-#>ImYHg?zW3 zZ!_syjoi)+##Ns_$eeoEb89zfG9^YNJ-NElupKvq&fe;iDCpt#31W&geHg+?+dGrP zo1}PE9u3OI7AJDP26@CT8VMXOxneA;(r&~i;#``Dg>3eip>-1SsIKs*!usL-y?)2xS?+HMj%b3o9XBLn3*~?Qs^Ot;n%B zd9Z!n3g=GUzCEiWxYtQr4s>zC={hA(a>P|^Z2Clz(x=%<0)&uAX8Q*-7zlppRO2So z5kQECdX#%d35Z4p<@~3OEo|=_?m{%a$r*86_FtwSMC4M$p_3u@-$Oe9XV~kXbpioz z&Ee4*>??>>f`9UH{Z?b;6EipFCsG8A$H9|^G%s^3K`!Fbm3p9@9r<&0P~0_#P;0Ihysb1oz>z(K|x>H4CXj zddIY%v}d86FRm4t6V2&>35xVt59lRDa|Z{>X!e*%KlUCxekigQ8tDOdsj&DhM(udqZN6t*Ey)Ck z6I_^wIrxWO-XF{Q*xT(nZ)!^fjuI6p!RYP)607mv_rqjGuI9YGXuOLilm9$epvbzB z^g}_!OURr3IK*{XD8!4m{KuxJALI+qx`Ykv`!SANOzcB+zzWzBf05~E?zLnVH8@k| ztCOX=c3G6|of8Q9*;Bdl1pRZ^+B-?sBtOt-=O-lVR0eO<#}bbxUJYnd$MITYO>R4c zs^_e`v_R|ye#F4*nb$#YrO0f+=|y(ap|T%3d7XEr6gISVGiKbOnsGgDdEKax__gJ8 zN(Ycwa2nSgD9Hy05hG;EQi1QKZuU~qp^bOrvF=-uG^h2}BYDxMayP3$r<*h{2IeIg zq2R;K251&!kh`ey8Jc(!8Y?xM7}LDX*quQ#7gvn0a>!IngYa-y+%q?_CV-^|Sl1Ct zC(~r65_=L_>RBS~68<|haHtMHl~3441D{+6OI3+cYaQTT$$&*jq2;sVUC&B*apd-F z_iWl@%S8dvLf{Y7kfNE=o$q(k-%V|G|IILHaWeN@JOT@XgT~WgMON_5L-ky9E*A^O zS?9#Zfg_jgiq;pJfsP%J10N_t)`z~{ITin+Ws^9Pg$3df0C4gRq)NCPwMD_tD2Foy zC5?z+Vsq(EL~uI#JH`6+!9ji9cb`baiy2>Paj%{^z1Li|s^AdPa= z1r8a$EMyhTyH|G}8OmiG@m>0T#x?nh?)u@uJo$f|d~?GS%8p!&pVy6U8dDX2Y|;xm zqW_l4mBKSiLFJwtbb1q{=avGr4AoJa7q4d<7S+UyL^ki*y9T&Z)$m)_T}QYX^-n(J zi(b@tHz%4$Tq&;2p7C{PL~}-6;oKO@9iuOaxB_K<2H}&x z(6-dWuWvb-Bbj#Cif|}n&pY5cbXbw!e48{|EG@XfP1j?Z%yOuPmr*YxbJZM1+#z za>GlJl`6Gu3xLJwnt*q9B98$Zmf7G@_2Z|W8m7pQ@$ZmvChRB+F)@sn+be>RY-6ly zQla^``RFIe{@Q5_nC3OxqVsOe(&zzWrhp2@VM&YtBQ!RWlWF_o!um`nO|La;_MQW& zG4DIF&dQp~Hxjoj>ITu$6mIH4TTKv}`J0x>9HwKswjGC6c(eqm-a*>mT<0));@~P! zKPJ~AlDvZvJIK+@K{e%_b#Z))BaI`kJpK5=)UKF2in|^sGE}87i$~}-E3DJb&n{rW z1%e^A>-Wn09_~njm;%z5crNy(s_0H^vCH~Mg44dzv@6Cbq>Tmszv?vw0 zqwJ*FGP5#kXC8DBCIvluddwD2?wTJN^L5O6Ad!vz+2?2%HG|S_D$6z$MDwAsaTI=r;B4nDz4;D~{m{En{ zMh2`?EKl_6hJRlh6slu_73|tXIlH>`WMo=--mr1k>GrrBW%BtBK`6p zSvuDEV+}T0dXM#RCbfYxYJ)R%Xj@$j)LHO7mEJ}lulqgt)C$356zs5iseQS+|-<+77sigcCvVb;hw@F%O)YD_08N z)H?vf;||JpnU|;@wo7`)k#H`s}Gv9 zS*9LgM-YRp-sgSa4fa8qb{*d}SKLGel3a~}>)N>Bb8?-0BH{#Bh{3LqSGX}h)~r*_ zhCs~rdY8I;=Ba5}o5dLRH<$&q_lr=6NO4@=0Q0IlDSUHymTsjWRUuNuGiSrIIfgc0 z0}LdYO0nXvAB#ZRgW^Jh%Q$@HN34n4s+p7DjnPpKG}b>E2kWw|U&ebBR&KAO@;(t< zSFL!a`W(9xfLfD5O4GU-=@Vuc{BW95$FI}UZ@J?fHcQrYHDs>Ghh|jr$HjX(FJ(95 zpOo+})1JHS8`;09VRr4gY;84cAGgB=YrD2v;M?t;K6xig$xU|Ej_F|s@!B2_^0FUq zy6CZ~Xbj3r;1p_ZxJmA(aqd~$>`{e%w4{VxbR1cl5O zT3UKKTKbxJ_M$6@MmUhvo5OlpJcp%Z3n190wqGD14_$mWk$3Dj6K9@@n<65D+c=%Y zz#r(aYF|}az3Ap{aYf#?3c0siEGizUmRyLvVE9{EoiiJT87&gTK7DX-6vIyaS#G}gS?0O>vg~O% z%7mij-mzz4#oY5ULn~yT`1HzMVb7sxMw=p8iY%5sK>=dlM91#iLc*W<^Ypj6Y29}A zipvyxg(+Zc)%*Hg9shj=!HDK_RcmGIk)77RRwOyL;{`TYr|cQq;+gcH1|1AGIyRuC zFS+^!=)5u2;+~C$i$U>7cRl79>!LAHbm2VtTgy$K$m%_ayF*t36x!p0sat_`tq(s| zm!A^X49Ve= z+UjEFS2~3&5h!@}uYqMWy11(J>W&)m&W~b(gO;;~!Og`7g-yr@-uHNu}e(uh52?J=l8_(yLl^3ehhqT3% ze8Z9SnH%?q_;?juiFhMW>4~8d#e(E{QLoi!NS$mUolKWN1MDR&US;6t4EE-E<=N`# z_k8OIT&oKIgUAVkN4kQy4Vjou^e zVcf&IGX*?cFz~ced~f#LjM4U=$S<6BiB>CVCrN;Y@*_rI2oXu517O7ZIp%Q^I;$de>nyP&tkd%FVFXsoCy7F();L6cp&z^3-3C#Ycx0m z+MR(dgR^>wQ5fv$n|gTfjRcdlJ0k8>oQGg&)0B@pQwxOHalGFA!*oQkoOTtIAdO}8NzJD5ZS)F)fOp++kZ0fD=p%1>BGpGlN7 z=C%h9J({%>Kx4J%`qGG!cWm}zyvvccHWRD4%JzSB5J1X})RYTd3}qlnaX1YL0rcG5 zj?t3LDxMRs;61`r8iXsi-Hp}^LS+x&j}?D^)vbrbmgYI}0Uy0X?(G zP= z(5KzTmZj|0%`d}sj-Te7S9)|R%=}6Gh=rWa&*YMo9m~C1cM1nVk0F}f$;MUs>6bA| zeAa7B0mf7uWvsW^D*M>byMM(aktnIVa8)uJUJ2a_g>v&HX8QY4B4FCu`6-j#Td-lXw0 z-HLg)Y$V2vc=tUU*f6Dh2AP`x7O46XC2QG^C<$fi@llK3|4nlMzzdLjf|ki|^pF2% zj#k||HN7Tzc(bPFSFO(y7cWoIPCWK&H!ri-v!wgWbfQnzq+nFTg>mIq4vnIG+d~wA zgZu~L_$4nJu6KQ9lqO1?+co zTrKc2yc4RwyOFkOAxpB16KTeO1aymfQQWr8#?C0;Vu2EOI(Fx7Z#2SitXFYm!?#W=_1zZ+r}qi>`)jkWiZu1<<4Au69GNJa^pq&hMX{ z`F3yiWF8GUN(>y>IO_Z6jbs*=?>3t1&VNw9d)_6cp%c57;mEz;R1Kh=Qf0ow}~r_=bH~d zYNS-1g&61=o=5nJo{w}EQvMl^=ri58ERLz#&x3=MmaKyecTD1obR0gjkRX{Ov9+^ei;wERnf6+Jt%O0>ENYc>pH~_`*X-a7~r56 zxg`|ZuUH&G-0MDKMaAJ|*y%?cGLpMj$G&s-I|9t=Sf2WtU15lA>}2hQ%th$)6torS zkl3dX+0Z@o^1CK;11JYcrL_(=g$O+(x8yL{%#-XRhswSGTWuLMh=pJf4iGRh&shQ; zBZZc9@Q^WNu}t{9lQ;JK&G$1pvuA_vD0op%pVWuYHu)o}YW1bHpC-sdF3M|B5bD7T6621BBC{SlT@Ds!pD5&FCV3ifR zwwvo+cE>Lj&MRk;DN60PX$#p!Ma$$W4dT}kc?pBrSg}7uzmkyI)ymm56*7~6O;cNVBjB%trBe%4AtUjgY8+XT(hhR&!YZux`u}YrtJFo3%*HWTB z%Bb$xs9}3)<7+DMm4B=BQ%=Q$&;+Aidu`P6iYMs7UKMy^d>Lc;T5xCP#R zjXkV%BII|acFw7@(cOu%f`tymFe3Tcr%pSRWpI5MyHW4!n}Na`C^PAY{G0v5|I-4j zt35w^jZ`so1^2|SbO*NkuiQF~w7%Fsi`cL}nZH;bB$jdOvUU@tAS zdmIT>3xkv5Q|9+D;*e7MgE0fj?<)8A@bvT0GQZMeVFg0+SbRD0!tCen+rIok8Iq!% zn@z7Tv=F$EdM|{*DY|~q}e>b!>q@0#@%XWf@Dyk^jkWE%yLj&@5_&1A~+O zZU0if4TU=Qw~ki)#6(02rT|QgN;T5H90^6_f?u+*QSV!j0$9y=U7(cllGpP5`})3o zA5TLdZ+NMx;`2_rBSMvIAl<84VxI|Q*+DEiMx+@$g>0KIM!@yhuB>C?)UfEQ%h>oQ zDU}Hq1=sN^O<99&e~EjLsL2XIQ<#gNpTl+bR@@^!cpc(tQ7lXvS58KxZtWsuuuq_m<*Bc@7#bdb@etkH%{a=`{R-282q(F-s_wXno?H# zukG0E1dT1eu9gXUMn57K!A}%|~zJqzw_nH_t`Q+B!fpFZ$6p3O-%YMZo1T z-=6u$>d3j@-$rStB`zWBPY8mnaTy56=l)PWg)?`NDjW7FqF$H2$J}*OO6-JC!#ZSr zjz_L^bsmR5HR$*9&%Nf9u28etnJ~U-SV&ld&?Y8|R6PW<6QJ13-FhL@br^Al67}+^ zvK5u*gpC5{@AWLH%=)2F^Ow;(ukP-4*LD-gC|G`*8WVzZ6-@e}( z#@vLzHd6(r!gK5UzN}v=-jpI%!u1jJ# z(PYeo?_m5aCHTMGgayl)s)>?>i>GPr%!%Je-$O|e3D+!NP;99WGGEJk-g-TIymtBN~8G|7Gj*znQ62? z<#5ORX*$ZO6%A~^OxTUXg$^(82$XKM;|Q2`-^6^8fz5b%rci?rEc8%=E9$=jLLB+z z6oJ?brAI)|0-!8)W!^ZX8sv_0>tvLUZJpHXYm`k?TL0KxkJW?h%F_i2->v)BYO!8J z#ADbl^l;(nsOv>}^Z;;im1&_YAE4SYyq-6|I++W<-Y5_!Sh+{c(K^Ny(weQAup8tW zGkG~iQ=uzVvhREwLG~FrgX?{BqtsKY)3o<_W$wW=PBBHIq-OUetB9Pkg~{`Km&{_j zXVv=<(%b#^?=!`oq`m?wgLqPDiA=%>fmp{_F}EoT`^WRzH*+*}byjRArgAd{rVw!w zkTP?jgL_O%9FWstg+IO(q~drRXrv~9&WigxM4ma|*W0}s5Vh#)S&q(EF+$BBwv|E( zKlE-@%5whsN>(|b$R&jlApSbAdMi`@Q;IYeN5UO*KHSob`!AzvU(d|?dn51I?BT%T z;T%l^z870U#9h=)L#hk=>2`}B_Ywbn!}B~%Vr{%1w*b@M;*B_~GBn&8V9Qkw9c=;(%DPl4$t&#m}cT@J>y) zChP64)nW>ZdhVy2hGZQ`2pO=+t}=Ce-W{dvY@r4@=E^Rm#nZ#!A_e+gm-K|b?Ducr zij!D-(*7WWkEP(tM|;sDU8=UgceJEQFYispzByIrW53NXrg|^*$rOVc9+E0#lUH57 zU;VRdY8*(?Fh={{K36ne>3O12pM6UC#)Eq4W{Vf*@l+xpcQ5L?b`g}i%U+z>nxg6@ z-GBJmKBi(JZY7wo1bszM@|!SONc-~}*Yj3KnW5zZUCOQu z#l`z-y0hZU*7#VK(oXc~J?TekAhtsY70DbP79l{pUjvt#tTH!oBoR9GNJ|Bo2=+ZP zpfu>)px*U7g|h^~HJGXt-IF%rN%p_)nq(ExTr8bi%&y+WelH~Ozc9mwM2%7uCYWJr zZSD{uW0(+QI?2R0rf*gX!c2g5zQt4j3L5{ZZClD3DJ6yX(?j-u=kph?lwo>;jdyVV+R(td*Y;kz`TwKqEu*UJx~O4q z1w;g;OB$rRC8fKO4(X8Y^eEj*cT1OWq)WO%y1TnO&bRTtpZ6Q@kM|mm^Xtes``UZ2 zJ=dIbtvzv=`YNZQPh*p3H8Pq}-R^cxN@c9RAB75Xnh(zR*jdgJ=>$3!4Cz#xBJ?-?h=R`FFAGl`O1@QBJ zf_3#m3|J;G8|4tk$qtHz;#du*FfqEwIB59aAoL#Vi*%di=!{|HT5M~0#I#ZQ6zgqE zt>x7gf5^a_km+|fM+)c>!46$2ps)1AzfPBh*_e~OMg&Yk@i;HLqm^xjA80cmD}a8FT|gvn9-#?wfmO- zNV+8U7<=44ud|cE zAW_|gi@*mG(gfVfS;$>zRhhTFa~A&_A=FkWdC{e)h67s^%b9#sHnJ?SE)6Gi6TMv1 zfnFEBEn?h$MkwLyDw_qk<|4X2Vwu~l?e{5tYOfG5tgINZrVL^hri{8_#4#Ig9F7yu zwH16*G~9BpPUNiR!ON*W2PmfSXwq5ZP=D-jto|9 z40-oHWT)fI;mtwZZd@V7us^zwkks7D+^&{CQ<1y=?Dp9x+aM3LphD>bOHpMTF`}KL z3b}aK9zvr$t$zWU|J!L~siifY8AIJhFWr0e=gxRCDt&@?^WClx)uUHb5Bq{BTzYhe z+l=-Z{)j$Hq2_up}uoKARO;x+YUrf2BC`bkTDQe~Q1eo3SdCV~3YUft~II zz_|cqrqS3{3Yjzw2Rp~fwW<*O>%%609NUxkh*1IKg^eZSyXS8=3)KT8#Lsz3egwTB z%8(pLZyY+^lPUk2>BB2Qbd!0}E3+L?2M(GY1KbH~=z*XcH&@L^J9CX9rO=K~$B4#Q z|DLC=w>MCkd>zIBrlil_zc-Jl+9^H{4SI}Ehh|pb3tk&sjxveLpEEYgbsOI_HJHt+ zd|OeSmbo63#oTD}Gs>4zd8_bNmvi^wS)fDUL7=#}GY^B@Xx}1I3I=DJSs%jY?(ZyV$#cd)3nTf#J@ALgddrxf`sc?QDeP8-?9kQmn*=U}iTEG0Q0#Pu^ zt1|ZftRHL*yivl4k=zqmMx_-Nf}f>UC75g)<_fJRSU*Iw^jlixtJ7lD<_sOCr_r#- zBdi&)-39SpIHYe*+J3u~mpExYTRVoDf^a6@u>!$o7n_4ZmmSrs=PDa*%!+0*Q2!TW zR!A%~28s#yvl@J@L`2AYg!eC6-dWtAY;NUdfo_b=_Yh@9Nvl|d#~i5Y?lSuRnTSPP z$w$fYaBv(}O*|flO%E+dozRIe55wkpuIrl&rjCILtQFjuxgRyXxpWWm^y$s%7EsJN z%AY!m@fV1RvMs3{M_nVfDA%x0i)(b=dNwv-;h}Mz;TjvRu&Ha=%v;S&uEUcODXhg4`<(iz z#QcH|oX6)O9bk@ZU**qVHN8_j>WtW^S(r?cCof-jS0!6RI`7 zyNk7tVBxEk6&-ABh#p8Q3?l45g6Z#KQMUtB{LA{^|y!qFO2K zxA#p$2mnE(41fHyyHTXuAdo(~UHFl2T*9$OEcf-CR=@mWpDj1+#Hn_*fRG{1v>>L3 zA(b(Ha<>S4i9!3j|LUm%o=6+H;&+aref!6IB73iC;ua=^f^?M_h}0W*Q6ojqq#=5Qm81MyTZ%3v4?P`Ykg0;{R`~5cfT$@ zlVn_j&t@J&u*qv0VjeGc*BslAcJY?HQmti`6(4bPOFCHDS(9`f2qp!YYwSf7OtK4$ zJz_3q0{LH#G&vZz`EQU>91Jb?>$L1!D29daF6_FUvw~Z#Izf9z5po zWEcb{>GOKTIjq_G5@z!0bd)dRds^Bg9E_%)BLuKpa(s9Y8pm?&CQ5lh`OKa(W(hk7 zF;Is^?|uEuSS=mqSD6mc3ngRKAO;?(&^Q*F%xjXzp}rsYS?wABMj?eJ=Drc~(=APb zGFtpuN+T_v54H^c?z}>?q8(Mb^xICVa4!930RNt9(Qj}wV!kqqPm~QXnZ6s*Qq+>p zY27@{8-cc=tQ`Br7#Oo!m}G0oaT`BWSMR`gLjmjO{bTh}c3#Q*;IV4$)vS}y9#A{d9&$90$6oeE!>`k3ugZk(r zFWOYjt)d^RLQw-#-oZ)=b0j#+?s*%~XAxk?Ehj!duhcNV?%Swu676du)YeGR)z-;0 z=V(snM|NAFvj%_x4e_r8Ugg*#tyX5exzx+9 zf3Q3}Wx|=;_I@zAWMjg&X>h^HDulN#VIV4F;_7#`^yoBaHy_)1OjQutejYIHeib=| z*92x9%Wy5#XYiV!?KON<+3Oor;Y+=-?|l`+nB|#5<$H2N;o#5?KWrX8zhZ-TjqtjZ z08Q1icLzdC>st#FKUUPsGxqXg)8^V0B-GT%jb>yPx&7?9HZNzd=TFn1Pt)Z|WsGjg z7}d@9UzXxz!;_RpK*xA6wh1X$7Iwikx%R}K5BdE)Die5}-{usH7f4Al&HTwPIEn9g zob4wh#1tr>#Nx;6U`v_(bNvGkUxg~AHD8*QsmX1^U4ncw}@#d;Le{v51cI9Ei_j&p+Ia^ zWq!96hEi*Ee71KujH-cp5{C; z()N5}A5koWm4y>yR_06z+D_s>N_~Q^eie*!@mFc#u3sZY3*PMf7FJ_4lzzc2QsCxc z{{j$0v;me}k{s?T9+FQ#u}L!UbpF5uW0ghw+Xhil*p3fRm6$sXxlq$cq!A=y1F@}g zL)_uwzG}~EW20CW&p%H_55X_Y=-cJp_G}_^>E|_H6h3khzp5)eyqk<2T<}3Zt{=l0 zpc7qec%jxZ_rkmhzZ)rzF&4$AY~({TKh=ryLF zfkyQ}6dZ zHp^o9ClgLAlsgU+JLa3z6ooJ%hTB16-X!es)^i6-QO4~7WLF0jMDLO&k24@kS%hy7 z+)!ogOYLTGK=NV%q2CX!AYdZ#<7B5UJxcv0K$*orZG?d*w~u~t z?#C2>QNL8`GnJ%P>@sixql#IV#aeZ(Vt)kNM@u$zRYI?x8 z#iFS(m45ln^z^gF3$;gpkVYj7NCP=1e08vRx)X`>gK2QPJpM}E`y_j*v*<%v$CiGz^&9YO-hoQM$ z4aXQ5nOq3oLn%)8%fH9S%XmLuZ@4{h{xcxZk4P{oUBJl?t<-dvi$91c*-+4#IV%r+ z^CF+%(ZluR$tMJMXpxvhZ9{Fr@Iit%6@~%@$a)F)t-;Pn>MkaH%rPH^zG2QKCfSGc zor&zaUKR_hR{qYI#6$6jpN-ZP&r$+ADm14S#LV)GDz&o7c-evQ47_w^{4?ne|KWns z<`YFIv~q{PUMS3#-nPCm_2DJb7Lq#x9z zxlP#hq0|*I(LoY8go?%!267Ye%v|t7ySny3_7KPBnwEgL|KL}z&WoiG}qXeMA%S#>7 z%OKIUSFTYE79&>_MZMKaq>39xznwv=ugm%oJmij}^2MJ>jDk?{&pRFPL<)k;hgonc z3m0w+X3LZ);)ifeQM~SjUVV-RRhOR0;5IzB7D5~k?>nR0ec1WMRTCO4%le*e*n~B} zGC<48KNqne!7q&T@7ve71!+{a9l>E$>g+U~oxB~IiRaZns&lig4`l}f?W^mr3dDm% z$v#PVJ}M!$u;2La+_hqVX@yT4_cLUdYm|n#TRU8>fc(kb(vq__z{p(0_l|b9aL_k| zvbH+#9`cU$bmmL^kf&WzEPFtveD+y%Jcswr$%ZypvP-4K%r55qYr!vAER9_I4zC5z z+8hw{=uD&T@h?DMA)s4~S(PfbEe)=NS6{d`oo9vZ7go`U%$g8V^IJ!@qU1g&U}i5=^dF z%qSSdoHzGiGtU%j8<6!-ee5R=!im!n8(FqI&8CQh%0n;UR-o!|ejMXNaY((?W0}Gy zzq)d@j}jtBjq8OhRf9lp>c2Bph^&e~!48BWz`;bM_cOu^ZSC(aHd((s)1t-lF(_WV zoe1_kWHNPx#a~!KmK8+x@NtS0H0Teu%nx?~PIgZ_+$Wx@b5{X*j3I*WLdIl0Ka||O zHM0Av-PlpZNX%!M^$txUi0nvC4vFuuI#G{uVK|G-zPMh#$#VR0x_;74aAEY=dIpmI zh*TE$2OvGOAd6KFj-iUb;c_aKn68HHoxmd@%X9JBd0j~VCtqYtTl}9MQS5oIQ9V7> zh;M;yQh(WB#*~lO{x7UK$#raW?F80hd{r*{ru6+ZZ>YxW0xAj6DqK9FAaJ!Ep_d!kzX9(_6*-S7L;P_EwZ zK^i%05nQCAnm=+Pb=YBq5=JYip>%I!639@64cn!E{(>v#u8N}4->ZhS>w6b1asJbG zRR;M7iz{?nIA+Cf+u!6E&+jc9TA(T0PfGh<`ow_|2R;w3-lLan{uB{XqJ9Zv)UwQm zq>@icO(u1m#d1WzFa3$0y6DD{mX^A9k^S0fv9`pl^V{6Cm*O2 z&~NU2*dkDq zt!rdIP)U>g*#C`tPvGAxhV0?mYPrUY69i)|)uyssmShCFX(hE61CXmyLrz0^u8YY{ znNvABQB*%%IyVeqZtqtg64B8uEXGK*kDg6R{Y<6wiYBsEF+f1@{VxPR>G&co@&!QG z#)Xf6>gUs>pDS1T?I-&)Nc_yEdU?>En>rOe^s;)f0ZHi}Mfg^W%*kpJ^@PgtFC4vn zv2RY^`Y&fNJ9aKn6Nn5r_{E?9-0gI}6VpDvmq4;sWzSI4$RL+N}TRy+)?NCGOBF?AY`3P&`^~K<*9FQ*_i;oQLn|1Uzf#Suc}~ zQE_`Kvt>_THnl5GeF*>lA4te(AhAF+4HnIEp09HCymeIng+H(%_l6c%`8-kH9)O;V zCC;!-EXdt%RUT1C(J$iZ9iXTON{BnkWD6}qKkEXm7To+q0MbTWL-Fq0dX zEFt2}gnK<)VNU2?mRe;9&1!S1AW@P{Ml?ml_BvU%u$5{M3R1babctzrvdnLf=A(0; z*`LAN7Y_z~o1yqidJcR{e$hH4g9xkBJsnoeg8|V5VoR60BME|>H|?TsxTl3f4ua>F zH}}bMnB|oCg|f2pD9EqY<9JG3Mmy7xI{ypg*Y8JA``MCZ<>O~JPtD>5njs#Ynr56I zI14S6T`%`bsnTCIsklWs6oY$)pfL06l{|I8@co$^+j8-2jhxnGXRPkpI>tY%1Ikwk zP>iZHFNyj3Poe7uz9MZ};Q*J%uFir1A&W>-+t)Ei4yGi9q@*B2@g8WC2Q8!4o1At{ zM^E$S0ff#IQi?jPeA=mP>vVsvxkzV=J!b)nmpGf$C*QZy_Rn{K6S~q)uXg2WWf~7> z9ypGqh)-wVW0%hJ0sMy!D#-Q6kPD&|X#Db_ng^ z7IDwg_lwA$*27LrPr`azkeN`g;O(}!iN0-aw$p^nR62SlI;?;J{Pr(8eN)ys3O%HQ z{68%re^eP$#^3MF*H@3QeBtn;kP0DdbnFNK#3;s#>|sS9AH2v>UNKYxY`lN$`$8tK>QjdPX1`E1Pz>a$`t0ajZc; zO+h?p4LPJvRpj9!QXv(=IDQ7Mdm#CrZTm-oA{F*^CIY)sibn5Ov+=V?^^ll8#M*wC z<+5R0%&1(qNpHy;ne&optIjua3J6hc2U7(F-3AgzQ*H?q%)VY1dF`fGr0Vp^PC1CO zbQt4B+b+INx~X2Dpq8%n;<60GrpLmG2Et010R|86N$S_!1rlZPP=E{zN-m!ibNr=l zQ8?;9r+(H=ZKybYV)rmY0E4RU*RB(456Edrt+#5TeA?=MtwFxCyOYA`T1uQayXsua zf8>(yhksZ;^L@N$OqG zj|q7Z8fP)i28ibu&FU}4AzO@>(L;f+2pMV|_nr$5>wD&`-BHpgJdXw>nvXqa&xzUR z^A1=35*~3faQF^rZD7Q6!ebQsQ0A=Adga)QcRj1_G5YfQ-{a{leppBh!fcxmeM zoXpJK6;|*fg{-(jQ=gBXsOjt&?@0)LL-`iV`vd>$s+(&(158>UM$%;;8T4hC|M0waSWH00;7VwS*U3c--UuqR_wZ>T3`gT9At1@ikN+eZvB#iSn3Sei!$(<}(tb}tq7;MstTgL|%d0(Z z5fL90Bwi^ht1AhcNKgx1f0<4MtO`g6$OnRa_qVWu=F4ZbbX{E{DlNv;EeNYRA=(K@ zPC(N=@lM7;ZmqyKseRTR`fDfAa^Q7Z`2Qge`gmzHFFnQ}*7MJ|6#5j}v3lOQ8ocM= zVA*RtR-{_^7?MGP1=t!sKFH^)YRJdQ!F{-Lor!Oo{=ojC^MT;2QMlzymXVfIKT-fW z)pS2#EVmWp;n35%t-gsh%EqL#_Y+5WwOBO3vJdj~gDzgt*A0@>3k|bGQpii8W9s6# zJSN8bAL56dp#{GIZa=6cmW&G_tyGRwqpUDb>^J_>H)6l{`NP;#;H9aNi-w7MJ_D*Wi3J+a$iN$AyVj zICfR7k7_J=A)|a-E96BJ4o5kk7<@XODnXbj8A5JftS82mzs*LqkIhaf`>NLPI=GeK zo!pFUg=LZx_PcSDXG< zzVnK&B#6_F^ip$(lq_xPthO$9Kk5rINx3gK9?y?2`2UOoUgzs}&y#+PU{N2EHOeew zR=l|NGakUT;?)hdBltwF=Dj??WKGmq{IVT!`N8YsC+7@{B$H`7x2o@P>!4;YLm>Z= zzna}Q*qo6FXhRDZmm59XV{Limx#GYgwQf?}Pdk%klEjXVC9+JH6n3{5m`m6sdXQ6? zLOFZhzhSz*uC{-`IpEK`L7z5QL%zNJo;7L3!8zVAJ57Ot==C4QS^@eR9XjJjfwWk8Q@9s7Rdxc)0i;;|$uz zg!Vl%Kb|UiLK^b#r$D%>?s}8V>&d~vZ_SE7W(iV0R*>@fT;@zFs{jfg;Wa_x&At%| zajoa0La=XL0u+KoS=0E4;}_X28H&GU5-UKlc2S?l+Sc)0>_XF5z-EGm^=rm!1Ej>s zXiludsk8lb$-#UWvbo$%Lgb{UBYHHPzNkCQlGk ziPtl0MOJm;;URmAZv-KbjR9LmdmvRBL~jJuv#j;&;d(3?44B}Ao_(L94F_f+O3FOY z(OWtPI0*kP#V09(`Rs6gEzGi+-xU;w@_zYbe)LYo3(O>V>k~?y+c*xyi4Ln*gxrP= z=i=z-+BsiVYh-Ct?du%Wg){ARQ}Ha{^@$3P0uFtWGS*~TlJ>NU+}V&dCDjF8!3+`k z6JJ-HyNm7J_2|KUb_+lQDN8(Ir_m9+s!r5{y3GrUd0Falp*k=1+$O#4=((|EZw!Hb zpoiYd^u3CkO${4qR#HyQVUD__fKk{DzVDzKtY(01WR+m+?47KV?SYyXZ7gm&`-0cK z-AX}o9kO#r18Y%U~oa!dq4+?D0*A|96{9^_IVnEB&K2{0=%IVz#BSr z@T6l@?{oQaRdUF5VldTBb6s`+@)&BkxUL!JX56qw{{e2Fx*3qXyPE+4J8LPw7>!yz48XZ+p26 z*T+PFGhlj!rf_G)6*;<=bqza*x+7Q-DXx6>V{c>Tdq2(3bdyBboX-A zmS?fj(^HD++90CMsha#dS&TY>^T(W+m{~dYzzCl9O#x)4xUW!wsw#l1BgZ9WH~Nsk z^Xr!f$PJZ{n_(V=Kl|h&!j{YafMKUz2vV!-Q)Jg^O6zd5 zNx6Sx`S4DEKQ=x{uG2x{!%NAiU(-AaDJiL9Zza^~ z80SCl`0*0@u-VMa+2@3Sbl&$DW$ zQM^L|F@43ay;sM))0qguCIx}MJ1_Wf@NQm^`wgH>tS7R&aQKuCrkb$QsQ&`$-~KQ4 z`h2p05O`B}eOX%W=GA}jNcVKPou1EpDy23xqeLgXy74u6l@|c65s4oDt{p%gc^uq> z)TuJ>CObBs(f6P$o%!l*o`RgDUFlTvOJPnUNwkD(GI}rybl(`6TuR(woz_?K)0dlS|0e)iY+De z8%S+(lUUGEu&0~U7x-X@h!q*D|D^-0iWZ0=i5|JT;)y=k%ybq>wr$sgwVQ$-;CP@mC`a*Ho)+top;KT#k|A zqxtRbhsgI7Jq*L#sHA@1*Hmz!V9MRQN&M*<63Xk3qXr5_K`T^+3C zUWZxLZu9pf@!WBBm(;sEN~Z9^2+nlKLooXPPTZIHvI+1Vy;(8&G zT=e5K4EF&k4nw&X~5JWd?pY5KEo3W)cV-UqR-{*}cg$2yb zW=c}**oA4+4?Gh235QMd=aXi>x36*Ldj3~2a_{zdLW%pf6u5fwE$p%4BFo9iajk`m zz;9AS4ieb=r7Zsl`fWR7-oD97X^1}A*MUe1Sj*qUAKx*?qRD&sSLZ!vkh6za3izt! zWpQ!Q;E#o-`?4U@sbn#9VMsN;+q&$UmUIYq9Bn(e>SDC%kA1%Is|8m+jl)7xr9F{6 z)8x#O?=;D#c^dq*`?6+TyUjD=2jiM+kMje~iXyg)imFQs3<%0tEqzm)S-#Kb+MXUs#SFESj3Se|T~6Ex@mUT3GT3MgHs; zhxofJd9g#w5AoaNyo_znF&vFTSLu_{@Hx(IZs1elx*F^Q1|JeJfvfmafF4}~?rl+?%gYBm*Dbe)v=T~rW?KX@l{bpT!rsWr zqES(`O*hk-SLEq^_Lp(L8|__l=J3f(0L5P2jH?a-FYQrVw5VG^PIYTamI)V#nk8xV z2}*mL!}!UJ+0DG@>!=K>oFT<7SNE(swl4|dg1HB6f^7L+u6RyD5^=5vas8hVnM+aZl(F<-xiZBu@ptPWgVX=e1w4l;n^Gxvv zqc&CKWi&AXzC4z_YC|U>-fO~iuqK6smw~4ZsJ6zUaLLP_dE_uhw;Od^xjkYOV~|s3 zS|u7_%Z(E5mxMcnfx0KuF8gQCd^;feLZ}UqXCQft?n;H-)UCSS=IQNUHnO2}Gh=rz zg0qdymX6XMaB3VqjQf95uE!5}t_Ri5{nLPr+<@EW{^40deo(KR&=#zhNFbIGqFC8F zF&zmLeaFt1Buhz$``GdaJex?sF%wH76_rJOs8hcD=&b0bC5+VT${WcPk@y=!yS*l@ z;k;X@xF?=QQP6~(-I#`=xC1;MO_`lpp;Thuf=fqsj8M7O2~_N;3*hKOzR}w|N3Jtm zZqOt{N3vy;tY)V>Jq`GuSuSx}gO4#*+C+K#Wbyb+6>tViHcVIMJ2N1PorF{VnV`huF1z^VP%-k`_BrT^tY%xew*a|EH z+}H%i?a_PaZ~e4rD$O?>qQxCVz8&A32h3Y1F+%5xvx)?iY9vR^1Q_>s=uJ1Tnjd%^ z>dGG7q&!fmqC_T>0T(_1KK@SHlN5|C3q-ZjS<`vfCDfszCR<{BqG1aTLzc1nSOiAw zX`%(sy@+qv&}m>~4CB%-Yup)Oef|OCulxOyI0a($8X4JQ%rZnPTf{+q;0xQ0bTe1a z2^l)9o-T<6c@>@%)#wAV*b>zGAMar&_8v5DQx0jsW&*cw!Bae*N1{UeRWmPNy?o9^ zDnXR_4&~F}?T=9dh;sNn``QE7Ts_|P#a3BA|5gj({K<&=;H#oK`I?B&tB*fOpKKsW z0a%fn8_afr6`7)ek2;)Dd$Z|TAZwzm%R6rN;&ID%G9{Djziz!J8Mpu+7ZZpn!3nM9 ziXZV2D^SRab9;VZb!tpJzb2MCUDv$M_ zUAi|P;5$YW0l)Jq3?7>Ir50H~51rWF4>qyOzlV6vo6W#OnM`0TDq55loX5&##k^xo z7Nwyx1p2^<(Fpn}3Zpg)tO*`S&$e|FiI*z2rlR|!L(L*Spcj5xc8|mCqta*sr;e$T zp2?^+G+*m1%*BS0jXQcOf2iFM6jvY}$a;F!1S#*&)jif%u<;U6V7wrAdYnkbXCR8^ z3xM%`aU7jeEK{X+kSUwobgBh~B8WC5v0HvnSmI#>fiOP7DW8jZDe8Un5E8!lvLGAawu07RX4w9}pdaotby3uv zw*j!GW}sK@gI#J$IU2F>?!!(bt;~Nt{veQj&JmUM?HBxSVPb-wr%MeVaVO?YI(}Z) zWqxg0cEaDVjYFo1)*_x!S=&gIfbBs(K{SCW#7v;-_U8FJ8+*sb9c%@&G@fOTix}RX zlWC{SFQk4y#o)cU19nu;61l!n)T%@;Q+?xw94Fl?NSMCj6>c)eCuJwpVHUw|`xzcC zjogOcPl*P&b)M6mr@~K)_hQ;pXB{oz=5_+${K#s}Nha3{K z@Afmpe)IRoQu}}H-A_JYgMhIqLxFC9i;V~e1>M!73u1;9>z}`qiLkJcpCT-O$TZ{( zd_`1Hf6LjhQGXcQp4H_-3A$COm5l;z;lt_8m&0ilM*T1`L?t+EX*+Z@8pI$RxWphu z88O;`sNfeNzuo_t_I{Bn`y-}b;0OrwWefO%#;?OYR=;xhy_BRS;|UufWGJL7nZYLX zW#Y@6%a64Bq-J9^Va}Z{-`^vO8(^?3DVnV|e1^U)`I9xO^CqT`kil?Fs&|_Vi*-FN z{1IjS7yVHGPxQ-kRD3!h7tUDCi=WExu;ow-n$PdAsi_IwRyXZHjEH$J103;sdbT4E z?`uw5AUcn{o1;p0-+r)o_VoJLS=g}opZxLIaLdR=;)EOb1(nI#sGy)lh#^#obi=J> zIsA2{ujWp6=>c?eQWF@kS| z_bm#}hqV>m7y}s@U@yeaV!R53%H)lju;M#uS^qC$e8dBJkMxr^X3!jkf3;ut1)R^h z^TypH$%cv^JY{vpcUf)7DIuStrr+ROT1MxpTKV_OX|SnR0m7g5%jZ>?1kGO)Irs^` zuz`m(41-hnD4*(P(C*`0dIB<~Zln`MkKSm>A$xBXJ6$zo9`|4f;W2*33UvhW?Lq*)M##;Loa0mJQ3> zN47riT>J3CM}%IyPPA#x>4v#}bjJ-ShzTj!PMnKiG6<^W(?pHZaG7kl-D5STE#xZX#xL02@h@Wz}>erfvnxlq`|}j}Tb1&03lF zx`=hE%Y~HOb~Onqo)ZghcsO=n&Sb#8^N(?15V~~bJ(bE9E{+UwSbSTZj)q2=Q!`3E zZ=<9;pHmO~+ZQFKXthf(#N?x=-Fr6+D?3^xH_|;o5o=pycgO?u_sH3#UQYAzjG+s{ zC1KJ-TY4{{!v1&Z9N-t3V|^?vZ~F~y_5jaM7oic<7}?OAM1zgR9fcuo@6V9Ob2)YK z8i6z#8_KqvfP34b@c7@&K^<@6^*>F38L3sS;Sd%^Ykrs#hblL$cqY8ap^id@`-+B^ zjE|*7lVgnS)H&88cDQJD(|r)o+}!H`TOV*e%x|{$J&t-c5P%>CjnibuU-tsW*Kv~p zhQ9`hO#^BPDQAHzXJt**IQO+oV#-rm>lSxQ8_SqJ<65ulWdtn}Uw6|+gZ`nKpFkw*QAuku;f@j)s4w|$O;PS);@=Ibd+ z@n0tw5MZ2Ece1SFpqrTNv7I&^=@Grwj}4XzjTn{`Gwys@5%E7S!1jr+S+-gM{>k8^ zL^eykQN!HguiPM~nG1e>M!-DazX8m{qvy#OlDIaxe{pfy_Me`hkqqQrX{j7#jtnyZ}&8L|v;U z&9e{%i_+l^MU`$(Ku$X&uomy)ZRHv`w+5QgbVZS z<<$1Oe=v-qm^RUMl5~HM8i(KdRIT@Bo*zy{b#o*0c~knu)#EbKxF!4#^gc4cF$58- z%HlgtwGQe&nRb951*a5Jj+y}+n%nJrmgF`G1@~qAt)8Ae3FFljL4>o}dUa)vq7ZrWnMz6SYuI-{_)j!cnsjWoa^1k#im^bZOsW$}RIG zQVZ@!8M8g$)~}_m; zKbjlg7pSsxaE2)W)_ywTIWYOhhJ)*6!1Kimv@JIoo*s^asV%Yy7%*>PIb7Klu-irb z;=lVHseOPJVxCUCOYCrW1~kN5wjxvwhAng3^Bb>%F)K2wv4cOm=gKtDyAm}#sikjT z`I~X+MEJ+;>UNH<6yCE;uV8y==aVIDzqWIp4N`Y8*)wXq&9pMrWCV5>Am=L!cPb~o zy(ITL^7mK0s_5`(GQT;Koip9%cHf!bi$xx~{5EBR5BG!JsDEfw z@3Tr>k0`|c4tC})-{%!om!4BETL`R*A$fA9RyfD+X1|MT{w!ML>(xk(OP@@|9o1*g zlVmOzubqmG*M~OrUeEkch2^w*x|l445dnm^^Fg7iwexxTw#F_Bg^HVzoIM{$_3%>; zS*k3i6k}v5dZb%&+VoWZ_Yf&xp+gr>RAF4>IutyPHw3>m=C41P6c99f{HRaChQ!vZJqAOU2-x$Tb7eX!B{p z85JO$WiLKg1b6F*D^pUU3S)lVOrPI?Cml$wWLnD_K+I(CIyF1IcH8*urx-Or;unM+ zdVLDnyo#5oAQt|JsB|~-LUdx6*kcK_==9jKT7*x*_{%(#cSV@R`;eSnT-YvF!T*nb z#{UPf=Zt@Cc(wdc5v;7?gn5>Q8N&i&+3}8(jv==+$U7D_w}7(W_V@ZT(DmTOWs19K zrhh<!gl8onIsh(0cc^94cdySsp zqO}Aes$_<+KkX9uRdh5oFfomG^jsqXC|UdixKb1@x~f#zp)VJXO81?`5K^2+5y*w9 z<5qm^8F9@(ZBD{RUl#BYi*706>vi2WbIsbYgp@ZzkS>GYIXzvsWKN4AXF~a`M~ti1 zyU!<%`oQjH*jL4@9`dvc7$n!D(xf;~ZOeG$u3=VeJqxo&-OZpk{$_$}DQe{|?Vj&c zcRAR&6{>q9X@lk#2EB9=?8rrIE4`YiHRJa*?tZ8 z&Zf%T1c)QJXRHOkfR>KpgFXA084rHLg1&PobX|#DWCx84!5S6+*OfZ=&)ssQQlmUw zWN|HJ4ba8~YqD=cVfP0ihs(_u3V@=AdDwmnE1f#^2XyS76{OmgGpECZ)%%%mGUr*$ z_N-I~=qNH#gyYJ`HwvhMQnm9D{Wk3Mjsy1InbJ6M72s_OsKD#t#tDOF!u7!fFAn$w zN-GwKuQW{;Hz;mtA8Kg*ijFPwzOTCdCY zLCOyz`Geg%(I3^*G)NH*dXRa)YZlDq2Q9`%FH+JrQig8V`_$7%ZQBAh*m_(6S# zXAQKRUKd~Gk_=WlD^6qF+&e~#08%wuUrwC10iE(xVf`9Q}#PjuHPN+D^gr6=OqN0 zFOHaMF)tzKvmbC$id>h)@rkHodrveyI}C?6U)#*qiOs3@)m@FvEcCQb!Z_{viI=xO zo;5D=9Q~nED1a8vOJGP^%MFNfqlXhU&6rry@}NbGaT6B7Xfh+$hP_#`X-0}aBvnec zHXZ%!X7(q`)?wXJlu}9ea2%DUilH+lxBPIHw>Xm0kKH@u+of{t#*^NUjp$>COUGr( zBU|{j0JGj0vv=mEF!DID4Q44L&tqosRw#8lzdZ7HQg##}4d+UZWjhKU?<f_kJPSq5iOl@X5*XkYr2E$$RZvPx*p(apb)wqK6-yJm!X) zYntuWPSW51>@tNGPQWUcFr}2wPSSPW$}w0|mWpQo;Ef{KAMsC@R}%ctbI0Yl|8-51CceT^Q5G&(>ko zmo**oT*dq@zH<8mVT8Z#P6ruOnO_JMwv)o~_PsYrnHWws|3to*fJNJvJYXy3{pZ9_EZ&`jCX-`-@fQdwNPo_ZTRbI5CAv z6)`gk?Ar{NT+B;;iY=|&*^1bWyA*+5uZVhe$ov8;!7JCUC1B?yuq?cNwQqh%1Xa{Y zK%1Hs$3|yr{YJ;!gI?`jX4JU(c-lgf`K%E|+O^q`Q%4C0vJrktQvc@ZnNd%FWcs)43D zdkj2ZWIGOBWVo%5JBu7}ec_WpFyLL5X>wU=LPrx-TtnmOJm50#Sv_GKKZ~57RUn>R zWoVZEq0DP@1M79(r&yh5M)o)(kiItse9gOsJStR;6s8=tIYWwv@ z)$CVUvhVC8#3SKhM_Cp%&o6r!NM1fYJ>o!#{f6+3ZN6Y#sGI>-DgWfq-RbpQ|B0YR z9GxlagKo9;eaGuRzt+X9!`TQTWW};`)>YYgSkOuf>_d{J3qc{;nTOs~b+ZQFobCy{ zkC}!GwH*UDgvX}VV|YqrPvwAhDivAU8n zshi=RA%p_odM3mgv}tQ_gc{LT-ED}%i*e3iqg;*^VYT|6}T_T`H%oVSckew5 z{_ef}<41J&eb1aZGtWFTXXYwGqyJ$>cj`%24pfs#vk%q?7mD3s3#O(W1;Aw!8}9aV z4^Aa+QWR`sX0isqx)gu@o3mXRHdS<{<4qa2IN{9_dV3xb5bo=?qd5P2%eiU)Pa{8{ zWigFWnTATpYTB(N z4TnW4CT~Z*B(|D-JoSFttAH~vKTeypEV4|ylBmq%Vk=)Vy-)js&DEkzBR@T^L9q!sDl@KS)=|`%?2eTq27iP%JsD z2D!1($!D883{0s0pINY`YmLjQb2Bqn8}`tIF2|4OQdbnGUH8;>)*3aY)LES9c>piQ ztX9|G#A?Zcg=RhZRl`%n+`gXFS+O97k~ZU*tGKKKKv2nz7whIc5pVPC zq+OM#jx@QyF}Fqab~$Fj?Lxz<;&bUb4qoz5x{8I0Q^k7~pKWi>n61Qu`jSA1(x7x= zVOV_XQUs( zv|lb9A;}3@Bgq;nJ+qy5o#sC(_P&_F#jGt2XuR>;bWO6Io)>tKy9~%ap7(+WALxma z6Zp&gw*DdpQ-xbtq)7=Uuat_d#`1&|{>&Mrx7Mm?T&(HN?PnZo2GMj@N3h`%eXB)i z4OiPLeL?_koPp_Ks~taa7)v>+ki7||XVNqmC(LWEhn{>cPmQc zQB+`W{?gs#G}*$Wx5F<#edsVL%eDU=MUzmQP`L#wGd&Z~27^Go^|wu%@~%2sMY1D< zhL{#s%1#SZQB8K9{QWsrRmpyKf*Bo~t6W=EJh!un#SxvElea@!rM)#*;n^#drLey@ z>Akt zZNY2ks=YF7`0^`jv2_iTIq%m;Q1{Sj*Ey(<`m#~ON_|CI+hvSUgRJz+w^z3u`A?M+ zdW?PfVOMG`4BmDP!MDFpHS>KIeB~=m#+tOvQZY!D&1B z+=iYmw`n`FV-LpfIsZIvaX?3}B-t@G3olW?{pW}U-{jP1fwUpE#NNlAB09Vm{q~*+U7q)pu@q?V_;v^#3kg|WTxZL@q32PM zHd_=96rEehna-~c?$b0hEsS{n)hi-icGVzPn=wA*a z754Y%$f1!IznF<=%?Ig8#c|92=vy@;6jdLBqq)hLFpV-r4{oN*_W-cRP|RhC>$u?+ z@nWdPfH#gwC}Bnu;@LkcR$K`*>q&cDlWibH5V~{QQ7ztyDLA4Sp}kKZ>MKC03x_h_ z|ENHTQUxOAKnZ@1oT!`2eRv^hbQsigm_7U?lGQTWVkI4ex`kEgXhKDhXC-?c_+0t5 zCwMSNvMDM_FDFxP_moe$%eI(u1&4di3vHW#i`dyaU_3K1crZ!6Iao6lB)Qu>9K^~!U}|X^C3A{ z7YyBsVKR%x2@0e7E!pPXJvYHcKz99M45@M~C=qG?yccqmT$gGR7w%;-rNJBos}4*K zuuFfBGcvL<0UP?G%I*8B-J*gvhUV%fam&A`p<+b{mI3=z>`0e#dsT+rbnD%FHOB;; z#dFEh?k#V7*dL|^LErsv$v-M+aH2ZLr-!+@*|jyPp9S51Y6cf>;YPJGH_du%O|(kr zm&wbh-izX?hMh4X>cR4rC)8vGN>q|W94Hy$@8y+slXUbwLsoM70-8pYoMXEMs+=D6RN;=>Bq}NOod7ImD>Z#i#Qb}q$cLgu?Tf7ugTBi4Y6RE!5{_?f| z$liu7@5?f$AeV?2CXqGqMk|R4#9N_jEVHT?aiUR;2N%Xd)#8#3u;Ix=^BAPZ2C(h; zEV`kb_dlew|AdpEahcp&vLj8THyGmmxiq(AAhnddSF13uQ2Hj)NG&`2I(FZt$#*SmhdYxIE2I_Nk`iL$)zoZT zTMY;ZNK8yrg(>|`o}C%Iu384O$x1UN!Y^Dz4}k<#;foeRqtNupDdy7{mFtV{I{ zdqog_qwY=&Kx^T*H*;W*)pePolzhfFRX~iV$98Kl*F&7)$l$+`_c~gy4-f-(*8@Kv z431lk)=o~A92^X>C3k-DxT+_sg2U94gnqf3i{jj7edbnys#$Aqs5w1fw%N%krm~vah7-45TOw$xM}IgT)tNHO!C(lRqat(pE4UujITjVU&I*fY)$icGGyG>$YY@{3KQ!%Fx0vtEN=K2LiTQHG&!SXGuSb z&4|c|5<{}prq8_1tOi48KLuT=Ni24US5kF&gsd{9bWsWhwyn+*1mVYS84z)Fg>dri zr(hx6MgIm4b(0x`X4HvHOcLVlNf?169ypH8`ncudOEtFr3~}1c5N{|%5t+?*)>>Ci2n^Daw>af$l&DU#fb2Z~SS0xCq& zgM9qz&{Ig9v5wrNlIqdo;__O| zpcFdaNQm*jc)Op)#|hg{q!WU_>G>_iPFij5{6;u`qW$TMvDVy9mMJKG z5cT|P=t;;dt?U{~6F;k9)gfo6W|@h^R=PJL3uYg4Lr~opsvy~utl+X)7W6*3Uemb3 zy#fyQDf?Rlgn4y9gHYrb<$;OW3-2<^Wgv*7H%?~UrEgS{?{XMUsWf}4#o~n0dkStWrG<;io_sNOM#10XXH8n z?13|0MR$!^?TQz*uN{bxw|uqb1d6I3AmoBj_wSm$E1(FM9e`&sGaKK?;Y^ePei_Zb zbG>G3b*%VWp1F0R)2x5?s+O!?3y@}mqmyhrJTWq)DQUAptm`>NoCL8#D-YYjj0F#Y zqA^_|WlcPkM|#hzox4qd+kR3Ab>E%1*@UF7Do&T32~H*UE2yYo^n5miP7d+&@NnpH z9{gU&;Us(o_+iy0)?8BQ1i$=&d|cu!9&aR@VOqiJhEXoTRIRi<6D@w@p>!o(cNZnv zNc9cf?cmq*>|&gn6{1jvS&@uPe9{?$c#WOL1c+&R>Zv?>v6su$0Kv$@KpZBIySLHq z@A>Vn6GyeAkxW5{eECKNv@h643{J`%QXYClO;zhfZK zpn3`XcaW=0AD+JyYD0dN30!QDU($6hFDz-(RhRW*99uiMrRgH$NXND+)w6OScll%1 zMxR{2G)t4BCL0=b=V))d5xopFZOAEd;m)IsXrYWxzYm&;(8iWRceYL4bnt zuwlJUWO+=jS|`+=rxdHFq z90W8HP{U_uJ6wJIS0X+xt$xHS*Ie5(7sIx=)fIl2sv!{26V}M7p!{2ZFPKrl&cy*( zxj*$bV?D<*%)aM4Hy>DyswR~4GJd#Y=5LKRSYDF%o6uxfs>%u8E-$C1Q}JPAZrdR9kuAa;*AulHkv3`c;koa+!$$|AHd+tUIMc zl_58n!5+=RLpsBqH$v6~wZw=yhG8|NXV@#2bkNNtFh*8dX<6E@>~-g~6zsT9Q&bGP zAK{&mi_5(B0{M@Pf{u&0=tUF_rYy!=wdtB%HKC+3w~7ajwH=kym|FeW&QtDaZKv`o zS+Q*BMoabwa5E=Af18)|u>-0IM6qUuG_dvtV$7U+hUCdvyF<;TUbo5aBf_VLj`{;%Kjdkps#q(LsAF>2s?OGaOB2CVO@bm+kPVOf^^Y#N{vwJx(A1q5ch^{_uz@ zCPxN$Q6EZ62k2v5{FlA(oW|E;7Ph;;N0#cds;3>?PJR)(q=P@%$E&lh?C?{87Vs{9 z#r3M1cpfv9PqD{u5#5R9pV-Ej4A7QrI2k6||I${{^-)o!fM6-P^To zRBqbut%Eu1{GVdy16im}ath)KHscV^N;GU2>TKlFOzbb?Z2G5%KWwH-6*;-oHRq*uU4Gla&|!~|HAZ#ZxE?|rSdN^H z)i12;Fzn=?Vkh}ZJrVXo80%5MSHN#9r|U;#S84Q_2~9in;s*m(Vk1$eXjp-7%6MwF zbUku6&1}=PSSFg85H!aRk?Y)zE?be zk*kcqJw9cHT`ii!daY#KNLV6DzOu@^7ss676lEnR2V_OZ{ZBvm=&W7LtjYZOn&-cx8&6CvB4I zQCDG!hW&f}?Q5e2Xs}hxK36?6AtXb^3;hyx@IeW2%sGA1aQL}Ku!E>E99xJrI9JrA zgc7w%$lIzzerkJGG&X3;V*!Taz)3K+r!^7+>+7X*w0NXaGVu!{V-o+Y zz>uTaEmAd{rN$$>IibwlaCoqJxDfRyfFMDxu=EiNuU-w}L4P2HKc!^kgb14!K?u`z z5j69%qt9Qy)spf6|H@=lZG0(LQ6ZHt8*(j4t#{oky@Y*$CX7XmJ{6)I-Dk6vih;JLe^ zm_$anRd<%Z%V+b9H%6vfhg4~B!|HUOi$Nf*k-knRLt`hAliP$Sv<)b=P^lwNV(;k} zk2>B4xOJ^41NXdG9u>{CK)CIecCrOd_#Vj(9{wE}TE^=N8;G)9jMLRIx77mc0oecPmxZh=X@V@A)SvhBu)5Qq8P(_!Sj0MM_^g z^x_b;nJ%*WaGTnDVzj%)={63jO|@2ETR0*vb;9g#@1)(Ig*}oaZBXm!sbwjjQ&rB0 z7b2)PKWxq2(V2J|uJ;TCg^Mf=T@bh9_Hb_I+TW#)b-+9Sbb5C&GB!&Kr}g{Er#-sh z7)9M%$C{fYK8Vz($<4~^&xX^u^$*0}#9HZDe4pck@ut&4jd()Lv)LQHlsa2D2@QOF za1fH9?}w>TCbu-uZcQ>hT$@G~%d%cn)ppS9lBqX;EZTDn4>;Qzk&<$bQmA&3SjHyyIat+^My>0CHvP4*EKF$@tOn^Ag z4QJewv)TgesGbQqY%+23@v-raj(-FopKDEGQS@S3;nQi$ix;Ipn{&6L`d2Wx8Vgm80kPrJg#bn0B6Bui2A z7lmzX#B1-`wR4VRmrhw*#7IpmD?*@5lQl_l*`f{JWL47O7wOg(W;a7?9($K4af`p5 z-H<-Z|8N6qc%BsV7+F<2w3PReETLJ~Pt{=!yoNV4WsicX=);g{X%7vvxfnew6P_bS zQdwR4lM`0!^`qM*u}Wdfi9c>=tT5I$bWc;*zd)o|jY!Gr_fgzinB*nAXqGpvAvF!) z$W_ie%u)d?Hc5OPl8Bjw6>B_R(Ie1}A+*&*tcOw{R88Ary8e^2v4veVU$GC6pZHqKnoLZy7>r{<~93aRKvY)~)3Em^_NMR<&=bd+ss($>}n zzU)=B;>OM4&coIn285P#*BQ+oWnHev?OAObIYoH}aS@%lc1DVS`G-^|YqW30ZK{NR zlc@(RNeq>reQL=33~lyb*n>Hp&RdE|M23TWnhRla3Yn0@hr4>*`E$6=(FLd1@(QfT ze@>f~p{7(N+jwO?JK)v|Aqk0UK)k{9ksjbLT|BAAF_w)5vo^@oAJ3|6Ys*nbdEX!| zwnjYn8=EWqE@Q~ReDa?+WjfX=;GnSf+n{;F`3x2C*A!*d&Vz-^MX@IND#8@razL)5u3J6ze zIPR!DmWFGTtH0#iC6fj8$AxYNfXS(dtGB(5n6smkd+71F>7Zeh#YYt$m#v)e!xnEv z7WQ$???`1noN1Gm=0UdjZ=mEKbSr|aOkZCQ4A@Celo)f%ttwTqYu_$Jkl$L}rIB_( z8Yu%)>3pgWySSYLJ7nePVu%e3d9ncSC5Sf=5Hh#GI1N-TgV*TD7<0ufoqZN1?89TB z-@;;IYqWRceFbT3sB(EcX-?L%JJge;tzlzLQLr+8kBhzH4kB(pW2Pcb&i+JCp zHTsk&z3c8P9NvSy6(93~dfuYX%@t4eYCD0u<}XDjw0Ftf()d61@tqYK0P+|C0qdch zNi*-j;B40W{8qUaI=t2ghUClY(s;qcXXMmjR$+(;>OWQSS(?5WGF_~v1|5Lw9PrN} z%)0?eZhaI<+xM6;$o^h#ofyB9ZCCITNur}2MsG%fpkB=Bf;U(0C;cM6UrFek@I50+ z@_qZfF!Y3gUxLaLI~*Kx;B_?)hC&D%!VF(sTnS-kEPY;m@Xy+e3!vtR6@+ zG7yY#1KIijeJL=2z3QlZ2gm=+`HMrx8aa9Dhd2NuYbQ0q!RW2A^%h{~K3Sd{W27q6 zxZ1ldKUsr${wLD;mMz{(f9}&_1ebHU=FUd8bvx=^2O1K@>$fOdLC}Sqq9TL1TqoY) zkZ$>l14>LBHGJ+sj^`C=ldA>5e3*q)guMrzC zk7Q>$T+nF@y7t=&Taey-#vAUn?oYDpBoSVH6g5MYP{FpE=e;X5$&-+9SH$RLSicZtMGEV7^&j+R_3FN zdZBSj1G0_xe59mYIxQvKAf#c#u9l4Fk_|zyd=?{#+Ul*_t1 zfn}wl01wQ)*}q~kwX2n^VNFGzcB>n3U-qNwJY=DMDC~4~eTNDVK4$o5y(kcJcCO$| zjM1Tv>TJAyaIp9eUGn|x)mfxe5u&{3-i?KYoASWBLwz&e zm(^(hqM-sy8|37w)IHJ&YXtzaS8jpYc(xq)A2d|MmvE8>{Dk3w!@+D`k#H!yht@Lx zovkTB$QLp@(_G@VT7z#N{y>j5cysgkSGPgQ_)hx^;|C+7-1*y_EQ=DTqmOw8bPsgIQkyuD0`p#nr)Yq@UOCk!-Y#L2g!olW z$0xR`G}Y%OxeW+?e`f#>KfsDU11&K0Xeja;;A>$G^m>rgu=swesNa%kYpjS8`SbIv zR3DdXwQ!UK2|=@j-D76n%3T z%;(cmN5>FAqW_k|1(tI;Iz4R0NPJ8d-VWOI0eAYUTy?AFjC1!SZHm0(3yps_B?D(- zMPpH{j@(^oV8{=NNl?tReH^_9aN&RcBEo;)Jrm7UNx3?8U`#nb*~^d@VlX4qDEZMF zijKa`#>p3Qzfr#*0WD}$LGz$NrWl?|{VObYJKo3Ju!;!lE8!L*#+>!a{^@SiIx<+K znm=JI58Qs|SMV$V*)L9{I+Z~h;!QpGGoy);irZ3YZagHLdI6zud+hoRGuzht+DC3p zu%KE~ZT|HJb;ir<2XJ5CURb9*;u6-XB!4S#eI@h+$P5QXEW5pz`3O|%d=%~a&*C0= zsS%M7g|)B20fjY}c41Nef?_;j;(EiHaN1D^Jh7B~hY=7EB#>l9zkd!E-h2JJWN`2m zc*r+Uur%X?eRi2Lb7qtQ{j(AglSl6v3hiXh&PvB#_yrD4C{<^0a`%j&Z;3k2tH{5! z$GrVBRSt|~R~sxQy4=s<^3wRW02Z}!P&;FZE}lFE|zylL72psm6kG8aDT zkOF64f#u)frypQ{81xIiMW|=jfPfXkpgZdFZtq2jiZDD_Hk6mS-Jk=X2~^BIjwV&? zTXFdG;oV>U$g!bmBi8i^W7zL`um4(3&X;RNmVxmF&Te?XV`mL6 zxOxHdRBRx);nC6Y9bhau1)nCjF^!0%nb&r28OMR*=fGFF^@?NMQ?}O>?Tcx)Hy>sL zH$3a@x${UzMz+~Nk)qtcWhnz@KQqe!F1Fh@^2~u5*7fG~i;7FJp(G(?3-8wZSm2`O z6|hOjnt=3^Gvc6tRda?I51vTD4?zk15Xgf{><@xF%FHn`f_rtRqZDDwN9{}YbfCiM z{GVpigFxZga!!#0H{Pj&QQ5?Y95uM3Qj7*shS7w4Vm>jfX-*DyQ}bm`lsL)Ko`sp zYw>r|2*^{vii0{TuE*i|7mjLn#bQ*t_|8np3OBgK8s8w87P}2!U*tO# zafY?V4<0mnlHst2dx0v+jT438KWck6P2_>m{z~=@L!cz|DF!g19z;UU@%n}|aa{(~ zTT>N@ba?(@ROGt%iu^GK2r5!Vt89Jrs}rvuU~tlvrS@(-pF}RJ3tdLR65SU*{)|`_ z%^m@nqS9_7wh6Lf!sY@ zx9-kl7rjj=0U>E(P%B)btJYdX%rN9L4-kL(D@#@Lmkx~&fD(*xkm|Ja!l+V4g)DkN zw3!1pY_HjD_Bnkn9l~0@&tR?=W^X&V4?)sA@ZWbcg7e-5#!TiL4nEaPW%*d9#&1!# zi<%I?8`ip!7SVXs#0(;|yPB34;oHaXLWDJz#qq2UBqJ()rCTw82JOtHgKJc`?`>dZ zp}(~8s~D8&UuyH*howe~!U%dY@HKn{>?>7qV^N>MtFmq*`jIiA=ei`hY0kXhFxnTB zUo5ngu{6+YlW7cYoq(HL+YUa0uUNUMG>HiJ85?yw_b?4)=)=uICq2o>Sg2q3H*lXG z>A|%e_#26BNp9%PgWI_I0Q{^|zRI2RXR{)d#9!3aGQPmu<=o#-~t9@V)N4r@)TYmAKl9m-yB zZk5x1@4Rw80@IQzmu+clP3Jqb(=x#z=sfL`T-sIXLMn{)Fk+jO;C!@_he=oAmGQj5?tEt{U8Y-X7}wx^(@ABMix>RwpexhQOpUPscvJ!sV@q>$%_Y@< z?4DRJ2cQ4~m#jAb!SjFmdkAoHhkx@gGjNHYb*2#k82S$HH71lOdyv<7j!f|s4c%v{ z4%qtep%=z3@G^Fv>lr)+Z=J#s1N}HKTdJ&#AgYUIEM+T3wR_U}VMJokBijZofGhtW z*y|tw_Ls7dsn??+Md!X21Wu$_m^xp{j@cEHU$366)gc03k6YF`JUsrJ+0@{lB_xJd zgoG5SZJrul5~nJ2=3Dlkkc7~Xdm0?F_`+pIfR2lexp;3%o@<#Aeg38ZFA4+OJ|*;B zhPt(jQoxzqr)fnt+n&+n(0#6=3h&M)N_cAJtL7WW0jk!Cm&t(497OjZ@wLqs@$M88 zEf`ppq&@fSo9QwhJS5wi*MYq$*Uk6JMJo)&T80>is)G82Heoqk_q~Qt#e5#Te)2<1 z?<*uADNMj0)T0UQZMk)^2l88_le;eO7KvMs#{CT*${3=jSmlZDax@3HPy!wvoLU@A%#I!xYKC|H* zC-mfG@i|Q@QStQ;dnNpK1M0-1y~$o2l+WUk;bEtop(gs-rax(-JK-#vHJpfp#Ow6+ z_M(FWU%V-F`r;@IASpzy*%bK?(+_PYwY2c$`PW3$U9Yxb;m_aPq*zpj$h-c=J=@f7pl-cqzEsrH~x*gpo-V zE-EGwlJ3<35J}kyK;(+jM|gqw$9q$G2$x-Seuj z(_ycX&~C(h;EYXvVXi(P={oPEM!@mVPR*HL{86b58?D#^vBXyc>ZK+ZE=?vc`?^V$ zJpW#ePj6V`x`P1nmJv5}oH(um8;fJ8=st43(T%Q?a`z!*n6oDew^%p%y;=GG8kqdX zyu57=IixxOs&1cy0~w&04|0Vs!?u5GLbpHpWq{efaH)Z`f49ftgf{ayHi(UuD=OKy zoQ$D!tkf}T(IN{CzAVD_mtikCSchRTwkV{Mf_H2(!ka$0;c~3gk%C0($jQn1Q$7W1 z^v(T>?j+9ptEcy#x1Kb5C^B&Az=_C7aMY2cBFl?qz)=cZuXSC($R95s8Q-NSdnGlf z3i+Bk|8ZOcunC(?y3g01x4TZGpPnjBM5>{1X}d zN|O&JK+LUW)ik?DNPtxU*1p%^6&L4lS@$}7R?|TN01qRT?Dq6bcsFCa%q9mhNnE>O zVjx-9O!iM`h(Kt&faTky=d0^kacG7MO+0N3lt1(~2%Z*u8wy>;*tr{tK@{D!N#V06 z0p|@owL7?+LlhdI1B#h7z=ALf-+S&(0V3nx=D`E@t{|id#(Nml@nDR$?f3s`0TKcb zzKkt_p{Uyq6eKALe}v^`$LBm^kG%2ii}xC?nuZL=(_6z6h)h*9fsx5D zuZ7kPsCNQ0FjtUmTMSCiNkWjCE^ge(F@laB!*jR@2u8@Vq6i{|s&%}R-KppQ0fOcE$8QdjU5vQjo*yeD1CbSo>z zA0`2AzJW{dAnG(o+N&M#nboQp3*0?-p#H|Bd7ZN!g!zvfil{WsYFZ#g)>TP9Q*9T6 zwUxNdH`~6vdi>=ZB@2g~DaSwIgOnUXpn1h4nEHH3BS;SRx6^T~w_LdYF5aDX3WU2h z=vJC}f{x48b*sU9I0MgfW9>!3D*(*v@*fNUlI7r%ADzZ#W@ zc(d08yG{lI4-zSzL(9PRzUncFR1e6CU|@sc_ymya0?BzyXQ-S|f@fOUR-1S7VK1F0lWI|ueVwgJZ$L6x;jiLd37($yeX zeU-v!Q@ScC>fO*<|BJ^GOqI3q(FHG_r3eBZ{|(y>CnD(Fxe5SB6SEG~?TfQljj=>z zxZoMXV0?_{np$0{fD{J{z>lO7INM1Zwg@}97`L)J69D!c6HMXZ z71*Aa3(&9i_6gAc3lh8>9DVdxe~zm_OEq=0y*K-vj{P>8*(-_~G7vi&z9wFQ|L+mX zpc4l>0UGd-fNl`n&`i{7JSW^ei3%5Z>V7Q=cTIBH40}QSy}g(i=pnJfNDubW&7LafwiGkID)kQ_ zBG=+xFMNe-L|P-p{Sz|L!4r&sd}{;4%3L&fJA#-*yk3}3dct%|t&m1H9h9Jeal>Vg z^oquyD-$>{C|N&xWcN;k`{cw~HQ!$c0|UK%ZS^i(5_8|3lFo|;;nG_;CZyaoTI+Sd z#@ug(&whHN2+n}IF2H3)R+^*J{BS)TSTVmoUYXu%Y4Wz(%=Jaalh8$bENUtJy6#(V zA4d3m81e&W4}1<)FMy*Myd^BV@rxGKNp7};z!rqZf;?c$y+4hHD-=K62G1VSfLu?| zTag0(A*jcRX}LAi^cEu5OhmY02%Zp}u=NnALpwQzD8jLSqFnxtnXPdgNXEI!re0(= zLVHW9*D5{B2H{i&%KZU*2Ot^(L#cWX`pL<5r2Y>Zp+4P%JHNBHpqq*=;j!!X<`?&& z$a`hAH4*uGI42|wzFXH&)DKJ>gnzeQwBp3YZ;%(;&`Ow=DL`^9{h!&PLIbiN0c9>0 z>=rIOe8la-a%37{+#GB?uzF1XdeIfye=QB7Bm*`kqG=w1*ROdwnSrs)H!GMYMV=+( za^}#zcoo(fA#`he7}Nx4C)HRvegeXOFarlJ79xKUP}&tO)GRFLQ!!&#Q9Ih%Fzb7h za?M|9c92n60r9dkmCxb2r@@Xryf2Qq>JgjkMp?Y&rlza!qJpRO7k$%K+904$FoO2fnR|b*qOppX#1{OZ$QT9 zvEL`10FSEy^@XQ}gKl>og*YBXfBJzhD0sp_d6X*j+<**j>E{8BhhXUlzydvPT~bOs zAt^Bk&;O413rt<4xB_4dr?gV$SU@0AqTU1eB|koz?EkS~={a2+wq^zUT^?;=lbB7Z z0QRwvw^k#P-4hWXkKuCRgeYz}*L9^HA0KCdxLt^}N~gG7qh{=?xUTS{=psKL{JEc4R!A`^A;|8xX7JmezwCPTSA!&(kO$JNg6eJN^Sz z=<6!#9qT1wTvmIk-QOT6z&-?0$8f%rUDnkd|FP?;L(Lm`IP0Z0sQx6qExp50`@~2*1(HDKMIv4yQ;J`7f@&TZ%hOS|*9=QW`pNeej zQs(~LhLSp6HJ8Jh$c=a9I-~k-Nt|ewG2U&!4_^xG>RqbiV}^p7mSNSzxG{4a>O(kX zk}#QYN?f$6o4QF%;I^*?_bEVWH7akI;J2wf;G8pVD6a2Mnmfd3psaElHdDuqm;E^4 z5{9ZL=~m{nD2(+__%uNHVRG4HvQYLqaIG&w5(r3k213rB>Oy~ZnYJg0<9mV*_u?us z4Q(4A;tZFT1~<&X0*K*N{akf_X(eYzbg%a+xtf9u5BAS|NPKs`!{M7r+2{xpFgd^P zVx5mCaJ!rV3cCU4*p+&YC~Wm~5mhpB_R1z>IIlulI#>)=Q5T}&b$hYkQWq`tvhMi8 z56)uq51CX=Y+6{3uYiFu^hAJEJ$BojkliXzk_r)n#o4r~JtM2(n;fy=sQdfxNdWYG z%otVBGcEZsVa%Dh^8^%s**aM>F{$|mczRP15S?GTegO(a4;k9BFZq95R*eoR{kKS* zZ&_(Y&SS78^(5o~Fqu#Q4oyVT){hytA@_{-$CJA!h{H1uqVG{*XN1J5zkwm*&^@tl zdCUT&X6xg-`&|#%p+zTfgFkxMN?+Tud zi-BNcW@Vec?YT=TN}z&e7^Xo%Gy$CP{BxRcTK_+HPR3<<^M^=tlcnoZt#Y?T=Q>~q z>Og&aaBDiiyM?v2T^s6V1)$Y)Z_nRX}|$VzZjVEvO1vKSkOX zflqVz6$TEt-8YcG4;81Z13uY@kjsS=tY-J*DDYZz?)DQL@d6mB0}yY#%;^c{SA2;} zz;&~$e#vmH;PAbr|Nnv9KR-}l3D5)ymP)GSZ{=1)K_zk5PFIN8sgT5bJ0SC@bn1WChPzsTVx-c{2#*dOQ$u&jgluI z;f-|fkqac?@5WL#F%25}-eyj%wOf~Yy(c){t+RQ`_Xwc_HLCjzxSw%4Du8=UOkG() z$#YA^Hr%h!z>Txp!0c$=h}rmx7sV^quPy%dxx9N!VXjR&+c*f%S)edCxozagbif6X zqDn-rt*wDcEMoUcZ23n<6oCv*5S-Lscja$Kf!yh|Z&&tFU)Jp@8c!+~9}+?bs4}}9 zmiLhx9t$RYTM7zS0UZx21n}3AuT)DY!AtzU^h*vvxc)U!;Ou-pp~xzZcocCKMEW0s z5j8KR%Y9qXL7>9FiCrb?G$GIU{Ec3jZhzo(4T6i~3q`9Od?^UeiQsF=P@n~Ha-1%+ zz-=Os0HO$A1L`+j2{+!qa(CH;yZ?5stOKrbgI~SSaba+rK>CaE0AZ*d>+QjzG;+Q4 z-Tz|?2vIW;jkR8%{RuaA`MONX#iu78oEs zKzL#x)3#_T7{R^Q3wlOwGb+L}gzN<4#_~D%_lzhF@izOzc|F-sM+%boU~%d0JJP5s zrBnk{QyB=YZI@x(QeK9ytsN$4&q2)ca??^&=eH7{vuzxAtaH~5Z=G;( za7VI$X+_|cpW{ReUz~Y@g_YG$bIobxp}(xF3!&$X&8%S76{3J@zKTC~*PoxX zdHG<{)9-Gy7Ay#U*BiSny!(7+(%XGej*rjlb)w!86BAjVD}9-uz7s8W1d`I(__ zcl>EplEx(m?j|V-5%W#$8>}N06)jF%9F~N?5dvvoXw^{h+7po~d%u?VOg2%D#*Z}AILk%q|<+lN+vT)n!yLVEnl>`Q3Sr$_s&VWtqcVe)h_r{KwI90Ob^cE_;~fD#8>eQ(CfYkaONc>rUVID$zgd-0yk(^ zMn-OK-K(KploF`uk42@WrCU=IQd7x^h)Sn+-Jz4?P(%;%ya=3w5zy((u=CN3v70js;X5#HLsdxSBM zn&D*S0{avvEuAXl*49UQZiV{gu0ZNG4#T(75ni8B=DzgUUJ0)GZ=N(hd5ooyJ$Cdb z0U;qUK>H?de@HBblq3NeXk7z^l5`H z|KK<7&--DLnWAnCN|4_S#Wj`>?WOqMUdZzwyuFC3+iUdPn{9yTFyZSs4YK$gpjuDp zim4~}S`-%N6O;1sOF4K^Y&-FS9ODRVxCK+_I=$n%JI>s_4U$pipCddwi{xfhoA7?m|E*v5@Hem3j~1gxfw1*0r~2WaCtD_Gpl zM8HD!yK~WfIg=R0a_w5J4R6U4~g~cXZ zLhypozqdD6w)F?%`jk>%XG|E#gLyeQ1qB2^QA{U8sv9dCZ`&>*(hD#YJhsWy@1Ccm zc>>JY7to{q0~;a0IdQ0`U*7h#WX#>3R)KG7Q<4{O?q;eY*_q96x;-hzl~3_H>|E+O z3X=C)97q#wg0+x?3-3yR=RYxc@n1(M9O5urHPKslXhLpl)>86+sJcA!=>tmSB{-;gu|v?eHP<1o1l+mOT7i^$Zf}eEm+%nPLaA!AN<9ZCmC5-gtTBO3G)~85 z;;3K9w~f&BYqHrVupCA{szj-5;7s6v`0lDn(eak@T~5k^B&bZ2>BQN2TaoSFdTtQ` z6#x14`U%J_tjLe@chu}Axv$2rB{^c8U7ntva-@>H1^~aXaHnuBsZL5(ODoat0SMQw zrs#=FM`zZwA&NS#GtLIt>X~o^X7{TbSgwf)K$KGz1K*v&Zj-2F6Ot=8W*N*MOc>MBQiy_y0@~-fdPNjD z_+kZtjP{9oBV|KZGEui#8P?QaYk)qu0v=A#ELo2RL4JO(&d8UP;@{R=5gtc>-3OHe`;(Kr4(h6^WQ0$P zlxfWZpFTW*>&SjH5->N}pr6&R56#P+wBpxI7D(P3JWvK8DH`Qt^BBFU}R+Hr7z8^4IoifCZeR|cFwpP*uAdz+#LB0 z`FmQNpeF|q^ovz*4D9$k&1Xr3-vu{q4SDFty?Fnn`q_# zc@2c@Qy3SICFYN{Oqj^cGu*Kb2nboA>F>wJmbD`!V~Ya;;}u+{U!^{JFQic?!akR! zXHbmj@09ULOLg}Y7BaTASW`sM3N$o?ew(mrGkxW!i(jNFA+BpBY{(Yh-Ni5-8aUN~ zieNbd}6=7PaC%0xz(BF6@{TYv>u@GviHSGT)zm_$&rzfWC730k`j=`?pit2ZwY8^YgFF*0etp^)a_QW*bPydAv$C?n!NK9qpE59@5Go!O9sRa-l1e0=j(_edi}J5=GdV9zb$_k9l7OoZ=<6S5yi3_^*nPa2p2?^&w6{Z zzdq@Hvi_&uWp#DcBqVomWu(+d=y1w2?Md*;$m+$a?#1!(YT;uS+u85Iw`t|qwk6=4 zBUb?O%(jg)>jEH&jrxN>Xk?DU)OeI$fU?)$d7|OC?zIx4Q$k`NyW=qbBl2MUEe*1ytj=+E(*08R}*_c(MoMr9AUnb3py7S{z5R2{A#l^p^S5BVKMJV)@#BlKP@@~}a zbnmZ^$qDW`u(Go5EOcpB+q|~1*$#go({Q}PaQdurtlqFIRyf=TKCn&ggf z8A?*?x@E>;8?CbptgLU~A(SQP7xOT2&O6$7XCC}{a4^odIP{dvy{bhmJ0~Yb(5>de zA?A@@^W%-W{@0JLsVW??%jZ;60y`yA8aMe3G`6;PA)k;_TpQFmYr=#H@?3l$4ZFPb-{cuF!NnI3z|m zZimyp7blY!?G_4dk~W)N%ZJaFJ7L)YBcuU6`Gn-I6stJluG3egZdM9Rb2U@whnf zcu#xN;qN%8{s*dRY7X}A`j>xpBY(N?j~GolEhaqvTjd$V#KKb88R1iu_mM`Bc|X50 zWPP$R7%S5A6irdDwA6N(eEmI$NWZu$ zIr$}E_lBLKQ;2$bYK87QUEH<&Y~0)jJ|vbrwoAB7CUjwboWxztjZC>Y1%huEY-MBu zFNURLWHd_+8t6%G!b~exQ z^YbC;KXyOmDW|2Pnt6{#CMhp(Upql7AAh*puVBGNBHF%gA3@ZjX<6;f?;FQ$CFpwi zNYwanb6PtQ*QEeC@J-rEkMjO~l6(;E# zA}b@~GUHDh5D;L$J~}usfG=szK^3;tL7Vg?FR!z^`+T*q(Kuwqmakl+-gdd~;{2fT zT(@m@>x*vVpK0s5?H>$>=Ma?a&QFgL9!$}}XP5cRXkupOGW8zKfbB)zJbBue zJX5asIejEVLm4@_2k5DFb*I0JU)kB&&51EFF#)<9E^IundEH5nqF4p*1N(k$yt*_u z_s2o2x9D^{@7wV3a6XyETPSZ(;@-VbRXyJ6)+HDhoIH?({vik;M8V(PPGBBnpNpmf z6i1Mb#C))}b|S8Ng-TNQrDrUx+(f6W9LpbGyh(0<6ZX#M>>jm$CQby;Rn;7{cTAk+ z<%`{mN@8Ro-1a|>O`cZYeKNnW@R3X)RywWjbh{J6Sc=?ydv|vthY4Oxizqsbtxx+U z+4&=1tJ(&>*uZA@eiFBDgmQ znd1dq?4LYA6~Tp^1sMMXz*GAR^Nx;=B=_Tof|Qm&K0Rb<+%|pt_J(M?X?er(MpVc> z-k`=Qhyr!?>l%}l_B{9URP&wTh;1f#ob3(Eu+*1XjB$DOmYMV~bSFG^J+w|yG-iuA zKb#GL(k+?`?GYp6m!cwE7HRDY^N}`k580a=a9+75U2i8*sA_6Hh7Xs*#b8Ujd! z9>`GjFTY+DQZO+wA@Z1r5#RWq7ht`-W;`&eQkdd!IV%Y^n}p2CAGd=E*uH-^@6_pR zWF@%=5C+}%;<6Yus2n4AKh*Q;MJCCb)dJR6-{~TyZA_XW+HyR89JkNc7D7(`*q*Uk zTV0(;{~Nl-f-NKrhw(R0RaHN$Gg_LP^E++NK?3YBU=C>;YxKB)_~>sGJT^LNjL{EZ ze=fiBPpS{5_e}Y)PVLFx+O0cS0%ScONgt~gX!$n|Ad5HnN)Z=@F zmKrd_^crg3qZ02w=aZxk4f~^JtiPTm2(-@I{K|zJ8V_ePWTK9p2!-FS{2clIJF^|!?KJ@o-`}3F7Q~t%Q;?S zKIgToNHDv6EtCW0ot#lmLSdenu7F~O^1d6nC*U6si1l;(OqJs$@W8NfT((*XlbjZa zOdVGSa&Fy!GKGec$RUoN!1G4`XEf^}nUGY==;-LDPx!7jsiG-|^HI75wSW0W!&LM3 zfToxzGr29gG-_6OHX9<8?VBl|a18%L*261Rpl8GtN61t+F>S4gX8Rd_B=5bCb8`l{ z*#wkj#=lbYH7fxl0uuR|q>wZ?Z_T6nMg9FE$}Q2=p?sHVZ$iG)t^U3~X058J+MoK* zo;`c=#Am0^cA%_`qh3d$VwuP7*w)a{&|%E>6ari2A7_RJqpg|NvJ5=83CIz;?ne(A z7Wc=jE*9dfPK8$c6@=hfHR|%ae{H5sbhKIG^yDO#d-I0XE+{FT(&X=W^`WRcyWvkB zn>oJ`b9Y#i;R0=$ys!i3de~0WX9trZ8}YC)0JU|joAhS_d?WGL1u{pY)}AR}H+Xn; z0`f9IqxacuNeltuadD3wH(%Pgz*7PQ zynVs^?z>?rX{VrwyE-79**3$Y^NSnKfrF+Q#OD;BEs<_eXM_UrRbbpGP3{^SDIgHH z**KORXVR)OGU`%GSgsX~t;5D-=LCrK#^`CE;pbnOME1i=ztXOGctflw?754FQtWnO zKN&<3=$fh&+=r}4ygJOfa&@5 z<)gV1r_UPsnHA(tb79Pr*L$J7$;!wK9h1-M4^?Zj+b(-SS#l&V$A6?DSW@ll;vRibYIwA) z#fXcZ?c|+aQo>epz_jf1)Y4Jso7cHq>CSFB~2o7?>(t zGy26hYAcc3SJ7tJlNemN1`UQo%eJ1VH&@mib4|8|E{=Mlblqb{FW_mgXq2`0kUYy{ z)#28os-Cu~j*X40y_3ITf`q}3*S?Bk+_*u)Ziw|qZvhISKmw-Jt|+Zyeea855p;pG z-F`sI)k*0qP$-GG$dx#bH~Nw}`^A5k1Ue`r(Rt+|+plHKlt!ybL7~MMEtT7pyYM>H zZF$PsXxM9wM$7fbb*G`YcbJkSQw0-X&)7L4mhePyq0$`)K;skBk@U+&vzQm5K>yJG zbHGrSZ_Kh%GmCq7Bp7)mbr1iehbG@SZ;|1A@kQlf&=_1FEp$gax4VHjVJjb<<)zIo z|4AGdOCPLK%=*i1bEvGY1H$5|?R`juA+j{9vy%ZPCM9!ob17oJ_D5gFENk;y(jmIg zMAFbcAh;8yC8Z2#x#iec%NofQdv^H0q%IL)9+ZemSEir0aWY~Njbj`6{rh)>H7obH zwWl(DYiB(&p>>D*&_tLlb|;{x6crT#9&iR!rL3|IWsg2v3|}6rL+!+X`48_Kgi0rc zt~QDk{WDL3iM?rW=H=#|pYAAVh~7*3I~)Tp4&Yb6SNgN(l$S@Y>B+@F8jLFR$0g>r zj2EUzj*X0pib~?M3;2oM_Wk?Ed%SoXw{}l98XZ1!>^j#gOCJE@{$1;^(Gfx0p=YO} z;_jaKB&`cB(dMUXIjC+t-M&%H3P?#d0FHozk}@EJ~M$j=fg`N7zB? zLHMV+^bgHcHot!l7;vbz?si-*eaHiEkyU!cF*lt8>N_|6H`W>}ug?^AZeXt4yO7s$ zT6otYDWArRGb98T z%2Umoy&e}Rql{pr^;Xbcdp|YxQsWO$0`9A8Q|n`u5yR6^T0qEeph}sr%py?VjZ8?8 z8yYk%AC04#q}$11`%46{7kCC{9H<>yd(M5zND=F*OcJVEISnYoWI9%U?(7H3D&!Sn>F>kYc z)5&eK0no#2TS%4AsuO_-uDi4)a84Uu6m@16QGI=V zS9yafyvVCH9^`)*yf{(X`6Y=WqUFT;kcr7|sW+wj^RM(+^V4-J4>9zvXx3{U13gy* zm6DLGYqgl`Cw+vF)B|D)6UjN|T@}HNA=0ROFQ)A#W*6K@hF<@AZ)LDd0LpJE z$iJYvgw)kFNNOEVZ#f2i@2oM%Oi#D9e-wOi|Bg*Hk9yEX&OuD*{L^>0@0i3qWb+rx zh~~a&mC2jh5e_#J_b*gexSG+ltKH{~S

I{9VQ!7aA>r-<5oj)X8mi`tXSh;}dKz4h( z9 zT)5?{K6X3aO?d1OZq{>9e|+qm)$hNNa_98;*6rJ~7YP5klJt}jb4zo~va(I-%#)jB z{CBM|*Q|N?G&yl7#XHC3qaUTTZOHh4{Fu3?b_9sW4d(>*bBFoD>CsLHuWeXZnB%0W zZ%gs3Htb1SVk}h8ufk*skr7$cv{K*nCp~L3+OiTpbF4+?Wv8d3ko)-hqLJa`+mk5} zZ#&jOT%HcGKU|LIdGo<_$ZlBs>E~BKQL?#6@;Xi-s-Hc3F)}_bYB{!UZ(1#-!C=w* zO3ELDZKU-gr;~wRLqh|}8HY#UZQa4yb+h@8kBkOBvPp%Sl2YemSB4E=g3syI%*js_ zesXbK6A(g41#8TP3*cND85o%VsT=9>;_=})GT8cdg*LH9>us;Y51pwDZ#A|F7CJh= ziqLT+xAgcBio+(~`pEe$ZJwTkyBd3W8(hsb2Q%6J=9S|%H#phDdKPt^!fZkm9~5aZ-2K{invOFH@TY6Iy(+m6p%7>J!KRGJ#yp)xF#vO74y-IDG6ciP{>HJk`7iS|kF&7YG0*t!0V&c&TcmklIr zr=+AzX}Pw&vlGRv&1e3*=UM>{)ooj<85au2PQiUaQ;Cw|DlGdp}gfPugoZZ-nocltsm|y~cEZ*D}pxZPHEA0R+sl#s><^ z#xr}<3t%#~F1*dm%H*5q789jV0&mnn=+NsCaK9 zix(XdC@&7yZG#S;iJ(_c{Dp-5jepd8fI;n+g{fX;0r%+FhGAX{dgBnG1Ynnxcy$Yh zPbLn8!h?T7Sqm8Z8qQsmt1NMQ&h^!NZj-JG*s`#>*RY@Jh>%|v`3M9mwj}XY>%kPJ z-(I+Esw$jZsH17KKm1!~fOJsE>OTXzh}8a%Bak>qd?)Dl%|Ty#>U#crS|8)l(c{!I zMVg3pAoPH8RwnDxFNQ>iJTkq?Ql>}B4#45;{5YYRH8#nC1M)oLTv}Qh z*bE=sfWSZ!h{Egsz(Tb~Xt^BxZB;u+BjCWYrqXnha0$U)=w(jVsSYp}_*fgskzW7B z>h>KC>B{8fce&&ad

_(Iy^gggY-{Xr;bek`b_q7tJy`dDjYqBL$C=xO;49r<_Z zO7yd{%oQW7ahfunKagC9GQG;Zp@DYewSJPzu^mmq2%uvNo*Y^wrjNgKvE3!GVH-u_ z$BjP%z~79f71nJ_lYhg76cp#B^6c5Iskd;PYWH*;Z&^-$A_!=mf!akOo~O9wspp3e zPlE1l6?&X8u&`JiZO_*y&QvdF#%(f_1IAFkZDdN>3J|ex0hP*2R7@;OF7C-qZRX+2 zVzsio98e1y<+sHh6B8C8o4p{1;N)_dzma)#yu3u@LjwaSWH{?n-@b{DUU~TNAte3g z1I14NzmtuP4GmAc4PPikSH1Z#{J5RssS4`0uBRt4byaw9~-Jd`yD%DAo}Ztr*wxcn%F?Ex*V5vT%(=kSE) z7Z=&j{zBPPu}c>DdqwHZn*)tlaj(k4xurelw|1zRGB;omSRKJ*f`$_yo(kTRjFOUV zfF7Xvz%R&8`o+LHuK3d?ZpvZpVaCYX&jxePxL}j=iCZGYgSjV0ZfeFSIj?UzZ2Yk( z9l;d^bqcHYQ}#-Dt3IWR?IYjM9Q-c(=I<1C-`@EZ?&Iw}K^KhN7Jax!ve)aw6==oc zn5TB~g1-0l;dzBhX8@?yn)>y@a(d?|hKMjL0F*bV80euza+G@x|&E=uqR5G!VxNALKdB%4J>{ShX+9OI`3zqdZuVvQAbLrrH$Nt*L-X08cTl-x+yVXV>;_pwyW`NWMf%F3jM0>Gd_*b9O(-1NVbYQYW z%R4{>dGX=}so)z3M~*JYNFP=$nydTNAvc*HS`@hB_G1>L4TnW~EnqCj(E~)Prw7`~3KRHzutE!;grC)mA+Z*~jI?!X?*wQ2?o1=sTr_$To=ezjRXDtmc{>>+O`R*Mu^s$U? zi2B8dc0&T@p{K{RS{S4JK!K#tHolUauo=PkJbcuYt_R!N#H4cg!9ZpfM)6+)iPi@M z-=%Npy6vfy?An1g;aF~^o2r5v;XezSWN`f5a_i~lZqw9A>3N_&WN$JNbb|13OB5I< zssog0j&#$cj+}&_cOLc^Pyr*}zkJn6pKHi1+Hkz;jbjT{?SwlpX~H`FrsaK^d^5BW zZ0@Irp`o~@#*gk2laTbp^NQKC#Q?iu^p+RnL$Vy#DbR`fuk_FPTzu~7Jk}Ex(F+?i zy-BUn%f4)qqkQBcqIWAs1r6(0de*eZ_a)Cem;C|=)XVdrBBLPF>Or`&C)y1Z1fy=l zxMd}j%>V+fbLSKiOn1dm49=>A9oSqQCKkFv=$2nW)Z z=@}WUTgUGurKOv*I#pCv!R||6)Sas+g1V!e7QMbYaa z1)l4UtlLBJ*?B%$P>P_s-@Mg&GcHb(o*}qldXF)}Pbe=X<^A|?z+Hi*p~jp=BO@aQ z3k;l`lF^bk7lMGC0%X5qfmEwz7ZIxEMpIu81It)~#8baZs>7h2tv zVm?zBF2JbF9d6CiL|{;InA#E?eFer(g*dM!h+sEbgOtE7r`2BU)%{Oi?DSgsAA`(w z&m))rzuizCh_3cj-0OMEAEr6#v7+$y^1joitPLsZaM-{gG_zdX%rDRT+fw#J_bqMX z6l775&k>WQz~zFVa_BS<0=8UM|MWwq$x1QwUq9G8K4FSV<_)6gMbPcs2sGy+*)#U- zxrv*2D=;L4mK!vZsIag%z&sRcR{nJ&UmY&=fN05O^(Q_lspnh#Qwz3bh$NTk?Sflnc{9@TCut#q0UM0W z%;&&H%`Uz$>W)8x|GSNkFW$K{P1rZr5n)xcOk=99th@_y$T1Gczq%%Nh zT3(x4?$4T)s}CU;l4s`T1)^MoCc#HF1&SAN5lC7p#PE5KCh02Pof`d^Y*ZGze+OrK z`^K6euW@f+S?q4Ij8WYGx3;}F;J?}8P=%~2W%8o1u0%Q`wE{xOU`dI*Zbfyq6VPb` zn~5%ezpiGIVz&WBzBP+OF1ThS(gaJ(!OcDElKm;75O^wJk!LqJ!7YGI$W6U?R}fro z=qXTtOyeYg-ky`=Ph3$^(MPY4Q+_Hgn)0MyhYi%uo_K*uj|=zHskEmd`%}Ze#C)-z z5f#H;3LMqd&^Z2zMh=4FIWIXZH&ROP%a8#bHCe738S ziW*)%>StF=S65e~=lCGpz!}^WK*PRgT{{uGt85z5VL;DNnCw%QmkOs1PIzaNvCUf~ zq9`pXF`h^T4R{Ml2DT`#O}F^cDkPeN;N-GaJTRbqngT#?y2S&Oaz`C2$jb&gg<>iy zDzFIOCbuB#;NakFtR43D_O2Gxek4ZL$p*{~Xi42bk>QGh<7;`s;W_k@bY^fr`g6^V zqD770x>^gBXG|6k_4;e$QB4GE&A$N#O^8ba&=v4yeRU@kMZXjEsbc>LIbL1+F9Wu& zSkAzq;Yej=WhCa<-VSVh*ZS=nKTwAQY2c3qt~2fRThQh}t{R!<&xDza9;dV9Wk|_n z9!~6}UMOTI2#kCUKusXcIiSiOZFjO<0Nw=5i~J8u-DYzrr5IQzrM!AUh~xG+cRhUw ziV!gH_t`DM3xHH@yYwHs?5}~eD4uyWEiH9GFJ(aCpihQnXmD^E4gByxSM5x6k*w?; z0|bruNhAPZ)KQSGDXUPcdiy(x74_rnT};8G5QV$cuMG^e-ga?n3pzj|ccjJb0}XC; z8ebx$Z3lkf@}9G+(W_UpKOz|hK~70o-&w#63}7P28(uvFUkk)=?$N+;sByLB_4nLX zkgFV0u{u=PjjXqb?)qUP?7Ik~myU*wTp|3m2w7_Z8h?T zAX^H!?Du98DZxJg0-uzam`|@6r7q*k@T%Kf*wa6-p2F_guv)+lDliiGdV6znan0aJ zW%O-sZi26+8&5$a-@e%k#G*=@5ba-?=Jzg77GW-RZ_bj{>~B6j!tw1O6u_jWYDfouBMI z{IGf6^B-7woRDzxhLr(zk@GxDsReYaO8D3{HOKJpNEgy>i9&S;Fm;y-?tuSzmv_7O zt0y#b7ZpvxVO)0`0&`JWL#%q(^!mRP*My z93vwm0avy@4@OS`*TriHQ`+u+du9lDt#!{bW(W>3{r} zwbK>EnHTt_9Hx5z)}Q2DfGx7%^tBw#Z5{g6Ng$Jq9^GxHJ+}HJa+`!?EWa=l=>*~) zfO;3V(Eo4mmlvgg)m(6=BT-~ObDe+}`p4o9n3h7Kd>uL&3hQj<$pvz9axhqmB+#|1 ztSOroks#^tcBFL1gk>8}Q!Ts0Ln1nt&z}->3Q+=g@W{D`RIDikv@+^VVah5*#$?2098bde!M-fTALm6~~>NFAHhKjCL{~ZwOw*7KtUkaY2Eq~#B1oCub#~ipVc^_P-6sMpLlROpT-c~Zx zATHNfcNu*MJP(h@`6-y|1drw-08KlxBw3OZ5+1O6dL&%V;J0zo-P=d5lON{0)3dUA05a9sfWn$x zmMO}tAnTX)`AL33j(bWX#`!=ce z^XJ3X%Q~ux>R(jz9?US*JiF{~!je@?#1i(H*>hF^hlJPQzE8ID{ij%~`J)UWZO>Iz zGly3-H8n}GebpI1OHzZp%F4QB%CM1}lXC>-q;~u)(UiNKrpn^t;)RWYt)>ErZXYyi zH-E_7c>DG(AZYP|Q0PGbR%jCz+=17vZ+6QSwkeL_Aw@#~(v%RzzHkg34mSLY)O+ku z@P52IWKvO}R#sjPvTj3J*#K~TQ8lW0_1iKNU~m2OX>MTwAP$h+{^HOX20}wYZmM!f z+gtbngZYxwc){dxgL?A!b}{3+>={o`j-L8Wz5G?hYF$C(m1HsS&t^Ea0^`VFCbTmB z%}uL)1jTLRP9-f5*9%*@AUYHj$WN9V`9%}t2;S^Z%x zcB@16o6R@g3nf%k{Q)5B<`TW@0H*At>)r#_o5?{-*>vD^u2}f|Il?0`)2En~HHL^7 z>3t56zR#(`&&J+sX?kUYKxjaoUZipzj9! zj$R544eX3oRaUhh!oh?We&3Chmby88`RTx}>$K>JBT#2IHayVCFm8DOnOaTYxQEuv1>pIDol$=AsH==zG^Uq8s>bJLj$Ui9XUD zdp|=p&*c3cg-;&b{OHNXo=x7hY%~4VTewh0Mh*iTJJYAyS*)kP0vG1iu(6jZx8lZI zJPFgC7#IORo^us)|4w}_Y&_cUEIc2+;Fnt%#Hlc)Lo!_dM|KMSm}03 zI8?>NUa8n|22#F>Q>K@bl27&Fipx(f%I;5l45XTbg#-61Kit$aPl^W*GSFDH&J2`H zY?P=tzvSoV<>$R(C9|B}IYJ7&z0jLL5LHBoeIWI1rfeWkCQO5I2#oFav(sGu;^DD< z4Vz|e$hW!~cv*PXFU46ko&c6RBQJI8%;HOEu$sA|bscFwe{iY1yxVv$D*xmG&9_dL z@_q^QNHt|H?n>JO2`y&A@(E8I==)L}6HT#h52c)N^FF6~$SOh;p(IvMnPl3F<*aRN zEQvMrrS8j@x>f^yR&92#b3Ybz_^ah`qUZ)7Zeah@ z!{_2u&ngrDg!Fb-v`ja-{CcO8C8cmjkEzbC#!`2T!ul`(O{b8Xu$wF5Wh?kL5b z_>IBG%fw*BPF(Ph?68*A)}83_h|)V+G?C_3i!q%73DBlFHa&Q;&x zsfbBx>xvjL_ZY2)84=1_N@UNSz>s0F?W$v3SGOYdO9~`0c1Dta^zehIyw5K~2m8u0 zWLf&6(OrNOX8eKhYh`S_1T4%#SDZr|sRN@l1t zN&;m7`sNg4NV}gujNE!tJW!O{_j_RAr#}BU4L)XhEQhHvTO71!s0aD(Y~I1qkj>l< zkVYP)q&qmd&>r>z=}$o{vHFXT?}iL=~JIR`G1moX5p!d<;TJqex>p*J()o?jJTII%tj`%ci4G@U{r zun~ZnZ_n8cK!M;HaQ*q}#WP?Yiv=sD@QJ#Za&Bzw=zi%$AKr~&u2%+ZF&*X=>R;xH zs;eKcQRz=n=bR);=X}Y}yD#%PE;cTWYTBIL?{p;-GXgFVybPJiLU^bc4(oqN54T=@ zH@Hc}7G3R#s03{l!HoV4v@kPZv3PB(CzzvT^08lW(-+KaIoecz$<2)&ePwA8?EkUZ zh2bPBJiG)vxj=kmW@buf^krt!j0PA&9(eHJH5`*01dNU=Iax?EC+yKY<=j7~tj8Dt z10x_MurQx5*;MoESg9M+K|_zapY&@ERi`=~uh8LDz~zA-1F%j0D)MtASbY&8y7n7N-lE;clJ8p3)y2>d zmf3|?JUhd@u07eeTEs+A^<#6amTo%TiCnKVm2)P}e<(Lqr&9>me~*}R#RXHKL)EQF zM(Sot)}Tt2OiJS;4Y?{ZOke)fL$ zF3~pT4~EgzzyB>SJsq9a!Ozn7_hQ1sxBHF>qd%fq-^G88Xu4%;r;a~#ijRzBif?g5 z{$&(PUGntgBJlI>uVrM5>j7o~uf0-^R}dxiH07de24`r(*>Pz7F7E5&rPy-%K{OW! z=jGKuJ@k#qy_UX%KZrhkj%6f3doFgFe3n8$g||u+%L}jY{f@*cH8yVg~@Q9WC(`zz6-wV$RpY~_!tyQLfol*S5B}Z~Z zzxauSL%cXx(W*M@KjY4d>rlX)5pAV!`tTtR+!{B)wulfJ^dTRd0hmIcpdXa z`snwm@%&p&EiCG*4qwRxKBUqjb-WL6uYE_Yxuum^dFiWPOi(1^P7|yQNkpF~V+5 z4`UJiw3Hkl>g{zrUelekS2p+-`qdNZAXp(G?vA(EAVa;ZrOskI=b_2j&k1+{4%5jACLrnWP7@zMmu_pGFXU54XYh1;xsjY z5BC-kK`xYNN+^Y2W~QOd?;*pmP6OD>6WKhnRJSzq36Yu5?0Ne<9!9)HW>Ww%lva7Q z&HA_M$H)?(Ts%Up$E<;^DpIzPrBitDp&QX35R*2TK3i!7;9_myN;KT1oqg3hq%FC3R}ih{&@&irJxEw^rwbz3THJ(${&5Y^K@Yb~`X9rG4VQA&9U5s(U9 zG$)iKFSjEFoEbdv)6=H*X)})mlUN7>%}mX1iCR{UE!%qMwfx2oC}mDq51eSZgw9~B zyT*R=$`nnpdi>|u4;v4GSr6b7DGUqUOQz)y6;e@=qN)0S4oKgz zIU6T^QGQMiq&&=p%-j-QdH3%#pLeA68K?hfxpEs;drN2&lDYbSZ)J@^4R7`3uu46n z#O?SFG&wTeW{8tOU*3f$l`Na@v%<7$Uu%bYt#->=Q`!cvp-QLVk8pXyb=f90{&%Ha z9Qj-V=$cy;gqH^ffrt`zVnZOqtuzQ$s?S`KLnEG*5pcj)owvzJDGMbc?Agy)j#tfN zQl}2@oJT%)6cM^Pwp_`Pd7J79GET0M+q=xJt=FPD=eL|&L?I7JG%W^-H`nm}r+&P{ zHiXFVS}BL)0Td5N$dg*0oRjUi2eA+02Zg zoR)*bU~W^7(dV0mY91UIjNEudMuHF$_E`G#>-C-b3iNffSnk9)!&5A+p6|>unzwtHv#n9tqMoPE0uK=3HaNYx{+k|6(zrhNfo7 z`J1)&>HF8cy$9r>9_XJVv8sC$u7ojB-xT>_+W^7ykplfXWQ8#A_^$vxKP6MV7j6PT zL+#5P)rkzw7?_wdJ5D#Fj&0N#339}#BBVxw1M(n%%hKoP7Et=iXWgdsQpmqvFGGQg zg5fsn7e~b9DOalm%(j^5!TR`dR`GHPjU9AwHG7Y@_=aGS3(8)( zVfl{XnnwxhyA%8YHa zvqmT}kJ;#(1fZAUrLZI*UCD0)y+~i*5wdRK$7HQ%)_OwG6-C{SM;V|<{1 zZ_3VsQ(%E=(x)ULN?E-Q0{EV@`l8n}u*TWAT`L^E5$qPjMVEG#M{>D~3pAkZbffX& zJ!mu@>VX}*9ScluxACUv)nNq5`_$^{`MI2a^QV{dO@z5$Z&O!@Au+}59`_&U3FtxK zxUUa-wCSy{A(jZD!DS_!qV#Z9|JBIHe?wT{eB(rZ{*0Q)DvrKB>G7^u9dUs4s9m9E z;^gH~|7MR5;qCI5b832enP!`sikgEPE+PR5Xs`J}i}--)|MRmH)V~Z_l5SWFM+o=6 zy;=d4%5F|kQ7@G3r#*61VaBJw8Q&usZr%llpN;(K$mq}$Y$yIdRje-@pA+{6@{=j- z`$t$<5ZKURW0P@n1)5V`iMC!Se7m=IbB-&ihg^t@frlr0B9!On>Ruv`(+>l#`#fVm z!F@)1&M(1<=6i{!@_a>^H-@$pO;HO_e!{b90*Mxrg%ftf$ohWW2T^2s1Fldyc28mm zn+ascNo?Ns75QA6py&o}S?gq~-mm%;VOKj*?TidGBqN^_%<=!)*Y5f7SKOar+wSx- z?Zp1KAw%Zo(k)5JiHJJH@BS0NugNPaD*AAdPQKz$MTWgQIExSQx4(Hgn*JUb*E0D6 zwIl=re2lI2_30lB^dNNtNFRgZ&)J!n)7;#gP4#OkFBXh7?2z)Gv;SA{!iR{{qLGo7 zP4=-y8s@Ac!TIztWaA0pf{xR6u(XR>Io74hA0HJS?-fP8 zNQghWW)I)GV&IcOan+gm;>u035MRi(Ok6a49MoRK3-g4(NwMK?)TMtziBXoeT%0H5 z0si(O@55*Gk&6F7SrsuZ%A2wA>MLaTh=^ihV}BU)UlyHBh>XF^*-rKs78M_w^^uD! zm?T7(R&PGEZoj$*XAb}0@24-hLS{Xddw)BB_HH^7Uhsm$97~a_le!Ld5~f{lZzC-> zd>UJb%Z>jVP(`$eC)+h%o9{t!G1#cqB>yW>5I0a<8x-lj63JxN>^ zD=@R-*TMTMAK%CLAh3)nkr+3?bYaz&5BhnHjr}f+-Y3H8V$1ooq74NZxKp86`|m|1 z|FW8enGxu)<0oSg&(Qw|3o?`yN!`Jq&@!FO6p!hA3zAIdoWU5;wz;ow-Uz?}2-8~3 zc{KxbX#XM-9>g{hF5BZ>lWEiPK~o7oTz_%X^5e`nnnLxh5ZID%x61lmk;bQr9aj-e zP0BjY{0jplaK(p~!Z54jTSKdu)QldfmJi>;BH^Rfy4kqSJ}+9(kb68+c#I+vM8d9t z`MNadA@f5fMpNA_i*H3KAe^~>L{n=IjndV5CS@6NPY|n5J*balo1Q*+j_V&rdL)cY z_Sdwv%!8ye(tRKcamYE-H&i9c<79YaL0b!^x2Vn7>Yh$7(UijLpHi}ZQLj2}SFlj4 zDu_VZ3Wcysg!*qr0(cl9r=h|GbmbqN0iRL;H9%yd4^ti<2p6jyTes!`>_7=9jFDh? zK!<WJTWwE4 zer*BMCE~AGf=T3zc!NWO5n^#&Gag>HndQK7&H$D-t(a_W5>1we+HECE1 zP_y!343Z$J6KF}h`u-iUs0n2nTRAaU+9xdQDMOGRI3~^)t~tdpun9Pz=}|-c@|i16 z6)A?FpJ~VuTeM!YpM`Y%s0pl>YisfAV-Bw1t^~Q6GiEBJJb4(oPm|#)0?g`h^+BC< z^U@JU-sf%JjUqTx}8>&P`^X-pNKzkO?Ki!Vrty-QWCHPok{`lCrx zLl$a=-25v`+L+aK)zz)eIghHVb9ia^0(urpp%{VsCvHMM@&~Z2;!nJ8z_J9%(sNc; zXfP6ZDZZ7F5l3=H6)4V8FhM1*i=eahe*n`3Ol~noNj)9BBAU;BtlKZn&jY*V3XQpr zE_~1^_#@&9!WK#7y?MlG3;eVItQ8%*nAm(i#8wl4O)KV)wVB;?1d(}?CcnNI=Gm-Y zgN`Axkf&%yae72xUQL#SD;9;x=^5#r{~i$%gKYdb83sN~UbvT+m*=yjT^&bOIS5-g z=QL)84C3f7zNq@EGeR6UL`V6MOd;YXcApne{gdn0F;K3$w`7z~Kwa38u9Wj>BD*S} zZB~kUCYHEvka5?*2p3L!mn9*lxAgV@TOj`KgZssH`+LgDU!0N~vc;{gtb7k99pip> zMg9ZC@psr>K3)JsXI-$4kSanx5HwsgsJegiv#Y$mu%gKN1qQ1;9D7^hEGAeTOZ`%` zN<)rwU!pf6!HMJHuso| zBWh59A{TDktP%AJ?O*m1m0y56F=A5Yx0u~&!Q==h3md5!jP(Ey4c=t~ZfeW^^nQ)5 zW*fAhsjB>7MX82RP$bI)s2qVvMs}#5p<$ys7PZZe)Elv+LH0(H-^s78zNfe^=DCsc zkLYJ0>yD8822XLI2zoqI``VFQWtgN=cto}y$OkG#Q2Q#<4Vd=j=)HRW9B3OP7QYVd zi&+_=W5L04-*Zjj1Bz*r1#;!Cy$(U0X*XW7L>$HesdKMaJhMOv`Ftzh1QJ;F#S zDF2>G7!^*Dw5*?6`}}~S(W5FYW=Ogf*)SQy?k?xg9v0{{1-R}6{JPW8b$d8MN?PVv z|Gry~dQElTjnlLnS{TT{dHPG=&fAZ1j`5wSd?u(LJkMotQFviTLaw=2A{oOs#uZne zT;w+qT6NQkbJeDH51CuM3LU2a<^Vm6a+zJ0WJJv|a)i+juli73xpoB#Uc@=Nx2T0P zF?d=qF?or=AG%dC0@#F($>5_bjf}i>y3d`+@4!XC2j>cdkusm9p{axwZ;16n-KxME_!PCWd!&T$aL(&6i=49EP$4|94xNeT^Ev zL*+$J&kz>}iTNRru&k7c#6w`*fjIpNKcKVQ-sDAf2U4`LydBu;m^ZQ#J9>PK6g2>9 zdFHMV!!!OQOOpZal}iwh70qRlt4xnE*A7|+G|aM-;!Hh4KJTk|BqX=-Y(}<;s`+5$ z6hW9!?mJhw;zkDF3;8>7DiG56JlkgV;JrW~Qcr1bC?CoM*T6CoHfv|aFcV<|2ap|+Le@|U;b71T^P^pPcf>Y$Kx zr1a0j3=lZs|05@CUa8zcZaQpVb}Fqd$ORzz%(3wfzV*>E#?f`d*`)&5BArpZG85$w zOb=hev?=g>KtwCO+1j?R9>1i-SGQjz8=-$t1iNSy#wC7wM9s)y+dC!F7GmVEzbFj- zrn}D-e7wKPk!#)haC}CNK-4H}s3z(v`fU)gI~!CI%v_%ckJ43C)e*V*6pMtI0JYJT zt@S^Eb{mD2%j}aZYS%)vl9G~!;y?q(O8c!A3YpIA56!7iYumW&wok(h4=}obB*6b> z#;e1iM4b^b9T)+?g#+~a^GsQ23e)HObAD6kKm2|7`3sZNX>drBK9Ah`(+9p^zox!@vYMwj$qhtz=RN%Am^m%4B|#^MYL4DRa91WZ>6 zHDm(c+t0+A8JW}%Xo}#Kn67+Y+KcoCS-hE9j%AZIV1SsIn4kbY+pEYYcRRdN&Q47& zo;O&dRRF_m`UvE;Va$-4uG`?Eo{jk1x1Al&vvMf^Vdh@)uao>&5EJ(!bcRUrD3!I8 z0DFFi!jTX!H6YkO#izG2_E|~_CF`$W14WRDVi{1J6SSCH-j*Q8hQS|7NoTOpp4?4j z)%rIH22K#zOP*V^ z0&6W26U7t(un>lAjRg4apj;Js;^Ra7Pi;nqSdczrQZoMQl1ILPBnwK<%%RKTD|`@; zQg7;p*`J?Y&xrD{0*6^wJZ}UsvwD#*bJT&hYdmtYnw~BVO=&g$KiIvb2GT5I3By}7 zF&Z~y<00f+nw*wq7`2w8hYVHbhDc7}AKKTubeSV%abOEEwty{6-@T}GSbQT#y4upz zgMmwdVla6ifPOVVQbo%;V@N;mWoz!hn3-l?zJ^7fN@1mQzM5`cMMb%WrbE4HYfTxm zME>#ZH=a}_8+^ZP&-ZyA z=W!gT^w}u!V^Ix-o2YVFgHJ6x73l~ z^yB3fJyi-a=?t1;t%(T{5y{he>Bcv2{+J*?_s?N2sVXoDi#tS8YCxxTxsP1uh?Ur= z;mwGg@H5LEeVGCsGz{WJ_1)d|A3w^h70DfGXl!Vobte@QQl&F9`||T!HX!>)k0>|V zV5Yu$R<$*jSJzH?S5}eC z{9`5hZ|@02UUdQlsS!?(dnB@=qM}Qj5$d<7sj1t_8n9!YOumu&8dhMSrDe3e@|Lz< zBzJOh@{WY%of=m%sS{~t-}`@fRKG-1&|c@@puL^l?A%<7@yg^dU5-Kdo9V;WGDb!( zM}C|7QS|%|5q|6P%k9@#)Ac$}G(Ote8joM^4`eq7 z+tAERg=2i22?sx0n3ZusVeah<`DabagV^Cg)OHc=iFx-YM%OTr@J)!Ui@d9=tBuXO zY=z{5>o}aq2u(wF4K1Uc=^QJtT7whHE7!+9RtMHxFbF57W~CU^t{w=eK z>){+$-g&Nrh%wv@6l>RPbo=i6Ip8?`uK^1yC1)BJyCJ1xGMvT~<9x;5urlWLy~Fvt z_pWeyd6hqEL?>9c*Th#gIkf1;OF^BsVR4BbC@E2&HqaJ0Eei+;PB2|eA_Es6)(@Fq zAOHNxa9sZ6v#(VORKC*VIP@$JIGz=(;iD_(|Mc{>f%!xRDhKE43 zf`b)|jg!F6+)(Ws8cKZm(g^Y$Jw1L#%7fJR!H30f0_kmbW+wW@)A)Fnefy3wJpAUD zJw%Xtz&;M=&f1!*SiF>Nk1z)ZMP>`n-&Xn3>;~ zm67h*)7#g__rY5=g%?D^1#BzuAyC^iTauBGkdTt{l?ESw#3v}&{N+n}W@hwdF0=@$ zt4&6~L^wyB;6CP9I)F`!ZeV~6Qk&4nkCl~`N#PkYGSbm3CHh+JCXB0bqyBiG52^;+ zk(`1eucRbSXlsec!{88+g}`d9!PJd&$cuTi7a-2 zgt+)#64~p~GMul-A90dP-5(<3sMvvs=8I7l(tbF`z>QBmImIq=b8{0UMbMJXZ{F-%^wcJ++0fFJ!`Z2!rwUHk`PC5)#fMd!Z{*v$HtuBFkUgz_*Ggnre zby!vSV%WhFmEKto-^?h^&HV$9a)H6Dq~-YOK1UhMg|`|i)Nb@t%dMU*5ovgn?1eoH^WxaGUQKM3cBqR)iO2b4u4l$(j2g?{z znnTWW1L7OBCr_Q4s-e1`UZKDEiz~SHx?>ma{5db{t&P=uez=(M`UQX;p2RzaL(*1E zZ|rPyb2Ad>H%o$=yho0t6g0)g#9U{r`?X9E+>H_EmXuVv&9j41`~3O!)Jtz!m$7c{ zhrLQl`W~ZYczxZ};&P;Zkxh);kz`&z5fOF8R(K45$=(402N8x_$|AMQebKbr5}L@3 ze30R*PFrO^-%%>nHwZtlV&=KB^MaU2#w}Mx`MI8M+r0}kt2AL@v|Ic-IUFZNPTN^q z+j>xPvMDJdy*44&TJoA_t|pl!j-)c0`&)R;_6PQpy6uD)JDUf#pg zlaJ#Pbek}k?RSC>0fNt!RCt6zQ?Qb%soB*f)>Z84dcM+kUrNfvLGXdWVV;98&++{s zlTK1nQdl(SJy)#p&ELJ_D_c7CvQ-*Jict;tf^rIuFhWIEnw6IqVSiZh6c5ihJkAM< zH)zcv2hmVdyWj>ZETl^K@s)*z^o)$ZNS9-()Zn6KhQ*3qh-p!4ZEz?xwXxyzdz74P za_!m^DD`U!5ZZoy|Neb`ei)n_b@ib0=9ZRXRxPZ_*Uq}m^!)s3%AM-DGI3o;rxy~J z_7y1gr`uDH!qgGpUq@XX*|5dz?tt&G#Im!uPn7h?e$PEAk4 zFdkMMkJl(@Xtv-a1am-~gd(_25aygHsc`xg1{uWJJ4b@5a|;Vu#T{RzrXG!Js;dn8 z+T0AcQEfFfhJy#+-Y()1dj}5|kgs#G!R^K_V8fl*HNkLEaC1}N* zNG5zQ<-LaEzuT(>XJ==36Gvx#CmL+;>WY2g_;Pz+_vb!7_sdU~)Mes|ii-y2E&ChV z14ic4`VX((LcqY2wkEek_=LzJiis@F=P8)jvwfu_ORs79=ZhVND^2St3twku3P?(B z9#_*eg-8X;xrCQS!Dk+{{+fpw1H>23YrH|#XMrWdohmO6miCZ@vr7;C=w3tu0cS(# zMUllezm%4ikdP1-e!|s&dkp~!K^`7rTTn1|Op*j}_1f+%LYkvhwxI z7wD`fN*5Ow5t>$hV6g@UEhZfFMt{(r_kJrZZ zNlaQ8YB_ORSV3U_JPZinKVzLR4Y*lAuUNUkCHg%oo4XYI+L*_(4a!M&S6jXGxkUd7 zTYF?=B$$@1(fkzXlTaby{0~*>#vz|$_(9@-{3Z92oW%7Yd_yiY@afa5w{Itrd4OvB zIX`A=a}$;S%IYf4u}a4=rIlod%5_T^{#a8-r=(-?5d;YKCc@0W=LRp%hoa8G87Jkw z;BS8i6?AD3Ni}kC-?!#m3IlPrkgB=YBmvd;AB}6NYS5 ze6J^LySq9kJ>|BT_rLr5Gejtt7WsbXtLXzf?`?1On_S6-fXY|>pRTDbBUMH zb*%!AN+#|Bjt;Qu*6k78v@gs-#Bi1VZh84%?B21zD4sv8kH`1qXyba5c`2#RGmz|M zWz8HH*g_~;(Sg6=wPk^Oxudt%3nxyo(_Q(vgT_UhI?g*7;^i#-gp@_zOIKU_+u>n0 zQZ_2GEy(^Tg^n;%eyFHuOT@FO*)V3(!lQ|<7)6$4EHbR)dCCTnu}23^?!FpFB8NrD!Fz^`X_52z(I^XSetIT61EaDbD<4^HK&YV2i-qCUH_IH1H zhe+UrZz9WwD$uF-z|KZ@)hE-Fu8*$qJ`NABzbfv&z$t$nUqFThs=Ur#PEf0T3V)$M z&ZEWBkUfWUrD;eLsjh36{Y+Sc^gSm)4C_98VN)l28~C8kxUsNv@#Zp&JF z&Qj)W;j%VNL`lcQCC+N=y7ov(@t-?j;6?5k8&VJ@oNq*rR9o= zre;{=BNv&6xJeLqB%XJ6b29?7xU0Usz52iZ0vpULCF~CMg=Ot7cb;S@Gi_;aCpcLv z*_5^4Z=BM`U+Xq$9=wFtD2Vq3lQ(H@HC+Rp!E=!O6 z^3x-S^nwgFDe8^s)3*xCLrSAkQ^%J34(N!N9*H2&c#sccMuBt%y-IKKLrNJrIXjRD zU*t%Uy~q=_4$1cM3i7CiPg9}rjh3xF&@S7p8LN^gsnd{3O-f9q1`oRQ{Cr2)#KIPe zqHh~wOJ~;z8Wu6-MA3%b9!>?0iCam^moGD)G$k24K~D}*TgK#-BojSK!z$l|)?e@q zj01J|Nvo%2K`+4=I`_~iFuMPOj z@b+ilh7z`Co*Lpn>+24A+DObeqPRRz4zvtqA}*S)-lL?XgzG{_FEe@{e%JltjSps8 zGZP&OAV40$cQ!ZMqy_I@FfsHEAOH#xTn#_|wDYI+_4hkDoy*!qS@w{tfhde)tKKzU z2KD_boDJ%*^}RsXovD*5{Poyh;UBjpzPZ)D12>)&M_cz&W2rF&P*$=zQlKrM^Yhs3++aEK667JsdSv@MIy%~4UFXg(D6lZPd+#1%k$!@DSI68mp_W=C-q-@#E0YW*DC5hTCtDaVth@YtBP(nitC0dCaKtb4Q00 z^u-CQB?Sd~h1Pq!Fs21jZ>YEH5k^&L7p}m*)-;b<&pl(5B^Z>0DiP7sr=>hs$Y*6D zxx{`!kpY{A#(kSe_g8VCB?&&a55tk{6I)VpV9igY@!s9?Z!d7ZSQVeNYPw8MaPk7*iFcq^8Db zZC7`B1Ke$&UxJWrLf$W5d8p;AxOj3f-dwU!{FZi;V(-D2Jz=9l=TNf?3}8+TWa64x zxVk>w;{y#RHnaJ@N~ zRBu!{k>uZJBT|tipf}4e5+p{yGLS^X&^&yhJ32lUm>rfZGUzY%zO?M=k;&K3bi@ZA z-m18J!l3MK%hT}i@JIm`t7FHGDb(*hdc@GjrEiIeKb%!6EcQ$(qE4R&^r<{NJX4QN z;AuQzVv44+I?no-;H%_;x1F2{3Jc}*%zuPKg6{86JIx)=F7<}A(4zv(DB zgh$P~bM6mOS3j4z78%ZMA}J}E+kN$b@Eu3TGT*HY=)ZA_mUyn(a36>;V(IDYdpFnS zqF}<49}&@xELJO9AFQQE#i=5bKQoQ%i6Nu2?UyTLR6Y~RA#LbEw$2^Ub0$PmNJz-m z!_vY6vn94+HH21iHbN;fTnW!~`nAfA8?VlpIXHy-Ts$s{R8-(XUvp6g@N72>EpP@g zvvZ5wbf711f`^XCnv#G(CI)jG?j>pO;^F1JD|glIu%e@Iq!@slCtoMQc3m1H80=PrMMX6fts=P;9y3>ZH2KB(NrBUG8V;DIK{3#|;Y$ z3u~&fC)a%IiF@gk#zcB8s7X@`Ufp~i8cMTvf|^}Yd;K_MLnMZV|t+Pa<5+IXU6XlOK?RJ%xR3(baow$T2Bm}tiv zr2a%%FznyqNa81Bs$p0twUCtHelfJ9N=jA4`(tDzsEAMyOTR8S#K}A(y*%l&g<2ux zzBqE`Xd#DJyi)+v@^fNrdD3Zk1T>xMUwL5_h?vXjTfIm~Xs_M$4*jh)t9v2O{vJLZfc1FZu&odJcadhlt<(-HC@8EO(2kk>7B2SJNKl{XH&l6EInO68YQ4R^ zJz;X^PTnO|(D!@}#y1-lO-0?K8eUvpjvCgh9c&gfsor$k$-6dI%tFMQCZL0kMOkfU z=WC3s0aw_*;cFxNUUpLHvcgZbgP0fjX3WD>%hL6#ocq(wMz*%2z}ho1ZvDy1B9R5| z^DIV0(w29if9m75z1A`2U*l71IyX88HfL(NMEH*#kwl{vwVa9&w;6h84h}}wugjJ& z?%$8_%qK4299iGv_QLipr!4)66DN>PUR+iN!W1~`@_ylVTR#fOr z9%$D2@?4O7SYuO@m!q;zOnG^E6&0nw|J0;DX!JYalKIVo-|Uk84qnR>m$Wpf*4$&7n`8n! zpyrvagG0{StfLRCr1?X}=03Ai9O4_+&J~d=+EhOhft9NPd_zo349ACqCx%!xHt!Oi z@=G&Yw1?UPb3H)vp<#LB$-zWaMnu>?f~QVfl$Vw<##Q{u!0erGU7tUH)5tn~PDUMo zgMIc1?$@o`C-?-%`g;4Q)sE)SJ69*kT4tYUc|BltMGNK>c6OPa`c2FK*}&`G!9hXq z8}+@X`Mq=Ys1I6qhjG8wSKn7*egD#Rrc5Whtgc&OuugJS48A6q>A)tVuRQ z1Rr0%u3@PurV-!Cu1e>&KaxLgPmxbiLsLgki*hvg@!uU)vtFUz1`ex02XDKHc(C)n z;^jMb=D>7PyQ$2KM;?Y+J$?SX1)iyPlLEd_hP{6N{7@k0=P}PGkY$0cm*%L4lgNzF936%qMWQo{)gK}t2%%j zn8j>+zC;{FdFnmecUsgmJtxNjfXUbx6DzBv%{N{bS6#WcI|*T{SQ5eE)`Y(0(Ofi2 zlMM1-^per3PsT#|{GVF(Ikc666B7-I>gTIdMSD%CUK;7EJ+d#P^L&DL{i}v+zgZOw z_I&#!I8s{1dqvB;p<$;kdwY0+n)K+$9eE}94E?-4p*6d-e+?b4CZM(`cshdNoS>R0 z8If}}BZGlg)YkDnEz8H|&h)x^2F3wHJXYr68uf5I4IOn%5DN7F? zE%2!i!h9Y*rP z-HGvWs`KtLk>+yl+I|CzoY48BHneWSJ|UPR&N(#krfzSj%=A~>DwiOz@`Xe$tCA9} zkU9vdIw#-N;A{Tca$n%^?Bb&D$shUAx1ZYKuWO>7hG#VSVk|9xO}I+>JbU<%V$H2w zhO=#g(lC+z^XE$>vf8@3rgS$M{XR+X3*71-0r~(CqdgW@MW!Li0LXV5jPf; zWb9~Pzx=0VFTGUNNX+RV02eFC0Rh!OPulB`-k17&{bBQ~MwP7vHagK43+I2rf%zgG zHg6TGlEg>M49~r-jis_OGd*pnas3~!?8d|LAL4BKDW`XcCBCgFILV68T;24UtbH?X z>A6ZdaR(si_7~gk?lTMSV-l>YMsL#6e$CB=6^}OBsiZGavieB}aw@B+K)hQPsLGvL zfT>)UY9~%*y4&BpdHj&(QSzVYwdD&r$HXhslf)h!=<4J9n?it$H8PO4xuO`mbB8ds zYI~tjDRuQ7pNYuZd`Z!@g8Y};Npi#q5EB#psCuWD-KSb$rx(oO-FU@-psbc*Cjlqx z!>9lLt7)%hw`&mMwTREuuj%)ZwCSU{k*V#VW!1;qw)X~)7v>YdJU*rRsh7{}xAsW%qi@MFH+%KZ#zsmoBJb>l00#~S zA)I7%;k1hitGA9DH?b&2Qj?RnIa3>4jN~fr5k7e`V*B$ZMvGV&!8T+f*Yx7$%jBL& z!TA^q8(k$~N24QdGuRK;ug|)O95F9Jhs}}|@;6tEVrnfTQ_}zbp3jhM z>c91U#yx3C<(R17F#0IpW*!dS-OBz-i-(isMCfj>umd3sGYWQ>#mp=CxR&joIE8ma z6-z6VcUkIrp~t_uZ`Mz}99@jcrqPtaonJqgzo0XQ#4%JU&V*$3G_~l4P zAD^I0IzN0KHJ$-(Rq^hfQtGYq?h7(UBGP9x4GcoF&WAmC&^GZ9UzLbR_98r$^7HbF z3yNg*FfH#ipAnI$shORf!*A1L+ZCI;Ri>{Jt@ify^c*R1yJ~OPIu|Y`to&S*qCaE- z%qp8iehgf*c-YmYnWO4wpsE_~D=H<$c12?DE702ZTDS$DCTly26c-Lt{>5J!xs58^ z^UBJeb3T9;f6{gNG-*%(ll>6v@z;}NSHj3jOV^d=?0Rby*PkY63utavKpWn8&bUrA zI<4psG$HZvse(%(948s&K{>ka!J&foM{+Pk{#T}#rH;gaL*krUk`|XVKDOG;MyF9Z zsolxwvas&$Vw%$vqZr95?OXAtm2j$8j1rQ(2|M4)^T0J$lC@{e&$OMc{BrjSSETJz zJG&(X%W?yOmX_v@S@#W;gs0JuC?pJEW3tKAM`Qr>IE3LZK4-oE(;)XFR{O51V%>oxy1gHBA<{O7g+-}@(x#3boTiz*? zo;$j1c)Iydp21@e-TniKS4!{SCtacUa#=!?djARi2gTli&(K10j?Ay|(&ME68h>UU zBUbQxX(=cUidxfeqqa{;J`5s(waAe02!r}m6X)B`Ep|{u(p(nYctZ*wXrC@ z&xw($I8c+31>-3v3j&+E@r*=pe+XK`zCSigKmFP@cVfoff0n7I$m@MPiwd7d$MHwH zMK(7uM6oNo4hJp}X%jg*V(>K$Dj)z;RWlM<#i5@4=RpFb0@`^19>52D8a!#|9;Fk$=MSmC=Ru~mct?2coU8xtVg zX!yQBf{=#THX)DnFN{Nas8_cX&|oO5xLPsiUYZJ+) zg7C1uk6pUJtwQ8~6_otZ=)V(&47KCW`r&8P+uKL7hqepU58>o% zPrZL9{#l=${cbG%bR-98>uG2ZVVwqS@j%ykOG5keGw3b(y|&<0$HWkNbLfe+dEFt* z^TWl+s=@&yj8$aZf$KXn1H%%wdlhEGS(l)z24u~4uq#~ zGIt7UYPnXVoLd;xZQ1hTrin>=U*GenC?{9|;sAUgmjd*XFi{DN5)SW$U4uJl?XP2Q z&g6|}$o)Zfkr%?)u?P6s_vPhoYYh8^3^&nAP7(w>h)zm8y0>r}lsf_*krFn&vND)J zT#7v0+!)BXzu5_W*&U#yn_F9Zc2^Uo!U0#FIes4g%m`sD=X2aL+mL*so(jp5q@krv z1oaoD!{!my=y$Y{07SqE%%4q2ar2#BYbFn*|GT| za{9DhlaM%CIwvuQ&d`!)+6VG zT^}d;ZW{QHJjZZ}q&?=DnVEQI6ScS&IR%BLSnrKLe=5+Zstu+q$;(6itE!?R=2HrUwQGjXcJf zfD{mz{)yY)tBDK8BT*D%r2t8oj73F^R5>m>dZ5fDqhQ8BNGNAv;Wi6PJg6+l1irXE zp*!@iwI&Hnkb~o8XNFcsD8s`~b#-Ht9khFzL8~J42>Kcg)-ZAp{o;n7*U+A({F~c5 z1DXS#A=>JZT=-pMO-+UvuL?LGAoY~=Dh5%3!yVTC@Cly852<)O5-#IsD@5%QWEFAay6b~$Lj0q@Fd@gFg(LaFx!)t znv!zrs7Q!CD}!i{X8Q*PdO3{7f&=^W_;><0Acj1x-C+Zh1JPaP8Qw*8sPXZk(|nwq z#=xK{r#<=U4!Pd1GlhdTVYDKnJZ!Z)ySqE+5Czc>@9gjQLa*?yew#4va(em%rs}C% z4RS{{V05jpW*8}96-=d|WKejeHW?KBRx2vIV`sZ;)eos-8GE4=$q9v$%Uo*iTZrKq(k)m~{1xQ!&2ux+*q!{67O>Eo zlFoS?71aTz-2^{@Qcy`3QWq&@bF=q{pFUV4#~%TqFT*-nB1|+un|g9)mT>d@l{Omv zx^@%!uYsws_`BR|{(OQoH^W26uowoPWUF)F7Axa{f`$IR|g^{BneVZpLGkZs*)}? zSLFsf=-w=(*O_u#ZsGXttnR4Y~o$B{@Mqq2|im|w0l?)8f z>QEtRPfkuI7%2TpaLawvPk7XA`v7DpnqpT>7~cRrLmf`kSfO@DtPpmSNFTij~kX`=%K6J)s88W`NNv>cxr7c+H# zTDs0Tb_^pJ&?xKc(@nkx`<=6_VL?F`-7cva8Yafa_jh&i><__yL{Lt6@gh3MCNn)9 zghPj$H}{6rA(W%!2T5Hjl5S`Z9w~wbE{#zhvjVB8La?vg_B)TqNH;h7=2gl0^)C9S zqz4A>h9SnUW&9jI5WnrszOdKFhTq>I)_aC=-RRmP@Pj_V#>z@ql={vDIE85JALM=V zQOcaO`?tFS3!wgaT70?9UEOwI|E2h1@`kFkyQ4GVQy0iSKG<8rW zA%n74?Un>AifH20pFvv4`M4BGZ*R7{kOAweO1c(&@lem@Na}^(X*Z?~6%^^`omn zWA`I=apLvur20JV&xVOY#bZBX7Y|9iWuhu;y<4}mayyOgic)Dpj3?1xnsssK;2V19 z0Hi(PacMKtk!sl+@fQM52&2Wn=DFBz@R>iy!Mg3oKP7U^M0q5zduUVbzdKFcay_!=V8bmp9P3Qx*dCfc|6bmb#Qjl(#k`USmx z?uUg*AHV-$&RoBnk#vQp>?M1=hzDi3nyHKiP>X= zc1{fpr@-IUGS=tEx@Mb6+!OMr6H_>S-5 zn;+ka!m{g6EtHm)X=vUkDGAxVxg{t__DA82Jb@=^9Rf9Hr!lU>KaPfC$Bm3=LjBLC zCQk=Z&AktAva&)O%>QE&56}cv?qH4W?^rM(FJ62)I|R$7r^-B;^z&@6Q9~Tr*dX4i z+5!0suu*#~{MZSD9Lo+7hAk|p5m*=hUP-NJE;(SN4(ggPF`&&LOMYo+P}E8#tYJfT z0%-K_wZe49egLRDYWH&>3VM2$d6OYpz?%6CrvBE{m|{@E2k8s=+3(!D@Pz-pWQ=~y z%zPbL{5IKOe##_-@UD!P@h}v{|ND|5Bf#|kUTs={G*)*mm=yHtvJv+0!l|JDd!7G& zAc>d&@BY7+GohaUe=ixb|GnCOZ~p)N=N*nK6o;-3q*6a7{BlM4t8&>Fj067<3x843 literal 0 HcmV?d00001 diff --git a/introduction.tex b/introduction.tex index 607de3f..35732c8 100644 --- a/introduction.tex +++ b/introduction.tex @@ -47,9 +47,10 @@ often fails to properly segment the pixels into the correct masks or assigns bac \centering \includegraphics[width=\textwidth]{figures/sfmnet_kitti} \caption{ -Results of SfM-Net \cite{MaskRCNN} on KITTI \cite{KITTI2015}. +Results of SfM-Net \cite{SfmNet} on KITTI \cite{KITTI2015}. From left to right we show, instance segmentation into up to 3 independent objects, ground truth instance masks for the segmented objects, composed optical flow and ground truth optical flow. +Figure from \cite{SfmNet}. } \label{figure:sfmnet_kitti} \end{figure} @@ -66,7 +67,7 @@ and predicts pixel-precise segmentation masks for each detected object (Figure \ \includegraphics[width=\textwidth]{figures/maskrcnn_cs} \caption{ Instance segmentation results of Mask R-CNN ResNet-50-FPN \cite{MaskRCNN} -on Cityscapes \cite{Cityscapes}. +on Cityscapes \cite{Cityscapes}. Figure from \cite{MaskRCNN} } \label{figure:maskrcnn_cs} \end{figure}