From a35c87e9f826ae07373de5146f7131ec76dcfb94 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Fri, 9 Aug 2024 15:28:57 +0900 Subject: [PATCH 01/14] docs: draft for JOSS --- paper.bib | 49 ++++++++++++++++++++++++++++++++++++++ paper.md | 71 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 120 insertions(+) create mode 100644 paper.bib create mode 100644 paper.md diff --git a/paper.bib b/paper.bib new file mode 100644 index 0000000..b58caf7 --- /dev/null +++ b/paper.bib @@ -0,0 +1,49 @@ +@misc{byambadalai2024estimatingdistributionaltreatmenteffects, + title={Estimating Distributional Treatment Effects in Randomized Experiments: Machine Learning for Variance Reduction}, + author={Undral Byambadalai and Tatsushi Oka and Shota Yasui}, + year={2024}, + eprint={2407.16037}, + archivePrefix={arXiv}, + primaryClass={econ.EM}, + url={https://arxiv.org/abs/2407.16037}, +} + +@book{fisher1935design, + title={The Design of Experiments}, + author={Fisher, Ronald A.}, + year={1935}, + publisher={Oliver and Boyd} +} + +@ARTICLE{2020NumPy-Array, + author = {Harris, Charles R. and Millman, K. Jarrod and + van der Walt, Stéfan J and Gommers, Ralf and + Virtanen, Pauli and Cournapeau, David and + Wieser, Eric and Taylor, Julian and Berg, Sebastian and + Smith, Nathaniel J. and Kern, Robert and Picus, Matti and + Hoyer, Stephan and van Kerkwijk, Marten H. and + Brett, Matthew and Haldane, Allan and + Fernández del Río, Jaime and Wiebe, Mark and + Peterson, Pearu and Gérard-Marchant, Pierre and + Sheppard, Kevin and Reddy, Tyler and Weckesser, Warren and + Abbasi, Hameer and Gohlke, Christoph and + Oliphant, Travis E.}, + title = {Array programming with {NumPy}}, + journal = {Nature}, + year = {2020}, + volume = {585}, + pages = {357–362}, + doi = {10.1038/s41586-020-2649-2} +} + +@article{scikit-learn, + title={Scikit-learn: Machine Learning in {P}ython}, + author={Pedregosa, F. and Varoquaux, G. and Gramfort, A. and Michel, V. + and Thirion, B. and Grisel, O. and Blondel, M. and Prettenhofer, P. + and Weiss, R. and Dubourg, V. and Vanderplas, J. and Passos, A. and + Cournapeau, D. and Brucher, M. and Perrot, M. and Duchesnay, E.}, + journal={Journal of Machine Learning Research}, + volume={12}, + pages={2825--2830}, + year={2011} +} \ No newline at end of file diff --git a/paper.md b/paper.md new file mode 100644 index 0000000..dfc041b --- /dev/null +++ b/paper.md @@ -0,0 +1,71 @@ +--- +title: 'dte_adj: A Python package for Distributional Treatment Effects' +tags: + - Python + - Distributional Treatment Effects + - Variance Reduction +authors: + - name: Tomu Hirata + orcid: 0009-0006-3140-291X + equal-contrib: true + affiliation: "1, 3" + - name: Undral Byambadalai + corresponding: true + affiliation: 1 + - name: Tatsushi Oka + corresponding: true + affiliation: "1, 2" + - name: Shota Yasui + corresponding: true + affiliation: 1 +affiliations: + - name: Cyber Agent, Inc, Japan + index: 1 + - name: Keio University, Japan + index: 2 + - name: Indeed Technologies Japan, Japan + index: 3 +date: 9 August 2024 +bibliography: paper.bib + +# Optional fields if submitting to a AAS journal too, see this blog post: +# https://blog.joss.theoj.org/2018/12/a-new-collaboration-with-aas-publishing +aas-doi: 10.3847/xxxxx +aas-journal: International Conference on Machine Learning +--- + +# Summary + +`dte_adj` is a Python package for computing empirical cumulative distribution function (CDF) and distributional treatment effect (DTE) from data obtained by Randomized control tests. This package also contains a novel method to reduce variance of DTE using pre-treatment covariates introduced in `@Undral:2024`. + +# Statement of need + +Since the groundbreaking work by `@Fisher:1935`, randomized experiments have been essential in understanding the impact of interventions and shaping policy decisions. A widely used metric in this context is the Average Treatment Effect (ATE). However, exploring the distributional treatment effects often offers a more nuanced understanding than focusing solely on the average effects. +Python is widely used in the research community recently with its flexibility and ease-of-use in the user-interface. However, there is no popular Python library for computing Distributional Treatment Effect from data obtained from randomized experiments. While scipy provides a method for computing the empirical cumulative distribution function, it lacks convenient functions for calculating DTE or for estimating the variance of the distribution. +`dte_adj` was developed to fill the gap by offering the functionalities for 1) computing CDF from data, 2) calculating DTE and its confidence band based on CDF and 3) visualizing DTE. This library uses `numpy` as input and output of methods, which is widely used for matrix computation in Python. The main classes of this library also follows the interface of popular library `scikit-learn`, which makes it easy for the users with Machine Learning development experieneces. + +# Functionalities + +The high level functionalities of `dte_adj` are as follows: +1. Computing CDF and its variance based on number arrays +2. Calculating distributional parameters and their confidence bands +3. Visualiving distributional parameters and the confidence bands + +It currently offers two classes to compute CDF and its variance. +- `SimpleDistributionEstimator`: this class offers a standard way to compute empirical CDF +- `AdjustedDistributionEstimator`: this class offers a way to compute CDF with smaller variance adjusted by pre-treatment covariates introduced in `@Undral:2024` + +Both classes implement following methods to calculate distributional parameters. +- `predict_dte`: method for computing Distributional Treatment Effect $DTE_{w, w'}(y) := F_{Y(w)}(y) - F_{Y(w')}(y)$, where $y$ is an outcome variable, $w$ is treatment type , and $F_{Y(w)}(y)$ is cumulative likelihood for treatment type $w$ and outcome $y$. +- `predict_pte`: method for computing Probability Treatment Effect (PTE) $PTE_{w, w'}(y, h) := \left( F_{Y(w)}(y+h) - F_{Y(w)}(y) \right) - \left( F_{Y(w')}(y+h) - F_{Y(w')}(y) \right)$, where $h > 0$ is an interval of each evaluation window. +- `predict_qte`: method for computing Quantile Treatment Effect (QTE) $QTE_{w, w'}(\tau) := F_{Y(w)}^{-1}(\tau) - F_{Y(w')}^{-1}(\tau)$, where $\tau$ is quantile. + +Lastly, `dte_adj.plot` module can be used for visualiting the distribution parameters. The examples of the visualization are available in the figures below. + +![DTE](docs/source/_static/dte_moment.png) +![PTE](docs/source/_static/pte_simple.png) +![QTE](docs/source/_static/qte.png) + +# Acknowledgements + +# References \ No newline at end of file From f0c4af66bba272af7158c82d0cc9da893d018536 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Sun, 24 Aug 2025 21:21:22 +0900 Subject: [PATCH 02/14] yodate paper.md --- paper.bib | 26 ++++++++++++++-- paper.md | 90 +++++++++++++++++++++++++++++++++++-------------------- 2 files changed, 80 insertions(+), 36 deletions(-) diff --git a/paper.bib b/paper.bib index b58caf7..851c0b5 100644 --- a/paper.bib +++ b/paper.bib @@ -1,11 +1,11 @@ @misc{byambadalai2024estimatingdistributionaltreatmenteffects, - title={Estimating Distributional Treatment Effects in Randomized Experiments: Machine Learning for Variance Reduction}, + title={Estimating Distributional Treatment Effects in Randomized Experiments: Machine Learning for Variance Reduction}, author={Undral Byambadalai and Tatsushi Oka and Shota Yasui}, year={2024}, eprint={2407.16037}, archivePrefix={arXiv}, primaryClass={econ.EM}, - url={https://arxiv.org/abs/2407.16037}, + url={https://arxiv.org/abs/2407.16037}, } @book{fisher1935design, @@ -46,4 +46,24 @@ @article{scikit-learn volume={12}, pages={2825--2830}, year={2011} -} \ No newline at end of file +} + +@misc{byambadalai2025efficientestimationdistributionaltreatment, + title={On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization}, + author={Undral Byambadalai and Tatsushi Oka and Shota Yasui}, + year={2025}, + eprint={2506.05945}, + archivePrefix={arXiv}, + primaryClass={econ.EM}, + url={https://arxiv.org/abs/2506.05945} +} + +@misc{hirata2025efficientscalableestimationdistributional, + title={Efficient and Scalable Estimation of Distributional Treatment Effects with Multi-Task Neural Networks}, + author={Tomu Hirata and Undral Byambadalai and Tatsushi Oka and Shota Yasui}, + year={2025}, + eprint={2507.07738}, + archivePrefix={arXiv}, + primaryClass={econ.EM}, + url={https://arxiv.org/abs/2507.07738} +} diff --git a/paper.md b/paper.md index dfc041b..d2885e0 100644 --- a/paper.md +++ b/paper.md @@ -1,71 +1,95 @@ --- -title: 'dte_adj: A Python package for Distributional Treatment Effects' +title: 'dte_adj: A Python Package for Estimating Distributional Treatment Effects in Randomized Experiments' tags: - Python - - Distributional Treatment Effects - - Variance Reduction + - randomized experiments + - causal inference + - distributional treatment effects + - machine learning + - variance reduction authors: - name: Tomu Hirata orcid: 0009-0006-3140-291X equal-contrib: true affiliation: "1, 3" - - name: Undral Byambadalai + - name: Undral Byambadalai corresponding: true affiliation: 1 - name: Tatsushi Oka corresponding: true affiliation: "1, 2" - - name: Shota Yasui + - name: Shota Yasui corresponding: true affiliation: 1 affiliations: - - name: Cyber Agent, Inc, Japan + - name: CyberAgent, Inc., Japan index: 1 - name: Keio University, Japan index: 2 - - name: Indeed Technologies Japan, Japan + - name: Databricks Japan, Japan index: 3 -date: 9 August 2024 +date: 24 August 2025 bibliography: paper.bib - -# Optional fields if submitting to a AAS journal too, see this blog post: -# https://blog.joss.theoj.org/2018/12/a-new-collaboration-with-aas-publishing -aas-doi: 10.3847/xxxxx -aas-journal: International Conference on Machine Learning --- # Summary -`dte_adj` is a Python package for computing empirical cumulative distribution function (CDF) and distributional treatment effect (DTE) from data obtained by Randomized control tests. This package also contains a novel method to reduce variance of DTE using pre-treatment covariates introduced in `@Undral:2024`. +`dte_adj` is a Python package designed for estimating distributional treatment effects (DTEs) in randomized experiments. Unlike traditional approaches that focus on average treatment effects, `dte_adj` enables researchers to analyze the full distributional impact of interventions across different outcome levels. The package implements machine learning-enhanced regression adjustment methods to achieve variance reduction, making distributional effect estimation more precise and computationally efficient. It supports multiple experimental designs including simple randomization, covariate-adaptive randomization (CAR), and local distributional treatment effect (LDTE) estimation. The package provides a scikit-learn compatible API and comprehensive functionality for computing distribution functions, probability treatment effects, and quantile treatment effects with confidence intervals. + +# Statement of Need + +Randomized experiments have been fundamental to scientific inquiry since the pioneering work of @Fisher:1935, providing the gold standard for causal inference. While most experimental analyses focus on average treatment effects (ATEs), many research questions require understanding how treatments affect the entire distribution of outcomes, not just the mean. Distributional treatment effects (DTEs) capture these richer patterns, revealing heterogeneous impacts across different outcome levels that averages can mask. + +Despite the growing importance of distributional analysis in fields ranging from economics to medicine, the Python ecosystem lacks comprehensive tools for DTE estimation. While SciPy provides basic empirical cumulative distribution functions, it offers no specialized functionality for treatment effect estimation, variance reduction, or confidence interval construction in experimental settings. Existing R packages like `RDDtools` focus on regression discontinuity rather than randomized experiments, and lack modern machine learning integration. + +`dte_adj` addresses this gap by providing a comprehensive Python framework for distributional treatment effect analysis. The package implements state-of-the-art variance reduction techniques using machine learning models for regression adjustment [@byambadalai2024estimatingdistributionaltreatmenteffects], enabling more precise DTE estimates with smaller sample sizes. It supports multiple experimental designs including covariate-adaptive randomization [@byambadalai2025efficientestimationdistributionaltreatment] and local treatment effects, with a scikit-learn [@scikit-learn] compatible API that integrates seamlessly into existing machine learning workflows. This makes advanced distributional analysis accessible to the broader Python research community, supporting more nuanced causal inference in experimental studies. + +# Features + +`dte_adj` provides a comprehensive suite of tools for distributional treatment effect analysis: + +## Estimator Classes -# Statement of need +The package implements multiple estimator classes following a hierarchical design pattern: -Since the groundbreaking work by `@Fisher:1935`, randomized experiments have been essential in understanding the impact of interventions and shaping policy decisions. A widely used metric in this context is the Average Treatment Effect (ATE). However, exploring the distributional treatment effects often offers a more nuanced understanding than focusing solely on the average effects. -Python is widely used in the research community recently with its flexibility and ease-of-use in the user-interface. However, there is no popular Python library for computing Distributional Treatment Effect from data obtained from randomized experiments. While scipy provides a method for computing the empirical cumulative distribution function, it lacks convenient functions for calculating DTE or for estimating the variance of the distribution. -`dte_adj` was developed to fill the gap by offering the functionalities for 1) computing CDF from data, 2) calculating DTE and its confidence band based on CDF and 3) visualizing DTE. This library uses `numpy` as input and output of methods, which is widely used for matrix computation in Python. The main classes of this library also follows the interface of popular library `scikit-learn`, which makes it easy for the users with Machine Learning development experieneces. +**Simple Randomization Estimators:** +- `SimpleDistributionEstimator`: Basic empirical distribution function estimator for simple randomized experiments +- `AdjustedDistributionEstimator`: Machine learning-enhanced estimator with regression adjustment for variance reduction -# Functionalities +**Stratified Estimators (for Covariate-Adaptive Randomization):** +- `SimpleStratifiedDistributionEstimator`: Handles stratified block randomization designs +- `AdjustedStratifiedDistributionEstimator`: Combines stratification with ML-based variance reduction -The high level functionalities of `dte_adj` are as follows: -1. Computing CDF and its variance based on number arrays -2. Calculating distributional parameters and their confidence bands -3. Visualiving distributional parameters and the confidence bands +**Local Distribution Estimators:** +- `SimpleLocalDistributionEstimator`: Estimates local distributional treatment effects (LDTE) +- `AdjustedLocalDistributionEstimator`: LDTE estimation with ML adjustment for improved precision -It currently offers two classes to compute CDF and its variance. -- `SimpleDistributionEstimator`: this class offers a standard way to compute empirical CDF -- `AdjustedDistributionEstimator`: this class offers a way to compute CDF with smaller variance adjusted by pre-treatment covariates introduced in `@Undral:2024` +## Core Methods -Both classes implement following methods to calculate distributional parameters. -- `predict_dte`: method for computing Distributional Treatment Effect $DTE_{w, w'}(y) := F_{Y(w)}(y) - F_{Y(w')}(y)$, where $y$ is an outcome variable, $w$ is treatment type , and $F_{Y(w)}(y)$ is cumulative likelihood for treatment type $w$ and outcome $y$. -- `predict_pte`: method for computing Probability Treatment Effect (PTE) $PTE_{w, w'}(y, h) := \left( F_{Y(w)}(y+h) - F_{Y(w)}(y) \right) - \left( F_{Y(w')}(y+h) - F_{Y(w')}(y) \right)$, where $h > 0$ is an interval of each evaluation window. -- `predict_qte`: method for computing Quantile Treatment Effect (QTE) $QTE_{w, w'}(\tau) := F_{Y(w)}^{-1}(\tau) - F_{Y(w')}^{-1}(\tau)$, where $\tau$ is quantile. +All estimators implement a consistent API with three primary methods: -Lastly, `dte_adj.plot` module can be used for visualiting the distribution parameters. The examples of the visualization are available in the figures below. +- `predict_dte()`: Computes Distributional Treatment Effects $DTE_{w, w'}(y) := F_{Y(w)}(y) - F_{Y(w')}(y)$, where $F_{Y(w)}(y)$ represents the cumulative distribution function for treatment $w$ at outcome level $y$. + +- `predict_pte()`: Computes Probability Treatment Effects over specified intervals, measuring differences in probability mass between treatment groups. + +- `predict_qte()`: Computes Quantile Treatment Effects $QTE_{w, w'}(\tau) := F_{Y(w)}^{-1}(\tau) - F_{Y(w')}^{-1}(\tau)$, comparing quantiles across treatments. + +## Advanced Features + +**Multi-task Learning:** The package supports multi-task neural networks (`is_multi_task=True`) for computational efficiency when analyzing many outcome locations simultaneously [@hirata2025efficientscalableestimationdistributional]. + +**Cross-fitting:** Adjusted estimators use K-fold cross-fitting to prevent overfitting in machine learning models, ensuring robust treatment effect estimates. + +**Confidence Intervals:** Built-in bootstrap methods provide confidence intervals with multiple variance estimation approaches (`moment`, `simple`, `uniform`). + +**Visualization:** The `dte_adj.plot` module enables easy plotting of treatment effects and confidence bands. ![DTE](docs/source/_static/dte_moment.png) -![PTE](docs/source/_static/pte_simple.png) +![PTE](docs/source/_static/pte_empirical.png) ![QTE](docs/source/_static/qte.png) # Acknowledgements -# References \ No newline at end of file +We thank CyberAgent, Inc. for supporting this research and the open-source community for valuable feedback during development. + +# References From b312938cf43948ebe6bc6bdf2051fc4ac7156aa0 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Tue, 31 Mar 2026 20:07:02 +0900 Subject: [PATCH 03/14] Address PR review comments and update for latest codebase MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - Add "(RCTs, also known as A/B tests)" clarification per reviewer feedback - Add DoWhy and EconML comparison in Statement of Need section - Add imperfect compliance paper reference for LDTE estimators - Update CAR paper to ICML'25 proceedings format with updated author list - Update multi-task learning paper author list (add Shunsuke Uto) - Fix Fisher citation key to match bibliography 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 --- paper.bib | 38 ++++++++++++++++++++++++++++++++------ paper.md | 10 +++++----- 2 files changed, 37 insertions(+), 11 deletions(-) diff --git a/paper.bib b/paper.bib index 851c0b5..e6c0c5c 100644 --- a/paper.bib +++ b/paper.bib @@ -48,22 +48,48 @@ @article{scikit-learn year={2011} } -@misc{byambadalai2025efficientestimationdistributionaltreatment, +@inproceedings{byambadalai2025efficientestimationdistributionaltreatment, title={On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization}, - author={Undral Byambadalai and Tatsushi Oka and Shota Yasui}, + author={Undral Byambadalai and Tomu Hirata and Tatsushi Oka and Shota Yasui}, + booktitle={Proceedings of the 42nd International Conference on Machine Learning}, year={2025}, - eprint={2506.05945}, - archivePrefix={arXiv}, - primaryClass={econ.EM}, + series={ICML'25}, url={https://arxiv.org/abs/2506.05945} } @misc{hirata2025efficientscalableestimationdistributional, title={Efficient and Scalable Estimation of Distributional Treatment Effects with Multi-Task Neural Networks}, - author={Tomu Hirata and Undral Byambadalai and Tatsushi Oka and Shota Yasui}, + author={Tomu Hirata and Undral Byambadalai and Tatsushi Oka and Shota Yasui and Shunsuke Uto}, year={2025}, eprint={2507.07738}, archivePrefix={arXiv}, primaryClass={econ.EM}, url={https://arxiv.org/abs/2507.07738} } + +@misc{byambadalai2025imperfectcompliance, + title={Beyond the Average: Distributional Causal Inference under Imperfect Compliance}, + author={Undral Byambadalai and Tomu Hirata and Tatsushi Oka and Shota Yasui}, + year={2025}, + eprint={2509.15594}, + archivePrefix={arXiv}, + primaryClass={econ.EM}, + url={https://arxiv.org/abs/2509.15594} +} + +@article{dowhy, + title={DoWhy: An End-to-End Library for Causal Inference}, + author={Sharma, Amit and Kiciman, Emre}, + journal={arXiv preprint arXiv:2011.04216}, + year={2020}, + url={https://arxiv.org/abs/2011.04216} +} + +@inproceedings{econml, + title={Causal Inference and Machine Learning in Practice with {EconML} and {CausalML}: Industrial Use Cases at {Microsoft}, {TripAdvisor}, {Uber}}, + author={Battocchi, Keith and Dillon, Eleanor and Hei, Maggie and Lewis, Greg and Ling, Miruna and Rao, Jing and Shyr, Dennis and Syrgkanis, Vasilis}, + booktitle={Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery \& Data Mining}, + pages={4072--4073}, + year={2021}, + organization={ACM} +} diff --git a/paper.md b/paper.md index d2885e0..8d35b21 100644 --- a/paper.md +++ b/paper.md @@ -38,9 +38,9 @@ bibliography: paper.bib # Statement of Need -Randomized experiments have been fundamental to scientific inquiry since the pioneering work of @Fisher:1935, providing the gold standard for causal inference. While most experimental analyses focus on average treatment effects (ATEs), many research questions require understanding how treatments affect the entire distribution of outcomes, not just the mean. Distributional treatment effects (DTEs) capture these richer patterns, revealing heterogeneous impacts across different outcome levels that averages can mask. +Randomized experiments (RCTs, also known as A/B tests) have been fundamental to scientific inquiry since the pioneering work of @fisher1935design, providing the gold standard for causal inference. While most experimental analyses focus on average treatment effects (ATEs), many research questions require understanding how treatments affect the entire distribution of outcomes, not just the mean. Distributional treatment effects (DTEs) capture these richer patterns, revealing heterogeneous impacts across different outcome levels that averages can mask. -Despite the growing importance of distributional analysis in fields ranging from economics to medicine, the Python ecosystem lacks comprehensive tools for DTE estimation. While SciPy provides basic empirical cumulative distribution functions, it offers no specialized functionality for treatment effect estimation, variance reduction, or confidence interval construction in experimental settings. Existing R packages like `RDDtools` focus on regression discontinuity rather than randomized experiments, and lack modern machine learning integration. +Despite the growing importance of distributional analysis in fields ranging from economics to medicine, the Python ecosystem lacks comprehensive tools for DTE estimation. While SciPy provides basic empirical cumulative distribution functions, it offers no specialized functionality for treatment effect estimation, variance reduction, or confidence interval construction in experimental settings. Existing Python packages for causal inference, such as DoWhy [@dowhy] and EconML [@econml], focus primarily on average treatment effects and conditional average treatment effects (CATE), with EconML incorporating machine learning for heterogeneous treatment effect estimation. However, these packages do not address distributional treatment effects, which capture how treatments affect the entire outcome distribution rather than just the mean. Existing R packages like `RDDtools` focus on regression discontinuity rather than randomized experiments, and lack modern machine learning integration. `dte_adj` addresses this gap by providing a comprehensive Python framework for distributional treatment effect analysis. The package implements state-of-the-art variance reduction techniques using machine learning models for regression adjustment [@byambadalai2024estimatingdistributionaltreatmenteffects], enabling more precise DTE estimates with smaller sample sizes. It supports multiple experimental designs including covariate-adaptive randomization [@byambadalai2025efficientestimationdistributionaltreatment] and local treatment effects, with a scikit-learn [@scikit-learn] compatible API that integrates seamlessly into existing machine learning workflows. This makes advanced distributional analysis accessible to the broader Python research community, supporting more nuanced causal inference in experimental studies. @@ -60,9 +60,9 @@ The package implements multiple estimator classes following a hierarchical desig - `SimpleStratifiedDistributionEstimator`: Handles stratified block randomization designs - `AdjustedStratifiedDistributionEstimator`: Combines stratification with ML-based variance reduction -**Local Distribution Estimators:** -- `SimpleLocalDistributionEstimator`: Estimates local distributional treatment effects (LDTE) -- `AdjustedLocalDistributionEstimator`: LDTE estimation with ML adjustment for improved precision +**Local Distribution Estimators (for Imperfect Compliance):** +- `SimpleLocalDistributionEstimator`: Estimates local distributional treatment effects (LDTE) for settings with imperfect compliance +- `AdjustedLocalDistributionEstimator`: LDTE estimation with ML adjustment for improved precision [@byambadalai2025imperfectcompliance] ## Core Methods From 29eb364f9f9dab1a85a107eded6880673927b403 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Tue, 31 Mar 2026 20:42:36 +0900 Subject: [PATCH 04/14] Restructure paper to comply with JOSS submission guidelines MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - Add required "State of the Field" section comparing with SciPy, DoWhy, EconML, and causal-curve packages - Add required "Software Design" section explaining class architecture - Add required "Research Impact Statement" with ICML publications and industry usage - Add required "AI Usage Disclosure" section - Remove detailed "Features" section (API docs belong in documentation) - Add SciPy and causal-curve references to bibliography - Ensure word count is within 750-1750 range (776 words) - Add VanderPlas to typos allowlist (author name in SciPy citation) References: - JOSS submission guidelines: https://joss.readthedocs.io/en/latest/paper.html - JOSS review checklist: https://github.com/openjournals/joss/blob/main/docs/review_checklist.md 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 --- _typos.toml | 4 ++++ paper.bib | 33 +++++++++++++++++++++++++++++++ paper.md | 57 +++++++++++++++++++---------------------------------- 3 files changed, 57 insertions(+), 37 deletions(-) diff --git a/_typos.toml b/_typos.toml index 66c04e3..b47387e 100644 --- a/_typos.toml +++ b/_typos.toml @@ -2,3 +2,7 @@ # Allow dataset-specific column names from Hillstrom dataset womens = "womens" mens = "mens" + +[default.extend-identifiers] +# Author name in SciPy citation +VanderPlas = "VanderPlas" diff --git a/paper.bib b/paper.bib index e6c0c5c..becfdf9 100644 --- a/paper.bib +++ b/paper.bib @@ -93,3 +93,36 @@ @inproceedings{econml year={2021}, organization={ACM} } + +@ARTICLE{2020SciPy-NMeth, + author = {Virtanen, Pauli and Gommers, Ralf and Oliphant, Travis E. and + Haberland, Matt and Reddy, Tyler and Cournapeau, David and + Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and + Bright, Jonathan and van der Walt, St{\'e}fan J. and + Brett, Matthew and Wilson, Joshua and Millman, K. Jarrod and + Mayorov, Nikolay and Nelson, Andrew R. J. and Jones, Eric and + Kern, Robert and Larson, Eric and Carey, C J and + Polat, {\.I}lhan and Feng, Yu and Moore, Eric W. and + VanderPlas, Jake and Laxalde, Denis and Perktold, Josef and + Cimrman, Robert and Henriksen, Ian and Quintero, E. A. and + Harris, Charles R. and Archibald, Anne M. and + Ribeiro, Ant{\^o}nio H. and Pedregosa, Fabian and + van Mulbregt, Paul and {SciPy 1.0 Contributors}}, + title = {{SciPy} 1.0: Fundamental Algorithms for Scientific Computing in Python}, + journal = {Nature Methods}, + year = {2020}, + volume = {17}, + pages = {261--272}, + doi = {10.1038/s41592-019-0686-2} +} + +@article{kobrosly2020causalcurve, + title={causal-curve: A Python Causal Inference Package to Estimate Causal Dose-Response Curves}, + author={Kobrosly, Roni W.}, + journal={Journal of Open Source Software}, + volume={5}, + number={52}, + pages={2523}, + year={2020}, + doi={10.21105/joss.02523} +} diff --git a/paper.md b/paper.md index 8d35b21..84af10d 100644 --- a/paper.md +++ b/paper.md @@ -34,59 +34,42 @@ bibliography: paper.bib # Summary -`dte_adj` is a Python package designed for estimating distributional treatment effects (DTEs) in randomized experiments. Unlike traditional approaches that focus on average treatment effects, `dte_adj` enables researchers to analyze the full distributional impact of interventions across different outcome levels. The package implements machine learning-enhanced regression adjustment methods to achieve variance reduction, making distributional effect estimation more precise and computationally efficient. It supports multiple experimental designs including simple randomization, covariate-adaptive randomization (CAR), and local distributional treatment effect (LDTE) estimation. The package provides a scikit-learn compatible API and comprehensive functionality for computing distribution functions, probability treatment effects, and quantile treatment effects with confidence intervals. +`dte_adj` is a Python package for estimating distributional treatment effects (DTEs) in randomized experiments (RCTs, also known as A/B tests). Unlike traditional approaches that focus on average treatment effects, `dte_adj` enables researchers to analyze the full distributional impact of interventions across different outcome levels. The package implements machine learning-enhanced regression adjustment methods for variance reduction, supports multiple experimental designs including simple randomization, covariate-adaptive randomization, and settings with imperfect compliance, and provides a scikit-learn compatible API with comprehensive functionality for computing distribution functions, probability treatment effects, and quantile treatment effects with confidence intervals. # Statement of Need -Randomized experiments (RCTs, also known as A/B tests) have been fundamental to scientific inquiry since the pioneering work of @fisher1935design, providing the gold standard for causal inference. While most experimental analyses focus on average treatment effects (ATEs), many research questions require understanding how treatments affect the entire distribution of outcomes, not just the mean. Distributional treatment effects (DTEs) capture these richer patterns, revealing heterogeneous impacts across different outcome levels that averages can mask. +Randomized experiments have been fundamental to scientific inquiry since @fisher1935design, providing the gold standard for causal inference. While most experimental analyses focus on average treatment effects (ATEs), many research questions require understanding how treatments affect the entire distribution of outcomes. Distributional treatment effects (DTEs) capture these richer patterns, revealing heterogeneous impacts across different outcome levels that averages can mask. For example, a policy intervention might have no effect on average income while substantially reducing poverty rates at lower quantiles, or a medical treatment might benefit patients at the tails of the distribution differently than those near the median. -Despite the growing importance of distributional analysis in fields ranging from economics to medicine, the Python ecosystem lacks comprehensive tools for DTE estimation. While SciPy provides basic empirical cumulative distribution functions, it offers no specialized functionality for treatment effect estimation, variance reduction, or confidence interval construction in experimental settings. Existing Python packages for causal inference, such as DoWhy [@dowhy] and EconML [@econml], focus primarily on average treatment effects and conditional average treatment effects (CATE), with EconML incorporating machine learning for heterogeneous treatment effect estimation. However, these packages do not address distributional treatment effects, which capture how treatments affect the entire outcome distribution rather than just the mean. Existing R packages like `RDDtools` focus on regression discontinuity rather than randomized experiments, and lack modern machine learning integration. +Despite the growing importance of distributional analysis in economics, medicine, and technology, the Python ecosystem lacks comprehensive tools for DTE estimation with modern variance reduction techniques. Researchers often resort to basic empirical CDFs or manual implementations that lack statistical rigor. `dte_adj` fills this gap by providing a unified framework for distributional treatment effect analysis that integrates state-of-the-art machine learning methods for improved precision, rigorous confidence interval construction, and support for complex experimental designs. -`dte_adj` addresses this gap by providing a comprehensive Python framework for distributional treatment effect analysis. The package implements state-of-the-art variance reduction techniques using machine learning models for regression adjustment [@byambadalai2024estimatingdistributionaltreatmenteffects], enabling more precise DTE estimates with smaller sample sizes. It supports multiple experimental designs including covariate-adaptive randomization [@byambadalai2025efficientestimationdistributionaltreatment] and local treatment effects, with a scikit-learn [@scikit-learn] compatible API that integrates seamlessly into existing machine learning workflows. This makes advanced distributional analysis accessible to the broader Python research community, supporting more nuanced causal inference in experimental studies. +# State of the Field -# Features +Several Python packages address causal inference, but none focus on distributional treatment effects with machine learning-based variance reduction: -`dte_adj` provides a comprehensive suite of tools for distributional treatment effect analysis: +- **SciPy** [@2020SciPy-NMeth]: Provides basic empirical cumulative distribution functions but offers no functionality for treatment effect estimation or confidence interval construction in experimental settings. +- **DoWhy** [@dowhy]: Focuses on causal graph-based inference and average treatment effects, without distributional analysis capabilities. +- **EconML** [@econml]: Incorporates machine learning for heterogeneous treatment effect estimation (CATE) but does not address distributional effects. +- **causal-curve** [@kobrosly2020causalcurve]: Estimates dose-response curves but targets continuous treatments rather than distributional outcomes. -## Estimator Classes +In the R ecosystem, packages like `qte` provide quantile treatment effect estimation but lack machine learning integration for variance reduction. `dte_adj` uniquely combines: (1) distributional treatment effect estimation across the full outcome distribution, (2) machine learning-enhanced regression adjustment for precision gains, and (3) support for multiple experimental designs including covariate-adaptive randomization and imperfect compliance settings. -The package implements multiple estimator classes following a hierarchical design pattern: +# Software Design -**Simple Randomization Estimators:** -- `SimpleDistributionEstimator`: Basic empirical distribution function estimator for simple randomized experiments -- `AdjustedDistributionEstimator`: Machine learning-enhanced estimator with regression adjustment for variance reduction +`dte_adj` follows a class-based architecture with a template method pattern, where a base class defines the algorithm structure and subclasses implement design-specific computations: -**Stratified Estimators (for Covariate-Adaptive Randomization):** -- `SimpleStratifiedDistributionEstimator`: Handles stratified block randomization designs -- `AdjustedStratifiedDistributionEstimator`: Combines stratification with ML-based variance reduction +- **`SimpleDistributionEstimator`** and **`AdjustedDistributionEstimator`**: For simple randomized experiments, implementing methods from @byambadalai2024estimatingdistributionaltreatmenteffects. +- **`SimpleStratifiedDistributionEstimator`** and **`AdjustedStratifiedDistributionEstimator`**: For covariate-adaptive randomization designs, implementing methods from @byambadalai2025efficientestimationdistributionaltreatment. +- **`SimpleLocalDistributionEstimator`** and **`AdjustedLocalDistributionEstimator`**: For settings with imperfect compliance, implementing methods from @byambadalai2025imperfectcompliance. -**Local Distribution Estimators (for Imperfect Compliance):** -- `SimpleLocalDistributionEstimator`: Estimates local distributional treatment effects (LDTE) for settings with imperfect compliance -- `AdjustedLocalDistributionEstimator`: LDTE estimation with ML adjustment for improved precision [@byambadalai2025imperfectcompliance] +All estimators implement a consistent API with three primary methods: `predict_dte()` for distributional treatment effects, `predict_pte()` for probability treatment effects over intervals, and `predict_qte()` for quantile treatment effects. The adjusted estimators use K-fold cross-fitting to prevent overfitting and support both single-task and multi-task learning modes [@hirata2025efficientscalableestimationdistributional] for computational efficiency. Bootstrap methods provide confidence intervals with multiple variance estimation approaches. -## Core Methods +# Research Impact Statement -All estimators implement a consistent API with three primary methods: +The methods implemented in `dte_adj` have been published at top machine learning venues: ICML 2024 [@byambadalai2024estimatingdistributionaltreatmenteffects] and ICML 2025 [@byambadalai2025efficientestimationdistributionaltreatment]. The package has been used internally at CyberAgent, Inc. for analyzing A/B tests where distributional impacts are critical, such as evaluating interventions on user engagement metrics where tail behavior matters more than averages. The documentation includes tutorials demonstrating applications to the Hillstrom email marketing dataset and the Oregon Health Insurance Experiment, facilitating adoption by researchers in economics, marketing, and healthcare. -- `predict_dte()`: Computes Distributional Treatment Effects $DTE_{w, w'}(y) := F_{Y(w)}(y) - F_{Y(w')}(y)$, where $F_{Y(w)}(y)$ represents the cumulative distribution function for treatment $w$ at outcome level $y$. +# AI Usage Disclosure -- `predict_pte()`: Computes Probability Treatment Effects over specified intervals, measuring differences in probability mass between treatment groups. - -- `predict_qte()`: Computes Quantile Treatment Effects $QTE_{w, w'}(\tau) := F_{Y(w)}^{-1}(\tau) - F_{Y(w')}^{-1}(\tau)$, comparing quantiles across treatments. - -## Advanced Features - -**Multi-task Learning:** The package supports multi-task neural networks (`is_multi_task=True`) for computational efficiency when analyzing many outcome locations simultaneously [@hirata2025efficientscalableestimationdistributional]. - -**Cross-fitting:** Adjusted estimators use K-fold cross-fitting to prevent overfitting in machine learning models, ensuring robust treatment effect estimates. - -**Confidence Intervals:** Built-in bootstrap methods provide confidence intervals with multiple variance estimation approaches (`moment`, `simple`, `uniform`). - -**Visualization:** The `dte_adj.plot` module enables easy plotting of treatment effects and confidence bands. - -![DTE](docs/source/_static/dte_moment.png) -![PTE](docs/source/_static/pte_empirical.png) -![QTE](docs/source/_static/qte.png) +Generative AI tools (Claude) were used to assist with documentation writing and code review during development. All AI-generated content was reviewed and validated by the human authors. # Acknowledgements From 38ea69b5ed767795040b0d588e460224f5d3d465 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Tue, 31 Mar 2026 20:48:21 +0900 Subject: [PATCH 05/14] Address Copilot review comments MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - Fix author metadata: make Tomu Hirata the single corresponding author and remove conflicting equal-contrib flag - Fix NumPy citation: replace Unicode en-dash (–) with BibTeX format (--) for compatibility across citation pipelines 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 --- paper.bib | 2 +- paper.md | 5 +---- 2 files changed, 2 insertions(+), 5 deletions(-) diff --git a/paper.bib b/paper.bib index becfdf9..f6897d4 100644 --- a/paper.bib +++ b/paper.bib @@ -32,7 +32,7 @@ @ARTICLE{2020NumPy-Array journal = {Nature}, year = {2020}, volume = {585}, - pages = {357–362}, + pages = {357--362}, doi = {10.1038/s41586-020-2649-2} } diff --git a/paper.md b/paper.md index 84af10d..6940ee4 100644 --- a/paper.md +++ b/paper.md @@ -10,16 +10,13 @@ tags: authors: - name: Tomu Hirata orcid: 0009-0006-3140-291X - equal-contrib: true + corresponding: true affiliation: "1, 3" - name: Undral Byambadalai - corresponding: true affiliation: 1 - name: Tatsushi Oka - corresponding: true affiliation: "1, 2" - name: Shota Yasui - corresponding: true affiliation: 1 affiliations: - name: CyberAgent, Inc., Japan From e53f6e56bb9674c65e78491606011dd50eeb31ca Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 21:48:30 +0900 Subject: [PATCH 06/14] Move paper to paper/ subdirectory and add tutorial figures - Move paper.md and paper.bib into paper/ to avoid top-level clutter - Add hillstorm_dte.png and oregon_ldte_costs_comparison.png from tutorials - Add figure captions showing simple vs ML-adjusted comparison for both the Hillstrom email campaign and Oregon Health Insurance Experiment - Cross-reference figures in Research Impact Statement - Add dataset citations (Hillstrom 2008, Finkelstein et al. 2012) - Add GitHub Action (draft-pdf.yml) to auto-build JOSS PDF on push using paper-path: paper/paper.md Co-Authored-By: Claude Sonnet 4.6 --- .github/workflows/draft-pdf.yml | 19 +++++++++++++++++++ paper/hillstorm_dte.png | Bin 0 -> 184790 bytes paper/oregon_ldte_costs_comparison.png | Bin 0 -> 69226 bytes paper.bib => paper/paper.bib | 18 ++++++++++++++++++ paper.md => paper/paper.md | 6 +++++- 5 files changed, 42 insertions(+), 1 deletion(-) create mode 100644 .github/workflows/draft-pdf.yml create mode 100644 paper/hillstorm_dte.png create mode 100644 paper/oregon_ldte_costs_comparison.png rename paper.bib => paper/paper.bib (88%) rename paper.md => paper/paper.md (86%) diff --git a/.github/workflows/draft-pdf.yml b/.github/workflows/draft-pdf.yml new file mode 100644 index 0000000..998b3ac --- /dev/null +++ b/.github/workflows/draft-pdf.yml @@ -0,0 +1,19 @@ +on: [push] + +jobs: + paper: + runs-on: ubuntu-latest + name: Paper Draft + steps: + - name: Checkout + uses: actions/checkout@v4 + - name: Build draft PDF + uses: openjournals/openjournals-draft-action@master + with: + journal: joss + paper-path: paper/paper.md + - name: Upload + uses: actions/upload-artifact@v4 + with: + name: paper + path: paper/paper.pdf diff --git a/paper/hillstorm_dte.png b/paper/hillstorm_dte.png new file mode 100644 index 0000000000000000000000000000000000000000..3a5712097faa16e0092617ae7b61c421c75f34cf GIT binary patch literal 184790 zcmeFYbyytHvM-7g9D)o4_Yh=|;4l+{>p)0wmk`|DA-F@Zpb1HE*WfO}-EDB6K?lEN z?|shM@4R#G{pGWQacA?+k%=GDWXOsF8N{eF|R#rqSpiiNGklOABQc72^aiB$VAZm`J&Ufc~GE zySonI9S*lwr#CNZLbs}d-yP)KKh}Q2l6FEyl6&2~_x3j(0hAWGT^IeGHV8?IZFAUS zOEKAlx?6;Im#&*%A+xZvR3Wph6M$(8HT_KpMhPex*%{ zIAf#{LK1g+;h_jEdN3WXiDWn*)9-|JNdajJDBVxIAzB^!k)0mN5=_S!IT~_7GiN;k zMmLWYYYJX2i*WxxsdzWJQ;#M6@@oa>JVzWv_Hwk&97Bo4oNjhin)X$9H}})~uz}e@QpxOx z2?5|;a5VY>vAhsgmk7i4CTv~n^Y?%^9}b1IndHOO;SJfZSHT+mWc5kgG6eG*58s!J znC&qfo15wEJZvODdYP|$o29Wx~6ZTjR$k30kFpj(puO^w?oQ^fKTsU#%pOIL@J{7`2~# zU%hBEppS4TkUJjdgfv(ok%-`V2V;Nu<2n(o268k#K*Rj8Z6|CVh>OX)E*Wj&Aw%Xn zmq;#$OBYHQ9A4Led>rh(z7E^REan^O6%XA&m_R&~cA3QJ^q5_U>{I}N0U1vUZdu>Rwf<%l&)eDYOZKI=>5|bkSvA3|rLqF-Y0zUIJR;*sLWmRb{v-8$Sn#13$JRP$Xi=)pZ^e^z~uQ@<&Q)EtM0i+gt-f+LR6 z?&pq>7rG0nHts5w#tWzpL&w$*jOaq{VO<&etj96aR*ZDndFj;WZO+0v)y)O%)oO1{TE`l=lDb_ zUG9Y$u&;7Vyb&`}?g%ax;_3+Jh{1@A2=9o#h^+3gZolUdKOj}Co=M;3eBOZyKi4GG zWVl4TL`*4rDQYlg#AgjiY?4%c)|295olS^K1SaIfgH>O%rI0`rNTlV`CAH<}B`m+` z<+`YYRBH=n-rHM=@nSDf|#!>5nsH?53 zsB4*5QR7?RkQ=UxjeWss!|_kA*}u|qhNM7rC5nG${aDUxnKt?)WCeTu`2`Mt(d!rd zT$atb>bof}_`WY%;tf>5a&2!;rhk0Wv2^0U{)+Ru_%(sG!R+Cz$tUHp$caCbI%Oa7 zT~!VW1>_eML{)VQ^h;XwZ+~1$V)SxbI8dPUKJUfvwUnvHSL@RW%TFp*|5YuUFPRU~ z)SZP;%K{?62J&`8pWqNUUF6H&p z%lo2izh%~94|9KKw|vHaqRl#ha*m>#e290>#9<-Vy<|kIyfUcrrzt^ni+T8q1&ZDm zBm7Htb9Qs~W;P@?O#BO*1HX=gB}k)|KsioMWQn^hTLzfv5u&Hq}1$u_!Nf7j+a>29;rSJ$VkV z-vxqi11$PmWJ~18p;27I`|a-M?^v9~4=#1iWI~$4hUsz{)$se4ZNQ!awjYsTb&>Zrx@3p&A;r; zQws%6<1^!YdH#{~y;6Ped&FY12t|RHN>q$?UV8dV2Of`zWOzNX&2thg+lce-sqTet zL3w0(>%>NuxEwM6(`&O}Ge;+9Y&K>da|?yXFYRmx`7Zg5zfyh}WLSy2UQRhr;{MuF z5#`MSWr6RFUeI&0m2w?1f+G!wPar!#?G;Xl9ndM!Mnn8MT{~$K>OaNo%7?o5evQsg zUrD#-^RDu#{!{JMqv!Y&VeTe(Iddo|tR~##gyEFrbhVvuHgSrtSi>xwsLmH{F_b;d zR-2k4*YweAL=RjIA!Au&ligI`8om~R$+qbm=OJVq+uHMUi`I@xy0tMatcO#E>~c&6 zE#@;>LnK1LLp?j#+jYZyS@j+(_t%-%i)3^-6M!1d*M?fn72ayF?_YbSdW56=Ih|hV zsGrX!{$S?pk?Dzwr1c}dQ}&IJi745;c~|~Zx2>m*)PJ$k23i#}2uj6P^k;K^#SXk} z(5tvhJz??DgfF*8Mw3#rFKgIoOY1RfyV};C?0@CgvBsVi)1$1cbELM`-)n4}ff)Li zE&cR4z}Uy=2`!6+MQ+ECQEJ-WX{CSgto_qwb;8o6JSfmI&#~lV8MT@cS}LPESl9Y! zaCaYL+p+kfxU^`np12-hPpkH>Alyz6)*@kT-O)&=RdXDsr_s%Se9>a(n#M7KmT|PAw?D2RprISN8!h<2rTXAb#i%>RDF((59}0h%CKBqO zJY@_d|6`PZ6kH_KY|29Q_d8u3K?NfLVco;pMFTlFObmxK_ek4AXEM){SNu%zhqF5 zkismH(EcT(^K|`t#6O)+=zm>N6GM?OpYEPNou6}2{zDogIv4dnXq1$vG9(FgNqPCF ztGcO^xw)P5CwrG6LRRZ129|@IwlfkE8S~#0S^gap>}hSZWcgmpMN3&p#MIuF%h=4` z#GK2+*5R)nBry+>C(_p3#hA{+*2d0R#6z6n9}*%@^j|iRf$kq7F4p1)$^7U}xz9dV>8=?Ej9^`Y#xW=k;qK&%a>*&j0Ti?f(tt-}(O? z197r^>PF+g)dT&9jeqC;OI{54cl`g&AO7iV|KL7(8wgtr_@AB&!Y*~D`i_M321#B@ z;=Kp*;WB1}xy1)Tm^fb{+UJ00*nfJ0TJq+oE5Dm)G7;hB20>Iekw0U~WshRM|6+n9 z{kedcQ2NawI%y1EQ!fTXq@)_ML_AvJ(30rc(W=S}>|-_LxBGF{;sd*|^T_;`@9a(a zO^WYvV3N4&H#$PJpf`liP$cM{p#}c$Z+xTLJABza(Fp!;g8x)OPS_u$hD-l{RQVsq zZm9w}$d8;+!x8_NI{Xvxq}b;FT-X0!@&9W;|G(^G$;a}1-&l!m0+f^B_NJAKvg6gq z`_-oU<}J=;OrLhlOp(X9M?~E6mPS$>n@fmql0LvBw(qjnko<^Gbdy`K8w7 zEpB7O{tr9;ZtksTS&;lfeaDl5B-W;b^0wC7;zuDsi>4)Cc( zYHXW-Y&?0R-P<)hdylTdOj8vL;CKWvYvW(b%lxx?lUdFq!jduUv#m$A_6I|z>BPVMz**#TP)7|;%cv)5kejk^V zqA}QF2C|_k8KpPB!yWnlcNCVXj~&8nGgNP8{qJXw|FD-I+BL&_k1hPREmSJ6H(9}+ zaxVyc)=--u*IyvQu0AuE!Nf^s_WrkvqQpvlvD61teDfw_Z)GG*SkaHPt86wAQ3mJV zkCV%N%&*e=#D3i_jovTCysj*~N((S|CzIXnxkAOn$L$G3b6pPs94~g%zuk*uFWYUs z8*FQC+Wj6C8o2N)$RFdyhUw#l;Uj-{48dJHfe)>I>o$Km#rg5l)1cn2GPKGq-L0Sv zh4~}GR-sE2M!thhMptC*oER%xT<-Iy4+7Pwj)+ow--GtHIW{nCnkSATn+0FFvoNQ z)fbk;pOOC?ZiV5rInZ0* zX~sxuB*{3vdUp~|!P3>;pCx8osoF0LxSO%R*SI=OZEL<=D8FAghDHIioww3OKbR81 zb;}aUt!?CJbx8+DGe1)v$Y(tyfv=N-SH({g;+B#IOqHaE9c@;Xjw4oE%ZdU5=&~)pbx$JBq z-$N|mvb!vQ`-MOheces-j=23CZg)B8ToUY3{dMKstxmwpR;0<&0Z2!O)Ut*S+Vd)4KImu?WV`6@2TU~2+u0+F9b`$0xvU=YJ^vMdGYP(*GzLWc_$k+(CodKF z6?E)5>Wz=S!fSi{)B4XgbAMKDd5n?u+nf1Oze`oLbP_~b+;y7zFsv^hbWzy#JszE> z=Q*vE&jivMJkI~l2s72t@0xqhZ+}}H)`4^m@LLBo#@!Ueyup()E>aU!I`~JAQe$KM_NT5>(y8%4%*xhf(z47F> z^`t3eI7`JJN}6^5ch2GtDAXUv2w`E^`~tqM1|K|4wGa7x!t^6A%+l6A53~9~S@DDU zTJ;4trbjAmEE{x$gY8munDQ}d%TZ`hZy!9^f)2$oU1EAFr zBUkSB0R0@V7TfL@yMDY8^;Glmolo_hS7lENSm1)1ZU~vu4|sI{=w&F15O;(^9yTGN zB6C_&(ZrLRxQ=i>&<0AZpt6MUUnbZyqkL&dPsf~)rN>N-U;3~Z!?tR zp%cz{R@1Z{4B8ELI|@qgsOt#t9N)C$BpC?F#yZ96AOYLVI8 zLFHbseScGl7y%uTkm)W^b%ZH@o&45WN2XP$qY=WN!6#R#TC4q|LLkA?{d>q^qU^)n z^6LHaaq81d5mt7{^hLK@NVHFUg=^dIu+Q&M)kD;R4x_6@;k3r9vwbS51mlnC>1%ak zA&}sOKJM7|S_Jb6I=|x;w~dyUqx^`56N~upI5YI=Npt zacw(gZ{t-8jnIdF6SM3X_)!Os6mck@Y0bIv%n=k5R=GU-Q3MdnvKpdqi^?RQF?UUlyzroVPsbrCjnsWI2J_fR3sTF;! zews5NM;wA0YLEpaQ=UR6`{gFUN{_zj>yv!GzA9=h4~1aA-RUIEHaKM)&#t`fdq0IX zE42;f)55{)dkytrb8^;o<7yks6PD{3`l$|n!>xuF6l7!s{jW3!)I&ms^#oc6z#hC3 zE2rl8O7~XHP9i^)eLf8er5X+#>dmrko zLFo;6(#e578b3it%B%WX{#1{@FEbax;OV6y5_=118tWUlj9&3asm$0K$ zA5*lOw4RfCZbPqytq_nXP0F_XLVsLGrNtn5x2wz<^>wf^gwJP!)ptVGXeuZnuxObs zS&La0iz1HY5w@dKwI(a-%K1Q=cSGbbU_h*i5`R!SVGS9Z&8|c9I87;82rLDWf)~bq zwJw)a-)pwP99G1+-Lsr;rj9)X!#e8R7Nzp=uAs#%JQwSJP9!-uIf>3gs|}--W#<;$ zB^7+g*1T%(Ox$g-kF$+bKdt3^OvFCmENajmcNr4-C~GAJ>qEzR=7d-ZZ%T}c7}7zt zlE-9*Juk&tbo^@6y-NMyDi#LZ%+~X`LyK6hwtKXD)h$?liO^*I%L+W(6ib8uL?;-**A#s8*{a(P+xt+t78~4)T%3 zlD~~{GUS|31aB}pWOakru79fGV!9hS3UgDNagE4k`x(_Or2 z2QTx}_3f-Wm$yL7+Eem{s<1v#3>Ns_TRNxMccA#25QjKP-tUZ|W+?F=d9u&%E zWMGZqHPUS?I*5`O^h2U~Jn0>ej$4v-%vDm!KbkdBsozi4g*rRkZ=74|_S#xnkF8ht z0$cMpdp4SH*DYo%hcUv_} zB!oqqb+Nu$bOe{Ynz=QoO2uo;299&KSz-YvVe{IxvWpOY%{^_p@Gtxq>sdsOEdu-@-S-n2iWtxb-2&B z%;S&ypv84{!`cUUzDa1So-R{tBp(Y=N=&e8xjVc|>Z|^;N$o=F>cU6 za?F?`^x2RncORwlXUohs#$3ULK7{M4p%uy|JdHEbq1j692v>4D zsN9;C)F`H;MAz4>H(cU4YYm)n)*a_z9wiN1lfluW0v2A^z4X&B&(39ZrW?^5af`Jf zqeL*@7JU7AS9$nlq=?M&+JSa?0&fM!LP$egM$qMCtJ-=vwML{!@vOeM_rh}<_j|s4DPpT&TXIvtwzfBBYLw>FQ>~Bnio1i`h zC(e71R{?m1;|HK9Ix=|$u)6I!#e;d?^y8eS9;*LMZw6xBKg5BX?b=NsK)>6%yEr-MR+`kyFgu~ABfsJ)-~x3tNqyV=T`hFu|ohQ*^AGCnE;T316}SLc;EPNgB8jrc59`JDi+8{)kF z!v^a=1eq2DW-cnrf$nb4Fg?xhMn5BwW5xHv?D@<(XRzC$4cKeLMo#x((r!+T1VJ zhQJbQEzca@71v{Tu|XYBN^#1@ueHlGb=gO8d0h^ zjGtzJ0HydBYw|kDpPBV>=&;A_Z|m$S{bt@&5pj?fx~1VXqr|gX6~O?ABMrd$T5;CG zp4r-WaS6Rux9F+7${^wAlVi9**Akx$gJ059_q}JwYK75Ldd67lt&F z0>aEj)u>u8TX&~NUWLdHU;&cy$mWcS4LfG+tjZmc`CMGiz|V{6aDfQ%O?m zL@2~RITPJYzN?#D2{vp1Lzi{a4EY)~ezH_$-Voz%<0Yn36<7<6xbGlob@`CCclLKJ z(3tEO1Yohz&`qRl`6^adpx#^!bh8hzM0yLZa46Tew#h9spCseCkkA-VU9`DVi}RXV z6Lov9HFuXedYy^P9zpD8^K7)@b09;9%EL{Jw(U!_-I|J`O#*h1hKR|IpU59abS7UY zo?(S~4w>t#^@)TtZlih(v}!rpg&I*vV&BtZ=q*O8jDQ=Ou5XoLRi)+CMpV{Ig*(0Q z2oiWt?aeRuIjdq1vmKiSsTOO*v*k~tBN`rEJ10*8ElAHxk<%o`3w07R2F9FrTEp&C z&{_Fox6GaeL%SmY8oKG3{_|9%* zW%uYov}4>&Dm{HtR;!(xZjFvvU&N34MMl`e09DjSXPAo#n0vVX+-rbCS`8DI`G+yp z$2hk&AXh??pYhh9^#+eV)ers@dl_@vGV$xN*ka_j*kS;&evBM2;d3hW$ zanY=P-L&PWrf_n5XqJn^9{7VSJ$WUu2bBD zm0dJbv#Uc|DUkKLM*Fl_Vh;y+mtkLxr_~f&%7B04nf4g3~67YtK;3G;EVFZ}Z9uGZqG8BJ`_WC)ckKZfG~aI~2Jwvp7s|wh6&& zmv50unW@A0nBl*xZn%%ldpbN({?Po2MW7x}436i2rr<*-djNq(ia-w6>w5{#cSM!e zAzY1W8Pj@-i+xiTh^~jHUOH~)yCpiQ>5&~EVC2swAAhcGEUn>6Lo@IPd_AoIegIPT zFxsrEfPBszr3@?d-_uTJeN|Vo-zxQ>`&iY&_8qfoEJpadz{AYNIku+T&}DSOI-4C_ zyGmqRq}c|24>RW+6AYK6sq-wUm;z>Zv-;5+bY!={06Wb(P)0K-T}qT|l4e9q9gO)hJ|dG}uzkCm$n*CF zP^d{m?Bj{Y7werEZQb9Cqb?)8FwtL;F%Iijhf2rLTN{PDL~so4CPOU>h4Gpu(Oj;O z=Tnr$du_Db0)}<=^@ccS4h$x2n#&tCNkcIDj|pal`<@op4eXydRv#r&7@BBLlR^u4 ze*gZ!Pb~~Te8Gd-cC)jw$rP18e2!!IfMb}m*XJnqM;10ye!G`6viqFN9Mi8yVW~aO z{h?6#EGOiE{N(5^n>D(Q_}+J=e(ldVS})JERW(%JNHC&|aBT@CBq*K_hvWILH8xdG z!F96=2INKMz6hc)er`f7G?KtRR}A%q_j2s*2|?21!}EewMxQN4%Hsjg^05`s^^_Y9 zi1AsLn91(2+EK(oa3wNiS-QZC0Uhu#8gbwlPnyt)mxQ8YX&M9TbuFhFn?iG!qA}++ zmf6C1&UX|Rlxn6uufQq(pd6`#M(B8xHG03!9AnZOpd)*x5=R0KEyl9uRg&%VOu8cj zEW-&GYCSX`pZLin2^8)Js$%4bc1Rm%?2WB_kj7zq_vl`{`782CwLPpHn<2Ba*i#e9d!>;aS;2c=Gz?E4wCOt*7JVM|Sv4p#9o z&UYT#utXyw-)h6T->E7jZ~ALf4ZQZa)28tcaeJ5Z#zMXwv@of7!7~ilo0w9r z7)!R9!Fkp2-0!U>Y0c_zQ^bi~@Z#dsFwZXPlRx{TnrN_fviRPM=?5 z<5s1Ve!?8F`6A9>3%~wa;}kvQqH=Y<9Rf^?PhA~h5Pif-*VI^}8|FfMgZG zE34qgu7w~?Q6jagX;JF%N7 zqUbH!)^(xg9_q@-$F)Ak)2P7n5&?QjiY@jegQFm#rf)b66CnpMNt!Bhp>fQXG1|uv z_TTp{Bvl*1cz|oGO#sVHWR{rz(7HOKh+UKu7tUkv-vsz&)0o%SK@mLn3^)ed z-2Gi`w_DJ!@y;>i9OX%vWDa=ffyGv5WfIDe%6bQz8Go zzN{;?Fsfg2=iFGVSm2#xG!~G*&H?|U z6;_fhH<|rB?t_xuwdku;e!V$vn)#Ro{H2(e+aS`?>-U znYw+N9i^0w4)}|_Jp%vpe%XF=ySKJl#2Yt`1lI$f&-;42K_E1D1VdHv7Q$B_ocZa2 z&HD!g)XCNNS#ZN3ieK~UOeA}zHeT&IXjq8#(iS}-p|py)Qpd3O$OP2%J&&RC4Hp!0 zN^~0nXbTJ$(3!wL9Kv~ufWd>&6?m(WuNXzXoRl@6{s|=5uxRd(5;L)fS)9cl5Io!& zaBMrnwt*%^OYJoJiaQHqWcp*8TS(Vvt7MRxBR&f84@xpe{cO2SbzA%rw% zwhXS+^E$0ku`(fx5#B*np*3Vb_crQhkLc+qUED6<{L)UAf7g<)PaFurl*M4XzWhB= z>m}C-VfI1UZ9Vhi?eZfjo*i*``m~r(+!F7 z*)ul`F)B5q1Q;nBN=^F^1-b*))!)smFf4jxnYU}et7@|Bn zXP;$u)Xh=rE3{(25eW>A=&Td3dj<$8L@{8n$bV>OU3{&oCZ#~0$evy!nw_TUprGe= zlT?0Fy8#I_q0c@1LbJEB_J=2!N(^*6KT1Q!zL#ecx`a*Knr|x-V9VT02Iwh<>g%q; z@K(7w3K~Zo>v}X2nD#`ZpQn?4IIOpsl?uA;9{>?9CV!qM5hKBfloxSr>9+me8s73! z4mJqdA>!b;dxZ^?r#-ifmyBa^B5gr9<7Fd?-Yy07x?h?jkL4;fSMDlepbAQ-p#yQM zZr&l$*^VWBUob2C9z**A?J1U?-1C0!rOjOY0QQ_nU+i4wCFu)NR8l33b=ey%SAm@e zow=EE(7YhkJT?f{*E%QKwDz=3HwOhVvB10^4*djJ$-?VFqYS*(spU}R9Bkc7*?Np$ z2$R{vX;0$4yN##J18+>~hM6~$7atr(_se6ul!42T%Eff=Pi@}_(qpxuYHE;?+9(EI zog^2rm)ZONu4Mnt!Y%oAd>lVT4W1y|fs;E(X7?mXI?RX?IwTFycd>FOyH-vd=c6Hg zhzieM=Vcac1NhHV-+p0L_L-nubOnlaGn`iGCQ)IqA9rPl{jo#P=44AdaOl-`fiEoL zwRa9*yhiY(8^h?sUckm_ALhe7=w-$j;3iSLW@O!8bLK0s2t zDLqw|yHiu?TRF3%c=i>qDP&BN$v2zC7H(R+WT~D*dhqzsw5RO&c~JNqrD@2zqF2|L zM@b4XyO}TIBEaO~o|;##kz(s!LJV+-P0H<*i%EjVkK1Uy#y#uhv;xv~*5^2u9 zqH8B{Hte%{ZIt&Ws?d4@)nNdqjtf07LnEUd@W?W;Tcpojya@lpT^oTs1y#qCA=fV!X5N5E<$G#p2X;kJoc z+tX5MVBf$HO|u2TD|Xak>BnDkFU9Ga1=K} zItc`PA~~CAw!>~0S~imb{clpVnO1)Tvd@oJ*e38-TAxyz{L>8M1!3&5A{JunXBzPc zI$+0m!%&p%;DFVEn_sv%x%-)*ht_WKHM8IL*~jO1F+CR+>Wz4V6(qL^jV&Y*(imGXy9{#)nT%^iOl(70^mGqxM%>e9OF=j(Ld+ghi*C{rqZZp%)^-D z^j>m>$LzllXqRnfgyZ9v5l8u~TD}$!TD8;*%@L0XFk_yVXW_(3L_rB-PehQMQE7HU zn^{zu=+J`(sKDCW4ebncY@e9513zRW)~D8y;02|>+4mwgLuov#er^J<8fs#Y9V663 z3pAI5(V3E}B8bP<&QQTHR7rY{foDeHxFd=aEJ0`@A?SKJmo@o#fs@^ioH8$8k+207 zN+di)qH%^myJ#*e0u6<**Ft^m+P>7!>;DP^OPyko-e_hwfh+6w;L8J zg&yUG#1sgfR75fEs!-2W_=KSu`QkGW>+_J=XN!c*@_=}c+H$&gVpWe{cc)rWk_~*h zGI58_ja4mRhd!jt3w;~wpvuJJqw1S7cXAl#<+}fB5-bZaEBb+6^ z?Z?m!77KZKn=9)l&myE9K>RqDU_x+dBd=Dt8bjDTFjz+7jZiI!@Qv(vVODF)ez^G} zgyGH#37OJSRj*_qTZelhdjO-!eSNfzB(#j!xm&b{%lFlvmiU7u_TE=wJwi!x2MK^s zr)S!GKhBGMeP0Fb&AgcocpvMWIoh5%nyTMuK{gI}=r15=Npky}Q-mk-JlN=`|LoW3 zbxMqxqt8(%EMJTUuXYU%gOdinYx}=WiR*77JHuTOlhC}$W555JnM;crLl(xC4*Ld? zzDh#?Y<4AvIvF$a>nA+D#9B1CiTzomV&o?80VjVBY?EtI6oAF<6A)|skwDw;i|W>YI? zc%!LVdp+zaLg)_*HFOky+QsKXVL8~Wbfj7fyZdp02V%eUYz$hxF5`l>o#M7>q6)<# zng+D^Lp@kyix#{4it@WzaA-%0i4nzA+}jq6 zfByiIN6M2jPLn(3e~mia&zq2j?`Kh23C)rN2MS$+EyNa?FEA72thGeV!kso_8#8}erMh7J8wW4?>mWG=H;~b>t4vL zygy598?HVtkb!+BY=Y@i+XoJzWH-|Ex*SWh*p7wWC=9bQ1ATqmXJ9rxhHSF5iQ6hO&8{a}j3TQw~G z15#Hae);Xe!d~_^Laxymb)R{)hoGT|w7+Pj1J%Ho)!xu)fI0YSt8ED#L~}dWhhJx> zgzVof)5DTceIB=Iv%&2}mNyr|Q~E+;*-R`}R%{$hH629udbf46{@LJm*)d=w+F5tC z2k84%_9f*zS{RUbKkT7r%4@;bsO+&XEyR{0IGw}*a_G#NKF4^CKyRt0M8Jh_n1OA9 zwZhmU+?Rzyn#~-y_{5G~n_@~+sl6bVm%RzjO8Ww>jAWA%)7A*5wgvn;6Z2`Sb9tJJ z%8h5Mz51KMrEXIJH>JX~vmHoalh6i!YpRn4wu$;mT)&COx<=KHCRuJTv{^8BQHJ{1iK8%7dfKZ=GDk*LACi*vU8Z z$o6o}@f!+ib^`Ky6>OdX@5Vfv>*`(!j4-rfFuJBb#J}j)%k(Mt`|j^3>)UfV)YK7L z-qqYgOJk$kOrIfaG}n|`)*ospV(I$LP`Cqyg7*DErR`CIzOrd-z;Nx^2Sh4PkJ)LQ zyxtV8K{vbD;p;y^<0_N=AgC3P^NyguK5_0GO!6%tRk92dtP4m zj0lZXoHT$07bVlkbd8A)Ak5;hFA24j&n-e$UoY?o%tOv$vU*FEtmJM0ChZ#+ydSF_E7ILOCKyQ2!P@%79+sx9dVPj(&IEfuBMzonoSNs@kmy-}f7k zNKAT7V&(AaR2v1@!k&va68Sihbg&n3ukfiNuZcna!zy#RRH=S5`FRHwa3Pt)xQtpL z*j-?=8Kv_orA8fx@dysLXr4?09g<9Ef1k!V*@}V*c67+j#~(y8>4>qvn~NU}=Q8TQ zDt;D-P8pK*c|H+K<8!Js6HUc^m3o@o@ttDFkGX)-@7OBrz8&F$ee2)-gYoNP{AbA$ z7P}Dy4{HSYJwJm=Fv>P9C_gZ@2R??&ZfXR)w{awx-^)6N_3prt2_DQk%AWykFu9;# z+@f|T-!$LNIPP4SGeqB+x4A-E*p*1fFxbk7)YoPEs*el32du~N1e<)U)*&5CAo7YA zdW^j5Sle#1<1Il%***=m_g4NhO%J#-h~#{&-7v^nx?fK{58ND%Jz{h5>7>5j{Zdgw z`gPl@Alk(OWXB$YwDd!(@>1{SrjeF1KJE|w87~>wY1VavQ!iUDt37`Df`nLXGdzsI z6|egNsOP3(jY3vNg=w}M!wuA{>?0%(=V%b#wD@-Gc~ajqNMne056e+Jj!JR{qR<-t zoe*#}0A_I;-rnN(t?Y`#EaO)V2@%@%C9+%K{=_ZJ&2k-t3#AA3P`V95jQ8~BeFwzs zOy^Iy&2oN%y^ItcLfIe7sRDtmM(eovWs){YK;hnB^QitG6+uneEu)OB_Aq?)BtIA% zG!8=>BN#*T=!Z}f;h+xntnbP|h<|f<^*6eT45Ii9<=*P+lx}3Y2>-%gSL^(MO&X$bs8&;a7FBC2~v+DXA<+sqB4oCTVgLg)9yGsmbAKBNW+XjHGkR& z(ApY6Vq*P-YT)s?{KbG{&Q_eZDG{Ow=fd=~!YLOt&B~HLW&E^mX^`KMh6*nib8Zv0 zY~Xd|Z)Uj1Iq$0pzilojSIMd1$Jz$)Z({%c5zbjGI!IPbb76NlZ4l+}l@zs?9s?)@ zZeicNMX)pp9BK%ru%!AK{o%5@dj%1eRoR$>HzXtJhg~ zMFdA(*pD^^)8dCpyV~zn`JR&KHg)&T?6#~=&I_qelq2w{NVhcL$HDrY3TuZr_>YQe7-e2}&L)pXq&K_(c@xa%o4pl4fwD)18m zmp^><+}${L#dd??)=0KDU)qIEkU|#{BF?)%pF`}%zUg6EgHu5-;fYp?xUYp=b}K6|~^7MM6Dsf&#>(KHK7 z-OoB7_iQz^Q+6krvHa1syn+;yRjQj=bKq6=is+~TIo#2NNUIrKTcwu_d#{O?cfh-F!HftB*C%hHZwuB5yCNQAu~bD@sa9Lhufga&1gS%>-sPDh*nceFU9(oWqpzhPMgG5nfPvp z__psZ96$Naw=0h^+et@9=0Us-22InOtoutil%BBNW4YqtRm7DVqVe&{E*5`VTjo<4 z(k15=Sz1$=?0Z%`1xiu_R%cwyXx3sDq1;{_&O&>sTgbK4sbicUE<*H^9%1R$?@~!#EHone?iM<#; zefOGzt%!TLQn(9)aUY+&BP&W+V%#EuDIbwQ?Vl-+PqL1EBYaSo8E`0)8*j1z4+ z9nVUeBM_$gx3<%G3n6s_p2Z;2OT0@q8_=hz~IWlm7I$X|$So{)^Ihn&~H&DM|Q<;iGk!i zP>s?a?|SPdcGM9O=!IaRq%nlgi(6(>i|mzrWj8R1r2FAT>p%Z*l4FA=aL}S!FTg$@S1h)2l9mudO2PwYcW~bg<-&c@Ki?tM4IA(TaouOzna)aR>@s+)PLOjBx3pLO84;BYwX z3;}xE-*Jm}tW*La&=sJO5@F)asswofVu*p^kWjN`sFZtHKonLw4P9o}&A}z!gm#ziXjLIg>k^TNVSGnNZkblN0#|(2$g!F;drU;{&Wjx0dHfO$W_~@;O>Nikg&bI zqaXplFynBPxjo&@9?#vtE;+S{iU7D2$PiJ76R{fa;5ZSte-#!t`EoPAGX-F(XU2Mx z0|60+IeD0?PzD}M98DUnl!tSelS%Dze%-V{pQz6>%`m?yl$yB&dN#;2o4+;}t28ae z-CJYdvJp%5dKQ0UI0t)tEV~@r@p?rM%6y}fJlVDsFjC}I6!B=_RoZH&xe@xqsPG%p zqe+E-jrdUs77q6=C0H7Pw^twiI?hghUf)IxM0O^V@0h7G6#! z%2TPIZ-V*ShP|wIE_Sc%~f2@TQ#q=>Is`L(3=}lQl)Fo#Om-nS)Cc4GC+?(y% z@g<{~ntU&{_36z#jM`=*siZT+%iPHi(m;8Q_&6Z|d8pl@j=G1}#N!%gOw+VMP%0?Q zI`P(AK3w15_zc+8xiPPQZ^pS-XGHyaMrk+7_RQxzKrnm{`K=TwK{Er&m1+#{4*`~@2xq?bUGWT1!MOm=yz?>wbqWv!ULA)g z-&7jbdx;wR*bnT74eUd+Gn$`!O={yh-@{Nf(SNF`2G;>PWoYTgW#QxEKdSpe$WL(> zI+Of>fCZm0*qnYRQbG72J_CT691MLM}fWb}G9=v|BH1bUXFxyY6q@*TfFP0w#35UlM(qq%A8 zH+!z@nk=DtJ}M}y=fX_GwYJmmbogU?`n=USniCcs=m*6*J}lh=7mit-BX%BrBdLye z81h8b<~l$8EW*r}FoW?#z%N>gqDS*;rR@>;EpK9dMisz7Rf%W%itX)J*Z|+yz|*Dg z7e_U3>5V#^fag-3ZKbtU2OjpU?fgjkg_f$6*6_luwlEw*gk%~M72({9-)4!0t+#V; ze$jWg1KLXiG>6LAp_q1=mhb)C9+0QW>5}1>sZQ1N=coB7?it$|y-geohC{&ZDX?t5 zvai_L)XGSQl`muZyw+IXntOC7!hXilh~RC> z(Z}7(kCcJ#)Yo+E61}Y+KhF+cRr`oA9afc%@^J``@*)!2RP(%E7MB>dvq~ALC{B-4 z-OE7EA8pYWSZ(%tlQU00W_e3IxCY^pRATm>;c(3zwv~%Bz zm{q59tEcMxu*)A9@fM-^;h&_$abz_{ITq;>q^u>4j3#l>HDrqftGHH1z5AiR)tJ}- z_n5VU!Mw(5bY0_SK>mjLlIyYgS8!n?Ri97y%RjRS^O3WHt8A-Q^)Jhpw|Pb9iVMBN zqzUMzwcWm(F4n$1`MxX2A4dr>uj7%LsZ>#z7`IyY=f8#tZMe-A`~EOSF{MRJ7F9~l zwe7+0?3QxwuT;*}6%2gGfZJDhPi<*JKcqW+og7eMMo|r1@Q7}w+C^JI5_~&h^q%|g zcl!BMIxUVw{6;4`k@ir4`zXtn+EE@j_BH(j1~T$zB78Svp1?NW?#=Xlb4Wgj-sw%@ z${vqyK3@_<*z!A1ZWoJ@xW;y)E=p~)$swtJi=Pf;a*M?L0iD(2S(l>J5rDvQw?PWL zu=ZS4r<{b&9-0+Vw)hI{;8o#uf=&{(%tN*toI*N$(+aN)-;iHzGpDWe@_oFmksGDB z6KLC#cOcAgn#~4`SElo9LUv=XZ-8Utl=Z>R91f}RKxs}LRTTRelWvJ&V z=1g{v;1EO}cyYvHjC3>Ho-7j@mz?je%abxIznWNsa_vE@ZOM+>d{Sa>jeA3ScF|P4vat!lG`=e z4`u)tf8VY=p5(K7UYV410>qkf#)|rz3t)|@`A~12&RA?e-Hx4BK5Y3FrsjLZj+8h^ z^ti0WZ1#NA6FUlvS@{_Vgo#(phi9Qx15D8|v!4^BQsbgnH=ir?5y*UimeA_U!88{r}7Jl z;**e9@gvQT< zKMrw1R0gv@PYe+?#rtUCE8TSSs;&@1x>l!ryY=Ly7LUB>`#P^?L^3{JYn#3?DPk%| zn7&v%@9n`sas(XU2uR#>-;(h5&#|i1^I5wM!YLcN!&PvCPTU z12_!~9<-Rq7^lYL7_A6T9Q%_Mo$%~+eZP))?-NecLXyeR+Zm?5o7k6s;T#`DUGouz zG!TsI-_Z+`)VJ$KjVbufTJ5e3z=MOg9~;LP51ccVOpvCo(!tX``@I9Kcv+;59Uh4= z!rNX49+md7q?y-`o*(f-C_VLujaLb-zz?=WrD=*B6E)jaOJV#r(92Xnw`60$yl0Y# zZDBNDCuODyS>!5#$GcDcShJ}8z#G%$*M`B0D_p1hEHBHuRJL%SZ=X|9?#H%m*hH%- z{)$ec%(l1a-W&+_$C~9|achlyK*TV5{}gpi?l)N#dt@-vFCn#_K{vxNyLwh>Tq`;~ zLG4;e6|fOvh)`xw$JC{l+f!Tt0oz~_Mh2m?Dk1=JQ%;e#RyCKvCgnccmAmPak@A=; zK4OJ_ktC-9WAq|WzmcY6FY03=98eyIZVoZ5_%N3!e@pZj<64Somb9U}Cr-a)(RYn= z61y=ylHI<$y`PNHnQy`Qxqh9Pw0e&B`fSg_m}yB!P4`%7!39ivX}fWRd6!P4xu=s^ znDdV0w(bC}`_UEc21)4^=>m1fDz^+KAeodj{Ncb*3b z8lAb#0cfT;RaIasscoRC56AWq{?M}m>GQW3MPxXptwpBIIPcQUT>N>IZVp7G<=3w! z48A00lcqw0iTdOEluAeYZmFKJ1%lE>p~i*L-_5*>rI>U`BM zXjVv2ADDhO`rI})>H3xUwkSu`Gm{exC%oPwZe@c6UDqY!!OF`Ze=aFXP>v(QW{T~SqE?)Kfa79;u{<$N*UQ_a@|bB_gAn3`*RoF+9#vb? z?P%QqhAHe5>HH`+qaErvO#3tW_*-7L;1$Q-tw$C|1xoO;bnBHy2ZH(n7w_<72YnTn zkHN%G#;;|$wkuISzvF5jDXqsVWcnlk_Txp~rP_K%=Ap^*barG&XY=zW_wMs9xq-Ad zZusoiip14h@u(6H;XLCD$QZasM9sK;A4kwJiV%G@|JDQ?y@FiF)Lw?vQc&J#IOXLx}uP?vcsMpHDHMycO|K9i@RGO{7?WNH{Ymzy~ z4!l<$%eO{(UZZN-ncc{??MYZ|iFmOfKkXpQ) zIR%hPC==Z_7ld*HOQ}x3_6gZ?dpQVyH;<8x#g5fH)V{cnf$LjIC{k-CaFl1h^HXO) zC&WnPnZ+vA{N`Jrm(4`oh0(R+aYioZ!PhgB6J z{y-@woMmht=#X!Cvd?~Jh_No$SGc`1r_1FrP2C|uJ#ZIB-#NHDsYbUtV-|791B`Om z(~vGqnbbwK;>`7z3#`Y+wsX7TZ3*}-LWo}N;zjl`xugXiC_1zz)vt-2C>vD|un6Xx z6{S9{^oBkr?8DeF|Hhb96Oi2hL)~YsM{`9+7)JnVd8iCH9M*wxNQeOxSE0!!+$mm1 zZ=IqNEJ}4)0bYmCQRR5YlY zqdB$D&)NeJCDgH#AxNWZ3^)G-Uz)ol?XLw1LE;I*O|_FzFC%h+Q9*2$iQe@r@w&kc z9ONRQ=(71}lH6qV z_&fnmZRkgq*gXnoi};i$!mw{TeO=XHJ2#(rmxzuju5W$gTl!dqEk8Lp_v6F|V!ZQnEzlg{`i7q&`T3$VztPKQ~l{>f1ofv+TRIg&%9T?CZNDE2D# zQ}+6UiXfM?euzf{zbZRuV#We z|G`j(hm!yXSo@mx#Vlt|6i~o#i2z5}G&_}(JWa&LQl|PX#ccTuUmfu_s zeBW7P=2f9&V<787hjF#fJ^VGZvVg7ctYurqXr2&pmtLK_cU(#DwoMg02OWjvHh0NE z(w;BZ)n_@r)LcvfcHiz)+7fA3RIP6z7O=%M_mxPxba&3C+da}i4T>1>S~f@V4Q9{_ zEPPKvB-Rm#7a{{sHQ^@Y5S+tdej=M=yX9U75 z@VIMXZOQXWqh9}Mr;vT0m%D0b&Pnl2au*Zl+vfrJ&jVLP?|3w9BT_0QQC-@T5tB=t z521Y=z0y$*QUk2QT~d5oKPRpd1!#S(-6SB&O#mm-I#Q~!1v3nKQ+;T-m=Pgo5|U#5 zP)%6QxdxA)-(fU4mY1=yw-(Lgz00^b;cU}CUamHQRfpL z5mS8caYbZC$)=D9>zu?`%I8ccSEj}Pl>A%iv`9`93Ga<)Ri=c;&8(v zchzb_wBpd5j@_L)+`Ja7fjvEWe7|nIJwIsuTh{?gXtzmACB9+OS+uA8+p`mOY0SJu z9f>XdI{dt?0gq_BsHaa(D&m_9Vn3a*=r0lTxT<&SiqZBwJ~d2yIhbJB`sOyAH|;CI zSu@~Z<*^iMZ+8sUsw?@LMRcW4rUf2g8c*e#SY?mPC{@mbs>>D>Eg> z$El~q?Bw{Vt{QhLJ(zjl9n%3R63Bg^zs0%xJ9Y!LwsC{{%S)lLt(x4TG zTZ`gjsZ=8li_C$u#-{u zbY$MVL`97pFPgi01pHW@5pfdt4B{|Lt2XOb zzsibw`rd359${AbA!Px$Yo3Z}oMsevDDLu+ua8ye6)$(MAqh!wa)qHwZCu*|JOX6$ zQvsUG=jB7gHEx2DqY27uST<)6o@UOc#%Dxv#dP?Kf25i!y{26Dy32`jC!+c3NXWa{ z$AZ(a1aY8EryiuC9?MpmiVD^wqsPwzsP86$1D9vcSuS{|^ykC|s^GX{YwCB5ID`?6 zu$!LFIYr_|V~33%^~ptu#0SpP=loJ;C4%t=`j5dMof(t5MJ9V{oYCJi^bj=*lK0(+ z(p*&+5LMP$S}j#A7&VJ);`I7vA{Oc4b{!M7x%T?vdxwV4nKGkqUyO6qADA;)FB2}v zgX+5%N<Gt!W=HuT^(dPw zS;MX~n1ySQQ?!d;97&wNjyNTOUeiRf@8~U$_<=1|x|Vv?MlnvSgV$-A2W-Sl`_mvu z>MJ;{Utf=0kw?5Tu5?EM(EpRfvF?nY9zWo?UFk~eI>lGjzE(^8KB@Gfo~xMA)mZ7W z%RQ=znxE_+=(egoOI0Q9(wXAl_LY$3kJ>8*yvouUEmJZQd|78!QCqW$(|PdY`hYk2 zrzFR`r2lJ=T0xmH$`85E)GnN6I#f0xap+~|G!YCWT`HMY)7r#7MH+EH|1e{r}G4%B= zgSIc4H_EjF8Q1T}xIOAAk+Mq_Z%eVON3RocKq9?jzM^DKCLA-t;{AT$<9iRaIqj?F z6u+&k>1Ip6c7gNb?e(&t^hmsQWSMRuMJ?iV|7xjI@Y1U zI2-zr&UOEZI3!=v45IS^s+1F2H1WRhmA=KI`C5nxhjA3%$%xsPT^6xpTV-)L{kow4 zN|gWMpI#k?_tVh%>5D+zbx%R9bN3eM|Y0@O?j=iCsBes?Ww?EYX zV&`#2-a!0`xzRD}GF`(Fv#7kEJ5fkejauoDVA-&wZ>Xd`S73WCf@NR#EmOH(FtGf2 z+CJQ|tZQX@xRKt-d&u)*xo(Zj(6fRuWr<|GL?~BKy)*dmn$ho2yz=&bX-0_(Qjytfat+HU_cvC!cwowi) zl$Px_X@3&N@(S}%k;(2|wo|BZM^Y5$#a)ESkLSCWJ7U$qJx)Knm-C)NEXoR>lTW%n z6ICw-?=3CWG36U#;}w`_f?vVISVd=oR1}>@lA}t!vyCN2!n^JLQ-EAt8=A|OBp03y zjf=Yyx^u+>V8%qYOWq%?G(~8l9W$^yHi^)tcF2>xFCx|4LS4(#xzec3w#6rC(mF3@ z>G_Se&vSxsiGX}2&$xNjX*>P8om>5HMs~W;vBEm<26n4rPj_?oU9Vqs19C?`^2DyH zsS+AYg-XfTCeX5{AK0PdkiVaxv%4225XjRSx_?L8nS0q2VX3(CwZ4d^sg1_i@6+xg znkHdOMfJtgCG>|VU~dKwa!Sm&L?>_1Es#t+7w5v$Smn->9r87Y7_$jfy)2_fn>It<_^zQ&|B#IZ%l)E0D6(g{w3wF#*+F5M5atum$WdDP?N8 zgSdl&URg&nCdeC~VCrMjT$&M2(&1atv3B}9 zMIE>S`hx0o7mG9Qaf<$>jKaK9!$iFm%o9f{s1|2{jpocj3j_ocKn#7W?jzo;g5FN; zhV3+aKLwpH;`AFi=SUx+r#H%?y`}y8tG6#?!g(a$Me-j+w_~iE_8FOCkk>LlU_f&D z0@ORGVW+yKFQiM_st{oHn)KogoN;(0p6<`{F0i^E<ghywUZ&5MWIp2fRg+Vv5-R9}z2Atnepj%a( zo?vXs(=#(!vka&pU=@0o0&J$&YAlyL`=N#A=ON$UF+eTq37`%HQkP{H;b9%x*|bS{N$-6ecYochQF9x1n_vq>)gm!m&Q;`%Qh~z)-kHjeubC0Pm=+-omWB zgoV7$x*Gv8d^|Be2)7W~c~U@bUcaa%wI)8pDn2{LVxLI=Mmk(2UwyYs^2(>hGNGMuA{9`qVibQGmVbAGK;n)|c7-RbvaJuEB z%O8EoIIxY>jN#I46M@R*8PTH>NhlJ-7JOT*C`X zZTG!;D#H30yI|WIP-uy1!bLY1gA2uh~*R#gBfLnMtzc<^)PBOLa=bs9VPN~tz|qPA!2ln zzeYZgX5}=B{+V-ce3PyQXWv33{)>Ru@b|{Ubv?0c&+5nOoc9l>8W1gVhuw-D<0&RiC2h`9$}lMADeyv^rxAyB^|8$HAFX$bp4l2ioM&oZb2% zLE&H5rOs>Gva?^iW8^1JB)d_}9GjuM-ZpG5a(C~%-@Z+irq{NCz|$2J|5i{f@o zWdgF99eaksEqv%OAe|&?mAiOi#P~O^#Vq;e(7RD zj0tl0hF5|bm44m90FSo=rg_gu9pGKtX1b$u&h41$Ds{~hy4NaWmR zzz6KC2Ub*@+_*cBv9P%6CQ)}bv~I5C@b=H2q8#(3CD!AyryG_$pP+UU7f8?QaxXvZ zzy)8N&J1#_K0)KRV~o`vn!$i`g-)naze+~eb6_3ZYhRd2RH#LFt)C4uGw<{q+n)Nk zVgSLJUll?n?i=fWWokzrCR|KIj+}7>fK~7QY`OdZ2hqJwoH@_Oe!9nWV1#uxk^`-3 z*mamon?K1}+Mh*;KfbQi{|oFb`al62kr&qyLym(ly$CV&!-AyWZLTlwRFo^34q4~p zC|&D(f%qSUB@A%Tp?YJ(+2SSlbH7>?Xz3WZ^*w64)i;N-p(F#Q~UY-GF6p`5dj%%^y6>`w=cGf2V^``Hu-n%{7(39FMxSRJs@$bl|X?0+VUA7 zwo;q1+(%^1`#5buV&fjzxBSJ=!@sHZ|EC$Wfoxm47Od;H4fn;a@ZQD7c$S%*7^!-8 za5j}+MQhi8^&QJMdUVzB(Xhoml@Rj}<;1aqJsY+S=lu8FAbUpVwG#)hJx^SH(=6HM zwd$0Hoi}X?HEsivRia7pe>{i2=zY?xd2y|U`bE#szo01XzA;#!V#P2JGNft^TEpW( zFC~#Zb#_HHId$uj`!&qp_Eo!EU;hdf{}v9I)WeEnl$IJD{zb(espn(otFlce-^@`% z)vzHyr)!llHn+lXP930Yn&q(C_-DUM7)cMphM*HhGYyCifj$1(jj#QwoCFG&GRN!b zZvCG(&wC@^;gD(KYM7%q zSRPFlfbkQT;J4o5V%A||gmTsG2p{Rs*F9;LY_-^gwGMf&XQV)gXIXt>ssFfHKQgE^ zZkpz`WJAKdaAj{rLZ<6L4vVwcvxQd*!#aNg`6~c}NxQ18z(LQ;|BxbUf{Cu|@H{B- ze`4{U5d8NLm{iBLXt9>I5ik40FR^}i(%&xrPxyaB_{$dmhdI`tF$4bZk@K%GYX1l`pHr*OJRepI zO+uwwfj1X}L+3;z`*-`L+|K~#@7F^AnBkiJu*ysxGD6ZbWER_}qxK~Cs~df{?H0z1 zsT@v2hPwCaXVNF;t{tgBRkd?S@>%#_tIU7H^)JtMBdgx)w5tWscV28> z6P~ZIeCW~t#jUn(_6%M0!pSYSe_7+#^76NMV^4*Nj*0wmPnRan2{q8Z_xkCazR~gc z-isT!dX>jYuH){co@YyV)s~~j5)yrDhcNm0e{{g{Npr|h+Wg_gUf=@iY;U%FU`A+s zVtl-|oH%1FBk&A42js6d_dIw3h^Zvpc}V}e7nYEs10i1+7gY>;(oP6Zrz;j8pSy^& z^28x_ocnBfdU2i0H1A&^KHn7i-@a>w8H|S5%jz`3vGB$Q-JIiG2@>8-K1Y;4;khq* zo!LLIs4Ibw{dc3p0|Kp0-mQHXcW!PL2KH&(=#aeuaPWA@v0UvV)UE zNV9mr%GN>SL4?77d+sDVF8vO~Y%5ZHZp#)B1B&|DYWZ7pXAggwdt_@4)#}}fTR2}b zouMOMdWjv4_2>Nb%e}}juPukCqo6}WenfwIxfK9yv<%()kr^?v$Ib^bo{P=Fj?f!@ z1v^-oNrf81)%&+5|1z66J%&E~3iWk`^XS2>xWQ{vf#2NFmjNwq*PC+R5hIc_<^nql zYHl~TReD7>Iak={%=o*zpgY`5jUla4RXZ*~T?nLu{5<~?D`y3=u8<*`aPf>nDhze1 z(;H8_HgW}IRK)3W%J!drD?>Utg-b8tm8ZW8;TNm?gq@X9_}!W6x8@q+WdJc}jd7mP zFcCg#;D%a3CUvIJ^jK>c^3PU(sf8PC+Oh$+^|asZ9QyeGhggY8m|9WBPGBOxI&~p= zqXENjE&dYQOsu&0;wEe;Dv2|KGe#}FYr$b|X$#-?Hw#UU{PGC(e9wJc&!eYk9pi*q zmH!j-@a8YKBQJoia|a?ZTI%rqN&x?#E9J+>Xek@L=9vr=V40iuL)}TJkfB)Egs)Z6 zf2^p{P48G^Ntn8q-Rw_Jm+pDpRK+~`O`l&es19A&Y6!#(2}n^`lyUloim2N|oFZ9E zR6>&p{tI|YT(8!=1;zx)3x#$?^`G2#D14t<4O z(E^KAsr4gu>RtSd@b2G4?Yu^Z)(Y)Mb&+r}w3|)bQu8+nqIf|5ZDeYw@?j?bRrvwW5B1OCGLo6!)X+7D`fj{%M$JNdV6p+fbI=?nG z{%NRLnB^UImIIcwp+IEwb9L&udK3488*v2`V;Gvl?zaZ9eqEg5V_wJo@}^q!oOBkg zl)(Di*poG(`X$RSuL9M1bdWZWO0I*Bk)sBN3(vOYbp@xLgh{>sv|h83!s24E9NJa% zJXf}ee06)f0Ka=Mnq3y{2*lbiCbu0gSWD3Nb(3=*C%Fs0{V&DnF*EOrc^|<`OyJ6Y zVSWQ?5GP5jD>|0G-m6>x3Flyz!3-0hMHlSDsDYEdpm4{CmpsTG)yp53o}IqO$Cn=f zy@cuM?>TC0FW-x1iJZlCT`+lk_bw)es};vb--|2I`Tp@mJ40l*MJn6e1AhtW$rIJ{ zswUXVj8xNu~E^rOoI313bx8HnOhiGr2qWTDlyT=?Wj^y zx@z$z-nad%-}CLCM?KhOTJwq=wcQ{Emb0r53QR|57ch$${(U^Sll*OUvU?FU1UtNb z6^!hxlPSXe=kL$_SW47m9S+NEeZ~6ly9zcToQ*_(>GZ!pgF2W?&XZ5}AnEz{f7apL z{^!YC!JDK^rVoKz^ltIkPfV$#Y7`w)7j5DX5&zq%{}C`+5mbENH3@zsJI)zvdto6e z`cKD#BPf|nM@y(+|Q}#K{%DnSC$ikIR?+t^dI}fP=%h{QLOP7?cXnQO_5b z$^I9a;7zoEhYi>C{$-fV6T>r5n?YIy`MU~9ug%5(X$7rt9&51K_l>&>6-z}~`kmid z{)V0C7j|D9tyYX;ria1}!!_RV=(I>aWB(CDd+zA=<~JA~=E{^RwdRTLOK8Xb-M!I> zTXIh`L^FIdvOY}ch zXCOd?X|UOc*QdFpbZeScz(V|f>pupTtu9kRSi&V|hWf;jjjQ9odJ*xWI}q&T$ZGhB z=e|&1^)Vjny%^ffaaX>7@4;nS#A@Z=&}ii%e|LYG;9u45OJcIc>F?+pjZwBd4@W*6ZjnXHgzjiyClf`1L0$B2IX)UEjONE0rWBN3~Rf*?AG+b z8I%hA^DblZqIVTy-OOw5{{8poxL4Dm49V&Sfi7xdydO~~apMmwO=eO~! z`?MrOWbm@=ZyVC2g#2UH7XoPPwpP_#GK<=|;N%K5uj1rQkWzmf*+z|HZVyK7DQjVeddVH(?Ayc&8uz5m`A>bDx8Gv6T)vl<0U@%l9YboBHURuD9$Oo8r7lJK()dbeId z(5wq=>I!-6pJ=t?fsNh?pD2P-@@P%2^`-fLeKM2jwT!MRK2?_RR8u$&iZ1grW)z6> z(0r++{8b(dn4+;tH^l+E4|X*&`W?vXH~2tr8q$?smTmMt7^u)x6MeI@g!HG!=#Ze@ z^z%vg^v3%wqHW=$-Kt2yw>UcpoWYV%LK(W;kDsOSfQrYP`Yik(>j%k5;$^?&_Fes; zy|b3)bT%eYSXel|u;7%hRe02$MZjO(tW6u+kW*B|U$sQ*B{@S?A8L&kG*<3yDn(PxhJ zVMCDX6V9|pcF7s&x|*O?JWekj=y%_>k5;P6U(ug_dSL<=hOgz&nx-9z=nx_uPXUiKpq)w6CiI4=U)r71Gp(-*51(!+bWMhnlT)nDHCVHQ3CGvIA2I2t$)Jx2q2$ zI7Wf4UChEx^YsZ;ls(o@W=7PdcRAY}=4!`^^($_!FE0Te(=T8zRI^UO6@l!L4qN=^Ku-6(o;;*5EC{M^6h!(h$Uk9-HOGa1@QNk5c?ezaTTCLjvG81L-vn!lXeyG!l%jUIt-+d_K4FV=L z_)_H#dnm=p#$&O_@_sQpxHEqLiEbwxYqX1nf>`&RM);7iKB3y)i2d*0V~)qwPt5?@ z?J-z>v)22O6-ey<;nP9R1QgAFl{M=xYe)Jak^{;ur774^H$i?##XgEnMmU;O^1LfnF{87>N>Mq=cm8kM+g z(KBjUcslgGp5d+0n7alK{s|z>?zj*AT9sNl9jfLm6&bwYN(A9(gZFFS{*K*7AW*+Z zE^2;=-T;?3{&+GJA27;afA9oyGW4~WMB*T*m`iLkJLzh6DnPo?@x5+N2yYX7h5fbaqm@CF<_8U_u1kE@aY5ToCya@hzx`De zA|uBcLosSCJQq#Lvx7c^bXd6X@PQf$^y!NyiXWb?#G(f-3`gph--$v2fhcsu z-mz*uJw5&S(xiF(>4P1Lr{BXv@5)bCS@v?9W^1{BRPEGe+w|z0)+E5;qs>Rz zax8e6+usr~`Yb%ZPK$&(_p-+bAro#@Clr6Mllyn!XTlQfD*vvQmaF@#*B6Ob!0a#2 zJ|9Lb(|2%X@w_XWD&X$d;KX7VsEEA>9-Y=NjXnEDEuXYjk+J;4^{x{6vO+w1HbkC7 z*YYc0t3Cu--~4!{S7y=Or7y+_t+nf9c9X2qG%4Za+^j;(?f?Mu+(KzK{%hiZ@BK@(=pJ7QN7DEHZ(=hL2tLb$? z(_alUp5u?d#~N;=RI=eZ%x+ek;ckA(=+aj)g7iNG8=pAP{e*E#=igOc|Dx-;`sJ1y z3^-?1e`QoIp=i*`)rFaTO=Y)W>pN3*^{X{zneqXYXHdMF~5oiB7tWAxfKA9=ia!J=eWb+2-&!X~K!9aRDr zrc*fjx#x8&?LSIK7>&rSXH_egrY1f$G&Bro&5X-O#RH$oe2=XfPvs+F&v7T}du&kK zb+eM>%5Mq>H-X;RDbNmew)XH5hA%!L@*Up5aWW5+miem}F>mTP+P;sa z*FKOq9o4*cdnSRV7PqKI$Kg57w-q;43jw}t%-_0Qi<_=%1Jgm%HUT$TD}qg|=K>8m zD{*nTot=)W4r!aZQC|&O*ki5;Ybmy0%+ch76xjIq)7Y z2eOLTAMq8}378p2FTN_aBGku~X2tR&1rqi+{zin^l*@?F2*t4<**1M{kH@58PZ(8G z8+-Ds{XO|ZpK+^EDcQmmik*!|?BVl1+kE=To`r>Z>RUIY|s>pmBS}GSj50k!jVT6BJ&>TycmVLPWA~ZsY*R12S zb+8KU2mF4%@d8={k^hgaw}6VW>)wYIK?Otw%g3#6J7%eO=ey_qp4@f;Lb4Ei2{QP~9$< z4sco4%a%#)lzH&%0MkItXmmxqtY_O>#)4ZUrzFi!1Jy%?PTqm0ih}d^1m=*Tf(Ist ztHwer_g<_F73_g|yhH^!R8b?Pr_|E0zr$irfRp2&!EiDVhkQru9o0j#S^1utiYmzry5*N zrB=}bUAhDh%}I$vNiQ{MMHuZvzH7dkq{#q86f+ z^;fIk*ECJ}r)@Tsrj9zw?q1~8g*VWujp%oOH|-Q=p=E%#oKKv9;9_L0YK3Le@;L#Y z#tjJq;n`|3R2An`#;286_-FnU8HGQ+1y>I6e$T2P5k?v)^%Fk z;|Fv%o-oSoXruMmobU6Zac4!JV6=%M>v}7Ni+L)uHv_DnQL8Nzyiv!t%zCPEUKR~G zMMR&Yp_S>I4LuylGtecfDA}7s^)Kx9{uqX z+D4MMui%)%#utQ6ovMLjK6W^8&gAO+&F)Y0^qfg?-%{J`qj`eh$3E_z)jDz3> z78#?34tUk%uraKf@01%S8bQxS^Y5Za;>N%i5ff41dH4ic`JxN2$9r8x;fryH`or^u z=Zr)Je%q~Ok^HD&yWnpRTAtZb`pXDJ^O762&&kGv(;nKO_O|=f_AnOh zOxKU46X6syI1mC;=Hwy~$@og!1$W(VLq(&Dj(JY0t!ut`;!>1L-!g-RdGU&cp~M)< z#1OIQ+DoPjK%AQE7;)=Z|Qt2hQ@whJNgDwpeG| zP_?U`5=`b!cwz+MhD2vSMFm^((rDRiSe1c~AcSwO!BeU9l%@ClY|oJ)QuX)V3RIKh z*XzoxIrV*9e^X#E5@05g{#CIxz4SGl@$VzWr0_#ZQ=xoUS%}X~R+kSzJ2HNJtz*WX z;3Ok@Vsex0g9CF7O1qO*&!Y(xf@mfsoVR3?-{=psUVy z`|k=j6QjfnRpXOu%rvsRNqdf^aC!n{k*1<8iXMHnvDaSo@mWt@ERPZ+K$OjDpKAD^JFuYLFQ%lcg4v~EA!EibTLnO#S?^$d z7Nt_?nB^eEzs6YUaT&l<)&x5N_ZjhkU|=~7vSl-{@rz$tK~K{=!=NJ4P@yi|fIP*t z;u;y=*)Euf0dTML-O&BHSdo0{cs#MI%%r4qFdOoV5#FTK&3~x8M`2rJ@_Y8C8AL}W%mPaU^Z|JfZ81(oUgREVy@OSE9EhQ`LVgOh(1OsY@|7>_D zSJ7W7{1^}ZD#2SJeo{>K!P%B)-UpAix6R@0DMI5X^5nwyYRG%@Qo5>0@F?{1uI(`1 zDGnEHQs_9qN=b5O_2xJq!JUh*&_;jp@Ko`MjvawmsmXA&AJIj`3oy}qjIa5Pbab=h zd}2=Jp?9NZmk*3=HqT5;m>dNTfB5p9UY1Fo4nK=6_LINkW$&AG8{_NLCycZP8y~V2 z;(@8-ndq}0dph5v)Kj){4Bk3^|2{C-)h8l{{W6*g<{$Dn31;V`-vc3@1O8E)fOKi% z0&W*(Mj?pJ)s0VOG$`>gH6Q8}zJUEU?KLs~kLB7wO!8JJQu{0IgBtvF`rF8h!2|D^ zJ@chC>x&Q7!*X|p9xt+>hE2E2e#5f`&)@rziVc)#4h_KlD_u5R=O4E~J-_p?QC&un za+M)i3IOZW!BWk=kA<3z&VK3NZU`f>y|r*HMW32Us#?eV2c6EQ2dqMD)oS%JnxLLC zgZe{N>pW6mgZ9?9&#wxUEh({tv(Dx4Uqq^5}E1XH)XTgFfVOlFa&4kcV0bv$iI(g z0~5%c^U9qI0vre;VdgwO=%fLYA6a|%LAwCK@yW-vSVHnYq|HsrUlM3i#HKhYNx!QY zqD`8{CH7#Kj=l^yMS0_QyNtWhT^oPos(}M8@WJ3eQA*|wPvF%t(!6w9{k;_g#-4QD zJwHS`C{n~_op(zCJfz6ChwhI1MG6c&hVLDDAB8pup(%OFi->$IoVQAqH_z`aF>r4U zEQU;#(y%5y0{iYo_N{MrVEEiCql{?zxiDsmegx{djF+D4h_L+Z^Kv>IPt z_9O?4#^l17F(FTB!liGQzCB3;xV>!Z4)=^#lw@x)sd#2&F{$8Yx|jX+6*L^`-w5j) zKJU4LNte@Yo}AgUFcK3OeXUC)2{KuD7@pd9b5nFE7C1kO&7!CO@my8HZA>~ptFfpM z)Qz$G#i1G~NTDNiW^(Sa7x0x6_~XXTR`oqMl-RU$)^5xst`P5%T6=JrLmU7ctTGN{_h}j$U{4=15=iA!lS7GR z=ml?Ff|9OyW@Ep6p}-i8LRi#`Bz`JgKFge0U;;F?8zBxO*-thxsZ`&5zqB>~1}{t? z23RDUd}i%riJyJ{wRlNj@yz2mU2hW61Dhi?o9Ko!-2UjD8725Q@YsXs;J8_deq@|w zw$5AnPmCb~KF*~)ve1w1tu4*RGsBYj@LjC4Yv_L*<^aCGw}T4toSVs6 zub;qhV)xg0e!=}7c}NncZQ9yIX(aFQ%3m3g*3BQ9MLX zcr#2Hd^sgov)_XCnOXhMO%8yYC=X#)N3JcO)R<}ZYxO+RG_e6&Cv$>f65WAY=^de; zNqJwbNZvfUW18=hb$~N_*E0#_2JYgxUwU>ac(znfalPqXd1=1RD#lr?wdai{8kT4` zME@<>|NipK#;ovf#mcXuTmNtHk%=Oo%`%I->51fhBA~0tDADJkgg8p>*LW!}PhJ^9 zNp@RP9*ar!hsqA`>P2Rk3o7}M zTJQVEt7k?loFs)huIL?>87jKQjXxm77@`3DH$y%p{m-*?P|v+Y_ z7a&#+Z2Q*umT!%IBAbmst#oswG%>gh!g)Riy|`F_qyUUY7H|i}0K+wIegw?)gu#00 zY=?vgfY*V9z{utN-UD_&^~-Ii(Emg`1NidNRgwI{`QhbBp9Jw`Db0HC zPZoYeqJ$IZayTv?m28*HHda{}ld9Z22g4_B4*Pi&gZ0SSbFZKc;Dci)qrEEtN7EaM z>>_`XNTbII|3ijzzC>US+01ut`?*P1fo!#%$aX-G?C>JgF?oS9o_BrscOpd&xU1Bj)O{OH0zRF zj1l}xI0ePA4uL>WA!S@O|Hj$qGBLoX;nqaFet|dtNe`aW6zbQhMYt7ThO5B2g`s1m zp<$+IR6UG4XWxGbgUu&hG~zx?%M=g+8K;*tA;yT;K1N6^xzEbQ6j8U;-J6wVcA_A9 zCs1q3)U#wDI?r+X!-Y$IyL^n5Tr>!teW#xS)TZxLFBRZ@p&h4EeXaX2hV+;2og zL{9Fady1wgpy=zeGVY+4kE&}A%(+}7PoYfl4i z1kNJd^}&eu48hTy4A()*3OrB*Yjz6%j|&OO&DD_ik)p@P$GiPXawE$FIc}Tcn2B#X zU?Tyv2`1QtOn(Bt6dkJQn|8)>?Y1)({Ex&oJ~`TV=#``$*=T(r6nU%Nklp>u&})}Y zXJ=We0G}DhtnB8UlL9DRwz%$nRqvGSE1)=h%w)216 zA5lA?;6^2k9`MdLL%?TOO8}0&83jkV0GjsLczdrDPvT(A{Sdhj^(_0hPNhxSI4D#Y z92j_0V%)uZs0BBRL6LWL`<+%Ak0OXANeQqpTo&RTR{;hMNAYM@hgouNf!;kV2QGVF zcw*VV6@4P`e5!;lkBou(7pI3dgSEmTI>sw!6YltvUkN6QZWgL1!GZ-s?X9lihM8 zAecF7PkvB^4Ik{RvM&2^xHV-%A%PS@k0FVPhqsZVe$5nWs$Ca&b zH(0Y4cB|}HUL`{COaLe707%aR*zJ_V_t5lsqbYj2Ub^zP^6TqG9Ci3Fk;6qz2~P7Q z_bDjK0OdT{jYd{0RDbUTK7mNCY%O1Y+REs0L`k z??HNF`JO2hXnD0-QGgDTXPFS3WfrVGG{%F}jUN~6jn zZ$DJAdW!)O`uh#n6>Y`y21TKgoPd3!k4Z=3oKqy)Lsh+Kn-U}q3+0#DuHjo}ZW4Gf z1cCgf*b`rxqAWf-(s7b$`mDlt8~T92-uzxxq`HAp9O(Ix8*R;Z-gknw1d}){Ci@{~ z!i3gxf&lG3Kg8Ti`qE4(BYE~$u6l`UOwSjqx}|FNMf>C2<;Rc*SvU#18+sy`tOD&2 zDTF)wIHjO@eaOKHkeY>;#JQAmE&JGLUwHJsM=1UP9#suE*;fH1ofS&=G2n)P;9|zd zr(}MM4G$&9awzQt5{UJVJW8pOIrrU?#F4w=WO^uW+RZ*PQV7^vK>^+YFx%x1mw8z@-jB zjE=h9xM7juJz(W=sF^6CNZq^mSb3jgMgSKFXXJ3b{?s#7g)e@$^?}u)kw+&jKqB}c zM047W|KWzIF;LL+;}A5Wo$k8dUf8bH-oWX8HfpDTy9J87!oIL6uAswH|IrjtB3P;T z+hc&*-~Igo7XjGd)oU~s86(yx(0sMDPj>6-XEgXEAb!#Zae#Cvyg;n?rx5(RQld46 z#z$?3&pN!jO_uY+BMykL*9NO%4b81n5MF`_9M%D+)) zr={f3pT2hYSS4UsN*UI-b5#fx+D;JLFF_2w8;xO%rEuF;a6!U{j?e{~bq~}7D35EL zcgzyK73UP34mT?^4L)ehn2x7x%0v0{TE7oNlH%?#MURozJ`Q%km?k*i2v0XWlSS_c zFVfC*r9zu#0@c58zy5Ab+&sNhbsOklE(wu5l!#tJF(w?rZjy-sm}x0I(|Rd73q)yK z_pDNwW#@OLJ85n?#~*g{M%>MGGSFifqDyb)MvBfgK*v6u&WY~e4lQic;bLfEWz+B* z6dY9=IjLMBuwlNjwN!%}PF2S7%L2Btz;}_J$4U!h5w;bq4oS2YB~qknoT31+M~j#2oYM z^FR|o1rPqw6iH$h`19R~r=EmZSa#98idQ%$W?Vr`hbnx<7mTS8npOmHWqiN`SM`To+?GGMTV$h zt7d6gjf|2|P9q&@8At^HBgsi}EUX!tUN&{ESOtlotZ*A?^P||p^x>)8lCI4s!I8qT z=o!N=V*~=FO9vy|SzB2&h9E2O>^J2IOGelTVEn=~ODQ6E^Srx&z|DN7B_y3I4F3T= zq2f;XV4%S6uz?$%#5f^mmP{K34!N8plbhH^*se~#2C3UU&L!EX?dW$5xXo(S(l#;W zrn?LJStcW{Wg{dN?iVJ6g^i-!RANWcpOo!*dOg5+Ik$H#mU5CdXhn+1e^y$%t8@}v z)^khHf@8vA&do&NF@%U-Sfk1=D{(2~w-m?Lc5Z5n7bAvMil6(?bg>kBjGq?=vsMSO zj_@8GnLYCoyJ&~;Cd@=m&Hqu#M$x+qk~nD%V|%Ya9~0iIqRG=tZWfeh7(pS`LZ@Ex zri0K0I8)albbFm&yg;qwMjiG)23zdjIOkg=D2UzO2#OR$DPKIa4^?iZ1laPyXy)h@ zQi&xoY4ZBA_qd|B6>?KyV+#o``BrxYopB!5M;w0a@-TlSdDyoSyn$htD1}kZ4RQbB zV!%~*ui)Z(UadrFLAL!$v;Cvr^?2pap@)9roRJyeGn-_akgw+J3{}Fe*pzQbX{T}% zdoJTNrHYgug&8j|@y6C>e13ljSxR_GVx6)Q-Vhd=DasI|MC=hO7&ph z-|B;^;0+8bpw>~`R|O)N)mywxD%Jy{pTFh8yDaUYq!CX@o4yxZ_Nd=)< z;>bvJKlqVaffDAd%f;XBcb1JB#;4T=HP?0I<@yOWE9>kzk%DGU8Mc7ew$)b_qa+Pg z>!R*=?yH4i_737f^L~booF@VAh5P>5Oxrbhf>Khd%FS5JjL)`GM|Sr1`>)5%XjRv~ z<)KGKN*|1QVPGF|o3$kcKB`SO42cwyhB0lp)eM9_e;&S@gC0J}+glgIq!@G^Lmexw zzvVf{TB&=4(Jt79%PrugllD9Vf7|Ft|pq0jrNIYIO z7N;4E*=yaZ!~Fi}ZIHL^U=(7Fg5UhKi_n&_a&1(0pYu`*ESofSPfR0e2i!<|-$L+s zW0%qitqbZ7vT>uMe<=8LgRo|S(HGx@{ql6?S+*lq@qP8lg@Bs99SiV+a;}gXY`PzN|jmw{JH;}&fE&l|h z6%B~G^6x1&VAMbS$kvMxt-XI>G0|H|Xe-1Ra2ef6Qoz99qg(rBP?OAUuX8j_{aRU8w~W@Rnn{nfh~MtL z4}UJnP@B<7PYR?kWQVjJyIJx`T=o9XgoH|Ej;0qRa|hoDO0w>*|sPlCIn;5VdsE1}J-`Cab$iDaPBh(+l>b zUV+-J@WZWsnG4qD8^cZH9|b9pc*^q{GIK<6_j8$l81>DCbC^yMf;_qB4eP~8x=G14 zjs|!$eHPPH@1BT5atma@ZEQb(r}V=*Yu7WUm^1WG;6hJ{DB~gD+7k2FrZhl^OT`XUawFMN2fsMGX z+~q;PZ`j7n{7Oe6jS#NtF1ux3_M(W*jkYP`;@Tw@OQLVr=vQ^w-<+#6lYnB-Kkq{z zmZLnx5yY0w&Csd3HFQ~#k z_&W2z$1?roio+16bBN?$0Fg;`HBY@C4_c(m@@DjUxmsCa$&4tuHd;pDKX457m%IT$ zL(Eovt@&c^{fW^+Vs_jzw8(V%vAN_r=g^gKY1&|jrucN_X+fgXWoaH!IL3aij60^! zs4rMcO8EuwWy?p`Z4T~Xf7jGkVSs5hnhf_IlQ70e-Sc?=9~SG1C*NK9UK24J-yM+M z3uGZ-jEZlRY0E?V&NDqTMF6tNOc2UuE26^qJZ=09<|SW;TY#3z?9U&;z2vFDp~wS_ z1N^%tGu`z;uENgp3`UZu+4P-sYwPnP5sKm$HacheX=(3Rg=mJ;56cgM%85^Jd7g5W zT|(`ycskz!dx2RF>j_mZ-^n=4l{bvk5*XvTL^z0|5U6*mycvO5FQ@Nv?8!V>lN|ZA z8*Qba+H(xjDDEquROg{gUr$1%6`rSOuvheJJT3^)sE{P@17)Prp$*y z_V$b|?h5-~Ugm6B-9_gW6aT0yd0ey&iLrR2wlZAD>hC>Qpyh*2GBadnhU7_#!u3Az z_$X!92m@lB)7K%h-K1jkhHlt@k0K{lY4fN#WuKKm6F-w`62Ttl1lGkq_U8d`7Dem+ zO}OdZ^blLT$r2|e4yS@4<%O`$y1<5k=H=;ST6n6U`0Y*poM843aJS+(^7D-mmO=4( zZKI8n5UmDyS`dvv9aq!uvWFeF*kaq!H-UtC!!vEvh1dzq?VWHMTARZe*`eKVR@B?|lTSIj*( zCJn3XRtU+CM%!`IrE%#NqxG~|R%DFU{%OP}3~NQA(zosup@Nq^A8-XtXcJwF$w56r z%H;F{DG?okG@&7y^qw)($M$M2&*D0>jMxlvaOz|IKj8@;@_jAxC$Pb|zf4*-Mczax zlw_;p*tT$H3s=!&m48q9Jt(6Jd)6Duw@eo#c1o&Y>J^74S${abzPg-KPzHoJ+*$H& zuN~xXF#%DHvoQaIgUVPD?K}Ol`FIfi zL&!q%wY%9G@v)Uj^yVCdcEd+h^O|2RCa=ZI_hT5am`0Pgt53EFu4oPGO-T^2CWWps zUvgP&Pd#mZ+M;SNXodDcyUB17LAHQwc{(*Yxv=9skJECLfbVfIO{)fnCX#rXZpwk#HFPIrtPrg;_t|Mi!;w?0YMe;~3qrC>!ND7b5_N*LG9+ znOVe-@9KdJ?Q9 z3mCLta_J7!o}ZqEQp{-0&LyjXot2?cP(73fus{tBhm%(K+I-%HPM*na=B4Oo`u!gp z5cxdFq5<5IYHetqJcHV^n)Q1_4vfDW!=EB;krmhEikfY4N03G?o4yQz%5cu`r9FsP ziH4DCs0?`N*C@FO_gC&)|sQY3Ap zNVE9NRvHNTO$NWEXa9!;e?TL@;oWmj?wQ`dLK%w;!HI0{z?UZn4`#UfBm+`&@~hf! z3vBADX|zW z13IuR0wl}4DPRX!-Ik*yR2Z=A0((YPZldThsFE z^7ecN75mUK`@`$?Vv010yRCuc%IU}Pd~}NuZN}!Kzl}864D<#V(#%$O-|}6|_K*_H zHiFO<^2e?Thv)C|*K4@XegTF>Ojdya<@aH2hORkPJpp5AMpx$U$p5;cuZ_LtI@NW8DL!ZK<-3YMZIa_GmL|=m7bx_QZMXN#2#gLOv%w>n=e_|5B9))n zC}2bw^_xzq?jQ$C+*yUTPqHC7ub|`r8_OcR#p_2(AOSf08tP@2M&rJxM7gH7U%&CS zBA6ax4^&%kao72l+K_OmVx{3F+8oxSW}f>zmfKI{JUmV%YEt0A+m3T61_xx~oVBgO zx{hef0o*}i`mo!I06=`__yswzg0T82aL`*j$W_r9NIe~2>oqHKwhzMhFo6qIE^`5; z0_Jn=F*7IP?ilJ}&o=!nh6|Hxo_60{eRKzEe>K!48jBi$EQE$$+GQgjiM9n^InZcJ zSC!&8;Hv*-__Be^NW~bk zK*wk{hN=L^`-ePg6M3!5y)W`FlQj-h@^J4S;8-o*y&G=P%s}xJg6iMRtH;L(Uhksy zyzJL+`{YBVhi&8ehM_hz0z0Xns)iUp&!Z#hjUS_C^LLAL6S!ky?up9sRPUK02YW;; z$_cgrVBS(Z(q5)Q;YQ}3@8QY)ku6Xh&oVE$)J2(BZx+!FL}b!0@0D5T27wUsXb*le zP=R(HSb3UZDl-G`hl3UB?Ba5)Ar`BbAvdZS)cGVNruoyi%8z;AOR;_eHx7Zgk>B}7e82)lvh^wvG>sU~f6v6fNY>3#Zs+ur zU0+$3xsFzS?^o4}B9k!)@td)f%?(+=o2N<><1F}^ulzRZAljW~~2OG)#kztT;LUp>$=cvW-~p>Fco9ahGb(`Z2X!H<}%`I(bG>Bo3-_w-Ye6 zy-THGPW}07P@ThAK%zpIW&NoJ;)VTy`WTJw0qD@*eZb}WB0olpTKqe;A0mn)7iyie zlj{6xg@#_T`Eux(iwam<$!vq&04=gJAj(?v$7?B>&@R^A5y0>Wh^qEnPH0QQoq8%Z zjRx9a#58Q#HmG4&wN6xIUoD$ms@kweelOhqr{mRVY30jG&`%P-QDG?N*Xp&Zj@Qmh z)E$QG>4Rf{{Vlw2Rnx;#e`GTWv;6Az=d~?%DXsSE6rPwUPbK?0p4k42y6LvKaA{)r zOHFg}Ulw0(aF4|80+*~|^BZ3tor{oxg&8laiM!vdJyfHX1va}W(8ewXR3j0st60Got3=Q(SJPmmJ!Qr7yUIjlNpLt)pjNp9R!qz>c*Iula@PN?9 zyGDHl9#7_VK8gx(_4wk6nR)Q6?t*uDb@o;&bKPzdJX>U-kJCyrQwprAszOoj#jOuK z#%O~YrwW;M>c)(^;!MK?4>Ak*sy5phkpLXYwtXR6AB#4}6&(1yhrKC3!!c{tcoH)6 zX=AYB&2*}^>$%!VIH?E&ReYMnDTbYjA4Wv{5?%i1RQh%atCm;P8ow*Y`=YCmuA-8Ej}xqA7g*t`MUz3=dZN zQnCFWDy^FKEEJ87I@9ab#Pzb<8Z0q6R?oUF`ko|Zqjf*q3f9Tk>6K>PSs@01axy&V zw$PEaj(v2eQI7NN)tVCwjcE%QI4>|ebyYo1tpRMa5V(f?&uWGg%lkiX9|Sl-1Sl^_ z-ZL3Jv#Epo>#Nh!nh9B7n+Lp{zDAeL+53|p62f-plZ3Ej+$T*4(Vz{W<>-o6uW-_D#)rl~TYxX|TgUh^d9clhgI3s}#3IlLG-^|ztZ(m+AFlr|SMdC5vxrg^K}h9BzdX0LDxyUK`;zF%GYq~<)7 z;jC<^aoWl)DCjO2iS3wk^`eVvLuH2ofQFfoO@P|4vv`ns4pvAL1Hq$fc*(06#$da@ zDyluvZN|AS6&=1g!)R7<(sR@oTTMzdtS+nI6ZH(#2(O|D_`|j8ZdKy6KGQ@x#pWO5 zj*}L&an@es)(ux7yzJHHoSyzUY>JI^{Lygbf*dv*46$T}*M|M%tUPgB7EAqT1#YsM z1Ys@tmrgF^H|X+Nget*r6<8U}+X{FD&jN~w-2!cJuW%c*1UuJ0>v9Q1M*uO|5|FS` zT75Qeah#SP)5^OW_FelKLGeU?7eh)H>@(qQhy{3QCd-H83ra}&d6F3>Ri0+4+o41K zuN)1o%)~tugDst=SZC2R_ff|Z2tsrTSOT=UW8Z7^Qu9X+mS=TCh)wlW;Hf#UuMoUl zq4D9+18799$JTpzrzPC%x_O90eFeqYA|p1sg!4}3k%S3UktGQP@y7f0v(Yz1Os;D%_XqgmY zhRwgX2N9pK>dd*g!#f|#7kNh<*cH64MbLU85C|r`fKNvKw>?NHt&|btXl;{ zCvsj{66Gz_IZ1b(bfa1e06w}11)l?L1b0xg{>d@w^}gJWYg$5Y^QR~)k^-?FKfb+~#EA2VK%a>>1kcOSc; z7Z$+vSpN0L9f0y|Ae-FxswekhqW}#`q>gHf;OI0eN%s>o&N$@nwb#+5WQ!1fmS9)( zHpohx=M(uSlfJ;(4EreG;-)gU%np#Ua2L3_H}7^{%FY1-qK`>+d}Eh7e|!uD5yUzH zcpw%6V$wY~%6FulrLOxnpsICa@#9|-Pi^#QHmOojW#vNqu}wW3hl$iyXMCC2V0Mq( z0rRUr)NiQTv_UbS;Pq%6H@3LiEr*Zyzhj@xL^$eZCf3V5_cyAX==-iudVs)LTxC# zg{91?XNQUhK^?RODntbDJjf%a@Hl!q>1)YUg!c-lV>mHWir9|of`lIvmc4_Hb^l~K zD6=Cub%96mOqu)uNe|Atg!+2RRp^qPg(wi{*?7D=7tfjydVc0Gj@3^9taMkI<+z!f zo0~iNkQNJ%B|(Hf1*n;6%v7&?=;9Htp_JJyaKO%Q&dX7;$B+0Rw|qzv_MJIiZglX3 zx$NEGx50$P;SU0Z^yiTr(0igrznqnE;9!Rp2ReUDF$Q%B>}l1*denH2-)0Zu+(H+8 zB|I5p^31FV_6c+hQWT*7@?;W&9kRBXJj0+)5a^w2>{bBE3knYfR*j}Eg}JeE!qB1w z`$2V3IQrr~DQPhzM`eE}NcQO*e*MW$wY$dxS=?i&uLrU7bl#x$uvrs8xU$85<{3mY z&gRm>o2s;-Rb?-$p|kRxDz}yF-;iCfExK0dsFPg9p3TGuw$t^I$>A>29r+b>i*CVJ zdva%(=xW}BXmdHl@6hL{Q2kICHyk1NwT52a}k%E&7 ziqfFcJceHmc!sZX;M&!PO}>7c-4!r%QH}tbMKqsul;er{?S3_^Hf4D*ag4s*CcXAtu1`c{@YiEd2dh$PRht;6Q1`^4Gz4Uk+I?U~N_J zjIj*fQ2R%7BKQiM!GTm^BFM4-yh?NtYqnr8Xy~C*1J&^LH?|Mg*xFgt9U z02FP`*{ToGus{Pczp(LOh(N>wOd0`WS5{*7R~AQc{I|WAl!^67OTEj z6z295br4a`@o2gLIC;y;IoR84k9W4}fGFq{CiQi+2Q+san3pv8v>Sh5$S=Ky5X+3` z^0=u+tqp(Byq-5HfG{f8P#loGxI#SB@qQVWqnaMLAFt?)PB#ZxfZzco=T*0|D#)sC z98RvMs^7`8`iv)a3!Mf};w2u;pO$%cdKpw6;TT;XPlaa%Jlz%u@iic{hls83)fxN! zR5v4U%k&X>75QZ`VT0oAeE#xBs}|#Oaf&&h+5$;X>8Mrx_=MY@HmV7h9UtnkqZWTs zee_9gz7i7#lQ1g#>)m6m(=Mf?rkA=EBt})2ZPGv;Y(&o6hhc?D^s|U~)o_h`^xP^# z9-%ZMhX+ITmZ-47VxnfGe0ptdZP%ibDnff`NI5vVV~%ugvoe$cYq1(HsAb%+!&{Dc zC&R_u^00=1V&g+~sBg+3;d1dQ+g-OA`|)p zVFXvJ2^o9@;WA!0cjzIQkE0~{_h+M!;Uc#;poDJVB{3y)9UuR7Y#8DoUAH4*(c+!x zsPS!JK?6ut{5El!_($%Z~k~GLzn8c6btq zI|fHaOo7z3_EbHOv|sGidd%rCQkS9wPu8ck<=)v=3FO%s@& z`9K7L%S^jb%nd4^Ft48V0^o`6J}1cR>3*N4R0To%-7P0-oE-XO zct(JH>s5@nSx=r?CfOD~7DgXw0k9~>q`(B&yBr)G++pD=tKtVa!wzQ8hY}VG^ON%i zX9hXU0^kR!3eC40E?u%_ual-_0)VJf2mwUQ$GQBSHA+FgU`of=@`%|#2Om%2dK99x zp7Wl(-w=0hii*^I70(k4%qQ1&{skc+NC%aj^xjl!LTDAQC$h-t@7sjEa5oFnx$o#z zat@iqY6>&YuT{+4UIE%W6nIo1w|#QBJ_`DiFei4qkjZ>fIz*ycBe}b!%oENm(=vuz zW2OwgtR@Kb#4$`SYm`{XG&I(%8Skx(w#P9qvYW^w&mqLUeDv;Sz60%s`oKXz8V{S! zkE42vp(IOC(ilj|NG_@!lVFjudx$YOc=CVnLd4x5VP(>LpklXGb< zhUyyta;VG^ui{_$JqHZ@jw7IN-@7-XtdNQ+BPVoRDzN#33MQ3y6z~z~a#k`6tdqS%`)A9wHx`iSas>Xt9ud`}F7 zvl}IgK>Ta~P`5JuW~_Qj^L@qs1U=A=%zaRMpDW=mkmE!Cmi{!^;FyeNiw(s}G#O(y zT0PO>0M(hk<+z${5dO<|tT^ROG?XaR{N3;|^CCk4Cs0_EE0LuxQ{mL+BLQl{vl;90 z!Q&yDr|>hF9f2Tu{*Hrg)8}N&vfK&H*vz4XIK5*Q^-{M?kZ-@*x!sa*7RUnHL>5I& zZR8h{NrdcC6g*^8_e23tUFP<+cv?ORR*1lF{oM{qR|p)TBy+t_?+`fIf$y^$(~ZK>{)UeN*WS4E=$_ zFP;s6x~F2X2DOM7Y)gUw9l(JSFS*841iox?$4T+D5E-mIy%qFbE9X+Fig4de^RbE^ zR29k!@y_n<0hcBSdbLk;9qa-FKr;|Zu;gmIAa4uO!BvCaz^N*;1~*@;P}I11S{XE~ zTyszUqkU!T5efFV(_=n=~)1Ti!ekCiDEFn^%i&>LqSc< z9zJL^b^ZGFBeWtebI{i36X?4dTU*jdH@-f|7MqSmRrFoM07mi?(J=VDud=k;BIgOS z>_q^UM2Z$$6D{$0HNj=A2o;*1wjQGbJj}P{6V5sC(pB>XF3wV0oLK7$1{w0~x2CJM zu^8LqfuaV4a*wr(i*yC@8;xa0wT2>Hmg_|#tiS8u{dh}ePp$ip!{jvOsNG@zF(H6LLY>dPnN%+hWGN^NtHm`o0DO*x`F zFy3*12w0Rj=Iep8UcnWdZq<^`5E$JcRXz9)$lI7kA_`5uPM`E&wSp9pS_ zk)`SQrGRhg9bPl%YuEV*aE%Hj4R7O>Sd-pUPe*bF40_LL`$P=h!Lso&g<@oYzF_cV z&<_E&EHeUwjB4F^GgzeSD@_x@8GB%LQ~aDrid9WgV>;met-;8B6RkdBHJF4W7YNNx zYxqBJ9u$N2rsV@bY|d>v|Eh0mF?n|y5SevfyyYI3mLYQ0Ew%5LTUt16@9tK$f70`C zh0pf6@VOfs)2TPry@y z5PDm;%R|54qG^Lb65`L*sC`s_hU(Wc7XFeEcnaXjTO-(0nibvXJGPCPeq3h}i5gCE zQ?@aHw*{4dcg^wtW+75GkXM3lgMHuer`28?vt!l8wFxQ<8IoqbBDul@c0c6-mMj`w z>M`EG0wK5oye0A^l!CwXb|0WJR{<Q5?Ynrf?fC}`n0sBSz0BI8zygvr*s{*~O; ziqcq2`$?hqEBBNfUXWcd z0sxwoiDGLvDiO41z{g7GvYOtg!2Q~u;*4;AikIaiOzV;7u?HZ)ff}gd7vurF8)__*Vp`81cX65hMay1x+Gr+164n3pp?{hhn82)AHjJXmW}>B8 zD!;291joonF6LLS6{EVJs$$c(IIY+2nIjSKboj^URfVxlfL!k*qoWmHNEU^Z^*kx{ z&C@4!E6d600Pf|;7=O#;6xBP;-qCSXhG%?M=Yh-Zqm$>J6;R2cDUG@|)_4+!7%{Fp z>lH3JsS5P3i>H2u*Nz{d7-dgBz-WlDs?s>fK-i1qpRCD5eYdT}y(_K3O>^MH0$Ll9YT1H{zZ0zIVjL zM;*r*t*lY+UW@AuK!|0XFZ_=A@o!s51b;Ble&Fm|58;_$gWLSt&unf2t$2~BPGbx5 z6J6`?J&-EGd)f9dY$>f5G#tMffCYSbq^A##wBLlDtf}^<#fo5lMwJMvp1LmZPZo*b3iP9@ z5f6bH$x+8XwPIm0C2V32A616oGUY?$y1Ka}=Yh~^1@27u6#UY{OfEO%TEMP-9|_@ z|Ei7W;btQzGdX@(OP;{Ul)-!)4eNDvsQn6wPli(`WwRcp>M_2Q?-)YLt;~6pq0d?R zh8@MM#vt+)@ftSI1@3Ss69)y$CwQDV=`^RPT$j2PLLpW640KC$0NuGzwSLbI@&ZqD zy+e(D9(SJ}h0$M4)s<;u!gzw^fNIdHW6HVPuzY+B-s{>%0a~$6Dn7p&E|{TmYw}>Q z+#D!l>)1^|9N+*<5&>YgCyMl`B#!sF%?nvwr?9CwcB0vwf5QEL!E@RraN!Q}kxM9z zav}!rF4jHkZOKdEdUCuw#>iuE#H%X+a$!@ zIE$Lz3%@}}F+lvQg-u7^#XXW#)Io|_(c^$AbU-z}`;^2kK4#C#HwbzL%d{od?|-S^ zkh~8$%IsYPB=@GpY;*+)tKRnCnXto(MS};2pX3fq)&t)pxNb@+O-sBDjMSH6pkG3v zgF+j!Phk1$s6=2CJaI~JB-O%B(Agw%tkDPi_%R>6n6X}P!mw7mAlpf&hxEDGPCBNK z=mPI2lF}}iy5PxYOM-EJ2&p}IMOH%pZOsz<7|lL9{X<6zT>zPL6^z1>Q+A0Qtrv9s zFb8U@-1j?l4j%iXMHOLEF~bg`5|6&u0i~acfL`kR1)%zg7dYPT+eOug?sco6Ex|M` zHwqX{?Z4j@U;Ub%v{?z;Typ!o zG(yxVK@}oaZP5k;CAKxX_wTRj7Cs?vz1}!2d{-h9KSo{>5}17yk*)k<;Nw>YkH_A3 zUc6|KF2s4gK~Yfr26>pT&Uv`KUDxTl;55gmkrI*nX6@DF(gJ~uRr$Kx98qeWte&|qT2xtg%v(5I7Q@Gs?M!uW!{sC;1M!GWo7Qd(8KUKTd%4fG2 zUX%geFZA9i__4rEGo=7gSh#WxuD{H?TCy9&o-smqqDup9QnK!vW?2rHQH;LW!)^@T zf6keWJ2gF=%4+!v?J+23|KA^ZXY}-AO(hU+)Gl104N+T+QbV^A$Ad(6K(ukB2f^>tvGxxnK z-glRGty#;p#xwtac76NX-)6NLlyydZy_xyisudpfG+_0bHbcajEEKEPSfud=86)#;}Fgad~%DRyRR-RxR|G z?6Q+%Jp)4rSN%BkT!tk0l$EoV6PLU{Wezy+yL!O@=#A_vf5>X03 zp*%`=yq|X2C0z7kP=dRO=#9YB1*ZBL=>SGkjzVFSu9HpOM5G9Nuq`;IP^;Wq06a1>o zRiwF(hXv9C4;MGHaCB^JO66i3eJKThS|n5O_%1@uA}ni!_{xNvHc6CG`T+}NI$X$7 zM^7)4SXPIYIf{mnQN-!a1BL(=%ePQy6`#$j4Cl4i9oFn$WUM5 zKlpxu=@-?g27M^=PWhq#R^7Kv>Ne%VpEk2Oqc{vwXpYC^$M{c%LMct3O}aVnXgJv1 z&U>OxQTVx3VilF|T`WpPfx!L0KMe>xc-U+K4mp00C*VHr7yT+7jjtrJV%)tIacH%K zIk4`arB2u0lHoymhzt_tX3@JLMlnu$#NjUP=%PA#XDb&wh65m5d6QDrB1~Gv-R5vC zWd6@F!3TE^qyrV&;@}ZobNSXCCaXBg+Xzva%K3v6 ze05#s>TxmnXM{D0m+Zg`D&lJm)|_<_9$lB@?9v8TuBU3yQZ~uUJ+OO zgF|8z*SNUFS8zePGs8Z9A!G`)5Zp_I9oAL^eS|8Eio8F73NiD{-6tuc4-#e`(6%mlEW*a~L)31=(m^1H^-Tnl=W%C^M7xM(&+VKSDb~ zwz;$QufNKQ8srl4=56HY$+kv7OUz-KYLotgmz* z0Rj~7_$TJ8)m-RLQCHCLMM^@WaYp16$WmFcEcDIK#E`=7&+_+PR?`R2elddzNT;-= z#@n%_^%hy^1Qg~k@lf2Loax@x!R&j5A&MbB5aV*Eb*BXIKNFM69 zO}TDLWJ!?fygWg~Y~z_s&od21wCo>7kXu1T8F{pj1S5AV`ybhd8Q{rXN&fz+ue*BS zombxqu9u>-$KpFPg~MNyKn|ifuN?GRF6AP~Nn2t{$Ew_-Y!8VkCQ4hC$8Wj9``wAB z!(iVd*(3t*7b9;=kNZ(^USe4xYu5peC5BfQnZm*Z$=jl+dC4t(y_~2%JkV0vWb1!(vCdDu}?!X zv&kw1p6IUjfe??z725e}3EEEK`raz_J#Lpl(}Obd^wKee6}GG$9jxE( zYu2NFvdzC_${oW$RP}myZTSYJ*2oS^fl=|}rT)gGKRyWumYstWp8loDw-3;DLVDvT zJrR}Cd!XHwUA$anmi6-?TfrBfF;fl#*;|MX^2$VcH3DYXnK5}6hPJ68*$$8-75FEs ztRt~v+Z?2Fg3|!ma;5mfGe|H%t}2ToR~j}>D$vjLYmO>?8yz9|!Z>q_OFcwV$kW$?C~G>Gy55e##1$*VYf*2hQzg zsuw)h;zlG65iYfKqy@9z#SEq&It?vq|f# zZzYHQ+LvmNbxWOgN`)(3L@qc#bO-INra7%!1ZdV{R2rn!+Pd*N9-c>FcLZ(+5tiCj zl~vbTb1&XsdM9IGyj8G65-ZdTRziOjWjDj_>|o1ndqAoy7bSsh5)KDEm$Pp_wV;ZJ zh;|58<^rbXkGqYW7Te=g0iQsFW&$cV&SWN^|D@J+?V`t}dD@2Svpxo%?|g8T-SyP$ zCD3B-IIPq~<1dhTofU>X-*0@MejGr*tnFdnUZSg+k?tMGn8vQ&$NoYtxi`{lw$vT= zS7ylwalVAM(8oE!)ftp|oZ(zg&|Jx1`4Hjv1~5tD$Zqn`Chk3oVI04TLE++FhU9?F z3mrBy%;`vraItqt)xVLM=9Kj_a%eR{_#ZYL4yHG>K za<~ucMK@3r-Y4~oA9(%(Hq}OohotP_jNjqJY3t5!K?_E!DH{(_R&+-xOgh1yXjc6S zzg&&KCosrMeT&k1-ukbnOpL^CbAj?6&LlXpb}*(8Qgq+9!t@869q)Y?8QDTWz<7te zT{C_awBC!P<9cPCgg%pQl)QW5Y=bvmIZ+xUh?(^brI026(>Uc|lasnPr)jNTSqC_S zg|j3ol9$p7_sxs971d6ib}SqJdbS|*t9uNp zcwX#+3vXZFEEN`m8#4T;&uW3;3nc2v5DNoA0+%}fK|<=qa);&Cq>R- zHKrG+TAjxDSgD_@z*s2vBWH>SwYVbh ziVlBozUowR`JvEUz^3jon^^q~eQ)KWLi)o9BD3Wtp-kGOx=gRwB=*oO?Z0!k ztJr4{-dR%k0vTa+4IM{!lgRb`1@rp?h?&z7^Rp-Z&Ml8PyI+T0zv=ou&yP-0Ap-u- zhp0R7$b#y;t25CU~J%nG@T8$yS&tv4%tN0W!F$=DrrmHf6B~kMGG)W zRSo1J`~ldrnCVnbKHjb#PXYq+hadgM40e)P5JZCxzLSl{v3i8&j4Uc?>f5N})m&mu zLiU_{7oS>7(VyF<3|{^{Or3f0AQe<9003cJU}+Sx$}cP1@EbN+*B%N>b08B~aFFa2#lm9cKx;Iy##ieY>5iro`|F zt^dBvS&%csPj7vt^ZmNlK(C~q#-S0gtH+n#Y;SZSf$ z(b3m8KgB)lsbv}mnmJ1$$wGv&&%a<>Lf`V4ZH2_-Cntl0cYRZW#zS(R^r){Srl@wR! z%E+T1Y7v^^IiGKT$W~jv6~%v-8m`{r`j(jU5u1Lu3Z_IVjzQ3ubMqVhpOY#9qC*`_ z0h*0%1p9-husB|2&0+uMbh|IkBK)ZWv=A$BVc6jQ+|+3ycu5o~qbnmRolwMwvR`~T zx7}c46(~1kO|$+Mh&z4DNR7K|M79Lf8~=qpzA0JQ+bfv?#vxF~h>iB2t*itMxD@J0 z(W^Ok`>?Fs0JKnl$Ue!m8S=r$P}ahN{ZA z{z`hI`KjRo=U>M^g52Hc7LWTe(Hhhw3HZaF%b%F72R6Z)mW)sbE35XXJ8wbHOkG1b zE^@KF7ae+w0$%^s1;V<^?>QvXnbWtfrTPZFFW{Lv{*A3aoBQ0t#+7G529*kEs%D_= z37IFiLX3O7i!Xos)*LKQGpDJ?msr9C4XEQqQMCtC8cuV1Nv2G^Ec)!x>!_(7Ui`va zn!W0CdR_p(ScJgexA+0PQk_2lY#d;_65Es3y36hlWN3C{e{S4(4z|=;;}up_2}q1~ z1f{}|G99ld6)Io$Wd_UKHLl}aPV$ypikDW1@OY)3yTkscxI$KQZtf0)X!OPnaF1zxz<<>MR~znNLC7+(+RQu`;~|8Z@E~Ps!G0fMsPfX@_fjI( z(h0gvCD{|Id@D!QG+q-Pf?%))@lSqekqQOsEimV0aS4f!3CE*Co7Q4hPeFU!xubLp zRJT+r2>Mc?AoP4jy3NkY#k0dT*NPBAM#{L(A=YF%U$bVBrcFQZbF)on_~+tcR}~SH zk>%D2j^Dr(GgA|1PC2-3!(b*jNF1J(*O8iLchXEamKwMiDq>78`LA1FR#lC!^QLnD z-gxc&Wj`iB_V%hQ6yw<8J7G`~PgF8m72@LJUKPie9$?QV_e_vlQ(%Y}Mv4B^)vbI% zS8Lh&vU;3~cpAm2i~Ik<3jY}FV-aM^qi&H8TQS__i9jMw-OO(ef9oi)AcUNUCZ6e* zYjd-HsR1+K3yaWaH1Sa3{Zh&G)8&_&O}e-Ay7wyQugb3ieHHFSW+jHc6XE?*MI#S< z;=yd@Uq;3D3?kw@j(E7gmIRhU+;+#wI0rGSl6 zJH5HhQBJAQPl5J6rr3?pS&LDXq)R*A%J%!x8f(=?b{?HUp-7B71<3;mLFVcU?7W;% zz#ZGcfskKbmn7%5Ontww1I?=J4BhRkUdf?T3s>Aem83@Um3`{TGLr%E#a+w9EN$iO z<7hkdCjJQp>G{h--~yp$wMz+I-#{v#X5AIv^XXIKDcI|BqsMd3-GI?=7PqLZ%%{^z zv@e+r4jV1DSWL3vU>1vnDZrag8Q-EgBH2?JQ=FMq+X~q#1ntnMySj!?DTCrZ;XDBCG`x$1q!bMXcyRqd*<8QZz0i?6D}H*#nb#k<)$E9U zU`P#cShp5>N~a$yb+D!uu6S-)xD*W*cuJw5b#!7gRp(OeyJ5yk-4@W)*l1YZdhFP^ zR^5l5YRBJ5XF?#!U^*cNmXEdCps3j4M@{oKftT*(5o+tT^YD94URw@|7!3r(HiC8I3vqhSN>nB%= zEQC^%O7{hbwEq8hrsI^>r}3FOTG3DEO+vRwieoI^<_s(u-ZASO>terd`jN1>j3`gX zgd6QC`24p<o_tnSO!5U0>Rn_8wN(($&?iLpN;3a^hGPin~`w zY6AQD!Oa@6g(Bjwi`b&NHQ^OFv_YaInDzL!$aCd9eCsOT_>V_d&aF5R!$VQIp2Z>A zJol{y%u%E`M6SM=$89Uj&yPT93S5?MwZJRnmt)F9>rvTK^q26re?|7`K1b=BWnc9( z-GnU={^hPn$;W}~YzCuPp0gO-;Sx$Ka@SFeZ;P7&k7oWHNP>gcd)ZeA*cMk(!y-;A=H>$@_33@Yw&2uHZxd zy;-nuOF-Fkc^L5agB9MgbVyc~Er6Vceb%G#{l*Sm$-9?&%P)0>h$Phbw;O{B1i6w1 z(|L1~o~6CLV6wJ{892B9&@b3Z^3$O(J!xO=X-{r9$cW+?!oI}L{pw2yecQ_^Z7@_c zo2)K6kW6a#p)4rYvnTtdEvn}~G zuDRA5yL`b`KJ@hj`VoC>@8A3Ux|S0Xw{QC&eEF(6*1d18><6?AefosK6bTKSWv5ee zp_^qB7Hd{laZ z2g}VDex%F4ZT}>m9KGFt2_5?J$F_FYP11qRDyutp(z5?|u^qmWt zZFlQlUWftUuXcn^fPpCMEBNN4=6RB3+tQ~qjcRwRi%JjQ0fG6iA8OjmXnc>a13}6> zKLswj_V~}$#AOM?gRrm2_4F!{;C|p*0_lPo{`sfoB>wq(ov6k_B z0_&Gn$S_d=F>$fWYhD)@9d>S^O3!w&h)Y|hC6EW*1|96)l8RO?@cBk--xWNY%Ti!T zDMw|s*@6_zm0D~YB)5O=tE~*yRRaeN{yHx&k0ScEjkEI_an>UfK7E3y z#@iAw1oVI)sGt{$QVrnYP4d@$z&Ml|)YQyHoqrMhpRYkpRo`XX=h2}3S7<@5u;NtN ze{Bt-Na%{=H5HT$PgcZtTy;4{-Ui;E%m^09Su10}lW)R^LDr=RZSp3_H8I<(^XHreN?dZOZZw8JlsV z5!7NHej3d9ZQPlu&Vt-+N|@-r+3Y`VUg=pWv>E8>6%=E4)usUXhAqE-;)+j8QDG3i z2ymzbJ!;#LA>Oo))N}Z$qjzcH_e?4Il$W#$5|UEB_y3ju%ZQPS-guEwFv3Xzb8vJ_ z29jABbzZ|AkDD?wL+_nXAIVAgIQ)LGYjA%xPUTb=Y-eiV(&6b<=16^Rp)oYAd4I`t}#>xp9Pej!;ZDhF$#w3xR0qS`keZz)~2tGCVsS2XUc2$ zlbzhA znf~;fkW&0EeRo5ch0QVb3a#U${oriJPPf>foA#IrnXj6rB9iq4%w)>4C#jxj-DQ08 zMy@KK6k?{PmVK3y@b8~5V_26*ealJ^-P`4x$3s3yF1gb%sXbWD@#NKlO+teIl`1?^rQy{`N);8FBX& zHDXR%E30{--zSZp;a`W8aljrD#r~zcJBnV+ z^L9L=6c6D9 zk~w~vVs_`Ya8(R%(Ef9WWFh}#ja>pLU2~Jzu*2tCU<=WWrDhoZsgdo=AoP$wRKa73 zSH^Cw+&!yb72YzR(eOEvj=qufZsxgv`7Ve%6`W*NZ4*6>RYS4SH#ZOIKRDumWs~8W z8}P2Q7xHr{>77ukqn&)>C!54vo+Tvp{SXOCM+E8~?+y(6_qo$P=ubQULcUB4rM`(J z#X^7Wg>iSUL`*&f^!VNU-XSE_yT@c+%lD>Ju5Xb&aCuc~Ivl&Xv|a;~CVm#@KM#St z0RtZI`HGXGV&v=Bud|bE^{-x53t`n^@W6iCCPB;m#o|{7;Y!je+Jjd)gDI1;MG&5i z)0D~4-_&juxSK}dJ|4cyvh>FrYv@7r%1y3SP5@yQDc0{j-#iOeqfnZLrWAF&B5M_w zbBPpaDA5*6-#^dY(AA=eE_l%yKlQ`Y>^jG*>ZRs$gP0A3W=}lX3zYU@oRZw?xU>zR z?e4?F29af*zRDMLs8=CJX(35D624Eno;JW>*;%1|3xewAQbsvpTzQ}ESEre-4*Jgd znm-RhYcNMhE5I@T0i{&ZFkq6=8Y|YCVd2rd@Sur3L#_Ag*_W@%=H})re4#hIh>i;9br{Z7?F zY$56(!A-%=OHzrNy57aJqX_+k3ch7lCdjqe#V(wjoLRSeV%6J5{k=dSj-WN=%y)u(VfK)Mm zawru{6HP`tSSWPgXRc9DVnfF&>bxE0^}D~ZuY4k6%@!~iYOGDu)Vg!s z;0OLRpW?nXS4Bxe$)7a;Dkom$STG32XqoWa4OmDmy=&uujyTE}h(ma2hwA2Z-B@SV z7n}QTbc;=6CDP(@qKO%ALv1f~j3%7w70HK*{CX2+ZtE2R2cP>-bB8Ov{z`h?XCMVq z0si|FcHaQ9QlLogOam4llUpm=eT14jcXf8!fXbJ`}i#S_`!4KPoBZ1VT5IZ9E19jyV9XD{6($PSY~L?_O3!GB{2d1K?Cp>-tbMx zNb+57G?720(QCo3!3(1lM!J`o)KnHI{N{ExItG_gQaJw&oax$-#LEYJ6(L^?P{P8( z8H@^GBI8EtRFpr0nzWnvY)Z|BhDn!4#pA^rtBe zm2Ip8VpTQX^ZylB5Xd0}|2j(3);8a80t<(Qvl<8h9@FsJ2t{0Jetm82;EOg?+leZR z1=unbDEL${BB=&HkBIl$_VJfwY9%FUZMYkzx9eFd zdDx&;KiC9K^?q*+Bvh>SZgK_ZBq-0y!48{oYabpcbEGBtiF3@WwJ3Qlm=q+CkCG2d zE0D}}UcA$`e(vT%Ahxe{W!fM%7hVa9!sO1`||BGtLST(d<#Hq1_fTJ983VbYpGgOf!Nw}`AD@-yBPz4 z_xo|bq3%b9c5MixdW#0|rKFgMbGJS-NB00!VC6+Z3{EfIu9DSNhqb+bIvZ5#DNJnt zUJwNC2~>s$OQv_G_y)8>;%e}?Vi0T$QUN-fq8VzFH*tS9mO>#xmRl!y$vp+S{4o*% z>wIZbY=N`j68+^WU|=iS*qIIo6)2v<&^2q8_2CgmL{kP)bzX=`x z?uD-k2zvyL;OXaARSh0Kc=zL^xTHK8h#FuFZpvuPBakZyoTT>)BV2iv1LK3T{F!S^ zw-~xIih`)K2r5*Q-^9!9dvkYXWL`*+|M#0g9qbM^8j&$R#@3pkSj{gl@9&U|LIP5? z_@y05z^5!&U7mh8*Tm)?4AHQ>tfTrSP?~#1ZT?hR=Y?@2x8%H!KYPB_pn-U?u0keT zm+W8P%>M&h40m>QsoP@B6A~s1i+qP?EWRsH3}WUb&b!nlZwxsxun#nNuDQl1XOP$J z_9j~L*C^w&TZEX7spA}(@@)2ruLs&yplkj(5(KhQ76+yhicbfutif851 zxR0~*2b<5j`ucPyjeusa2fvL~!ckBoijYewc=w9UeG39Wdq^=J1yc-IXssf&+Hjcj zai4P}4tGRrW$?AlJWXx>#n?W#tX7Gsl5bn*7LQ4SU^zO$uN8A&fYO(hUh4Ph4hMnU z^|92{D*&yEH=*M+4ZuFzO-IzUpKNs_da={1bF^&ten9XIPu&JI<$;FPrVWbzUCz#h zuCo6aIw-X&pZ-{17pA-ePq0uw04tTj-V{j|d+0+@cdI^p*c7L_P~I08TEri7Xe;-S zI{$kpd(PO#lW-%UmtBILsr?J}>jSmZe3a#XrI`g07{D&KpJ!;`<}t2Q1E4x9mL|Uo zIxzrr1dBza)+~j1Up8KXZ$3Pw*o3I%jym&|RLRK!bkKTE28$0_QlhIPO8ulJ!%~CL z^be)QJpNc^qb%s>JtOA<+Ehv!cfx1xz|k6}pw0lanN&L023gRM{7!JE`~@2{ohT0~ z__LE;Y<`VV0mo==NRajHo+A%0z@Qhe!+NtJ=ix9h^o|8!`y?x|sWmSgoO}Zm&`SoM zig}u>m%jj%0IwdOlp>uJ&$ZTl=jgP+k!0jfm5z>MSnY7t55oyU2y2wZvgPCM&e~hr zO4ED?r5_>)|0Fjci4s)?j-;@zZgiI`rfZ>K2zqP&7T}3kL#eQEY4Su7pS8vJ*!{>PIi`R;t5=!2PZKx#Kbf8oR7hA@StLtXZ`xfyRCHG{ zWLNQ(6(idRf62NClYCrcO}ZwYS;f7D<R_swRcSQj8d z?$5ox@3Zl<<7!*#F!0N1ia$-=LhaDK3J5bBes?viJ9Do=QfU9__TtW+f8QN0MLIs- zinkF~6~R?8fst}oC}0*zsAfffTqnn41h5Vd)AWdytwgWHQ1ge>_<^zj!4>T4xqu?) zis7&^!OzpYX79%FpYT1wj_pY&ErU%NeWDC}KDHVwX#XcLuqwuPzaVpYqiWrm9We|$Y>|Z>w+m*|Cfa+Buk>kN z!1rMsq_z_kW~-dIC?0&ro>hg^Unn+47iU0!*xu+|+zlydD>okB?mMmx~ z*!qnhnY)(>Zw2qe8m3o~45JE_0bvZLzh`8xz0AqU{|+LJc0^>40Cwc!?a4iff($fT^sPX-hMFt);%n#4C$XJJSN3V`rBFuV&AS{u z%wi)O!#=LIg{N%`p7!o<^}dUa#bNOE@lE?G_iboZb1%nnghzY6UjljTpQmq80QFlF z{{(n)oM34hmdlvw%sD>5Y~L=G`ECZd{%RmD(nLCNmU<1>rVyV6ccmdXX_tvQF=^qQ z;Io6dM~UH%^6@X{_a3rkbt~`QHbISO7CFR5U0>5K5ue&MQ}mpF`S-sD!!9h_1}hAD z!CtKEV9=LC^C#HMmT17k(G|~dX;-AUs3>E_Vav}eTV~w++RhIuWz~0hVl%!YgkxKfa1VCg+)a1@<@m@RfMc{YULp3=)QI5GH&yFx3? zl@WN%DsA}2O}`lWXpljQ@8rZ+b!=H6I=>4DoGRo$91^|p?=3-!DPvNJSRqP|(gS?QeZcSEcz%bYKW2N5cs=5(bi?mNMf2-vecrhh^c#VOn#? zLGtrjIlmYEE&JhdCX8uBiefb4LTniZ3Yo8yxrVgZrORZB-6h;iQ^tCh&P(E1HoSXM zdyh>*&h|Q8>F~%2VIMBz?w%IuF@KAR`0>)MP?a=yoV%s0gq@-^HT0&a_Hg>>f}aUv z^dW2V2lm>u!St7&C+tt-l^=d#aVT|84K?~Krb(Qor!-CT@9)8ZjjLg<05(A%yL9e= z%{cS0tCgIw-d&Fxf0yZ?j7tF8sCEx(3jgu0ptmsOcW9N57ta;Swr1RleF3fOZa__K zJ-&DF;};{T{3|}z*Jxhd$+5o2%lW5}n3crS`)~tvBHgvqxn|m&=-Di&kk5$Zxxbt8 z?Tvga(BJ$y!yuSUU;V@VMzuLq&T`#O=UauAQx}S!PK-<9Vg1UPWwh<*`++a`S$Ae- z{|W^ruyH-h&x3=^pq997^Tk$>#<~mV;iGv0fXe`cj4L=gLD|9xCRE{=e9Y^UOs&$5 z`6XK>TAYUetj1>UTs_;`^hARJ4n}Z38SaZOYOuqse*2S+`?9`V6{9S%{c~<(? z^EeqY`^+iS7I^Io6UK_O;S_jYqfy%}kuo*u)UvKF46gbNaq3vmi^M088un!H;5>i- z8fq4k_<3Z~hcX~TWy?^bZCeK82;|e7lZ!GdCFWTNNBuzTPYx(Ka+SZqcc~{W>y*K!f!4 zjMui<-1Hy2>J5{T&b_(y5%e6`858<`erd7=IO8@L4_4OJPVq3h%CU8}5%P6cj4L<} zJmWV|OsC7oQ=jUc?)zc_)4V4mndO=fX?P1wMNK2qYtJwbt}l{A>yJ_Qb=a8yW~?oa zRk2&3u-2sdd{9a9H@J@3B~RaX)5lK9Edm;kNQTd{>FeOL09khO4wlnysUwjf)w!oi zvubz^Gqh>K_Iv6%4b6Ns;(|OuO($`--5JIbKPXE-e)CW*5d=)8+o9;mydAgL1n z@}*3yxLuA$f9k(46hjW7hz<_~?cB#8W2N!12RgFVfbQks;2+h+kt=-Gs*ipxp39ZlWeJ{)8WHr^K z(Q7NTv#Ss)b52ftSAr&{J3DJL_#AQiT^caWzP1MUunB6tE!5+pr(qd>a9_N2V+ou|#_v)7}1 ziTlT-#t9PI5?#lVl#H#g zKa3J@n2#Elcx_ojQ%76d>+W5o%Ji2_Hc*@{%bqLQ#7qoHpchxWPE%A|aoIe7Ooz2miqB|-GHl^5Hf@6vUX_8`r`+xz zdr!T{7!!Dz$7s636NNQ*9a8=5xSvtVsbF^AC%?~8GbCn4)E`@i+jZ^#V1C`zo zCUjT5c{gP8W%bZZ;e7{!0AGrWwupp*t1v*Ji7ho=S>85G4>6*`ron6JHm<|=JVdb0 zIycV>1(38qM$Dv+NnN%Di3JBnCe*zS8H!7(R9Uisz-@vZat94HywoHhBgrs6NM$S& zcN{cR@KSu;@%6F(39YVu-nT$5PQb_B7w+vg2vU=)m4%H$*r;JK`X_XiCu*~=AZ`P?WUlU?(!aVTmAeDu+yJFUN=DoBIh{+`Z^xH zEP-5Y1S6VJUjX?|YGrj*(97A`Stz#=2ut3LmW3lNBw7Kj){DNB#)IC3-9kP-C;^S` z1~~=u$-&0WcJdC+s5?NJRd}jusCQHqqch5rRj(3HLZ8eUbn@*GW9i^#z_acXyC$_s zyumPLgt^53bxw_b$rgL#w#@CQT&L_;>xq|QEH#ux%y1b@L8{0 zadIUu*W;LqmE>E`vk`yUg&;2UDqIn6ob57uGV+Lo&Y23Wakc|E1l?F>miAtF?~B$} zX)({msAx^;y`p;=DtSFSlEJ49r$DOf%6QC809e;LPZgY5lG4M+BkSwyv&ImYrGN}p zXFwp9I7JJa*8pNABcdGuPDe}z)sL#-N64*DHx5fa3Sx?EfzXy@+FCAH(a`@$ znMvXk`~3Facino2_a7TeX4i6_NImSzoaRed?5s)t=p^}6RRyqg+73e@EzHq?#L%ip z9#X4wRPwo>kIgLY4IlLu)JL(@uV(3kd&QsIeU}qtL$3jK%dpmQqzL`G2*KAY0Vp}}nJUY^C(s(q!_ID*l$iBh-;W#c|TTMaV)De%@hQQ2cM3*C$^aQhnz-fC2 zwqa`!2tW<}u|iz1Y6NZD3$l!0fLTk}lH)o-Ar{!W4D5?dCg{*Mc{zLn<2@fqJpAesDN9zbrqZPR8i$n$5|#Mr z@CrK`>gUgE=D*dF<6=D8^&0(Z?@;=v=DbG-IvL>o{WEqY6k03Tci*chLQa;HLEHw9 za-2TnDU_p`BEi>))$ z<{ipZ65#7nDuULXw9L(~ch|{xG70W%N)v>0=y1&a0upyZl*%tafF;JL5?B{ z0@)8@yyZ++Nc-`<6UvF@UqnBh5zqXvoeF{%J@U>jU_W5)VVkdHhj5|a@pjr<=kMW% z1W#nJxMH*{;1~+$TkM7arHBHuL7+p6;MObUCsZum1-pg2LD9)+z-q{qq<5U)QJ+Az*G4|V-D{Y10qjt`-*e)cp>#)9ck9VTMBahrGQ8*Iaz+KTiq$98X z^hR@Yk||+qG>{;k0%)5>^-l0+7*jXc7!9HEtPGIpWW?k0^J0~Gm!`dTko#*T7kvh4 zek(A$6^BE__!T;9@o^LAW*lr|RSqELm6Z#XxuNDI=vAsQ(2bhOX{r95e_^Py;qLUF z*H>D&?BnapH|k+!&-xrjG(|E}9`B3A?4zU2@8rA78#q$w(49!TS^PBA9}kJxg8un2 z-(;RW4ku%K#ymmtf}H*`kFoOy@_sURIZ==Kj>oU?Q+=ZD!G9QB_(|w)Epv;lL&RNE zW}X3ZRrHz##*D5CzW)|W=R_Fh*8gM{A|WQGZjw+%SI=;_b_vY>49ybWG;Hqom;)8& z>&oZaZzAJ+$`_0mH=~291AZAHO-&ooJ@e$wKeQSj6LShV@XuT~&FjLqB40bk#QqI4 z%^v(SlzxV@HmbrNCFk}W=)3B;w(RLOuGdYW?Q4MYvh|I~F)$GQp}yzBK5SXc8(Z`y z!8XQ5M|*$&+W1Y&yn>apAP-LDjeJyHX(bc-NQSU&`0d%$+l(N2C6q?E1<|{C)Xbf6 z!s=~z`uGf1jS-z|v7^@?2iNBRkod%TMo93I_cKSU#u1c_k}LdJF4gZPjh_R_i+P%8 z5_UF$Y$2(5$>4#aA3opvNfU0L=NtLzI51nl_!T-+D%;qv5wnIWxXWQezxi%uCl(US z;R_7vT+o#(Ohjd&gb@VHdNg!&H*(|NTXj<6R;bUAD!ai;IASHzFF7|Q84%V-fWiC2 z*ZziG?Nad<6rsDC)db%!CdDkwL!b59TUx)OXAniOR>@902&|c|-o7+nEc)|{6-)?W|qV~Zk;*R6|M#mXLUUt>Ov_k{1o9t zDmTPZdmr;){AK_HGYBZ1TU&2x$>=Zcpj;T%)(~tteVuqyR|qXCL#_r(k4zs;Ho`DZ z%Ubj9W~K?Njy-T8I61Z~aws8pdlt~YCn$as7Nl>wr1qATGC@Q0vAV;1O(8%vSaDn2 z5?J`jO5=IqPtH}%68iMEJ3{6?Nk`rUWBLUvI5AZCeHES+A@Pr(otMjR;Wf-Nn2)rF zlKa;0uZ|N`5$_9+JV?o*;HiR2#ar~m%_vn?bH_)PTF(W8l_^lw$_w=b@e#q*^i%-d z_f$1DZQ^k?fE?vQ3o)0cFu*zLNf6dd?O zdh`d~#(%I~YHW)BJcPV@p#RF_4{W{=q3%ri3PhCrdzLHJ>+m(^JeOjm8klr0OsytB z35*LPVo60{<2S3n0@$d3yd3oB*65}v$v(j3E(I{QG0tNvPnqsaM~j*6*FXnSY)>S- z_}8yr+7IBbW@f{7I6`-5i(x(Uaj`E$LUM^2^_=k59m36|acfJtyKp%aM70EXg#qXK zDtm$BF|A;(bmf z`z%K=?T?~Mo+!mVxDX`@l(rS%@+M)2q$hIda4wE0l_=FphGQzjcA5?ror`TxaD!67 z$mf+tKnh2*hSmP=IS?g1?v>L4*LnBLM(|ib@5x0r9Nd@Bi^fBJAi+M!45=?lI^BY{ z19Amt8rU$tN3n4o1sdBTL1-gAA=QI;GqKi2RmbmRQRdt(i0sEd+#B!SpUGFyPQFLg z@%v%RYm+4RFMp20=1z+nIClN#k=JBa%_hW(BzfVJFI3q9ZU#@0od!Y)D$EAUhInKq zmrIXW3qmU|B-w#!AC3aC2W#p6@vUWg?er#8v++1f1Z#9B?RK+umG?i=dJ%VA^JA*} zy6gn8Bj^wxXWXFV+@Q2Z#mZMj+?uA+kVAshE!DKE zuKSydzH*SQyx~Q*%5N9_&K#9*BC5P~SddD}Y)!$tG~dnx9vlA+tpCl-aLE0t37>r? z2bG0NR>qbDCDuM#Ypk)IVpb(Xk>K$<$!yBrFoZ}!{VLsbxR@>4ANVm`z$Gt^f_oe4 zk&fCiU%@UXDBxP*Hb061oZ{c5<{AQS22+$5HPUaaZHw~7$wUyb7IGLhSIIw$j|1Rt zq8<&2vaS5M4z20r48y6z8EyxV?{Z!NVin;Ws8^;Tq-Hd){$N`&cwWgkRuP?@3d1#f zfYD8E&?Flw2Bc_KredT^qOQT^!L;gI0lk32#P%g@+0Xj38MJs`nXn+lnO%=NIHfJ; zFN-S5gtrHq4?ex^M;3n(tG*`IkDbG}VvSOKdpuIH2*?CiN)ouSm4?-;8un&O#`jyK zcPZk$(pub?qTZB)vQH* zy<5rgwh>F2k&o`A*Y~w^$iRXF_Q9bIYEF=p@hBAE(6v%66mDbKA}b`v!Pr3$5rnEC z@MP`f+?Wl->7@cbkN3|>1!9x(`yucc1YG#$O}Y>lo8dWu;<&rnqL$Kj6FY`hgFt$V z1FJ@X4vj#pgH-vJ9Rkt@#hn2PbazcJRbu4fogOO?FnS{76rpKt7v}A@kpxldw|>vml)l+IYuxQvdd>EIRDdE$9gl&Zu1Py7kM$S~;?qWtT$^ zMQ>FW$r#YhhG7my#fj{P-Iih-uw+z9eJ4ZwDE$#jVw-^V?{E4|wf=>Ij%FMZwY~OT zRUfNwpU@sj;mYRbc`_&n4Vfvt>yhob;%Ve@nMeP!$<6cdmW>_Jgto)LAbkg7Io)~T zQHD%5k{xAw!5qT;217IlI7GrXbuK(tfwq(3rC}H4=}I|Aed z+_)lUN8)gx?VobbCUR*uT%#1=-f((ZViYB$gBhX-%^WQ^b*hukr!ad6^t{y+7-`=E z=Iql|b;wuO5{ja9{O@GXiEcU}MBlZF@#Wwg9=C%+SG@`C9oPFQ$&yBtA2CvvR9!B( zP0*Yi#m+93niMEQIcc%-rKfVyZjKslPR_(;;kh>ikdnczG9)pVqo8qoZw1;NRKt9b&V(LNEkG1&q{s+#j!4L4H%LFscr}$!xQoEnB zPT9WFqxo22T3F}(#Mamr!p!>;HCRZbu=OGRGr^CJrRwWRduw))8|MvUV0>7v_ReSQ zG9;VlmJ{6#S(5M%;|w~81rl8utQ;NMBHR&S(}$IU^8~vBsbvZ0$*Y$w1PGC^3$kCZ zv^hcSz^Bx(AEg!9?eh5ecx`oK%bm~|W@j=F;tQxDAic}oJudGaQOs(y|6Ck;LB#jH zbup$I^knPOLa#5T?AKmon`C&Bl=p6^zTSLQR|q9+Dn@Ej!A#k| zLmLj?c*o1;pk&Q(RPf592Eo2LR9#{;30@u`sFui6~%1 z2qx)ylHjnw-tqddGLT~Enp@if<9Ua4YvF~PuU8%J{kj@>r1S~nTvdfJ!wq-ER}&5P z9YW4P(=t=ru4xSGSYJSt;6E;+?T~1$qtIe*@eWk$O{Mt#vk-`$Fluig%8#R!eRJ|( zxc*5~a(}4V$@U;?5A-M^Z*r7qT|lFOLPf+>Ea`A<@9H-yPCo{ZW!C(&n zPeACOL-FbGHsfT3oC0;8KMM!12pSLrtLn8^PHeI5=xz3}NmO88Ecr}Qaj@E7Yzilp z#`V+MRUJJyvxtFGu2p?0uq`;T>s22gl8CVXOC!>ABNui>yS)cTMWu>)2#}g=Y52TR z7dD&sX)?g#!H3q1lpi51sIC>EU%_Q7n{%OYyCSlQb$|qLfKr#qQw$ zjq=BTLqg2SvcSOLrn;GDu9}p6mlu}E0AQvR7pEFDDlOFy4<4Nj`D3^4Tz@edipj2% z&km*=yCn}?8lX*;2S{a20KuwO=%vEHLvMhj3s4SlL5NwGr)#ej17pI6f*mcHA!G-CN$>ylEjN35z%;sb(-qQF}>SFj6Qi4$;rtgA7i9-k}$O<S0KMVWT?N54V48`l-Rfrdn^9T|f})UK&E>1lwMm$)n2c8kwceU;++Nny z(KvXnT1zI!|IpIue-5Y_?Q}&p%QBrw|HFu74(tdO#6QV?KX_9dhZHdEvcjekg|r%HOE3bY&Q zj}W@Xt;qj;Bn3FzI}JqiQ=nSW09b>liXQHY4JO}jQo2X}tJ}@^YF!FR&)(T`Y&qn>;5XTov-SCnq?H17OZo zam!8c8sl&j()_=|2Vs~7iPu%XJzasMfFbkRA|y$Si1nYVOH0G-!Z6elQ2xGS0fecI z5N%L&bg?A|)qC$-v+kvan$jvqn_$cgVMA$z4L~X(tq;jS`FAfOVO&3zqQ{Mx7u!zF z(?OKI0=um{ICCr1EUpTlz({pgiIXtpY=jX%Urg>R{Sqpvv}GKj&-3G^Em;{70`2d? z{Xd@BaH^jjscJEc4&Xmh5QE6HLWkNl_j@Pjjtf&yI6l5sU$fS`|F z0YEo1iooTvZ?JupY*27pmWe@n-3A~N#kHQzbu-LLHZ7REGmVNNTY-B&7n>U}wYHJ9 zO=Ocbbou_n2cva|X;3F1H(UbD2-?p2iJY##8^t>!EFh+6RK8_;Vq_pySjuXCOEx?x ze@P*oU!b#8VH@8qCN>4Q3fx|yN*&=|^a7kl#lj!L`(R+Wzg_Do9aU&*tj9#ChfB?? z_+R0(d7v;OUObScJ+R{Cf7r1 z+OLPT^HKH*z%D=ycXoDm2#_t^12IUkKq$HjU`Ra9%Gn_SQ!7cPv=d)gfU2^lT=d}W zRP$>#Fpf~RR3hE0E^*S~PEX~FkSPWjc~SI5d4}x&s4f06ud1x3U)=!YX7qTf&~nU+ zE{To($-MKtmk&~aXFPBCz}Cxl%6B3#>~dXumoK^z=Ul31ijPmel2e9nyvXW1eu*{| zfH4%;+;qt*Gx`|B^DnFd(EiL|8WLsO08fjl&6{4y4tMMLVAXq67|?W=rz$_0gDwx>Po^G67E zTZaAYe#I+0B}38t2R3^tv0gnu%AWq@H5?0o+3J8YrG$;|X(01G^i9GOPMrM2qydI$ zZ>%+avf<1#gl&KZ*H-7%z%AOagNsswYy0a)#$Q3z1zBf9qnpqVU;Ti$dZF9$Ta0(b zI35!EFUUdW4{#85gE4z^0Zz_uBsWrSH+q>VR!Uq5@F+@`PevE z0zVMuFpNl9?s_t{*dhuapMNaWArTTUX0``e z2kzLdLkNcK0N9~d*2EgdVVNMCb};BaMNdjAoGk;s@Rp_;HYyt0Fd*cYZy@O~1maqF z9XErJ@-gHbdE%kG7IQk~6@uZaYs;g2@m*W>H!oZI%4eC`c`?2tlm^yZhoHUSD{BX% zp{3C0DcMQEsGCi-X-(J#M*tbTB5|Jqj!-qgHDM+O)J+Z%Z=T)TVN9m_TqFNIzeXis z0GCogK)o75LULQX8EVlhvTqf1wb(g0WI_7oV@#<;?bzJt?a!rxU*@;o4Zb#sA;Yh< zr`kl(u1=CIZMB4WZlhPKesCdK&dN8V0J#E<5#k_t%KC!RADT%cQxi6s7~wTHWlNXT z)Vfl#3!2o>ib`L8Srt|;UZk;b|2ukUqJn97fa!GFk@3g;=y*7}Bz<4K^VBvCL$gs%e;e3_ zU<^JxTvbo3e{Ul5%q~Cz$pk1f5{t5ZbwHAm@)L6|d`v(G&+dq^g7;-Q?6W5vS>d{Y zk$}A~ZKF55Z;Sz%-sq1m*1g$=p(#%WvypCh(>Gp$hPML7VjXfHWfi&DYle)$&jjbxLoRqQ09C$t!S|?~?<^^oM2v(!c-P zf@D{Nz2DQI?@6)%FcoNi0d~WlI_Q)>sn<^;ph!HX8hf*qT01x;ZMBkz_e=BR+-@!n zBvS4f#L`1R$0{9_eySfp@t>b?Si`a{mBK9-0DL5-L{Wli*` z_*3VcHlRgD)xw>q?-9$*tC0vKWlUIh9S zY$FLvm7vX~4P!ShcaRoh-v%>OXjm4TiuQ#(H|Q{TSoePGTC9*X)+&dVvA{%WB34uD zT?(jrcb`p`5r=3_ID$$o`=^vC6Pg%?Yhy5t{n{e(Q+<}I2xKhpiC%sg$cv*ZZZG8X z0~hpzU5!-DYS%+LvDNZFUB#dAYWIVCTEn_1@ z5OaV?N19MH609E#J;{>*36U`S6lioqy5rd^U#Lp>Bj@Rq=d!F3*y$laII+Z#OhV1K zjK-Ufkllmp%B+(@qMsdnU8HPpz0Efli9(hh1N0VZpAy*&q3*h8ighq%g2#t!DmqH4 z@g%?D5MA%``JN5tkDJ5H=R#HU(nJnYa*3~QKaq1HO z&ha*dWskoyAVu=m_`}jfsUH0yu|ovUA5OSG7DW}%&^7~3C-RAy88~D(-EKseE>DPQ znL+1U!`U%lhs6rlGg@hx)6*Ykyywb}hYj?mF43^8>PU||1!D~8SOeL>mWgehznG>+ zifgS>HDZEB(GT1zI@K;urjaYvxU9tE*x zv?WyY*vdqK*QDj~3o~roOOj;Za1eWi?OP`KCNq`v+UayY zOXB$))5l5Dhx0OAMOcJq1%>sX08Z!5VZ1GiwyN;PEcLgE5%#okY;@#jn#6(T@H0YKXp=W~UiZh!zb;hZ(EJdYT0 z$uAir!*Yy1C;BgaG3@*zz0*%Dcwu#Lx+0RH2M0zIHTLsk7+{!9rqQPOxN(($zD8?3o~(_%foT(Pi29M z2MGbYO=A`-#zX5R2({iztPj+4{1@_36RsWLzAO!ef?R9CN&I+A#nCeC&1CNNxu4tl zMvm>6F$?JjdQ>-mj`}V=8Yq#cSb`d(&(Gp%5g1wkCmHsnsJ9h!e(KzJ1lk-K&i20h zT%Pim8pPN2$RuwDw>9Skl;cNB-?58Q z(z~_2M3yX?urSXSq3zKjxBX-d;a;*6Ujl0=Jbbw9VCSj0XuE}O&`f|A_A`VoZLuxY zt{G@xyn;0KJ4*MbR@0pIrb-h}^rqeq+W#7)Vox>QOue4&btGPN47VX(Oe!mV;nY2eEWXaM2!nh-& z?Y5O+kM0)MYWBHhn5(h6fm^H#g9Q!+TFej@W1*^*>}AALUI9IH@1zS=%u|{{g;pP6 zYe5Ot5FhMtY1X$6m*AYXg%6tCMQ_sOLG&#K-x{oV9 zTrGoBpgjLW3xt%HDyq#=hxYR_f?Js4LS_>ToiG5vy2RmmSc8Qn-Bv?ji-Phu^d~8z zgV4LEUZ47VZ|mQk%2?(OQ9((EouyEP#K)zM%J9Za2Jw8QvXS>29 zAoieXd{av=C1iHbV*6KSY^&7E{BFYj(|Hk;^l8jA#&az%Ye>L`5I3hmcBk^>8`wN$q4mzzp>7mb@8W52Yi!)pR|b(g3$n+kpSAXXUZIa zVG1jTo&;^70G31+GnS5yyCEyjdznOE1h01=wW@S8o#$%TN;k~ch+sne$PEz*V17n- zgrGv>s zmCY$A;{&Cjyu5VW{KTyj~$PA zi2ztmNie?iXT6CYg@?Bb5htge(rC(N9OrRA+`k2E<~qx6vl%lN3Cyq!iHg!!CFDhz zmj?s$QGr1u+IQx{LiKn22_Fd*vf_vC^M8PD^E>T|dVBOC>!0E*%Jbo_L~t=C@FAvB zVJtL?@RuQ@6Q577`Pkp&B~90oh3)cu@NKw#vF5xeV7p-1*Keb}KvY7%`rWrY_^4Bq zIMK*1`QvmvjTsv|v}82NAo>g}V`=k>lzBBzHQHM2=u}gv^9|{*CVBf(&6CYyhrGVb zRv2H;7b6dc^QHR?@h-?&7R4Ra}t5^ zoLuQmo<4{U4eGOeb?+HW`*U<;0(o* zRtjHlPhLa^VGof&y29|o^An>{mru9pWHKZbn2UFiOKRd2yQBTEKr0<$&tmIPH#~6I zCBnu0#m~#G`biJEc-KWpTo&uI5F6Tl?@Zuxe#t=uOpQkj9%c=gP~utvUubHqaLbO| zk<)PY!CyQa9Qil8M0L#?C-S0xy>Q`yUbF%+r-V(4<1yK>)o^NT*hPlw^D$Q5f@+n8 zt`3334zYelq?B?;CMN!Z8}npI`a za6a)zqn2n|*_mh2cSrru)G=`q%(Tiy7tlo^;_P&ulA)L-an~jT$GYn%MKPEBQ^bz7 zR7YA|NXKUUs^Qi!oFEqC+$PxXWuoFJ(@x)M-*ki!j2yF^b+#|7#|D>l7PbHL^1b38 z##y;|ww;1swCxzUe0dLg*9_nyC^dO2x3Q4wi5gcKmvfgBrF1KoY*D{2=KixXg;&X{ zjDj$A#a&}np9q>(YzqMANa_AjR&w{t-Q7m$%OoJtXVnSdWbl?Ql#*9b!Kevb3p3-W z$ccHqfBrJS8_lh>hL$r8$W5Ad+6czfnzgg{okE1e910t~zC9<19sFcX1DzYy;5NFK zXA1Y$LRdU@A(e+0{-=h*(k;bQ1Ie5n8T|?wO7oi8bK~dFNM)`a_{s1}!iH$h2jkV0 zW>JPH<5iSwP0KdqJb)KafjjjL&YoxlD?VQ@Q@Yzh!T$2Bjl z5%v82@H3C=`Zqm4qn#ct@{UEjxLuzv`{tQ`nFK=Gwd!WgKk|1=tv)Q!{Y*Te{jfUA)Ek|$Ge@d1#D1TIHa2NvryPGfPz~uN)s@1*Y zA``-=16CIxO)kpvl2Mn6+*|e*Go*uCu$n;o?nx_p`02VwXwgj!oo^++Nx<#35U%!K zy~u?g!W!7yWTXk(V4Z2P`(C}xB^~F8lQFXOYzV#;u{dQW#ozUdm)g~sWw^*lA3+3o zmH`@VNy1F6*^gMdIie0qmGq^~jdE;JFXp&v+}pSGvEOCPFo&{cXfIg(e|)}GH%6ko zsQN0?jm=8)N^RjRu_0h|qfbCle{f5}Mvy+EjNqR_Mi(BMX4mufps1j+QS)2J-Mjh{ z=OsngNkCJ)%X3@%=8oN>p`8#$mZB@#VHbV#f>b)-NRS|uWS9G@Q>MRgVx+l5gqQC= z*KJ{K!JX*%fDbZJEZdM@ZNa9oe&O!f0TV1k{HZsqREYBRJ65FVi>j;f*54v zI{K+0)s?7!>SPrE5XtpKniN#);S$xNAmrSOy$Ktq zc|fBsMT?`ez!zAW{MJiareKO%WVQ$9vLB-f1sX!QfHaQK=S6$W5_58rpkI<%bqSFo z%=}67lAlh3Hpgu>a8k00fRe82d)-^~#IJsQJePhq2LTY1VpF54+BDIC&}JtUO1gRu z6Xu*g5%4n{wcd|EpxoHab}<5qxCSXhc_J=BpPbH&-I*Zo$fd=-xi_;0Y{kyol}Y7BL#F)$&xM#c_ekG?`yLPSLunK!tWIo_+*R z*pGUPv8{V0XAN>SJMNvx=*bn)YzMP!;dfd6uunTVxr(Z>l7}qe+4Nb%KYz4=ONW?10b-EzI zG*Wx4NX#jGnF_k^lwgac2wnPP!DY3j`}~7l;uoF0F1GCm=Lj?cIIHb#qg+bZ@?*>Q zk`zp_=^BiNEEU&NEeh(IU7qE^DTimUhn}>p<9!pf%8jKaG4$O6-|F`=b!eegmJ(9x?faOw6mLTKNnOzUwhVY+Ar*vd8pO^ z{MSL0FF#hge}dOqD{qHH^IolZx@0I0txt=k?U}8Ag|=Gz`sy3hA=tS?8+myOStZmn z?EI`v+*<^^abYb3FGH}Dqns(uX2)wR5{`pGtk$N)D}K~4YN}CwT%+n4?h-nq6OD{d zA8TnM8HZ>~F^*1e3!wtjn@a`F@IKm}T1z*|I)GXU?wVdd&*hlMOw=6g>(!H@(q>Q! zOgyI+&ELx-;*AOG6wB4GSylI0_sr>(Y-5vDgYk?qAC{_$-1^cWn0wQocF<-3Pt_3i zr!ecn6Z3;zdjxW#21^P^8JgdOejE1YqjIxYy`@RQs2%7sdR{_;udw4yq4^D|WSd_d!!Z1I zLGEGXKneF!qwx0%*Vg|OpNNQM_c*oDctM|y>T{2d#YIIbHP@U?NuJxcv^fG3d0jiA zC!gkM!3BCq|vQO_4&AWPKN=~3pFKbtW_a5SL^$@P@F)oQjGqB)yMK$bkuiWaE@gEWw=QkZ45fOKE_MShS#%|M3DC;IKzS2%LNs zH6K2U!@P$fsm8(j8uLOc_Q$&Txn=#b&e&s`)r-v@6A7@jI%K$0gD!wJ)6c66?4yFK zg?B0GE(O&bf4ne*?73X=cVYe%H~f5B_T)EJpvP?E*DHJgcq+9}K{HZ+MX9n>l@uL{ zAa~$|n~94KiiIec&tM%{AK4CHc?eVAOnN)NE}ObvlJVAhr z@oOC9GG#B5UQvVHFGxyTRD*v(z^XYu8wO3U7pL?I$B;x)V#?_-x4>=8&ROlH-4{!L zP#@;=Mv=I%y%53t*TE|c7h&WVZ)S7V5@m!0=WCS$)P;{(E7b-n+%QTA77=RZBmV7q zY!R=0glNX3L}H&4mZ8V;IXrt9DM*x?!`T zGe{Kse)%07D~^iFQSTroukQwnZP$6+a%N)tDgrfD7+2JJa5X#<3%yT@ZdkL8Mg+Ii zx9=&aeFz15bkZ|x4a|&9=%gPMG!CCblVaoRv3QIt?!77y&<&tI03RO`uuK(qpe4@s z{{_|pF&4hR?FH0(<%B2rfRjB-j*gxsU6M?J%i+O6Ggas0=FV|D`8wJ9_PFESih2)} zOn52Exs%3ff~H&Az@>{!4v7e9Soy|UX7btAR+Vg^9pU-0ZD+k=t{|rRuB}Y4brUwV zgD{rWH?oBuczzfLUC4%(*~cw9Px2^YezJ5?C89W29TOv{z6|Me8NNEsciniExG+?~ zDd^$bB+}i_0ZeYPcMUFlMuT?heM;*J&q+SDK1J-85+L;a z*)VU{!ch$(PY@0t_%Np)7C17vi^VN^|%?GV#k}J;6G_v< zkv?pdD;+yW;%b&g|v^>9rnd1aZ+QLgW1I`)A2l}In zw*d4}kSq$~7H6{tDK`QzsNdUhh6aX6fq-|+=kaE5XVD_tRfxHb+1KNCVL<2=5H^3* z;||mwTFB5@?C)Z9+e9jHRH=Dvx+BW62q@Lk#ARsl;?mb#Dfx!pJ6$e1W4%fK2y=d1 zH4b}qoN!4}#opxeLU0~$RpI;Z z)fP)Kz#SaVj=DS#mWbkoalWpB2szh88u4EIc~V-KPQaee%+hinnFDV?Ap6(--jh0B zlhK47#lH{HLzu(E(5zoncCT;m%jQnSyf@@sOON1r%fnS^?*5l`dovCpM#W@dLfl%X zqw#qu#eDo5k3%x(Y0iNB1}Cy1tQ#wJ^kyAO2w#g&E!CCNP8o_qW>BJ`1fStXqq;MS zezZKh@2L~Vhn|nMtXA&M@>JpgI5xHbIkGPy2}@-4&L;; zriJO|On!#%c^yt1eCAqp?R$xU0%k(TL>V6Ro*QuyXBlaSRNNe9Jx{nH!g5<@+!h|Fwnp4HzAMkzA$)X4|812@@Ow_smbUnr7Fp|;4v48?! zzXkyk324@zhQUBQ*poPtt7ji8==ie=@J;KkSE{`Glfmg$J8$iSC%n~Ota5(m+rB(V zuu!;HVV28IT+8qhbt))pGheTtnvW)~teUsk7M&UhG!aoD3l{)U!6pq)_xf;c9Vn~g zm*DE7Iv;SbWsRx1KR#0L^Vbt5T zPxSH@AT#`u6Y!m|fQtFvi*|_X;5b@7W>(=X> zvhTn6pz8a=^gPZ2;y~OTCw?w3_cam~^5SUJ+&%B#4GmnVfP5+53_%I-yk-`VWA8&o zthFjhpiesrPcqX;d+D#?@-C<~APzyA&(|1>g!DpmsyJ=rXzC*WX0i&8@XwOu;D`Be z!To`hb=>-h~t3Hgwa^-9hfY_P(*MzMjd)-X6t&2P5YsJ$p`rjD+6V3lCvs5g)vhW^A*BIa>eAJ$nU7F%b7 z)A>l#hswZ#)?6wX88QN4sEG885`b=m6DiJkIH~HL=->JBi&(ip`PT_qLQaac)GT~7 z^br591-kLC|=%v3C#y0-L-q) zqWV)~Iy8C-6OjL+e3q@g%R(!An9Q;X#nw2&^xp2_JcQ^g#fGrBf*udW=Xfy{+C6Gj zsgbdVMv>r|YExYzBlG9S15#jl@|eld6|l#dy(66(GF#G%#FEkcI)tStDlL?8JdetJ z3aLD|j4diOMa3+?)}D;xLt+fb7?@)ThQ^*a1sSquv$H9pbQncvUuTc3tXz#!&gym(x*Y zInR`;iYV!Hh7zfT5`>|sYfcRDo!%S$D2N3SAvM~K{JA;T`6#;l9l-W@_;H{Nf{QN^ zt$NI_AD6xoJ@w^i<~q%u!o=$4L|Yj~9k~_zT+KEUM^xLWBh}ACO$p(KF9JK0&TXZ? zI(%nBU3MJb66w=y^AiBE-_WOqdNQiTQ$0s|A)PgF$BX000TN@8mR{#|zHqWGs_g5(}$R4g5h)&ZC0pdlm5*s!3_ z8+`^(Xp1)7LiaJM2dCt)tr-Xj*-d8NdoqK0)yQ$J8Fu@luEPC&f*h*+NU&HQ*7EdX zRS+gF70hz8XbTyx*2GOuCAf*wKfgJWIX_4IJc6iKgp#<_>rFCYg>T)t62&A#qJR`7 z?Xd^?emk@bC1n)hHwT^Izs-3P$ZyVhXuE)yi`U_uwY7{%7LYW_;8^!#iEAFi10GuF z^_^Z-of|^V)KL4mL}sT+N48e6?z>gTP&gxH_v>S4r+3Cm3K?%G+)-9a8(fdI?CW}f z&o;^7Jsi4rAPcDfnkTkT{#zH{t^?GK^NQq@N^F0d9>${n7%u+?=#kk43yqW z5hKZh*i0hbi_t2L-?O7H3y3pd7O&k@i|rwr>^3+OH$0~x66w6gwjHJ(sNo3mDndA& z93`W##Elc>5qKz@)&!C4OXKiMuv5<^l-<~9;w=hI4`*=HcyN*2#<{6ddl0rX6ieyB z!(_0#d!>1YmVz2`_-0Nl(UT1q5M&#v<@VW(goaUd%ym!EDMXXgpgn0HT4tsublcWy z$p8{)$5;>Bw#PuA!N&VN{p`Y6;6oHRAEb0T-yV`BlKy8gh&c5Lj&2$R9!=uvxY&&^qhK}pNh2tX~-vXz?f{>TD{k=@h8(=t< zRlIr}Lj7{B@djRHWj{0G$UW@wYxFILB|0Fs-flV9`^5J#SR;53ON>~T&fC~~Hfi$s z0Eh6-8cx%m>hv5_ zQk?I@P;N|@#{{QtDU-K1wdwMau?`}OVJsYViFO2WRjC)E5^Kz~Cdgwj6U5p3UPwoo zQE{f>eSqUlGbFbR63!!@T$-7_IgxoTMVY~%YEjG)AV`u3$`lc_KPE0gy=+kbbI$oM zQDQ{)QKgT)t)WaFDNsOua$y5559Qp+ed@Jk1TjNiI4P;?6`}VUI`2}wvR+HOC<>w83qhB4A@%akjFo ziBF*N^!_`geAbGgI}UPi@99_R4?z9J9nX{$oYRNm2l4Z-z4|wVVaNNJ$>!^822V(+1x zytoJtkPF}lXm1={Bf~m?vC6-R#!G78{-=Q!79;EGcIY!!PU&_hoEH1BsL6H|Z{3K|+DhMRv>n(E1r^&IGX6b)XO0SS^&XDaIFP|xvSN@5C-C2VwLr6}eYRrR#l>=e6UI@|d zvx$J3NL6A)w2|JvAk0YOeataZ#Q%oGT_$KE4_zDecc?zY31ebn(jx!@do$}b<-qPr z0ZyAMc8#mAo~i~@Ao67Mr`J&niAM9SaM$XghD0QhOwhTl_Q}Cvo_T4d1QeXCdp^E+ ziDX1UfL#8-OiMZLjUQoo)vq4(bknvqA2xn0HO7dH!G3Wdp0#CI z5+3(WC*w_mvcqByJ}7U#k02(_qYzyPAJ@d63%CnIiyeluRh8B97I2xVm_xA#@{EE> zzt+~)%8))>54=h85Ma)}TiHwC&~Yxe3uX$3db!YEC6YHK#e}v;e(;V3Q?|5Kq6OVe zM1GQfaR`hv1z+r9UNF0wa_*A6R^SPqQo>Va65R=V89_(Nz!1{qg!ug)u{$DqPxhYC ziO}El7~T~9mm0(Xew|Jx-SP-1JQjO+thHATSWQ}&b6$E|^wqkA=a z>L*u^e82=}-1^;Yi6mSK`T)8t12cV8 zenFTMn55WPP*0%g^mc)OYMdXp;hBQqqu*hf0&}u1pBO7Ov7&;C72Fh}3?-Iy(j8eVRsN+93zmpOdg552ET^mN!vw2PgGv5pbTW`49hjBmNi^6G}v(AGOJ zwsrRt32HS$aXa=&5yCISKon%(N_D5B$xz5iZE5X!U2()>ytF|xUEyWY*Wbh7lW%D2?b8iwUp@Zt%@TXu3>G~=1RBVQmd%2%3)=DldFTb{01w2%8;IMEdfdgcgq6_d{cZgix zOIkcH8@&@I;MB?@Zt!o`;k)IZiDTf5OA@X5(i9Xs&s5MHhm(VICj77yXhybd*@vM8 z2)mV@-%<$GbLPvS(!>~QLFz=T#P2{1{~kN-lwBF+@87?#ao$s~va!)Eu2?=vdLrkO z;1z=#fx8_KdY2X1QA> zpt>$gvK+CDVz~!ISPn0xJNI`jsA8=9HM;P%S5*Q566)BQJdUCWy`BQ3^$LX#myxeK zXo+bDh%0{o9qZeyQvxxi`X%E@lqEIgcKkCcVapMosUexVl04693Rn|)MYb6z&W?g_ zM&Km}1tIDv9V2hxANpHh71O|Rfmt?TX8=!}5QLuG zhS$j>R-5y`ZYzJqsTL{F`S+Sxid|iA#n>q18GgZ%n42C_I(3ESCTCJ2fy)(AujAP^ zZJ+lpx$V1j6OG+UJa*AIA8H$=h*1_>a@fdGQr`}-9^UcqbPuHKNaEw-vcJG|*l?F; z`0UidSn}Koyws~DP#n{;M zZp(I4fLmmNiu8B}FXiZg*R(u0QgZe0<)T@tdqJRF`zDm?Q#n)A6{EO>1YG8p-6VnS zLH~}ZlNQqXrY`I zN88)^4vRiQm8@TSke`No$T06dVnc63S@#_suh@#AWqm?T9@t7esn*9*lp7Ad)~5UB z5hazgMPFXIRw+#gER+gdXvk?w6Q{xW2#YU%`E?^cZ&dCh?PhjAWoRWXy6)Pa<&q&sr{r zCcF2{*mxl%+fk5|mL=BOQ&j(YGGSe@ON(t;PhR%-Iwl7X`Uq6}<&cHB8m51CVHZX* zJPZ*~b`ado6bThjcHd+nQ_@Fh(91ou0S+@F7^bw(VQ@S3U^M+@A#Pu*N6?bH+u*_w z5(Y@=h5T;5_lNoZ)4>KyL=FNNHNqP!{#5$G-XZCF-W9hh*??rlqDyB6qd~J5E>BfdB$Nl<*)xHuhzyzw=L#rH( zt0bF&O#0?vSh(uZ90GN zlX+a|T^m8bd}IYE)zZ?#nswMyMyQt+`Mcxedkv32`R7Pu^2AbHypi>I$4-BPfr>K^GL4ghu?Bhc?8tSzUZpa*hN9U zGJ`ZFeH09YZwb<*)%v%quXqh}26*&(?UUcMcHZ%i4c!7}ssC9#Wochiwk}G1MCAGE<41C`fgWT31e%(X;{?)djH}!YS^TAB!*- zbuxIunV1P>!8dVEO|a3V01tOXHpZu-~p&7(j+ zeif@GO90gsrK|8K*&(!08RjdGu{5Iot%A$IP_ye`Pmh?*-h1&_DbvNoCNSM#a_~Th z)-z~`bI+g!8#i}kTe+6y!AGyu`XdWBhVfskt9Lsy-<4BkqWpopfndrH_&}2+=J)0V z&#|$kfMJDlx|xnw$W8piH-3J%Kn}=P8x)#9?*`DFYk$8T`uh*JMIY!hGc#ofNq|wF z+qV@upi&sY606p-tzEXcvw?xLl-@!0rp{)B^b{+n_pZt58l=N5S+*x8&sc(h(w5uW zx$J9*usp{4soJ2=O(EGAAVu$K1VBv~b@=%9 z;ZSn~Xy|D&I`t|jd@X?e{~QM3Y#2#PF1tV4%!qoouV%t-UR+bBrY8YW(`I8P+cq<= zwA)erQ6>EzV5p)jH&WIvJ%7+G1#c)~Vh zh~EQVZ3$UfS*2>9JGkI9VxV58+*dip->8XUgKVpprbL)xd6GYl!hu%EU&p%!R6goe zQ8z-tpyUWJpGo+rj3=7i%85!YUJ~y=v)XQ48^Pm&FAM)&XU$;4xJWq|XDu#jdtPsZ zCbBFw+Gp#`%&O{_Ge<>2T_m%FV$NED%L`ahaqIHZGuJl^ae|cf3XNJ5K^;lo!hWK2 zd|R~Qtb~U--um8?#f;1_Y|F|Js9MBKA7!X$k*~}|)@b(^zb^9=mM0bTtQ|40W#i+c zpPrtc4PC*dH!^quVD_h4+;Xm%DCWAZF=d3b{N7T%#;VzcLh4_)< zyd{V$I?+j@P?%6YD=p5raOV5G-x8F$F6W;uzJLO^keUq$wtzO(fasywg`K7LeHj{b zmM%0^hlOu^YBO?{yD%ZArDLO4ervmC8)>$jc|VnU7`LTt*;0tG9j4)%w27VM9B!&b z70|dG`mn=h+4hKbq^I@scDvO-cN~CLbIBRid9=1oIwq{vP6#nP`S%3ffMK8)5R4u- zr{?Ag*O;;_fQx08Y+Zzu@@JsEZHLso(e~{j$|v>}reHM@Hj_aU9MrOBFuMI13o3}; zA4R65cH33Plg;8TUH0V;crmaI7A$1)m|VHEOnx6I#prMz_ANlnh8f`R>U+1mW*thp zl{I2Qy5#WMLR0f4Q? zPuPix#bYYFg(;92Uj%q~FgU$vKhi&qd$_xr$0XKB1c`JJeU~XMQV)y*H8|Qh)=L2a zFWu2OII0rTXfy=_=LDR?R+c|kOyLNGo8QkaYam^^Y|*?@w_0ege7;?(dF>4}xPNDa zwy$8op3%g-6hMFdB+}mSYZ@8)^pt*Q+^oxeQOxn!aXq@_s%nMd*FvVLjt_*DD}1<9 zP?wX>my8{t1*SmMXP@3b9VzQFaFBdurOrW|X(zAS{L#kIk#EuMWKjW{C^Ah>&$<3~ zfL$0(Tu(h<;)t0#?#_iLugm^N02O8kC;~dct0ilD8+&cIHYI}QoZ(bDjhXW=vrA1T z-zWr)klaH}Ew`oeEtEgYYb|Auq)!ni(t= z@UmOx&VR2Xu4hl^xBH4?DA008j?}wI3dQ$cUR~MCG1xNbqCo==>7;VFTB>7XYlk5ryb&%DOpdP0i?2J8%=K4dGK{!VJ(WEpDD&2Sny-VQ!myOI$MOA!l)LHs!n z3(^}o!v7=c9pftPqy6D*yC%1rY-6&UY}=Y_yUA{9CO6r(ZENymTT}n5`<(MUpZ#v{ z7hAt;U2CncI&SOj(y9d^LTBpx>#DJV#WWrSFoUwhk=)0C!ZDtj35h+4S*8e2k6-{O zWc$syU`9>2{{JS?#b}|4bH)$dEY0TAqbx%?^ZBr$eT%xV4f|Nd-}b;q zo*#*GY;J`p$=ZB}S_{UA3N6Sg8@|nWv?i@K15O@Tm%rPjcw?wp5j&m9&}RVmO~z7& zHnL~?|HtkM^v9R3cW=+@5vqkB_E()!zzi(?w1eA4O+i8Ny{i`3V3(qOn}=2YAoRzwqr1PV-XLOVR|87%U@ALpWMqVH-@#?q)_Wo*fN4kNB~F4W+9Nrr zCP&OWe|U(M_NorPVTl~y{9{V3j1{7j%2z~^&czhZ;AVlw@RavLaghcSD+l^NB0-(r z|0&ysc5vLLXa@KGu>7{x8J@i< zBJDm)6x15~fICACDBz)0%-g`Zz;|dFuvjYGAvgnvz_6qAx6%FtlPoFDJXBfd! z^=eqv0Pcnc+aY}A<-^LxHFp`3fQk^=_$~a=S6CW5ku?FIqt|l5BACGv zZd$icoS!kS9(wHq+1%Qb&iI*`0yNqM4AYE^6fvjl|I4CWCjI!%zzxx_F#iY;p%3g{ z4Nv*qv-!FnkF|?uZQM6DjR3$u+Zq6=a!_ntwE^yvY-pSQh)FVm zoUEr9|a>n*z6iR!5}H8rE94LG)K3!)yOjWq)s6jvz97xVQ1qXe{x`v zM8I*bB(bMhWIhmr_+e12%jhnRLNUlU?r;n1e+fIgveH81F+MR-f~-%Tpvd~YMiWRf zLz!>m|9!fHfQ(3b?`o)wVQci!SwrpCZmu-2B6vwdJf#fB!EO(A_}sLBQnOxI#e&SxOGgxK#Q7 zJE_7`8`|3G4uk|uGQ7_isu^sxfO^){JR_`2qaFgzYF9$g&@AKw0B=8riqjc~=S!?2 z&m;~+b~#IwP(v5m0?`e@00-p=4e5MtDT>+QFR&~kO?{_F1dr3&3;%Og>7b3ey9V0 z0A1P0Hs)jyj8}W2gRoCO}=lv{~V+4-HXYU~=HFjH%8c%(h2+6l#YvRK=%WY#J#iJXev^U-Yh0EDG`@V+K$|E2;W`yjV1*ok<8-jeHqVgo zfDJu?!cObj>?}gE4Hs#~-CpPatWXrI;B2{#Zo45-Ep@!Tyx6Hd@5&4+^5^mTzljk@ z{+ua|b#uwIn(9`v2XGm@c}FqAlWwcR88^Z&!>-&>u^~{wqOpnjfBJX6^FcDi4)4+- zrcoFT3XN)2s?r|0H6zt*`SFlXjXnby$#9JHEV=Wr*6eTyx;z%V*uUA0&CdtFYgs+L zWKh~YHz_FqF;8-Z6`U;*)GgUNxs__ztfe*Dy^G~rOiEiAVZotxq0}^}r0l4&0cai2;`aJKW8S)dpGn=53f5yT@{%F=pq?5eY{i z8H8$JM9IC0aqtnfZ@-dIgJbEM{V+U_@!xoVd$93w^85lU0jM@#HXhhfRs8+C2Bkq+ zmSmGt-uI_7o*#aXA)G$ZR#E~ijn5MfeiSY#kF#l=1=bWBB$hLhJ5dD~BNT&D>Wgt; zeDoBZ#nIyi)83EIO{yd|ZZK_zzZXh4wKIHVln0XgMQ|{ehGuHC*)bs6IUm=*{+ouF zr8c09xs2YNZ7^6BCa9SzOAV3iFdQf@lEIkBow8S#siPbSPES43j^1 zcYH5(0#HD@(f-pm%dd)YVXf-Uv-vS?zdZx1&6&>@*?d%yMw*ey79pNmmwSdEM|v>; zjHJ1Qy?2?9hV1vuUI`5|r78j0Z*)J4`dSx1jR|?b%V+9uXM1pMEg4!RSmunStVK&W{#k2{{l@=pE%S zuU22RWKX@;@0sg%gPVS&jdH}Hn}lnpj*LjhG~i|W>`{;;0IUQXq??b3(~nfG(%A^> z{IM|rgezo0FiArNJjiA2XY4O>!-*4tjJdB(eXf4M*lL-%ar5h8SZDKDJ*K4d{olnH zSJ#u=Q^)4Z7taYB{d~wB6k`73D{RNDZoUpwz0NsG0fWxW`Jx}= zjRCMmND<5U@dG0^Xg7jSdv2jtaiTAz&fsNr$tC-%lfAOj1;8zk5--k%R&q56XA58GBcg%T?;9^Qt?(6Hj1|oAM7>GP-ET(Ty z4BoXNxfa7I1frSAYfqRM4i#@<_zzWW#U-h|#6Gn%19@X{qxI@E1((b$8Q4=O{Or}v zKtp5*R}K%Z@1DXQ$p9v zcP7H(g^Qt7_K%rACl&4$05lg5d|T<5IXym@DuBRm(K9;qQ1kM`ij%_?XX`Hc3L9H+ zeW^_Uj+Ure^2H%hfN*&*(0L49g7EO6b*V#M>b^Su&=Y9XI27in1g_v+mA1C{Dizus zX-n};5ys_UNcnjO4z41n zLM_~i&7~ybk**A#w!fS)5;+AiD3X^Dh;L!$!G|9PkcV00VG!5d_*6)VQ?;IS->>1$ zu1m`kHCMziBGw=ui-*cXpV6rzq?bZ{Escj#N)~>pv-E}^Q+3wW*bQrdqOpC*q9qxc z!@C7%{Y`-fSwbW_S&dHdj~j>=jUC4D4cD zhq(%ZO2`GyDX78YC5fZM++wF>UREv3?Y#e6A#wl55qkQYc1JEli5yG`!;D03sU@_t z$G7Ova) z6X+G|2=*LZHa?^mqk8Ut3Mnq7LcUz}WAe^qzkb-uZfIJaTu!~eP7x^1yk{cN#R~ZB ze%dT9E&4DIlVN5!9p5NN1mEE9IP3=Xu&QX8TpYCCYp$i}9%3Q0#wl4y)z0x`@bfSv z{Gxu1jA1TixNf{F5(is?d#i*X<#u|Mlgmnn@>GSm3JjnL2pGATiF}&YRRkSDk^cZO zFfi(T@lX(fdcUp7u7N#eI;-=7i%eL-`n9Cp+YE)xEA-VJ~GwkS{`yDA&s98o0m+qcD~eoPpIYqdN@Nm zG9lKmNq85ztpFBYg!tlZusXoERTrCE^?ZId>PJ5UY&pxm@g(VxQ|6M{KK_mn&k+3f zwMk!(Lu}~J8=^?S(1rA7eiSX>vPh?4yi$;^wXfA0LK{%;yI6iX~Ne$i#dZxRhiM z+6lvME$9o2sOOB$gm_vLf!%!Sh-5t?$XbS3IwJ$I$l{p(bRVC9V0OkIRFz5|E%hzI)@!M4qY#9N3Wlw;`F7vLkwNdwp_I{2nh-J# zG-3kC)0SpKf8Xv5-qj*mo-=N|V;gtbfea|?vISyA`?w?(JT%6))A7yHDq@%CT4#4+ zqPPh~?z~PDjqn00R#s1A89{}}-JzduhB!KE4JYB7MsX=zuRsLXb5kWXsMI__p-@$6 z0XI$(MrM@oc+^LwKi%_axhEr|TdMu4ym_*H3Jr(yX z327(OOyBuXfh%n2z>>mx0r91;m@|+kCTWHfi>+6QJ?&= zA~Rk=O2@Cdzu^cy{XvUL)Dfz=d;KA}Kdq$!&$wCk;lh&B0^?62D3s@AkUX2kIQbkU zxFF0hHb#64h9!Lz-mL&g`A>mUPA@f2z@1{+=f4GA19&94b_2_|ox=vf*FM4d?^uwS zFzQ*>S{^35ol>*k_6QkvH9c|*cEStIv@l(&rawUVRRJx|c!|1vA315I#u+ zMJyuG+hRV54xl%kK{-rvE?$p4oR)AMVr zAShBhS3b{qk$3@KhUh-VXjY?BIQK`y%MNP`HynCAv7R?yifXQmImLD2 z)eoWCsJo&8U6j%+5e$lRq4sI}*btnuxhR|b?3aS)@Q1bUph>SvD&6oUZTTDL&e8?F z%Ai+8xQ#HC(w(RP{v9ma2fA$uSKMGMS|K&LkXEHH89>xG&T>46e7~P3z}3Z5GA8sn zNr=2fH7|vDHSHUv(2nN<1{2f{H5|kUDka7RBq4XskbGS4@msZA>#w@0g$i@o|D-Jd z#`SjXDR^F_Iva0iSgS*ilf&>QF-50U3Fm_YnM7j)6B6woGufTN5yBFZM%wEbdY{l-jZ#+ zlr>AZ#%iHi6zH`EY8R^g_=W`zZC~&@AmF+!=bLR$0CPWvPo+D;CS3{S0J@qW6;)g? zC?jOmGVT@-=QxF&ID6aBfCftXX;imR67bWm#_<~YDtMys`W!Q5ekm%2dRhOh{yyNU zGxnygol09s8u4i#Z>J%xG*^1wf|if5@2yUe$WOB)3r0#BDNOObxT}Nr z5v*A)PS)a6uZ3Tk+tE_sIXYQGhS<#`&JdzU}4Dydua)N5o zDv@=V5tLF#nT$Ns=N!&n2flS|-Qj=vd^&w3_DQrP7!38xvT#ipc2y>nwbP zIo(R@8OqBWI8Wf#u$LD_(wXZeGvb9Rfdi4xgC^ox9C}R}h;fi8Gakycy5P7+a*#Bh zmsgxjN29cKY^5fS0nctfMYf9LhJ>bPsYrLm_PVgk68A!KcAt1)B+3%BcdeM7$eec_ zu{4Q8o2jnl5#8)SGB2Yh`(+s;(WRHuqrBojaH(1w%92N}?dbjC{Vkg|GCey51u|NY z+cM^&guT>^ELZds4~l*M-92Di-%-uC+EMc@U`;s`#~EaS=gu7(yvWU5SyiRL2X9i7 z&PxYMss;;MjbI_X$eaXg&2;G0N^@Q`I4Lu6XyWEH^~EdG+TKNgku}8fZ3Qdr)=olO zBn$BQ1qYk(2{cPNGGC80G=*oR8a^~qI^PUO?{~lSAg^*6Et?n`5{My<6|&>d05 z+GlAjvmw=7m5e~=gijCL{!sncJ1~Q=NM6%8=Ut1TOLYIj)5FBDv9-qK#Qsh{?eSy? zebK@!t?}?R*P$5pEpQd@B7Ns!nRjwsaMQI#-OU3V=0(BUx-TuQN{s$3X}V+FmbNyj zeA8k_xe{Q&Cq<$CN2x{(Bl-bnzusE!x9_I%OP%LO#pnO1_#aNR?BBQnPz^>J6pvSY zX*>e9`}T67fnimbGe6la&gU$^QP{7kEUQQ@A*;_6 zvtgSqaGohr=el@oZzFRho zqO_KN#ALYl;%XSdI)GZDm60oB>8zr;g3?Wd^?MZMzNzsWcWIW7QTFOi0)MOZztrU- z05n4itO)N0#wvND2nnYM$U&*a`1af}Fj(AnV6n0kXwx7Twq$yq*|_!*Cq0NdNy@bk zO@777%uEWz{Y8@cxve8uPE5QC|5nIO1*ql6zLAq?uj7r?$m3mN*l5n|KLO5|3GX22 zxBO3KD)>RqSb`BHvMcX{FArn9PF?F8e+OD{d8=OiO8zn$IF$K(W*kTEJ;0RP-+)4n z%bTmETBX9AqW}_>7H1w;;n$(;>OKSBW6@w$RIk6q4Jz#nX`3XW6A^=ZEF&dl+iK<` zR+}kvkXAPuAsek(itXaVVf*`cdTmybD%Pt}Mv6*Kp{}34TE^xfsB#j=e=N5@} zL#lV#g%k|N82S}Maly?atzzSf*KCl;sERqT>qvBTaQfZVh_r6qk`Km1ktNzqu!EJ(XcKEEe^%F!)NWe{ex#=*6YSsyx1DYCL zedHLGSZX&DdOCV7OlmsD(O_3rcg*gpDMGVE z6_ky_hC}jA2$%F*9-uE)SG}D~0fY*m_+s5_3im|ZQA*>l*Tpl~5@eFaFQ{Q_329MH9Nn-17z>R)biMm}?j=aF z`;oTvwaY}0;H4@0-gaF0Dl|H5GLpPM4mG?vc+F+9FVX8E;|;M4ZhPxnY+l8>TmD-K z1LATFoRPjfALf_R8=s1rVVk6QpSM`_j07&OxMXkxj5Yx5qsAG>A3I~srqd&dM$&%i zL8W-&fyYKkw+!pe5?>R-BN7R>A!kG8`%N!ZWBn;J*LM*{)Ub?9=x%AhDiKCTRvL@OLO z7f|kc^{n&Lw?FwJzhlo~%Mnk);2@9bmlyQfKbI}6$;jlnW?`-4F6d+}ss*sUod`j$ zK26j?;td7UR%69tN9KXSQ3%2pd>^G?8u?=EPXyoU+OVMhQA3f%1Fj%Vel2yJyUJ7( z@;HgX7V+>IA${}A#pW7G*TzB7&0mdlgoFR_F@?+6&M=+_!V*t|TQGIpCenuB9FXNr za72FIH+cS_9)C0Nyzt(nQ$^u%giDhyRBjbGt8GCpjY`i?g-Wg?G^wQe~a4-=rG+wkiva?~LCOimNPsX&z{vagH=9Jv)LY-m*ho54VIu3d>RK z8zfd~WHC(XQXb2kxEYp+vYNBNC~PtQv9*ez(shFth!UHYXSF^CU6;udRV*83D25n1O(zf z24!;vgGz(X+i&E5N+NM1#FbL4s>OTc9dJ3L*D!ZsvQXfgn%Ti!;(0^hfyc`8uL^L8 zI#Mu6**b)~ecj(VrJj^si*-A5u@Ae~`UKn&BLeixa)?3Ois1OF<2Z#bt-5NT&McES zOJfYfB}5|op(_rY(yvLGi+}As+ZU%4mlmWDn9p-Ws|qDhm#sm(!jWCMx1@U5LY!qmV4T69v^QvPEimAyvjF(ICME<(4sh(cuq^LB`1i4s z7g=lii%_)*WJt)$!Hzg?f(nYspKpH)Rk8*Px9~Jddlkk&fL%#1CSk4Zikq z^zV^Z3V+Ex#UM)fbtvNo*-RA43jYc{2**oG5G>pVEDs_F@=6$rW)I`(GHY&&{|baV zJ4~h14Ws?3jSEBdA%Rl85}l0@Mj81;>&v|wtN%w8*HMfpm^u~*WF$r&Av_7V>Ck*! zU=5X^AC@gbCoq!u1Ap<^6TUtJVR|JKxwXfD5I+LOa$OKH@-v3?y5Jd^W%vWh`xVLa zAKbPF5S#n(q^-2g&ohSJJVDXb83^<|uBmLzA48kMzY2D#K{YUU2aq+lEY% zcM1zg49BT-?eg4DzNtFq`TI%TZ4DeV zc+qrK$ZP1v?Rp#ty`LW30z_F&`W_xvM%&PlV|GQz;K4kx$1i|Ge;F;s zv$e`X|I9Z~h}7XXuQMBiGIg(5Jt#XOP5(}T_t%?JxI9#zj`ob(6iz4l`uhBxi2S3h znNSDMF&!5uVrjO6K}kNC(kQ}r``G*ZLTJPtx3Odcv%kZI;zdZH|IKq=PwoWJHe%mEtM@; zV2EU;($*V4-Uz%{~-031IFeK`;^KZaz3T7Px8|R{ffuD6L2+aA7nT&_^ zk|W0T6CW|8B9?2T`xv+FQJ7(1eVubRT)qvDlI;0TYPZ{6|WoF`hv zOZ@&_%x(>CRZmjjH!qd{S@3?5U$k4;un#`}=*a2`hMocurfgj5ZWHyF%0=Uy}5e5jP z-_ys6<7vo-q|a{rDzN=^I-k&TW(z~I3_*szMxX&CuOC$jcZ=khSPN_DT8#p~z@#Xm zFU|GyuhOW>WGI&Bcx0B$H0qf?5LPR+EslT|xev%r-#N+6-QDI*rIm&`8=axr9{~7^ zEAG;|Ud_}9>>Ard<^t@HwS)rHh3lbygof(WVUegx{3DGV{LDD~bVNpTMk6DZH!;g^ zBS}Qj&Cd|2s|#Q&R}>oG*rz|McKr8$(LxXwPvf5^nzE1u&;54k?j@#<$s>)?=GOiX zlV<1mFcA8H17+Z5-ePzHEC;&5tnBUOA&dLMQO5-T%?XYG0d|N-XJ;vafq_LZ%LE$U zP`ba=qtD$Co#mOQ#8j|ch*fwNHZ~bEMxqNOSbuo@fy*!AZgu?whao;^QKa@oX(X-i zMX0GCuyCsWLe04Bm@r;%3}$3|T36 zovPD|4rw=+(?qxkitH6~&Oa7lrgRMNBk$anj}SysbtgC z|8bn=RTQKFLQng@Glw+5HB9!^?;Qe%Fh&wko+C@f5vxr(P(Rv;m%OtoPUH1w*rmaR+984!>QOmo{^)qX;IZCS=KClxCP>#x5_yC5q#;w6dr_{@gNUJ89B0p-Q(V+FqBJQnO9f*N}$DKyXcHcP)wdLG@(p*V%(oY4EX zV&&_1cHwKR2{*?okoMJ&C@2(S%`08BLa!Vh){>y(F`Mdj(FIvD8`hW zv}dF3R7Q-a)Jg4PwAN@HWrkGT7~Ajcc-+ZX7}J@8{=7f64li=E*nf%&W=3XAu5ZWU zM&_tUlAKk76KO#O#s|iF#42U;`()jIx>aMp#r(k4V@0oozgiS1_5_%BP}u3~*(!WY zzMgQl&Akyk(cD3d1bQ3Z9zfn15HZK#vD&`7e(P;~4wopQD$TAENg4+12VdwZt0DUtCoG zcXoKU0#m5}9&E}1WDx*ywC~?;y~`0%AV4DDKrYlfjEcyW zy27((CoYgL!9>_`?&xi*F{LkEwCiMkYUc9Q%MjHq-zD6eGGLw6Ud5CB*{HMf?)L!M z4R%Z+`ZSLn2;oSD6c8Z{CNV-Wdm9aAZ%_ zirKL4(T_5rN}F;Eqf3vI)?s-GqsK^*%gf&htk6FNg7NLI3uH|mf?1$#(*Il{?AUOZ ztx9I4jF-U2e_g;o>%Lgsr2KPO9~Y#U_XUrS`s&sikVkPqfHcOczkSV;8=iKU8X3+- z&&d8MIhjYi;=6>h;?~D`73w?N!1Ni|-I_Ullld(Fb@XXrGvS>MEm0sQfO&O=CgfT!bn}>@Yk`KZCuJ< z8Hq$Se`YsZbiBwrWRkTkeAx)kChYp?=?`l|>$8HL&W0}$oxD!dS9yuo{mRyBPt7le+n=oPu z7`X6~oGqX=n?o(r4Dn0gJ?8ZARS5may)0_q<>2dwWVJnA4slzIBP2-SrteFHO#-a- zb@*0$iH944!)7XgpXXOKWv{NoB}xR5dOEK&{rE2>|(8?pWw+VX>37+8vg zA6j|n3QvXrCal$7;;+<74WPxbsp9w$+K9oP@B>2KQc|b3Dx?yr&q{h?JHtSQMTcwF z)?K9;^6)mV^i;J`l{t$&OX0iUpaOifoXAkVn^MO}?Dk#1uXscv8H>Jbs_}fm&y7-P%f5Hak)%7KsA|A9uOI7f|;AA3#n+ z0M*{2fdGT%akb%l@Z(`)1qfQf)*9N}H^bUsrRZcy+1q{klscqzTl~EPwe}Jrrk(yb zE_TTJa7Dz>QOc*?i!n(y1F(o_!0($AC$=J<4-o$;7WLL)AjXbLn5%J;wKXJrP)F{h zf+*lKA}7bV6L*cO6U$9M2`kTFVFr$21!iYRXcVk6j`hudF(Zh1qIyOoNRQ_mtfdOH zs01IvGp)114EzK4dVRDXpPP}M9`F5dKC-{R{|Vh(fSc#1ykpy+aQ+`pt&u{45W3rT zj@l{P)pMa@ebC_19&Wb@8ULKrF|VZqB#tDlXvmeit`F2a#tO_{cO4@QzROthQcnh` zV5WFPu?~oBHK3DzkNBktMV$TS2<<2Cc_pfC5Ao0PoQrMe`^z3eyOutEJBtgYSqTcA}@B})32Y%A@SE=>*T9e-O|Gpl0?5D||J0DS21t)|o z41UVFh=4?z&k&y>aH$~LrwZ2BFZ|!V$t-V+L}++N7gua?gW<1>ZrHQ4ze3&mjYo+a z8*@VECdLGRo%;|B4Nj@W&gZNEt8W{7AcgyeR9cPuV!DK{3G-`U>1o793WN z7+?P?H%o0`63~1B57@6AkI_wfdUQ4Gej@5wIzMoAWzEij(@loE6{g^O?j*U6f%c4V zrMu1lbQMEPHy2@h#)u=!-}l(qs}gy~+&v?Hq_)pLX$%!h$ovdY4kdxa)(ifnO@P`C zh+iKj&&FXsL%c+S$qm`H1t_70?Y3QSygVJZEQjHW3i_BGTCrd2BS}ApfG()Ezo?`R z`0FKO`Tcr?qPrla7(B}o7*=afycNWyvZnPcM_)d-Evw&tHsIVlx-e4JvXVg)+U-q4 zWW5;xfez!Ry@vXlsCE)UyIf^oCB7!QC)GkbR5;~9fjNttofUPZCzdEsQF3D|W>=Ve zA**x73}nLrhmIg;EHOJy)LMm(Va9Findh_LCnCwa%^lgBn$)afUTBFxuU4av@X-I!*Km`CXw+!7+oK8+pQOXBGuQVhJuYnmaHI zU@U<7wpUS!BLlHzda>GQ*Q%tNes~w!}2c%=jT}(2?#0$-&VCzSXg)ONCqrnZRC}m@d(VQpX?S?nIItb+_Yd6@ zoxN{=pVxRIgVpNqOq6@_5PTYm-|eMoBWZ;whk0~^C!3nOo)1Kz=dz+ByLHIBueM{k z3lkhfdRcxyh$VZvJQKyEG!T|7;R@&3Ra<*k{BqgKb%Fj)RgO}ok4Uq?Yl?%?URXE| zTzn|R;H+VAcxO#$h99MYnq0`)aV;LXu>yAq_xsXJ1RrhS%I2`hYZyQH?xEvOAQr8b zXX}&DqOvS2pOao?+(zn5cXRXUxdG^N_K*eF*C~HV3N*t!tc*DsCKMHz$bjdXZO85M zAQ$Av((=D)f7ht|6`9C`%5HYK_VQPz%s8D~4Gd?#Ul|%i2XhF+^Fm0*sXSme)2r59 z|Au}4JRl(~@bj%9y8C@Ns^Cs5X6*ZcsUIggcPZRAA2fD^W)WVTbINlmJZG-I#60Hh zZdG1>T469#FHV!^b`J&6oOTUr3AzjvS{_l@nb1aYDQPcT?#1&U z1Xkz?P5x4K&X<;PbRe8?m>*4V|J=pvVDkF{GlnCl#-_*Xt+msu2cwwYDjnJcZy?Xtwp~wU z)-qC*U+m;Q+qO**WAu1{ec1`Q)F*ljXoU1n)<^uj3g%s3EoVS8CFu1Q43{CURZXPc zsTKULD1U<2MVwUT@2>{23m#2$pMcwWIVxcAK=*<$wZOH589r4q-gMVxJ2LoilBhEx z-L{cxUQ*a8b&f{7moUr;{He8KhHy@5lR2W=e;fBw5e;t|`aj))S-O@nr9*eR^ey%b z{TxK?GdO^VK9SFdgl}(Eo#pWTjaATxI{exY}74SU72@dhzxuP|oniiyxfeR!Vb=umhz;o5L)WymmkJ;k_! zvAtJ0C&^wL-och| zj$S{#rm^Dx0N>rdQ+9_WoV2wj|5i!f-Lh&G$PnbLxe81(bnqh!Z0RQd5QxS9u)u>& zuUzQ%c;+Cc8e+bXlgh>yD^AHas6YqIFVsj?cZGI2G%itU$x(akIX`~OLh~7h*l~UY zS51hq>1l34$vWWHKsKbcA^#Traq157PORPR$S)V+G^>BCqDaW$C%WKvZpmki2*MT| z(1XRT3pe%Ki)}L`i8-+{HQ1{C+P6W-`_h%R*mi7&rhXa0vfG${5i`2brGtj* zWBJw78|o2EW?!Co{O7efen6obzfx0>Jr45*y&Ijr?q5M}%{(wAR4{ZrXH!%c9X%^` z9m}=W(|UK3oQ>%)?URnMAx=4FqFGt|k|MfzIv0c$tK?XUWl@}7BZR4m?V9fkR)GaS ziT({zj`?d3)RflVwo%Q}H<2oFkqI#Grh~ZtJ9+`6$oOs@@(R{1t?Y<$!(6G`R(m+8 zcwWShbB>dwY^g9C2*+KN+)z$M<`-l>*(-$93o=M=pIEBB=6(_)wttO)4im+=3?R}52hSKF$LL$JmXL|z$LRH45f|vVfA}qUN1}=D?pHL? z{?ZwIQ0rnZYgQ(*5wZU3SyRhWxSC{`n$Na|^yw7;e)Ov;@xLKTeNZ7uVaLLd$%`Oh=pJLNF#4oBc-9R!>g%(oI)=FDGyB{D=?$`XMz z5KV*+`TQ?cP#InV92|n@G`ijR&XtH_;P;2%VG6ppX;&s>c_3Z@P<@$WtY?~5%280!Z2>L8>cYWFyEH^ujbYi)lT=m;>k z`)-0|iJ*+xs0n4rU!qb7kUmCE%}Q3o_r#j@OA*aR&*CdDmhPXRe@^523j|&! z0*sOV^5F8uk)3SWs`{c5rP=~ds&f&?JfwIfVF6-b9+hqDg4iaoF`G zHw(#1&8umsZnuanrhTtTIllC%Ozi4cX33N!jv5nvr^hp1E_CjlE7evMdMdNrAQsE0 z$bkPI-79ec5E7a^a*L2jr>5?@VZ9)-B!AOSo|gjf<}+SK;*H&`rA=i_p+ZJb9pxQJ z)Zfftz@u2dt~#eA(-)GAiW#ng9t@&3rn)Xy3Knc+O4}(PxcrlUM@lL?s13keR)#&O z8X=ZFk`#k-HZfH>@qGP*q!8H1#-T3s6H`(1I*9l;1ITii>P0(Lk_gj7TdX^v?xFOmQF3YcL>Vl>!p?DH*NW2DQ($ac<=ua+scy+Af3F#rlv+R znmJ|rZo9iWUswRVdE|YV%xJCRtCA{4OMd9{<6qRYQ0 z$(GZgUwj?uFBnp*jei}3^OJA&dj?~;B^Qk^)#2sK>#|jIgAF0~+{6OUF>@rk;&yQT zxZx{>!|OgQCbznGK1RorP@W7bz^|l}N)^rvq-$hSGP-7BKF8JQrU9?!Iff2Elv2F@ z^@Hmbfq;y-ec=54?V3+Gs_lIVGKHfQxNMcpZiQI?rofQ~B-0vI#Yb6ZFl=$NF_mDNLWV?F&kNa~8o{5t_pom~zr^7R z{o@RvsEpm`yd|b^cDTs3Vj}joEg%|K<6DD@yS!9|xuMpA##}13rKpX~im!5pkRs>1 zbTwSqcpGo~|4w)WSduvksK*0&i}rj>-9ztB5@DV@^xvN5*4KerL`<6DOG1ljo|v@Mb_{w8ccZ>T6f z&GqTNJ6aXlp5Wr*4GE~2;n`guJ^NHJEMYekXQ;sJPNV|09&!jphcTT?2VL%-cW%_n zm*mP_$2G9_%s;I`iW$~2#=p?(wtjMb*@~w-SVcKIg4=>>&diXGFNskUIX5UWz>ZaU1>L&#{9740llweIWj)y6>&lvZ56 z@~gg@#$R3ptz#bi_WEiOhi(R53{&2&@ZQNh@co%!7K5=54&$K;FxcYQW=n*V2sb2l zv!lObUQL4@#uToHH0eHeBy%xfLu~lDJ_*qC1G*xY#Yl}x$N}+lIs@FUk>Sq&kEXW_ zi>mwnhi8U?p`^RJTco=|x&&z$8lHoS?iLN`v$fy*V`{ACuW0p>ezX zI%yT8!C%Lw8Yf6RL+O)4X>K5c7AK5XYecVWCoA%%r-qBg=G*Dfl%9nn^W6gvnwIk( zHMh$gj?tUCdomS}0h-F)C3smzhCI#Mz0Thjx*br~Hybyfd;a!z3}BJS`$&ZPim!=M z!UD`?MJWt)A$9nYi&~7O#y&f=shDe@cVj>bW7Ho|iN5V{!xa4J+Tgg)eWzeuyp!80 zx6}e#16Et@c?+ztfS&H zUWVPvUqow*_^mHZi+B!hh7qr10zIKpyrfUdEy^LTVp2uXXbwZ?JiH1hQLVy}0L;&uOdPLVQtA&1 zpS5Z=kb3zXOEjql*aua_B&^$4;ts^zFdzQv2?Y8!AXeS#S@((4qeve;cW zmEMp_i0iz;50RCJnX`)D6j+QFK^eM>*kCsI*m`*8Kp@W|tvT3GWoW`}mPxa{nZ+_G z%;UM@-9mvGX*U~atzF$K2T%=E_}B8e5_U^#v=6RV)1Q7j7cY%v*YpOL=YLCR7FCD_ z%|@~K>rWlw*-hi$`KkHxHqpOM(rl`^ln1`veH$O;oz7JF8NDkuAnc|WM8#25wiU?p z=|rhB1*pTqAlN4fFt|*PyH)V?ru8;hYYil%!C6p{2Pi&_9tg6RJ`=oftQc z<(N3YTDHV4;`HshjIxv(xMW3T^}1QQ2$^{%v5iisJlK<^uIc#=bQYdLJOBFXU+M76 zRXdpZbK)y_O2j&<+2{SDdwWkng62jYI+64n9X(~%oaxt{Q0j@P0kpA_Gbhv|`n@9s znoYoh2=}}cIBq~es4fD{R&on>|4`0m=HOj*is8^;NwA@=x$}LavbBI_=O{zPm?L|x zA2kzVb3m_fFtum38mpb?_+kxwOz3^&>UY2Nq3(VhHPdv~@>Lr$bsu#JS(pbuu!td% zlE9{k$otCv35cj(-$ zh)Z3nXSwb7T7ij?08lH1rYJy&AiPTh)!kDXrVbr43764GyPNlP+n~rd*XMLzP5p6+ z53^_7|G4BHcoaC`;yXr3;)T1|m-lNS7Eov_baJPytPC4eyo$SMdcFNG%Y!ER?S31x zIXNNiS7FMWbP^(5&C!&(s5#D>prxRZHB~T|JL1gwm~3>Dl26XfuljSL-ZD4Ee8^PL z{oTU}vMQ!_-{^_E6kR39kAPRt!jxGhW+#N+0VEZ5<9&%n4fj`M`Y-R+Ww$a06PclZ zB{{#iJ>`ZDXDzM(l6H=Un*MeUXjj1%OXHr#OQHj31rPMre^YwVN&ElPZysll_*d+m z()*2J(4W=tu`uzCwD?8u_r6G)-2a-PH=y>QS4MEqxs1$cSUl|Q1w|njqp{mMQsIQx^*Ly!i=`nSi5!XnfRIU zfgI4XW?UGfhnn~|CSHihCN9sr>es7M<$huEFb-@Gg=$|ep8?p%0(unSRv0S3J<`DM z9{=_giCb?lDf?;WOFi3RuptCCtG_coTvu*={fE>V1ru+(g9M)<#x%bDawfIY!f$__ z`3&RTA1U^`id0_k_4EF(9=3eJrTI00LR~`p*c^t8 zk_Hs8e!(qL?EC1W{tSySkyk3m9Y;NuD||3L3f9;S{?pm4DLZ??eop=Ag>K(ihScD~ z3K5~aVD<-Ux1(7@!slYg6P%5{piiw9$Rtsr@%(t}oigk))V20NZ2f+v;9gGQ(JH6D zQCyHC#D-p`fFK-8ufD7>GQl&LPKG-3_Pqt z%kCT;>+40TD;M|e`a?qRG0Ft0AO!N!IaB3gioylhelT7LF{h1{%H;=RV(XbczKUbe zQ3e0sQWpe4NhUtoqSk!8=M##`uQ2-T;q|eP`yd3IJVigEV1ifnD&|gshV=UF;13Tk z2PD1BXW4B9t&GXUt?IHhW#n_WOZK1Q7nRGAV3XhY;sgf#n$}$NeV3|+(Q7z@(7osH zw=N9@h8fhEU6ICS$`X0csy1{NY$2Y8g=pLc9=TJ(QMl{#^tmh`*cSxOPPVG>#)mu} z!UPw{(RZOIZwHN#h)2acRs}$Dg?#!h2CB@xpmeSuaY((02FPl(8-_3lEfYTKot!6!+43v)Jfnd$rFj?FsKHabOpt zE4_98O}Y)8=s;SSd{JSX0(IE%%-Y($%lX$sjq{I($u$*)DkVrROZv##!GeGH33!T@ zyJeg@t?69%P+jw|rp#XPc|uPF7Cj9|xZjd)df-_Jo=_gYI1f0$N@Nw6j@HE#-~ff) zZ!@a-Ya~FX5@+??q(LcR-p>DOz~iQb+g+Y^1Oa(q9NF2zXNC!RPALBIM{j;%6@9Jh z#gt)71^v$VGtf%rtrYwESOj1~=(3DK?{jhXalR9t`x>&f+b2PRT+&@N`EjAF#%Xmy zJF#M6ZCD9Pq|j1}kb+ty+IG4iDi)6v`1UMZ9#Mzgul$-<(tG@HrTpH8;s>FrqWs?f zVMPm!v-tKEQLF!NF_Qy35<_cpUP)lSBJs}U^8s_1laBP({_)$WEd7)VP<1#T_W$=- zr27U4(6VjgTrq{z)sl7ER+AczeTDzBH+&Jm%}?gud3n!7BfZ-cc8pzzt%F4jOwi#Y zDK@Wr@_Tk62={M!F4_8Hf7kh7i@g6qEr7*>r!}hv0gc1u)<4oojVl%UQHw4Hine9U9PH&Wge5bd{QvVR`2#cVx(epN$|4E{R( z7G0NLA?6?HKP0Q?3vr@$qmq(hx$?|2R7t_?q z=5mP)trsDKut-*)>2HD$um5XKuK}if>{0$nyC@hQjV(+Oe(Iqo^>PUOT)JV(`nWMU z%bVXp;UT_u;iUb4Kk$!j)ebXqAYa_khg_jBb^8!ft8YfVGQw|CU4oq3O>bzqGr9+h zN~y{TIjTyl@q)R%KD=TV)}99|3GOaI;ovdaz2rkU&un^zlLlt3?!Az(bC4-dJeCxl1Yd-^tAb2y~JLi zlHk%DfQ}YdBeg+NMcUO2y?&-6%Z{&D6!@#p_P(%6P472t554AjI#!QCQGkjGAq#ij zK*jj2OD9nFr6BVHd+3Al|CTns3)3R`KrrhnlECd1$LFhCQGCQ5=tNVPVfIH!$MQ;JfEov7o)9@zlSG7R8`r<5?Y>=w_`s3 zeO-GW&@Z_d9QSe0@KBn#T4HDP-15h6nIyUTtUfW`cq8q?Juj;XLvt3KOs3&a+LIy; zxDMSG&U&|PAI4ER}5 zi&c_0;0rl>M8Bhl`%+bf-1f6x>(`b#N0BvCIcT#rGAh%cIb9pi>=S2)m#XS zJ{3Ut2+B;Wl(8;2yVs&cV{drcPurio#CoomkO=RdF>ACya!WqD0c)p{&a}B{7!;4A zfJe;u{z>0sN866oU3R2^*%JE$ZzekU%(xa0(E$K8HrF~Gx3PyJqX@?O@Cnf2w=A6r zZ7!ru02l&VZqF8cSyY8?e)2euNqKvfu%zqiJ;+eT6^j5$yuN+~6yr;%eeiiLABOAp z*LX#`xi-soV=Cr%1>e-@FEq0xk@LPdsg+@39T`%QY7Y{Q<~Xq@Y`bb%6i0NYx^Z1i zpv8rzzcg9k4ZjcmBc{hIMmi(&d#xSSZfU(4J z9i)fLSz)fSSjE@Wz)P1bJfkVA^G9N@fFdZhT2o17Mhku%3-0I*9~sq#^S2!pB|~mi zJEB!aLY;{8t?8T5;q0GbTFpzz7WE^>22WI>aVgvz#MF4PRLvS7C3ifW3(>F~JF!2z z!cy!7ZE9$`qjEdl{5=;HE1aKlZ~ij=02S4N$G@-&_N~sz@f(IIoF6Q|IdxDA0wux}8W*UhAI`Ku@g9$VXBik~~VcW96ldCF%rz z2C`~|DyH2r0H{MChVz6#*DamM^=|P5p|(^ZIYTYJC^W2Y?+=hu^GE@IgGWph)<6uS z6(T^1vs;L@7-=4g58;$QAn=2LHa7E|G*@>rqR&GkyKBg&dnX|Jh;=*qTUVe$4F#?A zAky5@cIcVT+|!^JsgkyqsE(M%8L)m_gQVa;ZCvZa{UL2VB2`N=?|p6LHRJ~g4LC4* zfXw|aA8{Nx_C^se=o%Bc4DBnz6XY&_D_U)Urv4m+*G;UXalF{eGa^+4zhRuH)^aQ6 z9O)R}Z=TVbhHW-2g#%xU{f<2Z_|S&zpGoUaW3Llkw7gy)|Mu49M^J7d$YYyD{gUoD z434`qZShjiG`r_0hc-`CD$afud;y5wZC8;hfG4L*D1o1`jPAE+C6osRw&a^d)5yw( zSxn9Gb+pqwKR?=;V`8rhNjy#XXqf*WIRlsf1W45miZZkLg`sUn08e0L-yDeTB`QDS z2C4RGMA>v-yQxWsrn1s z2MDm31y0lDV-oaFM!>c%JH#(;aF40yO?$e|9rL= ztc>7JjPR*}I8=J(F7@KMU&pk;LNLFv5Q-#S0ai5#8fC43J~aXU=RkRQ z;%%$BgL+rtxsCGHR7_ytDZfCmq85JjGW%dG!)XO{^FhtzO8xZn8ON6%-nv`GvorpX zxfYGG)M2$b`|s1w3Bs5e=1{I0Kp|z(#Eym$Nxu;rGeUeZf2{SLHqIiE4y~h$E&FQ% zEhr6h`!U6gT?qs?->3*zk;9XtLmC$hov9~j3_)Xhk-mYzR)QA@!g-E^G2{8xxr7r% z=DD&Q7YZl}M=Cg38Hem%CxR!uR5QIi#-oLAcFx(q0;*@6B0o&z|gm#Y6?ziWqEA6khMfjuU(l@rkX{G(Q) zye{+~|9tRUeBFQD{n5_^h78ZKfVHdv$=IH7ON?_JeZHV1J26bVsg>>xGFa%oyS6^x z+jE|OscG!UYFrxOwoo<+=G4E2Uq$gpqJ(F)^4b>qpXt_-c+R@FDgfv*=exZx$m4c; zok5S(+AgzOPx~6ybo}w1n~Q6s!}??NQVQ4|9R(q{lA7ItoTTQ8$;|5)s%YyUh@R~d zQ?adC-OPr}c^kz0dvt`bh;}BaAe0#@lX7c^lOHx>XPI9L)4??QG=8RlH(8j``Xh$|-bXE{KS1J8 z*|Z%JhRaay!9Y5dzZ01WaGaTx7L8E7qeA}+e+&$Y^Rwz|>B)vT$2kHbsqk((PlLSj z5pGv%u|b{JunA>_C5l5idROF{V$^vCFrGWRyc(Nuk`P~Y{LLI@Q0+L1xvl}H`Ao39 zACH(MI(3O@b;UC=l+XYUJ$RU1@9_rQ(cBYg^~6>$PGL_qT#ukgMMT!JVULBw5ouz# zQC1t|yl8iYE#emaQhFv0`4Jqo)5uYjH5+5vz<#x2;zk5DN+8aDCf=+ zk3*=g>Ck@mqMb5#(gbRa`Ch2-fJCN^qRExZzDKsE-V`=x44MqHM~)Segj9sFu`W2u z=<}IP2?Ss7QV!#L)l&dz)Id@>m@wMC*|9xCMWVqon1tt%*+HY#Jad$@`jf<_vJIQr zkr~(x{gBazLeGto^;2*Sw}%RkPg3y%$m2`IliOd#wxPl$KXE#k9ILEE;(jR3(b7SW zGorYpm998BDR;~#4XA2vbEWS?7yXmYa-9+UK6q3bob{36Efd5y&vPJ^w{k1IE4g)U z#cY%J9}#xK+o~)n%S`32qhx@BDxHKfw@SlziU)vi93DOuEF7@Vnaj3~ zNEeS{+svO*KV$P~77T>Niyp{d{T;UDqi*8@XQ;MyiY}_Q&n@b)c8^tqVCefWQLO1#8{X+6dwJ@Km~LDR5*fCdNAT4Y^gq4+FCpR?3zO z2*FPYVX*n3pxhHVeW1sh5u5@Mm5d%iPb4!luzc1cz0qJhv_mx&e&BAAY&J-fAG~k_ zxE2NM)9g0BSebNkvXT!z%aqetZ^i;#SVV7YeV=BOa0%$(1H7YwlcIyp|MsrhU$&ku zE^a#m4EC6;cvd=CnD$lHdp@vm#*NHG0wbqV5L5q=FbT*r>&2FY@i$p`yISgn=$tzG&+AcxpwAubyJy1_vr?RqO z|6(T^zuBZKS)vej7$JUAzV-{A5=mv$IQe3X6z;oBoldqVmTKi)zs{;jj}W)6xMf)R z1B8A^(_ChO=MUDDL&c5BDvlfG*vl0~>7a4L{YcK%yNmNnjX#qx>Y2yj97q@~%O&UQ z?PJ^`NzFz?=s9i;>31BCCV!{EBor!HFiO^A@W1jalLzQ|7us0vhQ1*lAeK*~K_p>N zEXL#$tB2F2moZ~6TTE5jHxpq)eOUHUvO`no+S^5JRJ=PN%^koAABX8oOv!{r$Dv!y z3W8(Z-Fa^gDV@SBI;*t_*j&5W(D2pZ`|b3h=EO->Buc`aq%P2`s=K!1z&RqQFlb5Fk|Nx?*J%T<&r4jKWG*28(G(se zz|lZe2h+rk3b+hq%#g)$+iPKAyU7EJg6CviUl43*az~a`TA<39{Y&<(1%8#|S^B-j zHoNlHoZHu)P?Q?6O>V(Q|$ zq}3`VEWeb9`jM=LSbhpWfA+%@P6n@H-+n{{R*DLrBcPs1Z?@3ir;^L}r|0<(W9KBD z3YCxL-vKnbIgG7up1=8NlU}bM{?bsu4Jig?g=)P1bk6@|KDPhJLR^0#HBD)fa}K*KOVxO6(my z4{My=Eup`RT0PJ@Do1LQ#x?b->SQ~ohOp-AaEVB@FhATk33Z&f#5J&tTxm!K_Y{G-5dxJnUM{1d8NtL%pu z6UE89Fs<~W659zLX!NS*=fHOq!)qFXAScP8r1%mOG=rMJT|h2JghUz`XF0<$*qg8+ zj8=^*7y81z6f31wjRF@mOrI{{?^9Fg9~V853~r?l=>FRjR@u#~a>|`JiVc1*B!Sl1 zjwQ9>z{)3`bJ+~kaXW<6tecD~lY8*uzkIHw$$}{rNJ&iN_x&QlyQdhVird|riT0aH zk{@t0jo5;ct|N8#a2cpT^V0Fkc89w|P99t(h>RgZW8*IU5HxkOyc$OG(I|lSOTa)C zpLIRzUf!o&bquO-!mP#+&*{Uf?Tw#dr=N~!+3yVssTPB$aK(W6yjSi@=#RvO7?u_2>(rUS7_5ek-*7g%fD|+Lf;wc>%3JGS)yUy)wyG3a{px~ zNln58%{{x#ztF*9C{;Wf%j0q>(t}9-K;e0p0ks~f3&SSuP`8G+C@vrI2OFSNj9HFIuZ!JC0Y`GBEZM)9(7|`$BUGFwW-lX%M1sG!u{8A%evju z<7&D~_E@tb${e)nmz5t+;2uTMRl$+r77zZXsoOuE^HtlPQ!#sXxfiD*h;= zqd8L^R+GVO#DIs3iUs^5@rVum* zTsYG`7irPnS2cgG7X*0&1oOQez?%=(&^xg=4Owqp8y-Z3%c#upwmP9xVR>11?;WP2 zw5Sv{WmH!$=f0vDlafOH;OB2rlk@A=DI@ocVhBlBT80XcA{xi5QbcSujcmrsE>4lF%;)j{2;~Q)WUAcZdlf&q6Bh z;-B0-AftB?l9)zoRZIRrZQ4R_o4c_fzYAR*$G!*j9wwh+`npsr#7sliCX4!Qj6*0IJ;(tF9eMzHbK8@qb9pg2LRz8ehu%*NFZ%XtRi?x4ANdX&&o=MW9{x0?SnUZvM)DWtdf1J^@!Vu6)M8iyi zJg^19xuny){3(LVL!rMQbbDFs8@&u&T67-Z9%4}@N>$!t{KJY*$XixYXAOol)l^v?d_fE3tauXRk_7Rte_qQ`WoLL(X zV4zgu)qkHP^t&+_`B?IAq4vHA-}|w9ao`1bteZAenj}>mR!VlNW+#oNfO?E_>}_#0iZ^GX*0de~U$he;;8rUf^W1XP zReCZ-+0n1)-zgzXbj1r|jtMJ1)T)=DPeR22Z2fXzsfVoTvv3Q<1>&=8`I&`^Uk=!( zF987Wds?j_{bZ#$-#@smDsR&Ad?E%o^0E4U029?@c?{T<=rV;kVvD7U?HNH0|M^+M z2BnMR|DqNh^FwugjIHs3;C&0s>5KLtz{lU7QBL&tm-&h`_1Kslm^jBp8aHs3qn?*f z5*8H9VMN!CQ7RfBRr>YFwO2m;b_65W2>z59*kD;M>!E!JM=C+n1jP{E>t}8CFD-@7 zRYX%)zvNS06|KiUXP+%EbzMewO?vDpA0%&eqoTI_;uCaNBT`sRrGgc*=?Bcw=FUi+ zw=ULHnQ7xH>|Zie@vAtzm%vm&+g%|_`AL9j3~vFUrS|Jza5P0o8uptmu*Q_R(kJI?RrDIGZ%x2Jmb3nGv7;OWKk5QUb9R8OVC*E2 zJW5LIXfexmINU}ix{w&{u{+BE%JGAST@kz32F-ke%X}JQRPV=%-1T|HmW^In%39-n zf@%>~I}{Z!F*r_SxEL!5XKUWph^jORC_Sj$DTiVt+~O_fxA~ns$$$M0Q1l^mTPf0t zGpZF*{f&&V=Dn(3nzJ+!eK(swoHj(OE3QKlw8IEB2WaNXL;P{Y+fJS6tvf!ieH1kT zuu<}yzy(m|2mC>jMDp_hu(rf$4H0rQ&uF@Ec<0hDpQ9fD%#gc^)7xwe5!9m-pu!jd z57|*(?Rv&Q$n)qw<=Mhtw*AH!_IM2huGGE$tp5`kjD7$Mt;u)rW;>}>bC^^wO+y*q z!E?{AQOB#phx2vq-=dp8^N_v~Mffobo!N$shB$n|G7X@T;y z*j8RiJ;G@)G&7DRbHOA^rKAY%^z`4N+0nfUR4@!8uF${m>+Bdaoq7jMbG>CjudO2r z=pdsulBpf)@SdH%O-NCEUy-@-3wg)%rRpd~Hn%AKLh1<$B>z8>brJ#&i; zW7v~r>f54VURLe_M-m`07tLQ8yiOJk#ll~&-<{=6^?GJXX<=Z}t#)bVBsX|G=JBQa zoavvX{{rNoiu9o#t#3XNOKAKh8@>z2Ec0uwh1KKF*&=kF-SneK07u{B6Y7%SqbKS; zODEx5lkr|57Z0LL9uTx}oSxYjl8;;ciS_N~&VeCAvZ`+&m_uoAl015qZo~oKsc<<2 zK+lw$U9>qA$W*tPhAn&US;v!I!G8NKmio1j6yx}Ca$0$~{www#a>jNV>)!AJYP2O( zytSYK%oI1vp=|{4ouO){9?V+?D1KUcbG1wBBXSFBJ{$LR>}&lv&~jF8%KOc)=6lh4OP;vBw)v*5fB094>4m< zBN@BTQEmwlgivK%l9P3bNPt#z#Mq+A6L~+>>>7I$16$T7W%0rGB|bq;2FWVR)B@`E z(>0%I{?FkX#J$=z(X;gF`7-MKd2GbVp3YU7PD(lFAG~yE98t9iTk7WncivwmTvrI9 zS(jubF@J(N(cH&iY8O~w1?B}6tukM-kz!ALe9Xi+779+eTd(ht1wYH9(>PDn=W)<; zTuX;?AH`G(KUrHPVd9MYn0Eh2CwRS_EO*od?{YT<~Fbc?2dk}Q`M_?4XYb=RoOM7 zEtvXwM%x~xLD%%M@^vf``1Vnpq><}4bjwE7nh|?w$5$uN6r6+#7XKwn-(Aec zcu2#kTwex>;c)p~Z7L70PGBc;soJC|Z2aI~Q^sRp5Lf5NXtDbDZk)Xm)TebI{a*jM zy(&2Oju21!OoKl6fTM7U#;s*Zt(0kJtvM^e~5SnS=(6K63D^Z;fFq%J!N= z^2HYy!T(p$HNowG=4J~ZvI|=YQk>-AfdCDxBn4_gl>0_-;#umb+ z&DQJXe~!6^0d8or+7LuFx4ER8!KUD zRbBDn*ifvCeJ>FzcM3G=7calI{B`Tv`S}Kap$ol8pxcIGQQM2Md$OJ#UDHohtf?`7 zO6b5V@P=|&i+bW{Om3Zk9l=E`%N>Q!$$uMyzy3~GI_+)mFERI+q@_N9R~ zThShwd616e1wedFqOV11WkG!^kQIOuNQP#|RE*|#-~YIGp(JzaU4PV=kXB|%*Zkd3 z`>u7XqcoH>EtDLnNzg@>`7g1Lca(+Mmv5bo@k9)_9`|2L?!V`FmW`g6jW1)tg%%>y zHZ>PiNHI^Jc$rg&{+gtXP~$h;{mN*%QT>5LG>3|}4@vD5(q$aU#HIi?aWis$1sJf@ z8K~=Hwb!LS()BO!3_E?zGK)MOpY4lnD;ZAY;R0_U>dWk*<^dMFf45K-h@Z0QN*uK{ z>G!KbK4K{CvHAWt30piT(ZN36em-0wSzLrTx{V&4>`i@DmJ=z{CrqR5k@(+P5dr2_ zZDHoG(-_g#qR7wA=3Oir_LYgyj3_nosSQG^CE~qb1yE97Cuy6619yS3jXg(1BUzGIxv)u^2ry(z zYb2A_PDPU%n!uEWf=J$LLhxX}tB2_^ZF}gKy7actG!D+9K zuft>@p|bb}m8^-EPbt1yuvb)Z)P7u)0-uLACl5cxENwjzNPP+i$rCBrNBqSgKrXR6 zL1=~v+zLcTiobPAZRR=eBA?wpzqlFg*~^)y)YLL9hDrUs+)YJSrP8Ao9cgM)?JsBG zw0dSWD2FoSi5%H0F!tz0IGg z(*wEMe3Ty9*r&BXWJ72vz|fu$qU*<}hnHL-%Ck3%?aYVS%;p<#2u56?QPLL-Acw@T z(C(YU>LXTv#U^7ycUPlc&-6wWV!X-7J26s#{wKA5C6U6UvS}pW-~%1QXp>ByfS)XD z{9ks}URpQ@I<^@Mm(xz?IwK!feNY5%^5+79ss0 z?RgFrJg3hH4Edhqs3JC|jrEo`?TrN$N*fE$hb)vuaMzG;|UoS(U?Zk|FsUsUEl;aIo^@MioZLpEQE&ZtD!EUB-g}q%FQrG zt4c9>kA*6ZuYdKH=#;E+|mkrJA8~yKL3nfS11$7T+ztl7uW!duCwjlu$64fsi4dgAv1g%&$c;e9r z&R(b%^_J%I?ULh*VPNw~-5RiwlW|Srjy}K>{d{n^ZsMgF;N)XA)z$196m`+lG5RTZ z0ZH5l5O%<+G&tALVE7*5+DN;tVj4pGskf@Z8y6r$>q6qxo)oHR)T*a6oxXJ#)s?Ck z)8_8KfzNZaLm=c167nNPis~=#4n*VgMx!|WJDb>$ZK&tVud)&!&b-wTgAIq~gixeG z0F#3QW0dcf@-v@O+@!Nb=8j*u*0^4z3mB*dRUDKb`XR3i55&9);u@a%!Z_4H7ioG{ z)HGkLqs*p@#;zX33l>J@%(+2w4|qN?v%|ExA<8>cEqJ?opP+BU;pPLiTMQ3E)$H+{N$0-rKC^mnRFvM= z{I2cO6&be7GCoDmy?X%C!`;+a%W###7ih+I6GJSBQ8uIFP;$xPqZruE+>{@0h(UT3vYs&6{eEGdSOFl}OmaQpfV3}9u>7TQ zSSm0*MvV2Yb#CUWA)rry-j4qt+7<~9cWq!B6RQ%<2heZ&+~dVHZY4XB<*tnOlE6eZ z8{=f<@+B`9#nk)urtDs9MM9It!Vc8EfFGi2?`a|gog&aU+f!Md23wQOD=kQm<^Qzj zqUOb;^8DS6a=TFV!Lk{(X8rqP_m!)FgHMyg-+G~5Y-N#(sS@Rh_>BWmAU#tXd4A-w z;fwy_N>iL%pEbV_t7=-VC7r7=?)sYrXOBM!AfhL@kNbuefz6s^L>)>hx0)s)^gz}8 zbcRA>A+!NeYvhX$&>QbrH^{1-FiZ}iQ{4JvXb<9FjC91^#UW1E?8%BoR_M;l4ZmFe z;&(g#ryQZqKC*jH#dtcX;Ae>70s|Uvsn7X4SmqR5hWKIhp%TlGn=4DLq^hOKR{O?! zCXJLtn&2iTN*IVj@~@Qfm=zsA@+s4@6^ZM5?^*u97}P+!_(TvRxZ>Fm`>$hK9ds+V zKUfoXO0P66Pt$S7rFq7a&ZsFT^({ImpT&9akG+S=upmkk1iC-*&dEGr{%c@i!MleA ziv;IRM>;s->mHuLDoxT=IpcY8GvT}g!3Di`U80-*M2^k}2-g4@cO7o)P9Ma{MtQ5r z&#D%+1u2`m71RkzfbqrA2+M{YyZ@|-(P|K0O-~!R_41S3soTNK!KKL35X%8f`g>To zSi95@oU?!7Vr+AJSp@Fg*4 z>U8?pEZ+r{02fdMwecct!1UhN(oJ{1jnSk_k-8a(Z@FcD@{_p4=H1?+gI_naI-?_! zM}ZIXJ2$z2DG!;%W@Mgt%a+>uz_mAuROPea*MmNeJz>mU#z{oA7Faw|{{?$wfUmCP zXUXqN|BEal1SSCaNS)Qmw=&<0ywd|ZEhX6ndP;~^S>-(a=WM0m0O>1LB^~qO7i6{# zivASIEg#{p7Tn%Ob$v#sX;t%5sKrzeY!{WtyS?-sdfNW_&~?@1pAuH_$K4WlOlZs; z^F;!hC)1^r&~?(b^BO4 zcK?Xo&7-n23lu5a=DsSf+0C#P2-4f7rjFY*(QjU5*4(H>K@;b@C@7^?Zm35~LLbKU zp^J~G00^w444z?<=+autli{qV+lY3XEzCNtVKYnW6rKI;OH5&L8hH9aU5a8>J+>Rn zo_0|R-ql(~YH-12VGG-@gwaV2IPua_`uwSv6r=iyZ9c(jdZ{zIP=j=+T%YV715O^m zfi~d2DAOJd|ew4}Ks^g$nc{Gjev~Ex(Zqdor@%IJ~V4d{_;uW|kA4n@O>a$!x z4S2Zw6VU9rx4viLR4{AdN+_g3$=o~tTHxQ5U5XmKVCb)8lHuiHTFeBZ0#>#s9s-0z zcbZcZ0QlQrmId#H(~Ot;(BFQt;zC};v&YpcszVpje?U7a#2j=Q)74bcwMqEOz9gMi zYXX2VFqSD8GBD)Py%x4MShvlH%qkpOvA}G2B|6mo5?Q3-BT@Q5s~H~=Fc~tegTzt1 z>`t|6W>~Y#&58rRAn6cC!LH8&GlR@kT3@wv)Dekw#-n1P081O|7JxKg13S2S<5R@$ z-}%_4zn4)U-BlYRpt9B~Ld9%G*_ow-jcO;~4Px4rhPq=U-~{A!{_=Gu4OqP~zOni6|&T%~TVem$%RLlYYs zj}If2VoiTjZ!^HxVCH5V3+*p@MG!u$KP^aSr1aA^R(b#WC)@T-`P1~bja%MFql1wu zZ~$%gClkWvm8bHNkzK^LpaL=tFw0B<+$j)#|1G!k&hLw_Zm}&t=TgNZOg(doIlF~N zL1!MGM_y52u}wTkSac0ugP$1Ivf^Tg3m8eLYd%S7`#K096BJkYk^eJ|COQpHzt*>m z%&$i|s{K3NLOF8`K@IH@&QvN@^^EB@Xz49<6Y%I^%SY8n+wS)<`)-MNu-iM;5sfef z*fIRN1B)hcxG5^3?VSG#mCqAdr> zgk3L`LKL9zhwxK)K+-Guwc4-|eGrQwid}EecTD&YVuI-AxiXIFM{x2y_pv_59@te* z>Jy)3xyEq?;jloR0_-W0FItT!wsGeB@9r88V13tKuh@*qC6U1H%_yp7C$kvkIM~dh z<4a7U1&IYY=X0TKzX1*hx;l#9Dm#+l&)bof2Ku1YgWW?-MEBW=l-C zaeU1F7%!hBWiWW)^0UkMw5?Jc&1W<4ZNpUeuT%-%4%f%$#?+v*gSCO-b#JuiB>UYn z;a*aJX873ZY@x4?G){~vv-fZ0G5)SC{d2Z`rM1N09i_V6WJ8<~2CVhy26wRG1v@hB z>pH-C2%n$oNq=2GqZj4R`!m_|L|?)p9>F&5?`y2jN~BUMZie|p=~YG#DAjC#9lt+T z5Zdc4^4%e3G6trp1Q-P`jIrXd&~3)Y71Pz~kpf~bHLI)-(e*{b)YEyKWNdz<)KrPl zXS-riZ;^J_&<`jfuxS(k+LQOZ3IMIs2#}3s)ciEP=HvmShxsN1wwZ+08IkSG-=Xw{ z{bOy1aXPE-=PAP6r6mURM-*$7ajSo@e(oDF}V+nA0y@1*__NXvzQ3$`FgI zW$nKDDRi}NqWfig>))*SW1tdI--k*F-qV}^E}(;`Ny@l4iL6sB2h_0f(()>tPWLD-?f9pJ&X zIH%AbyI>twLnSyfD`$K2E5!Ul!tC_Hfw)k;p#MkIR|dryEm7j`1Pcy>gux-WySoN= zcXxMp5(w_@?hxGF-95OollR{4{-mgys{7sBcDnnVFazKf`z2K?;gLXK7z@~aX7V>E zj$8X%am|ow(0F?aKIRqBUa|k8-X*lEg@uT1B>=IK% zIR7k7oKWG(J5}7`fKwS;jp>UY?wF_uY&l&rQ%-IDJntSnv^D*#W!BQ&hEq_YX2w@} zkl|fZC_*n(v!`G(VTVuybmucR%B@Sot)M55d%|G5_JcwdoVx%5Mpt2C51w+tNtIEQ7UL(pB-D*lLTh-iseV=X)Oe>;azxwYg1Kh#-K6ovo!E+g^eH8fx8V_9U0?5xrIS@X!~y8{CDA*AQ= z$^GZe;xIDb{&@R#~wy>EpYXMLTu+x6}(YCVKb&&^3cZ zwy}o0m#4i!`w|f7+qkF#eKFN@wAvy3$Xl!s)y6=Zeb6pMcOGp5#G4&07-kdc0q)Ho zFpLQr4OW<(Z`%l87xygK*7@QtdI%)N@WIIg8Qj=?6|PDt_xIRl7LCa~{u+9{>@e*^ zp2aNT>Dn!ee+zpd)961U2)vq*(@8vldKpvJgsn*jnknR)UDJG>@nr@VaG-_b2l}hg z8sn+<3gsE=Kv&a%m!wE10M|QFO6{s0kEVA#`{lIe29*xTcb0cADwNcEQBdd_dg6N{ z5Tcy;mhA!^kz-jzqy0y&l)OsLuXfrmBw`d9Ozh3$OPx#?x_My+)NGUiGk>-{RMqxj zvgG?5tsGbd%KBYs{C2U3kttJt1^MXIj8Oy)(;!y6JP>4C`Jm%iGCRl^szK5)zJ2`<+VQ>!=48sfGDa9#o(#6xsc~6mul1 z!W0S15;zBFCh%3#bO|RG$UL$dJsC#f`Mt=0ChLm^?1yG%kkxf^sqkhEXof)1d>8r} z4*q={Q+o@H4}*tnE!4t@vzmG|HUDQTaG8^#v!SWl4 z$1fM=AvVzF_LB(|huR2kI{XA_u3EB)hwv&?%(kQ>p7WC*Uc!=JABW;15aBBIeqLE~ zf_}tou3Iec)`TQ;$($09mWXOUL6ADKoEiNxv=DS~@5Pq=3r@QVuB02oJ4UR>+c3`; z%bNV22BL-$J5vEopR0|;S7HlYg7M%@#6QAmw9|*}v2e79ceM@p(@_e$w+TS953_v< ztN+DWwC583^op;B{e3bgKEOCYU3JSyx&V663Wm=f|Jj6bwuc7dcmO&Rorgs}xVYE+ z7p%Z{$eiKF*<1K(t#fYgi18jjE3K~*`$ulPrpJ4#A=WyrA!gJP;k=T(N>_zY0tG*h z30}4fUWhbluZh{-uil)yl&iiK^26hrOc#&#*3l1e9?%e;?I7zrSL}S}=9fp$X+Sf{ zr6y0-8I{Bb_W#yMhyl)daK#vv0|Q8j-!!x(m=~jR!0b}vmr;gmJ`eCa=sUj8qRt}F zI6%K1S8^qTjK9{j`@6%pK2<1`ziInU<6FKaeuU3*oYmgH;DH z#&30h8Sseg6|DjWIE=z3P8dM~>0J1sYl=xY!FcaVZv-%6!P;;bP)$%svBXIdDu@Gb zEQKbeQNnFET-8Sk?ci<_=2qXMblk>V=Qp<^MG6D(D+5<1lh|B*-{IRRaNbPD~N6GCXPe{P~{ zND7u&5(OZkr;W9>ZTYlI*}=l*eE$MsVipJ;876oq5WP+M$A@e?&IxKreFK`*x6t3q z$sXpZV}UX1-#=eabY!)V$Ud%&+x3hhK}H9p0R_I+d6%_x6-dKt%JZwn4E zgRHHdu?xKkPZlm%n%Oi#cE}V z{!nJPQMS61VO|1Pt;JG!4UKc56ci2VGM zx!j@lH|AyF0|Crl)5tk%YK{Fg1^@G)?FD}(sSKt_zI;|`WlpM&?6)f{5_$U*sq!)Y zBRs_!93`g%a-F|N17wq?Lp@lLsgP2f(pQ8RhY&L%=8s-n|Ge~<(N3V)xfxWl&=Zj! z?Qc`@gLA3mX01?|+pl1B9cF1evdw5|BU3}Yfr#MqH@MvuZqUnbI^;dRVwzmnI|VgL zP&q4RVmD6zjZV5y)PGr`MAn1mb)H07RSZLbU9uM?lPP7=5@#qJfy3J z?BMY%!&>9^-A-Byx}%;wE(*BJT? z<-7gt9@&Gg4BM5hq(fs#7PNL}ts7IZvyv~?VUUEgW5~;0_4Y{sRLOr9=BwkthEN1g zY9CAE;%sDm@9zV+!m?L+*?tCA&kRuQ%?{oNX9@p_!O#%aSU}D$qxa7cEBtogEV|gG zsUC_B?&>yxOx$8qE8GfsJyDB+rbR)O=mm%485#04h4@`BxR9l%3FI0qZ;J*eSh$f*E z&Yt2yL|;m(5`)KGtp=f;o^KfHVq+@1^J}(sjCuHFf24aiJ9UbsWR@|pf z{V?8icH~vc_6s&EO8K>!o58WCW0|$7#EXo5--_11zQPG4XqhO*-FOc8{Mpv};wo-T zw7qEb^u!BdqK0@o^40zjD-^Yp^M;~acZukAiz30ShWyvXq=#rSXm<{6SF*&Oce4VY zx?*6C*L(bZ`|`83p0k?H?9|5^9k&Uk)WXS~4GuLjxj}zl#-c61!<fiX^h3ioaJQj#Qdd;I zR7`Hb4Cw>!4`^gK@z|(=YD4@gu=1_Vxt$OXRW{(e!&$PW1&S4Dr52PVx{~EcI;}q- z@A2cdm5yf+W-%~H%-F=a6wX0ncBUrD&{3>N{qW@0flfOF+lAV8Fm5K$o0Pp$28a05 zTXWz31{Pa2W_Z&DA#O5!yrJ`qUFRNW;=3REK^J+!<61(O$+R8L$^P{R zI*s>luLy-n(s19YwX?pxPT+~BNCy+D+SuH68=#dcelf6M_B<_}gEYi>#mE(~1JA%kwt!XFGfCx$MdE>eg&B9(W9fdiiovMeR2>kcmC z02#=17YfFcQ}zB;`Dnm}v?6mi$h>}n%r3LM^5?O+P-+zduP^0b$&_7W_}XPEsEfI- z_YqaF<27^6s_GgV(dLp~Z-=!w0O>m8>@B#WNoo3jDK{9z{g?2@un>*qpVznjX?Iv0 zxJjPzaFwsab6_zqv?~M-&m_OROeqk7c#!a_=NJ&d%83FT8nS%JQcN{3D?h*}|3 zk4)G=c^Q}{Te;4eBn(!elspEmdnoQb$%c578f>E9g{{`Xw5vVcjj&K$mMn_DBb8=x zl6i5xFVa{Ia@@VE#G=se{L}=dttaWaMzhY3*UD^<-7ULQqhn{;ogzj*Iu6Gjc23XifN-Qm2f_ zOm6YtEhYcNsy zLfV2;CDb`3S#S=|c#OK-<6KrxApJ)zpVz!0q!$7rO=u|N60zTRKXe%8+NPaHks>S9 z>ff79%r$q~B=^I&%>S@k_qm{$n<$Dq9Cb8!e`((>XU7?Kab|Q}9tfzg-A%>P=JFdr zbMWJ@r+5JtvAaCrhWNITapP*if~7=J;{3ig9{e8>U&t;3te;}c4&Yg|l)6EWytl~z zc9h0&!1?BuS!Znf0bZU*^7bKJK2Sj&*V&{`{D|54R4bw@!xd7cy74rJPmSZ4e}%GG zk0T^->k1SZ3)l4tHFpg9Z@0O@OY)(thDlB@?h4vB66kR?2|GaTkjg>|RQo(_pj5r; z(ot0o0~w;eNDOSTiddZgB36x_4CertyHOETFmY*k#$7}BpsgZNieaYt+a9U zNed|I!ov_E6Km4E^2+uJ`hC?Z0o`Lq`Mz--B_fNxA!I;E&DXB^_|NQ4_NWBf_Oehn z)A3sZPBRl^9p0Z0*K4>&JzP1GiTGEF^KH|Eeo@ph4hCB)gR&4FKeWt!&9+ZZdefdh%;^)NwnRLAdcGY|J!vDuuad&LI5!()gm z5ZLUjWBufxydi zk@&?)r3pt^-y40K1Sf@coHGGdb3iaosxxq9GoT8^KZY!j04iFa(wpp8LQJLVzGqu$ zkYjU`-Eil7d-@Dv^79=t3|f7@ySthhNIQ!WijoL?6VjtnK8!_*=uf|-rc4D~^{tzm ztz=aTY@Mk*+r!eIBY277pcX3J1PyBN5L>Ybz~SeuKZkvPUNp@{6m{V*K)C*tI0d@R zq@%*yr9fkymcW~8uqClaw$weDiWrs!>nRzXc|9X#bJ&{#cPrgb*>QNA_rsiZgnB-b z`avHuNTWV%M}3eN6^E`x*h`6Om>YSw$G3J$bhd&mccgGZ29P-sK8SyFtE2CHTZ7Bj z&9hn~vVd|&-xh#%!aHb=;?hcmG{{7syrOKpURG95v|Jm~91zv-7rErFxh&rYazXm+ zT-;*D%oIEzbK8QBjf)E9{u`AdsAd9D!Wah9w5Dv0(gA90Ca4i|WN%d-ksCugfFo4{ z?TkD;P>CiPp^Q0lA3cJRMkcwYwZ|q2Ax{kx3jK!9p>*lA46c7-&5W6F z_R|0&nNcNgydV|Wbh`HPCg&tu*`WXtV(ai!<}?$d*Q)gu4@}FvVy3;fPy+pjXTz7e zik`ag`f-p(5$GrM;iB0f%34CoSi#E|Wvs>M3#*SCeQ}mvJ9lIE?d^D|&qwFMFOkz# z1Pc{$PYdAYC1(_n5ayEswe}cdR3QW6;52%P{2iZ%QQ`GDwxjA!N_|HQKSUX^Il>b z`lwc=dt*RigMzctf2pocL#BdFtoa%+oo#^XwbFySVPPsU`T z%sY@I)bITS@CF($se#l7|2=6CShNY*PZAIq;a?_-F(#>+lt#o01WbWNtQYKIiyODZr+j{h@3h>ihz-1=}^b)494E#{#dL%4+) zg{HEgdZf9yn||%}aev_+G?diY&Q;S8udQUVh^DqyX8g|!`IFj%vJ?J-Q&$vD3z`^a ziA-J!Cr9^*G{y($g&E(UP_qHnf*u8!g>R($by4L~5zh%+-vH5(_kje-j|MK0&ZrBC zqhWy`BoZYB19Otr?4$__U#>8v%oC9EOg5|?6)$Z!<~uwyNMf6u@D+Nh#*|(XX^Yg1 zVFnc8JS-)nqko*|Balgy{-C2BUJ|*-Vj?R|C08rV`~nOi?>l~DLhQ#!RS^Wdr1wX9 zKu8)9m5plCw|d!YVedxmIz11au8e>#f6`RTjSPJ|N79?L z3{vilzlADB3KAz|Q}Z$!VwgZi%5dBM5!<+i-OWL{KL#bpxpYiY)BUJ#cq3j6+^IYt zWCXYnK|=mZ-an>8W^1-R44Q1fn-zm!n!q4n)$F}U90btLH#XB^)Y(*3MWox~R~JK6 zO@zu@hIrkxdH5zt%CTt@w`c>?Y07>z{wOO`3RiQgKppv(sF`ala`6|!fuEB3i^-nq zllRtgg3fyOgF#xGmlY11%QcbsN9VF3QZNi0tr*f#9D|&i4fTg9$sc2No4gW=1v1ms zizEr4G_qgwFTzBzp1=WJFwP1Hk`E5>VDNaDbQ#43@R326cl7En)BHE)+)W8d)mGF! z^-}-B4I~_tesjHh0kfygEmn!u+irGW&k)p+@b>l$1PqFgF2AE#^1gUEgA{a-2SI_B z1l<|f9HhW1?0=jBE(CcTShh(iQEpV9wzXhfF@$uBPDzY8PP5)}sLne0a1MOV zIlUD8j3`I1TV&0R&JY5x=6N<3tZLwjRb9L`wr$c`d1O)0(X%fb8I}lva_|U&E{ef~ zdjEb(reOx`plG~Hd&4&QgMv?rAa;TbS?ipp0BoUBeQB*_vN8G>lW*HHi!bg63jBM_Nl4Bl<-!a!^hN-o&xD4E(O1)pqv<^>s3G+PW%=Avnxzh?0! z?+%JDkYSaHTz8JuJ7HS0S+85(f3M+2uPu4y5ig2JYJgFs?FCh&UaW!K#Xrqbak%TO zBj}a9B~R@uO-l-%X*)wxf_VUEi5z0hx5Ae&H3zL~n~~E;weM(_K(EQ{wUQO{GGszO zacJ}fsRm&RbgYTK5Ae~HV{Z}F9za7TMn>GoI166u<7{ZfySIgtqM1 zTik?!qp%c~cta_^=xjIuU6RyLyH{y44eOEMOE)KV`6$U%xE@uAO*wXOeGhhv*sQ^y zMR^Yyh5yF|00ECk2gj{Q=<10-kIGbO*fvoB$3izrM9e%)e=dV+^>AJcj~#ZL>Op4 z{Duvjb#%AfMK{;oWfxuT#x5$^Oq$#FFy zY>hUZ4n3!O^CpNy7u%bTbVc$J^bewBkuNL^)P_IM7B%B+PuTZTl7<+FC&m_$i!hA0 zy?o}>Scz88V?TO2U-<7!*%pNA5JZyzrqlX|^kP@NZx!uA*Y|+@ zES34ER3LTit2Du5=DwrQNd6Ey4uC3kvz{6clqteg`??`pE)%vs-h?KKJ55cD7G`UN z>U@O&D=Au+B0#ML1AJc@8K!UKK~_U5BBp+xEiHw?>tpjsaBs zh7Jbgmi;N)*V9hY+OiCXonqgSc^ZyHOGUl~eZiMMQQjh%jC+6kuU?@P-P&x5ixRpK z&J=vcwZj_fR0A*XA|EzErwWAu$AKfG{PguaQ3M4&`8Rkyi zq*Je#A3Mun7z4WbkvzJp9iSwOyRmI0|M@v~(-h)X#_WpQ8ts`+|9EqzAij}M8RS)y zIR^YICPWR^-r_-B7+CWm{inWWRja2UhJw0UDcsGQu(wh~ub$=+WifesWnC}HM+PRh zlhBE5V-qZ(;YGM%s6nKkX=LC?)q)3KFaC=JV0G#DCe>OjTYy#wGFo_hx;Lh9R>9{=3gln+lMvL0h$>O6B~k9# zza5DW7Eee<{|2mUcrKA8OMYANF06FTqj7yd@T&W}S~Skj&k%JbVyLma*1F!$xstnL z9zYF|0ExBqL1WEu+4o_{QvA>|M-r~B3(yV;(I$X*>mR0*2*J9A^NeRNNt!Kd&XEhe zhg6ww_*#KoCiPL-*-%8oL{r&+65eEGh^~gW44a}iu5%F7m%f!v|CVW43|W#z#)3W( z=Hf}{We2kQ{tz(Os#Hv%W&7>mgOyTMJ7W+hYwe=*T+g26A&3Fk7QGro32Woq>V3zHRtk*N)w^Ve*0F57OZ4}K{8~%wSz%>28kF4`Q5F;oW zt8R%BRIT~tszB)dFqiL?C*}GG?7V~=q1ed+T)v_uUHMrL`u6$qlajWLev{p$K`My% zo*UaM4=q(%61cjSL5B}WL@6|ypYBKKP#g>8vUiRa1i zZ}9GY8<(#%Shi<*Auc=b-!c&*u5?s{D2&ern<{+0R8={x_;YT|@lSC8Tp2R=H!2;5|8&?w56pFOOdRP@<`&nA7%4$9iv+FN|^Ih zNf!*~}tHYbNn^-rVgF28-e zT^{fv$iZ9x2`c|5ByO`pYykY^>|S}UJhCZwgebQFFV8oVi8p){5o%oaM6L|z z0t6&55*k)-W`4zShNjCWJZHGe{L;dzole4zK@~jz%l{$&2TfI`zShHn^-QT!-MG7h zDg!hSXwL2qma#e8)zSJOoOHC#huV+fD-=utjTb#YK#gKWb z1eLAtwf?~vq-ZR|38&jZM!I2SuSd^}-1qadhwY22a{w8sj%r{nwZKlTOP+ajf)S!Ebk6%w_T; zMwsvedZgYTn>cECQinRh@4jYfrla0$4h&;ksJMo+AV z4p9%x$TRkiD0ngS!fQtf*%z+5P3-NZ-t*k!}3 z^F2%)rE}9MyE)rzR|kp{Nqbl(wBH~?toMa9-cw_zt(Cx3jm+Kfy5HCssSP6F)l7rQ ze76bDy7t-&MsvMsu6JY~J1V_0;=XL5%y*K@-m4-mWk%3{kpq2vU)GSMPz?eSIc3HG zZJ*Wj-tn+v?l9j;1gE-`I7mveVYsKN#+ zK?O9pS)n3S>NM7Y7}Jf97ORW}2`2c`pu6Cda%qjKhCJb$g4a~=umrqii)h-3b)q6t zo12>tZnHl-I&MaRN~J2St{x!KOnQ(2>g&O5p|lwWTeMpjNP~S0B&zv2yt?`eL+q!F znC~u?uaNI)vF~X*k|FJVW3`X$5(Vj>Cldz;MZUKk)UmktfJA@&mWK7 z)wKy*S83(y)uRj_y9}d!eM0;>H9pSZ;D;=RhF3wOUF)kpkOPX65&G<=2eWH_i3rt= zc;l{3h9Fn+U!(y80uU=5RU~n!~=_EaNJ$5=CjHa*z6^rGQR(F52dGFzG#j|gDV@lFzNHKQ4Zd@&bjgrQ+ zT;S+=D=w`qFHa}M8>0smn`L<%atyC-ZKYj#A93GQ+WAOJl26`bWM`+v`QGKfyp(AA z<*)Y+HeP0i;DtuWX|)+!YQK$p}IY!&L4+%A7T<1baZsz z!0-Mr_$uNqdEX>us}#K)k~)D#s05I`WF}mA;HbO6ZfN2 zZpj&uAtK7osLziuhUHoDv(vcO*bf#TkXJYaCi8Eqtgq~b4Najf^EDeAr9)T}dxZ{W zsMBhxqH#%>(x1z_Ux4s1Oc`Gyi4xh7g3VK|&$>U(%wXg*GBUniGjimGK){ntF76@m zbh!=ZEa`g5c3p=s!0F+AVG5@laejTUdEVH1*ZG-|p}gzg{knd{SNyu~sY3TY*8LU* zBj5dC)m?P&D7US6sgR`P)`VsD&8&(tfu)BW)~dneTVGotX-sM8WXVB=K1!WNyPQ;Y z%^vH2kBCbFZ4v@U)9WJhv4*V=z=g=1P5 zt3-1zQG|lGHN4 zkEs8VCv$ot@Wd+J;;gq~g|3v9M#*a4zy(rT#+xnXlXmgMs(yL$fA{;Z7M_A*lvn%; zP~bsJ!`tWnB{@Dq6P}-;K(MjeLAn-#?A=8oxo;l!B{Kvk00};|R(Gi_*o9~0~HQcoCWSqAc5mI{a_IMww1E(CHjmXC~{)FYfYWIIdua2<)b5bHuGJ>R8>C{n)?a)Z5aE5+SW=R#fNxgO;lciDYAF*ArR?~Igf&P8{kWc)6 zEYo*#{#C?YqUy9zTK(h)A+gxs9H4W)M}0e*#Ott|8Jz%hr7cD zmrETHG0kC<8WFllEHJ7EAwa1Wm+Q&+!*&f#IEg~f0tFt_RcG`vZ60OKSf@LSLs^V@ zIuM;5crA=*hW_;(9ThAm9CI<2IrVVlt3yog&nMX*CwtqGiwIYqOYO;wyA_p%R?%;17NY>%h{jL*=KHP40>u~?L_4W0AZ(_76WK6`gR5LRc z0WK{bw?$3BJpx-0@S9-dGUOiq94m(@=Iob)4=G7k9M6+yIXb$HZcm=ANCsxO9OcBG zoZ?keti`RNx?2)%JZ(GPc3~EH+DmTu{l5vfB)UEkUWVdZ0rlz?a)2Odk6HfD;Y{QH zdPjzO%IeBb1zWpr(zABAgF>yX1RjZe1rp=nnVhf34zD(iKl)ZI8p6CzON&|cckFO^ znBsF^!nS^^rr)Ks#>!Eenb-?+!%oHR-k&YGtN|)YY_Wi+}NVR>@sFIOvr4K9&3Q>{(1}OLU*_(?cHefa}sUmo};K0~ecS#Ns`YX9YV z`5`^5Sq^4dzY_q~r%k$RsS8)@DV0mj#yC?VpO5TYDdqf5s(qy-j`%Cj8!@Q|J}VJU zxEa@;r(XuiXeS>hk0ya5go9F4R0V(SJ|;=ST2PGed>d^ynjFgdF1L_vqRJ2bP+cPAmOPuKA>`^5et?pCNULEf#1{2T*u zef}umai@09E3(jBh6@o127CE4sXQ`7T?Q4DQ;_TCuLeDZFX4S=I!rLbAqPZ}Oml%K zYc-by_r=OXeM8~@XlMF{eGI;GMgShT>z_N|Mj!Bn^%>!=Gk?c1L~h~7hulRL1X?_g z0QtGoN1ZLIjo!Zu!H|FJ@7q2!zTSrtoqR0*>&)l2xcB(Ri^VtT=6%4|GI?^JNpj{) zD0dmQX4IK^k@*(umn31Bxa#*(?8o&Hx#n|_qCPMD_9Xmjbc{Cb`q)=yM2WOJOvU8> zc(uV9|K6CHj7l@?wU4g1T#gDQtnN`hEkd4Stk_CF-v2%3=kBJz_$kHVep#tyRxmm| zK$VaX`H&?W=auKuoW8r9> z)i}!SF}{xAm3e0hQH5Mw*!;}SUb14|_)-GMn_yeo>vF#+3W8Y5M<%M?a8}l@5TZbj z0Ak2R|1SBmTfol6&f$<;&(L+lVw)YMEMdlCY`{PHNv z=8ga#c9??au<3=7@OCboLWrj5CVOHsfT~`>LdinKHrXM&LK0Gn9UJ53@*DdLA^ceL z3(+ev&i^{D@o4l5Ql4AxOS4}4KMpN^tFUA?^6FI%{$ez|7Sw{7^juqCpYQn-5Ymeh zlnGR?;vn959Log!5#b<%~fjePdu#Q*4Y)70E~@dSj?@_P#NbAQ|Xz}1Dqm2hAjz`u!H zJ4+QnkGz+6KDBWesC7 z+ASc?I4JEiy5eZjnA}hP!9X?5 zxsZ|LB~^4s-_pYCJJhC-zp)}RQ*kglXxN%5Y|uq+Wgq zd;*uG6-ZJ*BcTR2urEZHBY?<~Rj3f>9W$y;Wl8U^C(k8|byr9MQKUV$na1c;MB{^p z2jAjF8!ih|u0X{PWwg5@AosQpkzqKBp^KXMsR#CsH-ZdC$lF9;Brq%(0}>ZT-J2p| zU|p!X>!l1v8mU17!sm#4OA#V$jRJjOT@)$Y_o<)Ql1u|b(5~ZeD2z(>$6c|DpA6gB zMOAh6*+Z3XXBpp~-sj8G_m@Swm7hf`PTdEDw^xzAG zR3!_(4j24FIVIov>O^GF*8-awMUM)nNFF=bI>z(2wpI_+LB1!w%^JYjP3%fk;g*Lv z`cA!JDSF2cFaG_3J3iILd$FQ-61+OR6<4@-79p5oSZYAwH`W_Scr^~c_ftH$7+J{` zCcn}Kjt2q!Xmgf~G+L6SeyXSF5`$7nP|&>43+}q@54;Wk=Pf^j-{+$rgWlUN&Oeak ztJ`|3llbQQr5jMQzx(s@>O;;5akcxe8~|r{UZ3k%to8#zY;LJ1VF$@n8YB9wv}MTSj`N zM&!c^NK;-i=5&x0+N`08`ii@32Na|!O9mQtlEK_XsUn>kk#w&P)KkMiq5efX6{?~y zJ2w9|7~l>J?XiF=At6K840W71`#Wmwu8VCXoft7LfhNK&U4gaaEq3nKa903P8qGJ< zYxcm|bt>SDSc@hBcp^HvntVl-{Kgmc*DVul*TG}0fU0WEk~2^b^RO0bHkI0W6@jY@rpQ?A$G?;A5uwzJo zGGZ?B*wO4M--mjx+n6)QMFO+Y&bwJ@uD#e^)REjPpDV9EO^@z-t?t6Jc#&Pt2H&Sq z^{%`TgB8sd;zX`14^)g7qHmcJKO6?Pj`%-4o=39ZBpaQdhWy^dk;~}yy1a|}!cn)h z<*;F_i}|^qJKuRb;A%G3alPYA$!qd&Ppm!DP*F43SS;M}ZGYO?+R`SDPQqN$S?#|aBHXj;20%pM@lo|iGaoC;5YM?MgDRDgg|@Di`bHv#hgAN zOG;x%0%BTXV7ip+FrbvELh9dsmB60mKWgnO&3Nm;72Um~M+~$r`!r)3+vo`A#P~be zS!{3ah9KJ=B3#l*Gc|I3k8X?C&>NDuEm0K-ItCr=b%=zgmkWO3Hb30XJeyTwxlI`9ta zWIqj1CDW2w3`n|Y+nJb%{=+0JI*gW-VW0ncnXGy0@I}D#;(FVSY`m+kjM|l${QfJt zx3rn>D$+rEa|nlPdm>_)W#cE={ubpA!8`gE*I%u4tH4?__BuJ(B#Z(JOQ^)X}V zf_t+O)&UjXjqa%>Wb3;Ct76K$z*hP6AWF~-AcjWaTSCQofxQrS1m$8#x0%m!v6lGX zIHQ0e^%%bzhcJn+VtLsG44=(r5ptdZm1{sK_bTqwd1FRW^;!ASjRGfzlLg<`#tjhO zBysBfwnvOJSwapjat8H-I)3>1XZAB5vc5T$ICD4@2Hl>9`A-2+46{jxRH9~K7gV!V zP}zOOtp??Qy)ii%%gAU{Me2%+p@5W?<@<87*Au>$$IlT&MH5qExhoob=Flz2xL6(9 z{Eh?z-qgyl8|3GwfD%779Mda>qVR(W3MK@sS~JEHGw51E35hJpgCbD@`&qxxUCn+< zb4QzpB>t_074}V%ZU773Cmemg>{vV0#l+xfv~9r9#e8Cql8ADXgI%nx_XbHF*O z?S3)_FIlNi2608F<{1cPb7(F(w7d(;{}iyQJD2*}dX$wkmc~#H5spBN!IcSj^hGZD zby%D)uL?do&7ID`T8J#-Z=!N)TH5!bybA)V>6wIp?9M&HD~>KPN1Y!QN%nctknog= zQqJlm&#ci)a-bG_qZiyX-?wn^XB9^1+2hz%#u=EANK!^>D0iJbFdipt2m24Hmoi!^ z_2h9LAOljhs&+MRJTZo`jj|UbjX(*rVOEvwGJTyN_bF=jSoK+UW2Yct^F?zvT>SWU zFl%EDq+mZ&moqg`TA_BLCos0GX-EgL*rD>Sjo^kEz(gSAl>uPbshQp(P0-)HD$%CN z-QtplE~A>rhPnwCu|$6VwN~^Rh{&0|9^Sip$~h|*6z3IAH)VvO>n=#B2pr_`cKO)z z0)`4?qv`tXVr8wVvn-jQ6fmX~?Ih;^w!%p+m7%-Eh~G zL*SiK;LOiprhapwwiKn1k4HZA1u<#(*v{H-xatU0!j=}zZeDs1mzUB9Ic1h;WoPMQ zW3;%hYlHoD><^8?)S;FnrV^tqxqRf3B5VEjlm(|_LklU7$`gnLpJ-zUT^2&-j+x5d4{Dzuk8dC*tyzdc%tn*Wf>ABY>aDhYsF@~uWd95xJgKo z`gsF%r&7`LSyQD;itGD4QJI)TS`%pQhEV41()f~DK*{*N(y?#roh zaswyk&yx`yCf(=dAwJg|^7o8vZq68@r<(!T=+YLRvUplg6+)tWtyf!bw47ZsJSl^o znLqHk*mwO-H!9$`zhu*D?I6BG#2rVebuP+YPxxeMdD<^qT49*&sM47tvRWHmi z0|j2Yl$6_*eoXPE`2286g4X}eDd%H*4o_a(?U+iCR*w*or}w^~=d%sO5F!4!4s-F? z9>XNIcO^^{$AXau1uqRIC=P+sz8rFjK5ZETQ=hyoxZ@xkn9H$e1R`@C36ZSDHzvMi z7~T=tSjd+hXNimRz!CrR_4TB6F``8S75j$k*5cp5LA>e9#{l`KkL61)pXh4PBnUj0 z>Yc$`Z`_sE*ol=21mUhHjuaFHe4`))wlnzh3@?1daM!64u(io#)`y#K2w!!Uuu|NM ztI-sy^v}nSkO?w6?C)_H!vGA42Ac}349iA|UGuL&JMxZQnfZ}%+9Asr%7$ZJ@MpX8 z;-7qK5~Z}K?pj9{&3h>qsKPnIP$&sgu}854a;1hTY{dZ_Fq}kAmB;q%O>H3D28juR zfHsD_)eEjZf&=Zug~cwkG&;0pWU#@~E^Kd!~iN=rXl0Q&5g4~rwt zKlOZ+QGZweu(=I~@BQMQ`mT{J1&P7bkY}{jf`qkZ_(nI@zqf^SjLP>4tPKRW;ZIF( zNEuW)R}uVDx(mmdWZs@0w6Uw^t1LBKI1lBq-+R9X+dd zSutAbiOBTT=52d>pU3>B!zxK4eI6n2v1Mag_bViF39@zbUJzI{lS&$MQ!9+gjMNRy zVLi;y?U99Tim-o3$=@%0Q+SUf4MD;e$awMwyV- zHf<6a$zOGh0=bSOBfnLGZc1u2lQ$M1DU!^Cj=(JjK4jL~d~s~#8y|yEJ>JotvmsWA z#)%J0@=^YBPO)ILnp4=+f71ZJ(TKAl`>o%+1*`vxFAXK3f3qoMLZn*0a)t*_IcU|i zp~03(m`pki+DsO}xeEml$_NKID3g)IMl+r=4>z?Aktx`*sXPZor#Qkl>B~>?R6iTH zVFlG>7x!x}ql*H-1Me0uGwO7f=LOwXF`g5)hFxmIE{BLRDz8Z(yj|VqSybkxDW_e; zyl=J{wqAmL3L|`l$qHSDSux&Fre>b{Nr|4&M=)(OG_U^8(s;;VDwfYSyLwkr#jd@yJWfKX4u#_L- z`LLS!c!; zhLX+H^CD3!F#@f8^4-NbKO~B}1fOA3pZbel>1_e!kh4k^Z+A$pyNP++ws7=p~qf{b%q2 ztQq^#&u7gEewH~tVm-I^r&S2y2pwu zmt~PE*@$flpOA@BDZTY(uwT)R>hV_T`>-wJ?Scchk@{}1BpSo8)xq3j*BZYbDnv>m z%q;;ISS8e!B7WB^He4kY_j;H~!?p5IK}9Z^Gj2@Z>|O;@3juFw$}Cn=e)$rojeluw z_X1WP2&y`t6rV603xvBTJ5(yN-sC@Xx#G{b&pm%|0nhQT&Nzc{Mwo2@n9mWu;ZCL4 z?T#2=E=+wOjAgumm08AZkThNgb-Z0jfFUB18+R={C5SSGu!&J7O`TP1sWCdr?Dj8h zne2#@y>kY;oyaq#z7BoZT#%5+$ByJqn1W|)m0Ji>Bk4Xg(#Fh#*5wv|#xBH~;i<9C zmnNTK1vAM>oq2|&h1-0ssoaRZugozswmpP-}|FxcUkp4%5y16D(!R2 zDD>!Ff9-1xjTn069+n8>x!x%Bq>`)p=yAkA%*6zm!X5Cntyj^DT6 z^b7L4^8RKT{tMJ%75|GwJJ6{COAx(3B{K^F6Q-zuf z)T>NcaK>R2JFMFzrr5sM3B1&xlqYUw{s=FPi$;INYYzu0H9*J6Zu)vOqWx*CL!`~J^?cU+o$cy1QjVO|wPTekN zT?yS!u_d+luRO$%ZsS?dr)Z2T;U#bm)l9P(E+Uyrx-1DlqWUXn;!zxooAVJDiJ15@ z=zhPvxf_o@54o}Y(0y8&T{S4=Hi8q@_N)kqu^_s)kBpt+qqX=AaY>GdDIOE=MF!SD zCx&zS&}fzo`#9VczNAqUa1D+KRehtQp_nagst1tIX}v}xOn|+k`nui!m&Y(%TY76* zx0z^`MFS4D?pO~|+%q&D*L$=^tXkX|Y^?c=c&-i5WdYXFRq@w~XoyUrzEQIUopV2> zi?li2x%`p=$u9nwSm+ehX|58Ek1ay@y-;^MCnYgiMTw2kuPG4a{1H{jVyy2hi;%K? zY9B&5vgXZAqAw&wjAEO2%kZe|00uDEMkXEH?5)i#7)qoNMBOZ6F7^we!;M2PG{b#M z2x)-gRPLMBt*-FyM+RJNvMka8Xp3Cal0Kf4Y}pJ|X)kDQSe7?nAd{Bk=S48@A0Z&!!r{aTim5$ zJvL69msoP%a1s7x_+551F?^Sx8QhO z55~Y_OTNQ5lTT5oqt7a~)Xa!WALX%nI7O&~^BCDDz5~KoR_Q8d&OhOF%iddzyN5Z) zwFySioxt5n;T3;-6aSzkjt-77dZvZE&U@JL6chHmmeM=G^oR6fVnL38ye91SM<6Im z535Rgp8Sq#v2WYzjdDok({i^I+RL7RG$1M~XiR|Z;JN(LRAM&Y26ZcQ_!bcIg=f$w zk)tdy2-7bu!^>C8qb<}n zp>c)*8v{M9CeUNWG4VKheb2`&#Ho);hdjoROC!eMh{)FN@vS~fDEfD%$Koy|V z9`~>)LU;txP4wtX{b|MZJA8WNaXYZJ2d3&9#Tj{26*5lcDLY3!^om$kaaXAgzkOJQ zKh}wDby9J7z2SVtbY<(4^(-7EAPkIrPr9$H(8|bQ5;TrotA&L$AESGsDlOfEInd05 z7O*1lc1!!qdtpCh2T)Oe;>7MLb5Z}2UXAQQ0dUbNzhDboxy=KS`guEC`f_zfmWF+JYghulXJ)M$u&^$30iYd)nxeKEisA2ZlhcT8N( zR2Ty^SIION8o~h&GcSR8>dJQkE>qsx4KZsxd5?+UR^~bfCAYhb);x^!^bv1e zsl%%Rku1_KgPpX{J5Kguw4iGnU#n`QleA-M6i8wnngKoa6eHiT>80=%hR zMHvr=;FF?+o>^BYYa|@Y-MYSO<3aFAK8-e@DQV#`f_)E{D))N5DBajUP zK?IN_SG5KlyVTjBVOwdN0rZ88<06NzJn@1U5h}%~q^7WPo=M;MjAIUy51}}9J~%J-ZtrHG2FsElWgo8fIAq|I2g~0)$$=D zr1@;LYN7XoCCmO6sQe=JICG|tytt5~UwND5SYd{L zcz#KGQfIpDTP*5seugX*yoZ?gvxfExx<|&vAL{r(8nf43$=~-rYc6q)wz!)K4307jA#7&>L<8&l4la2Ge!hWyU>9+IpKeErTCJdyA0{t(!u-=J`7lwMZ0&`RrML za>-9lZEO!AXjZC|hE(XM&0TFpT zq;J=Jq#qfw+=8`~QM=6Oe`D0wrsf5HcKr#mE zv3g<6yKUZBLUIkv%1oC8xKaI9r}h>fqI_RVSs}SSui2Dc83rM`$WyZn7ABo;WC7oh zZMRXqjPrzDhFH2I<$lcR4_@2CBd#uf+IY{wxLiLhwY7gu-}sjRvjby^#!Wq8jQ`p7 zRHL3Xd|>fuBuK}hr?>|097mYBZj+2pg}dx8twqT=hG+MUx^Fh9^kS0Y^!DH)@`Wi- z53a=4m}mk_jl4Li3EzaaoOqVic{#*_vWCY9V9mxn#zbAKLUWH55^wJ_!7wfho?TCg zW|gbgBKaz{{o$zk(D!s9u3K!TEjybS75oO{PYXX=Wn6veqGW2fOIA$%;ZPeNaN9=r zb`-66%^hKiPZjKkqPWL3jh*E(8@Pb84ndK^wJ%Zh9)q_U<=7RYa!wl;hzwHY!=L`R z5xXEO+P2y53`CsPKt;+88CV_~dNP2%!R2-20Z2lb z*4tdZouY?)dnyCD(`ppE>}T}i@Gfu4@~9}V%gGJ#-srqHlritJ*4Cb9rk%q@_~Kod z`2`(GqB#MIbf1Ev=q-LCWs>kquwef5`XRr1Qfm~hoX%UW#6ujM2)E50?FtspbM4*L z=%`RpB9=$J^j@nC;_(N^P9qQnP0{oN%M{M`Zx$v!E(HYBp1uB*>W*1ej1V^Kr4x2s zL!N>O8jw>HcprCR8pC49E7)RQS58-RZSx!)HSZ4bRI8PXRPgg$L^mq=GX4B`6RS)yCm|(2NY>O5#l_7tQmdL{LA@6iD)V#fzWH9 zw#ME;=N=mlx6!9K2L5G+R;cH?YnVrni+)JZ{DkXIhhkxoDXT@G0&{>wgAAj8u&0VR zJ_=HYEjSjTbfrfgrIP%+LA5{)#>oLpOB{6`Kg70EaMO42tG1msM{Pr zdc>u7;&V#l2_@H9kuLm$25B(2IClvH>4}zhlR&hPwccTm1VnIzW$bp>f!7Xm?lyrl zN@K1sNlld;i%+j8MT41Lz8GsxouQZeNnD^e2iB-vA7`_pqL7RQ@GeS8aYt4baXz3C0~WV6S;lV|nhg`t%9x zhbJG`n)q;OvL=c(`J9J%!dWLgO@sx$9OSc&>b-VWY$X|G4(8FK02wGo+Y-w?E8nMYI^VC?|Jv@I1~Y9R=`gXw=N>4$`y zD4EEy(cT6VwK0HW#nEZXA*bvSX(Gm*4=T^Xgwqe->a-Sv7)J5KOazD7)jR8I_nF0- z=({(8HyMbzj+#rsr|AjNOX(%1>uc<#h2A6$;M1PFMd8|I3GbL`5%^DRqgU5r#+>9I z1G%RON2I5UC}%(mK1cM$egvdY;RC&*gdqLFv0-#*B6 zQ_SrPjtb7201CB(Lo4#WIQGI>@~m(5*jpn>R6WKE#4gK|=mVXWOilb^emSIZk%V$K zP1l;$O)1o$`2I7%WP5Sn|$~U z5AFiWbtXh21_}t0U*jaK!aCVe;bFPc^R$~R#4565qm#G@imOj?&%@ff(#&Ge&`B`2 z*Iq#Ae0kA%6h<8d$|<7W`qE=7e&MKSPSH!AY0;bq(fk6$td@7{2zs=fe}Z_MwXKN> zqaImd^mJ?$Zbg)U*}qnqjgR#w&NJ)^o7Z6}mMnVMg|4CUHU zB$=VWa~C<%pKm$hXoo*emx_$+n{3GyzC;YDRT)yiwx+{#73exvMdiskmcu-y*zJ6< z>mo_T0;KyN&I0oN#q9-okQM;rDKUso87(95wZ3Dt^=>fK-*n7RIsY6YDUzDV0?vD+ zS*Hd202MrApnY1hB(ZoZj=Ui>e^Oe2kJU{(bq0aeFtk-iIaW}>1>T+hG*%Dg5SaUO)v*Vp?wJs%s_)WJfxKZyW2}Gn$AhDyD&BMIHd*4JRr383JhFy z^;hxM>GA;J_D4q_Q-oKQsWz#hprBv(y{26t-vV0=TfsyBsf2U_dG|yG!LxFL^9hw`@kLm} z3fzkj+Hkr1MyFE1C-F8`hH@w}rEQ&D4lERbjROW37DkXExuh~(45T@cL$d6U{z|!H zSwvS|VtLxOU(-9bo1v9dVX9+{9;bXEh@&g}Vi$ozo#>hef3)7USLng!O zbS6H=9jvUR1~&7LKsNjb2cs#-YRdM|#HTv^t*D3>98{HZA>^EIM?J+MlhIAJgcfPR z5KY{GAxj>F`^JXucn1- z>J!8to;GEgmRNnAxBWhb*iR(d(Ix;Z_NHFa_5Zz1FkruHkny1eMa;MOT0`isq?qAro|&gIJthTT_tQ0%!wewod+uh5(L80vJ6dFeiuyFrIb9M1$dGq}I8- zo8k`)&aZOH6|HH0=FMuXKzBI9;ZiAqvbSCyMH(!!tBuVvqb>n27mxY0F$jz;)#xMk zl9>rq`gx^mkq^%0ybXd5=x}=|SQNRGU2*Z3rcyZkMLL77YkoBMy9x5k=*gYEK%J9LmkIY_kzyeS!(K(MR z@7#b#g?S9o0`iMCymQ4y$0LT^E3|uwV(dkhL-}0Y@g}cR%2H76 zQdA^6Zt1b?g9ro`xDO*s42et6N2g@!V*YCWS(a%BcOMTDTJJ0^9R`AW^6`_vCJFRTC#Eq-1?&ZHU_t;29Q_8XyHc|80pWh0>U1`LO; zv2|;_-GJ)4^0X%|=qSA4M`UYPGG&8MsC1!9I6J8?0gg&r4MxJYEt^j%Tz=t{pH!e* zk{eCFVX|B?$~Ag?KCOK1f5>B_LD(@Ryp}>TZ9d$BC;sas@56j;=TFKj>)*XFe-o{A zMZw{}mPO?JUVdO4*T_Isjw{o(WcX#jmoYT(t_H)4W5a{bxO>GK5is|0v27V;t^ z;Bcv=8xgXVZ|s{lFg)cT;0Bt`-=Yb2!Zdob?CWO|G+^v6-Ewt4_ueB?=Z&erJByFA|ooR$8#{wH#(z#OH~G z*01x$RunZ4aDQri|8m{qb<}gT!`(3Wa9VXbk>bJw!M{w*GZ3cLEgds;)&Q3vIMjeZ zk_R`MTYlT0`V(TZZS~2hk#IBAGZN86izmv0rgoI{&vY}67mjZ^!x0lOXnQ1BC!m-XXkN@Gc-fq|Skd+uJHTHJSg`hQ2BfsmV5E+_j`U(Y%QiDJ zE}%nyp8mGjVIP#r*qo+s>@rPEAi^U9h8#b|L4Dlp2m(tm(Uae}kK?P+4_bo8f9WQG zbJd<?huD(}Rto%U zsm?HTknUC`^|(35LFC4L%|QUi0{go`Dvh9FQ7MzO#?<#m)i~p7F_&*YN?T!)j|XiStp(!_xc5JYlGil#@I*slfVMM=K~e2Ed%Uu zEp{~*x5%jNPmB6eo4X!2jJM<(9DIoSEc%L7FfI@aD82M~hZ>Qe00mMo>ko4JCXOT} zd!O$VcNzIs($jF47o(W3*k2#)Atj-3xD`{f$K}`AI_--sh3l;ANFM)A2T#Eft#EO$ z&@RzPIUxG+_aVSv(}6ww&q3|GsU*|%H46)7F?dj7)@K7AapVUAu%pvQRRs8Uo&qJ3 z0R+rn`|WM&%{=_wA7X#zZ??hE((AZ5evFe8(m1^eqizS6zDQw>)XXKAa^yRw;mDg# z?l{)@XE7RkW|}^=GeQ~f-a874(nkPKRHU^+)1mO4VTUQ;0phF)dKB7FSN@N(;rAC_ z4;$w<-g<9h{oBAJ#@s3S+$M;HSzHMRHwqu0*a~+q~>B$-KY! zYkB#5GDbkae5NN5x+sF>uMRlL?I-T^<4wYAyc9YZ&{fIOfABw^-Z`|51?CJEB$U&# zr~gJUtEJ*?_-|}?1hhl)8ND~ZQI&FG-E&b-zW~RS=wrg05c>-Zh3YE{g}UYPFuya= zB6oUf5Ia%n;FVTG+5Mg$fhVcIE=A6yjkGAXlWO9nzPts`B(=zo$B!Gw=c~KL89Pq8 z=$>z92aR3mRwe0HRdphVF}_8rF=-T{n?$}mvaIa0RacerlB}&)2u@;*ph+dPy}4=m zmn7sbwmm}s3hG<_$^%U|V{uL5)f#&Z95a(WC3-p1ZYg~EgzfGKeHc5y0gsOgX zNEJrM9OD)ts3wK|*{QZ*6uk*2Rbilze~URie0Z%1*b*D&RA*EEk3$X=hixMrYj&`d zmP*VIv6p7cG}kE?o`pGf43Lj1MwW*oN2Ssup;tA`ed8dStc7Px)*Y1mM3+;*;&fNQ z;T*p;NOeYbxcWVHQUqa12tv0LC=quD9PO!0SEt8a&{|DoGpUKxuI&h1khR^lqy$Kl zXezPST}#$^;34CQl@NcxACS?oN?Pn!I5IS3MzxhrvO~Skr#JSdVIJrqxzp(H?@kn3 zSp8?Xum?vKKt&iRDUx!zWB_h>{toBO9P5_iJrJ{>+-N2%I7p62^}A%%=U_M6b>YJ;9g?H=AnYKl zv%-ys;SdY%J`Dt-1PvTl-L(2J>&~#}c?DBcC3`&Qfyb|SlsaPyuxDh3S!$e}iEpE?;!TU0l`_eO2}nB{@p6b5T_Z1Pc|MoNVL-dOuwyf`|F6V`2+Rj#b&{}vb0qC(3U zX`iiK8=hGId6%$)G6MsChge=Lx9gk<^S)tZ&Nvv3yg`Q038dRhqeV;F<+izT_1*{r z^`7q`Mk<=vnwI(idAUEfpfN1xj;3g_w9`lZlmO9+KzLAf8p_QTOLz|Y_^4kWS9kbt z2ouuIK?4lg106)}wR1wyu$lEp*^x+&HTDP%pahK~_a)3xAzxFp$a`TI$^!wU37++C zt3ew-h|Um`#SkW9$UlBIGTWVf5H~z+Sa zG#Th9Q(p<=f2{VMwso-e`fHUg0Pu8Co=7FOv%m%b0!7K^nVH#W1Eu$(+ajTJGtzEv z+Zurgm|0{h{?qa&`k?`4r-#lYC$YuRYr#C0qI*mT#XIn)i7i?2z$6%NgFQ zivia_t^{@3M~L-!c#59(X77PHc!dj47`Jjzp^iqD#Au$!)<+m*IC)*FJ=j~UrU>J{ zvm)Ws#5eI_i!D@$c}mYtqw3cL(0hjg5^FUV2f3smD0n}|K4(2Y1ewzxHWM#xRUni zIXj`{C!%f^H90IA=XeOW4&3KSti8lNSc@-Mx+38QsCx0>w*a4F*(7>iem&-%?Qowu z6hlREvE|P0rR?(J(|Y-rE@S%Bh_xR*NR9)vkR{v2{RTtCmdX_hRs0v=`4ug*&CJ#S zTzg7IDb~eu12AH%JvE7BX2kL|&9zg&8UZatN*dQfNjL7|eg8YGzu)&aA&I9y9NQ0Q zP*?!hpe8m~Xo2vjON5`J^{?$`lUdIsk?RtgXfZ0ypc=yrEzV%nbBrums#Mw3j_LM4 z!8ivyAkFxsp{9VcPY2W=LDSegC+~e;)bKel`nixE}}<6bHHKqENo4mue|6Ush@C6Ml>w z9g;sQ1|w9gWVvBg>DJl*XVpkRzk^hu_hK!L0}%q4?g2MD=!LkIh8PKVyOz#RL4;5h zx0$$VsH&~J3wz#MRX)l|eLkP^sKfF9nKLY2$49bsmeFf^3u3A@SJDk7I>yYif)Htv z6EgYNC?HQcvH2LN<<>M9D_HzX@dSgH5W{>2L;c}le<%SwiVAzVf{C+gF8wl%?az`6 z!ryJj`)T!z8EB9zPu;wJmL3Y3Kg7a+Vp4d?wE@*cgQdl{3@=!Qjo<$FhhdjkMKLr#9{zxRI6-FOoemm$f=tGvS)nL^@wJAn z+-(l)6I6L8>l3^Ou{vJ4D0fB6f4uGAdE;g@oF=QIPIj<24U-0j+@f`mYMX9`jyAa_ z(`*DN?AzJF!Qxxa)^~$>zTAF?eSZW-;`Cd|&QPlMlo%1zX*;CE&?03hgK&vhTBQA8 zC|gj3bx+d{TVA^5Q=d6~B?*50Sdw&$^lIIIdpTIh0Rc38)-QgSi>!;*9SlqV>>tg) zdMFa$Iv0nRS!5><6|tmLL$6O3ss4z{KVH0D_yu=u1Q`N;5FDY*N{)8tPmUyZ!9yXL zRy2g1H=z;VT}v5KQ@E^^tY*pP&y$cg;UT49a7zN0nQGfY=auPy`B~TqQ~~dos34){ zTbCo++_CWr?-I1?1>s7cbRFV#L^HD)ZUThR+5sgv}Tjao}A2anka`q*SYC z;^gxYBF{hH2*VEq5rZkSaG9G;p3-iXd=*oxNzM^~i^~6stWp{lidNd{3Rb`%`AL}K zR$HoSfgVx%?4Mk(!H30Cs)$4%qCZa?i^`G<#ICYTe1wzBgE7tjK$m(1MR2rD*y@O) z&0gvK$Kd{Q5o@ZjwOhhN>juM5LRAL*znzd8Uof^)om0+8$rEsMGGR-pVtGWQkuv|l zl7EpDkj{cwqKc}lRVo0Aq9BTAXO~L8r9z?gL4&|dvqd^M-fO{BYjIa*M(ZD5DVW>&kM&c)k2dAs z^MX0g(@z1>fVKW*xK>7@PKc2w8h2ehzjfCN!t+cPrWNK_|eys-F z!$}rT#2`}KH_EqQc0y^WwgMxN#R+J5;dlSaGQ^hs;li}nrV#*M2eO9y|+ zjniobS6EC!lPzUIhxZ9p73pjQXTK4{l*jL?9!7Zkzh<-<0hc}0)f&cZGVAQ#GZh45 zEZh2_go+6FbP1Mz*6Yt?O+j0Z6PD~*9U}CZ?bPF;N$BTQc7Yx7BU@~d<9wW2DOd?w zmG_i3+(!f^|Crby%I>S&zL6OjVAJ+j3jVjYad!3#_|tG~0dBC9 zHH|9hN@z__;yXDHZXx1})-~1j`R{@fj4+K19*>7TRlz*}yh?K^LB1sq+|Nbu5*CE$ zzjx)FZ4|TeQ_MBctJX|K*`$1sy$jFQ9z;!Wj9rm>(HBsr!_wJW&ICJkZ_H?7ehG?v z0CoJ&3Y`?d)cv9W@S8A*if%2yR`<>cftfCQix)#}o>?BP;8lJ^h79+O_uOL8M3R@h zrnovzrNf$HjTR+XHNOp?nPLbn@GNn+h)tDQv#3KDa3)mG72g(0F|Bay7<2#C#MhQ^?$_8`+ zO8YEUb_o=v-|yNjBotenpe|9Un#j;kpj#VH5x^UW6xa8RR33tfo&hVg8S36=y-ME? zel#do+2NcBh$$dTeywblmBLbFgNFeo$TmEjtNUG3|3`HagMXLSXI9*5=py`hw+QXW zSt`FX63Dm$Wz3lDQVZY^!atn9e3fb=Vas*cE8z+Zqg$A*g@PoG=th1bCWL#$>FYdC z13;zq_%QMqNeMoXygWE*4)o%OalN4y;3jjUF=wC zw+ab!9*7tkVL2sp87;| zWjC?6;g-U(L;O3(xw=e*iVC!8Nq{;+xFQJ~!Qw=ueQemQmq&k;{qf+xdQptyum!N$>;uW;Ue^1))@+XAiP zA^|I%!zF0CwFYM-kjG_;FJ-H?;jOWF-zJ!yNGST4-$>T{KdK4SP!3;!WdXc{4L121@x-A?ByUl`;UUMn#B0O{x4+rFK#o<2}baSfbdxWCu zBESxKACP{&+Y7rI!aCSB(^xWbOB>(1NFK2`PsSi1D~`3B*o?^$t&e!ll)O$r&*x7F z0V_8O0XaG?g%&1aOeAo}bI^{&bSlSqvWUIsNjgjK)-Mz z+b?o?i6_XkzKrn*YSQdEGNt@3w;!OMc_W6U`hn=^)c_d0MJEd4M2|K@z)EVQdJ5eX z^eDNi|37;WthI8jvUkn#_*WH~z!d!aVrV*w2S z=gLIdZoH^T#iiDIvrIL76%CbVS2lVRaU1d zX_NzC^g!C!l|Tk#+5Y?nUtXM~TdFmO$Y@vn+E#=_ER#0u6{=LJfG=uiSc+$5Bcp~i zTAYmW>CK^!N%E93w*@MP$keco_renz#MPTZevkSYg z;+)8@MXyjrRZqZv&Aa+?RGo9Q%Ephic;~Ba-tIj6M)!e0ug0?#fkoSi0PHKz@a-UQ z>Sn(-!4H9t+Xq92|GR_2UTX9J%vq*Pmj8Z|QF2%gIj;~G1}dF2xmC=_XqKLH!J32_ z?XtI3?;TnrZ@jBuDL<_KK7SG%6v6nL?CGCq2F7}b0n5Mcf@@%D5eW^F5Il~Cj5xo} z7$jdUbfJH$cH!kOU_Jj~2qDrLcCDtAEDgc4c`Mc#Z0N=DRkl26qU`>c{jpHa0rRM! z$YQe=Ax#8u8D;9Lr3Aqg5`sVH{AYZ8P(!bXNMHCNc$aS=JkfMFjbO2Ro+Q+ogC}jJ z<6?yq?P-}3=e9ilIR(oiwtmf--i~Xv-OV-Nb_0F>3HSf83>o4z@tCZDldfw9( z*YEaP5UrvKClMiH2zYURNMH~=nw%IyemdMWigk$%z~H+bh!!-hfx{so`CNW^sJSb_ zOyR{B4WBdtU)`mVi(S@_>U+N))ds|qL^hB2-f>c1o0weVH)nb7LOgs@{(YgOv{>kb zO8jL59-}D&pNVcSZ|L35UByicJ*Uf3y3=SyGnB=dOljf0#Jo}8WtA7M=4DAkB*t+< zAme80Ii=5qY+XYcgEQADW#XSP{ykiOe>_WjD)L^QXpD2>QH)Fj%8y-^Ml?sB*=;W| zr29JSPe<@O84`Ig4x>%IEKcoz&+e~Zs^}mHM4{_(6DcClDS$!>$)`w&+S>SBTzy?u zb2?HiJ+9AUkHj^!$$Lvhd8Vh!KWcghvZb!3{yU}LZ()uUGqp8M8*kD|yI$A(*adyL zXj@}WArmP`)D!29Uu{}%el7pvU)lOQ(He0qITf=FG5Z^X+mG^gPL;mLWTlT+#(k>A z7Gcv%??D~QkO^wG`85CH1`&2r`e(F%FAE}=gA?O%RwXgGaHzB2v3#b3p>wfPuc*{m zac!P^<%4B~++*t>`E)75M9zVD?SjWhj4;DJlk1_c4z(}BG2rt=VamA?!Oq`f&CtJ3 zs(ZiWWNge&|Mw^Uk?MqOPK%?yr41RrJKRCz?SF)54wk1JRJkf{I{^KqswfjeVR2tuJeck8I#>Z!Y3k>z1)Oq3) z7ejx4=r3)^!POjS(^v-=E9o>-^DsgM} zuS6=+kz7bh891sHauNVsz!dgHGY$1Olmf&!c0C5W8)hF98G~k72@K8~cH}#M)XM*l zej|fb)D7@^>lFc)wXb!Wjyj|@by9rV1Tk!R%RKarcGYUvf1vAZ5<&hk3jZg-501Uq zxNL=-0JQ1pxuBg&rQFI|8W(n59wWS==4MmAraKtDmAR81)D}v5Qu&bgXBqwxD7SP@ ziFja4wCrkkY%41nK)NZ}!Fb?}YzMPA0_s|0+?$l$|7bEIZBCb^*!)8R5_71SkFx=$g0_JkWeOYV zd_;4MG^L}q>b`DiF3n=;+l2qoey|D>g6l_T)Mv_)wx&BAx-W8QrDfG;&Nzx(oh)rB zHW3=(RWB6&EHr^XY*Mt-0cLSx{%F%Y)4g8U+%;T5h16wF(MI2sr)MSz(-%Hf-4k<% zrVBJ@B52o3C;r#x#H4XScktH~)}{8**Py+UF+zs((f5nHaY`PNaasW@vY!;=Hm{Tg|Ow871@y}92WgM%)e)Np`24$ z$J87<(J-zv1tfB3-}{B}D?8uI!(AsKxbo}QQXHGuj~hpMnhX{)f0QIbUMy5q)uz`8 zk|Tx$P?&qs%Fp#yFgnF|zH+cRswu@a_^9k?JY~5VYV|I^@*9T67b5BZ_VIW(VQ|Jj zd%5aD>kcNq&omBtqJ~bODi8&Qvn6)k?1cN@)BKk-`SGD_yPbM&XmUpH0YJ4fN5wh? zMhuqgQHBrW79QXY1&l)N!-}im^2$o3BEnw z4ZU9RdhX)p76SX@@O`-9FzyV>+Dr^JWz(quu^KdSP0!2(hvPr%75d31qL9;%q^8p} zNBoT_#m)6?b1>N*D|=gZ*^$}Dz#_Vm#a3U-V*8_8;fq^o(WHO%44HwWm{Nws0kXuZ z-@s9ZXZ`LS=4Nj+28=M>hs@cVxFeWRLa1^BT3gEBkhVeu{jVKnt0{dU9^Y%6<+jAu(; zt5V0uJL!LTyj%}|mIZ*3dN~a{|HusuflZ-dm#;NWEFO((ny2@kgnO2|*9-4*mjh;K zgPn71Xlz)e#%F1+5(c&F^^y8*DN27lqbA*n#aLhQ_Vcg$uxm-ygNZwnwGByul6-a! zM2AiLWyYbDZU8FMblW`ju%YyPMq(L4)`Pb2QDp8KT^a|Z6x z@pYCHd47`C3pG~ojzKD4M;T^bpSrL5TnD34dvknyy4!cuTp9lOvMYn%SilY#>J& zVR?CZ{kugr7L4EDaayqXph{t_1*U|80)F$xHOp>`$u(X5?#r%oJ*dsW4#3JtfbIFy z{c)&0?4p1P|Dcq{gQ}rCI!o9;V5xt3g;oCTOx}g7GV+IW_PxH4$O6U}+~aj`+Oexe z`&EnEM->EVS`~Fo{H{sd^rPtOHyA_JA2fYR;!NEqCJ&yZ_oEj<&Rn&jd(3oCu)b`z z$UmBZNXKE#M&`x~de7dl7T~aGTQ7~O?Ck7JAkg;f@i;#)RHzW|SoEYF(cilyPtuYG zYS;F7<*?$h+rxpeW$e1lR=$)%Urafk(v)_vTA+&|T+f zIv*A6N>mnky*J(Z%)b*4T2_j#bQtYeuhQ3PcFJ2>)yY2Q`&>ap2k+eZ7p>c~|F{M} zB$#zmyo?n(lheWpNrmUqUe6EN(}g^o_|E$2x+?pq$bU1l%l3^IrW%sfVlMznxqa>~8W`=qWV-D+ zI~pFJojk6@u6R7?AxJ9uEqjc-)@eK)*_3qOVKqUW7{ohA%v}CX+cXT3;%f>?4@rpm zp6s)F{cmg|rivD&8;bjitQ+g=p^m2el4D=6y0S3$0gP^J{pXkLm{$c`<3A1fgQDrS z5NOdVUC#!dg+BLQ$KX6bz^02CLKoNwO2m1FyAiQTmj{?|IBMMO)mZ~jn|OcNQ_n`P zwiv?(G|r$&&*|}<)R?otEQ~Lm~3ML<0Pt4XMTeU6~k(Y1-0zZEgB~~U(6fV3v4fy> z)*<+!5zR3XdBY%jcM#)i(K)B?_@^|ep|o;P!)eF%8}n%l^sk6;dOBopQDrkFd&juZ z5hexMp>zS}uma{G+6yPa+Dz=R)E*Lc^HBQQ&gZNMnl7)*;sg-QC^8I*$b}_U(5k2HsZAG%wJG6{lYlGWs}-_@mFLBdnV|a9Pw1g>r!8I%)!uSLTaXsIFnXqj7%)F@fY+XxDNOZ z^tc)-r4m^Wjea<~zL}LUcrPFk6V{wBmGkF1R#Z0ClZ+vhkDy0?Ld=hyua?Bto8k#^ zRTqyj238~Zb7qQ51)P~N&gqsR42VCk2Ht|xsPp=+=9(jVj#FR0rtGCy-b%{3UgeaG zSVnns$Jo&r#Pi`wx62Fasj;TMf5jbyVCayJ2GT2-k>VVzw76~E0#vYOlu_V<828-I z|E@Ni*W~^%eB3E{!sV~rBIpfkAi39aB>Qq#eOpx!HYWGm?k`G^MqhJzf+pzbQS|7M z*EuhUF-u8}>rnhMY`=Ka%c}FGKa6lD!ex2V+yic!&yefYxX*nSuNC>m#W)E)EK?3Z zkyA8m4Vw3>Tkq5vCf%Q&MvcJtxb@~g0Ji_~s7T4)Tqz4`R{jv_IOC2i-9Fkskw^%QNzGi+TjHKrg zj(v#m_|{q>>FU8tKq5v(zn!`L?ED7b!%O8uill8b2FQoD8v{w#qq{g7-)#MT|Ig z;jo|3pte5jcT7T84wm!N@&hjl3e?Qns7;N-_2%olcnjjH;^3cJxHRpvv(_>JV)!KM z&qb$d+K-)_(jD>5J%)?FxTB&JqbH%r#nU$zXh$+R>}JV5 zB#UdFTQ2V>4#&@)(A`?C&71a+n9<9eeHrZquWp?~65Ecm_X}nHQafr$a^DjnQ zfvst5{fQ*ZVDd|$knkswKBM7X=;-1>wC95qXodQ5Lny~7gRxuGkNO?vXyLn9`k+>i+SK(ug`H{6LiS6`Mj zmyZk6)U!+4&(2@J(JpF9w%~_p!VV2k(HI{+e?kFd^wDSRFB(s}?b#nU;8t%J7%AK2 z#97j22$hLy+U5{b{c+pk>B_KiB>C46<=UVR4w@<1d?|>N;9TG<RXY{it6%Kz!pX&DN@KlKXv6g!~ zlkU_%tE_<`&(KhsEMY3Se|T+%rxu9X>sEogsWp zU8zV%en51T8G!{ISwG()R3D+;lU6eBS}L+04AM$h@JwqeUJ$j|Z`bgHow4xip_iAJ zE4ui(=znl{VR@@^~8lxhj>TkgJkV2Wm8J zQxJQ(8Y%#j4ijDQT-ogSa<$i7;HmJ4(SCh6lcJZ1>x@uu_2D%`I-!o#@e-X1_Jci3 zXhE0{&f$<>q?0P+M{E_9^p@wwUaS&tA)=B13$e@*`k^~9S&1YVUxj3&x1-RIOGwyD z_Kwn=`_3bW;rbD(qCCSfib#V9kY*Sw@3&t1Z*C7|6=ifsxX;gLh{yk7*VTABE8FCe zc$=SBiW1N9C#CR!EwD9*hX2NhE&6L?+Sp{-lkQLoV{o?8^eZLHbWw*F*A+!YtP3Wm zZC87^X(s+&=i_N@q}htx>aziNH!H0nRc)8sjhpydE&z&A*{uhr{lj?xN56#bHq7$= z!5G*3g)ZRwu_eXaj z_v+llQ=QPNo%aui1YWfd)wohpwMpQrh;h`5G*A?`kSqO#PAfQ2E!X%JAP7T)6sSCm zQ(Qs9@XY}0m@_nK1Hf@HfNiLI8$u%n@kMT{VL~GiX5VF!oQP&`cL4c0edYNK{hc^O zPCx9i%TlTMXc0QcmT6B+_y`Ff&TsYu=uMbE>W_K!GloNfo#?beX_*F6)c>`r1Q-nQ zrT(rwB}$G=HPkhQdnJ#qW*iVVQy=Huet-%R<_Al;W(_7&#ADvUL_!7ZK5b8Prwsy! z(*7TTn9PHsFqo$jCw0{CbaOftj)a@WxQ{__=1p=}qUhC#_kqvM#;79CtViF|$}$(u zYw)#_n$+2C=7xaobAM+>BuiXw7IXtphnmj!ptJEF0oItdw}M9bgNrjZgZnuR@LtC zv4*#vEl|^nSM+ytd>WOxQc@w3jA`8B4h|m$9LyLZH$4SN# zKDGexB0f*w5#6E0$9sQ{svbIu!bqNxe#I{&jJIQZiwb}!n4KaZ)Ss8(W^OdgVe?ARD2p^gg$QUMqLoBPN%5s6s2Sg$PAx*^VZ|#wR)|REIw>L z*yJKILc#AXsICVxlV7K`IcmA~*tw@a|Jd$cZQwwpoEZ<2$MfPwOHx?J6%_O5S?b!n zw&BRMtIhp$+*M88uw(uRCsL*ZWuOsAITdM+YZumyNZ5r+MU?2~{}HPBgFb@&+4$im z*D!!+5BSwg>j_KkDq@>s$XhPV^go)~b{Odt@GwqnCfYBKtNa{Hi&M z&3esPf?H(_!33gZrIz17>tRE&x4=GI&0Y@9rJ?yb6BxJREb%~ z9>?FdM;3-4GbSW~=t@;7Z8Z;HnBGJEVz#_z`{)0FDFNPIIjq?_rK6PL(bFeneHHWX zLW@3p-7}gbEw`Jihj4O!kN&rmfqYn|AYcY}g{_FiHAlbu`LdAduS`K%1cg?)pvk)x zf`ca=)DT&jXX$&aGLm1OC!1sa&x^3|8=_O$Wp*i(h~a$t_9?Hp+^SLgr`Mm)mCzAE z2saaEnwBm@S0)I0cloYE_W#u?#*}lL$|Z!qiNrU2J@WqRn{sNXees-;y<&gmsei4q zza}A)zLr}9?w}{`wwv-N9Q!+G|JQEZ1Xa3{s|9R2eHvp|&p4IeA{eRtA z4H$HF^FytFOx}M`>i?P);=qlcY8&|!{jWQ-0OQVDdtUsHIrX1Er<4Y`F=t6ttM320 z^IL#jeyjg?dgnhM54gS_2)J>|lPTk}|8eIB$n~Jc-S#J*_|MV%*Y!EWfg2B&*;VWP zwUqu!tp4^&P;io2a=S}e$QS=jHvac9N>K!Eyf*piIp;sV{+}0BCkAE%akS(Anhb-C zz>VK*&v5-ac>ni1vm#mS|DR98Pppfl!(Al$ZN6Tr@m&<(QUCtybMK+oPU-DAoc@TePy~R}?#IT)b{nGVu>WnigN|5g03H%P zHb&Ic+`O{<;ix5Q%w&6&;YE$9ye9y}75)5fHrs8gfoS>fC=tRC*q0$2;4g(*jrGnI zjGsJ*?N|Nb1J~r_BosI?l+Dno6Cd0zGY8%T(Sf%xS3LGsv_ z1O&nmWZhXTcNRC#Y*r*FbU8k?9sQ&k_(C|d+&tJi(er070Dzs} zeO^yt@m+3bGx1QgBC9OtL@_YV8bD+`*gK#+@ElfW#z2JYf>$DOp7WSjgI=*z1 z{vFpz=?C>d?SsAMebDgmuo-|q=pM1xmv0D~)Z*|z-~proL+4&HGXQE6tia-@G7NV( zsmI67(xuMkToAn!CU`>DBYH~s9J}W9G#ee?o}19ZB2K@*9Cgm0{w7jb)a~yJAkA*+ zvpEuw_-eu?9ys!$O3d?F`#P;@(-3@`lCb(dEDtB)Q97R;BG_9PeSohM7InvGO4xhv z?J|DHg%*7fI~vwKXHvh=+MO`!*Vvh_n{0tB;hn{g2j#QhT1oo^FGFWvH}lYVZ{|5G zyk_b9_>JfD#gr7vy29-_Og8M-uU|E5sV`^vD!h6uU6kxK^5q05-oWLtLTfExe^{hy z?N(j{n=5`CSV2;FLRLfIL|}YbEcEQ`eg@8(KXoTW^Qyl1CGbk6uu10npnH>%ri_*r zFXn79ARt}HTH6!g>ARpfB+KPdjLkL#hWe-w-u9!EYN*?H19sn^FBb)s%+2in*u)L< zWYS~m{!JbL1ecL z5}m%+WUc}*JPtJCG9nIhw%;so;*N{CwOU?Fexi+|Lho1KJTjR7J*+rkC={9HA^_?F zmq6y+)Ri0(hll?&pFZVF!i0{jw0+yfIs$p%lMCa+>KaCFdKX3{pFaJ%dACbTY`zf9 z8QUPOJFk{8;8$Hv=+CNV-mHMz3ubf0akd|aqQGgjD3W#aW_qAXp6ALjMCiX%*=RnVz6PX(4m~E>jbAS58vizV5BaGf#Uf;K#o$aQNW~Z859H5x+R)M}TTX{zJwezlzY!1MRP@fiH70?b zfr0R5vR}rNth2?l^^PT>`tC$xQRe&V7pZ55IEA8eYRA>lCT}zvi*Z)3P-eT{6~8ow zeVq#M&UU0JV^iCTugniFlHt06Lpe&oP#=UqDb^DtV4 z!Q^hs?NHq|GAu-{{BB{i+xhkAV=2_uS(Jb@i#Hrf;@@AO490wR9WPX?D&~+gig`t} zd9HsWu4`tgd^96{2E(Zd4)zba?bQ2$zDcdG)lI(nMkIl=8FOmx+mkzJmF1^w2eV;k z9vJiEyNbd05a)b{JQ}W=u0c&7OG~!1;+p;Pnn6D2FZC#Uqg(pVP_;4IdO<=P3Y;~| zK1PB!s1`k@&kxTeDgshDP~1r?KYf3pY``&b`7)`u*pcC#{FlYAgS5(aAD`;=*QS$N z9fkXVZH1q!7FDkbQj&W7uuOw+Xg2ddy-Ouz5ZM_N5A$))lw3NT>~aE?sPm@@U2X4PCVVRhCoWY`Tu zil>^HqFO{xJ9mSemjT5inrm-5aU1gZOKgyT5H+m**NmOngqGMRlhV23N9517eX@bg zY^iu%qUCG;U|4-Zj3r;ofr26F(><=Ra8f-Cnyj`Ue*~b^M3DqT@HJD^0{szy_!rq+ z^1~oM_GQLy;D=(9&7B>*QK@51L*a})muFv;IIjk62>?~VeNq~b%y?c77j&0?oUMJ% z_qXl$zkLiAD6~ia+h+UUKIV4ecoG!ccf9w(=fb|y;wKM*aoujJ!!)!=C>De|$(7gZ zxh_Ai3v2zXKY^($Bg1S`b^u+iZA~_r&E504`+j_y)t*0T)vxoyAah`2MPEw&Q{{E7 zpAjDFMT4AlRmE-plu7%eatHA^(t`I`x>0%?tykeEJ{uV-RLq_uCCR`M;>>rnCKMrZ z+Z4LQP@Hs7g6nST=P+e@C2 zA?piY-_)ZujnoZhtVPc^-_|+uGhnc&QhLp?rNxGxm^6OGP?Z~pvlhqs5@4SZ= z3m)X~ozk~n1xhTwfkdD?h-S(zN3bw+xJ{MuRp`PO1@QWvX9hi#;I{&U!=)DLyi`SVKNQ<_GuT%NDQ!>r>z?=}7GTlw_) z^O+_$`SLPP4-C5n5W)}J+~!7u7Dd8{PQUu8HU>~eHkIfw^vOB^At}9e?jR8|!odR= z2Clbs0-{z_+e~ur;5ORHgBusXnXYj+QV!+-#v=egigDJfg7@@3JGGuU%>V{FJ)yU& zZgr10)5g}F$`TiHIyULU zUcPdn|A5q;;%I+X)M?k#9v{w2N=TT~E8cbb#nrIZa5YSu;T&8^%}1hu1s^p+pRIjwj#PZByZNZY| z67_@|*v-N&@OPda6%-a3P`Z!Y;xm_89S*6ej8`4}htYZ?-1=u4IIQUS9TP|NuM0|4}98)TLiPb~Y?uEAo+>1F2(2JbS!7H23+b^Fs-rZhV z&`wx6k^Akr~BQ)|IpOJmOfGYTC$ zv-T>IGher~??dLm)kPM+>L%WQWe}xt zuYl5;X!)*wLOVK)wCHB>?q>19dIq;CSy)*MPK|nSoIPO6ikFn%?BNgA7g01{-bZ5n zWo&DNwd~c8b4n7G9l;;Kn^?{b&pQzdI2Iejjc48hTubpG%-Sj=K}ms~<9eE?*vv5u zY@Pyv)g}1`a6NDDs#cmhS1uFS8`cnBlcLm_Gqmdp@GXDaVW81D+qY**bm0i{+GK-+gd!bpU!5HXIh z;Q3)hs}qiEF4q=h%P3!cgq#}Vu;x`O9K%fn7>A}|EQfd3qjyS@X_4kKLMiMhJG%B@ zw*9T@8u{trh+k&9X+2EfotMv`LO&jzVra+sY#O@~k@VAqgz^Ca z^tIu6$18#8Whe4O>6sqh<{e^oOkA8Uh29H-Z9*Rm7zmVa`(uQu9?u;&dbY+7X+MI^ zTa{$JQQx7cvAzTH;J$2B&C9o&vTZa$^6#j3W&Rg0CzV>&O9y|K;GTL9D(bIIIc8n?;E(h6DrLHzXfmDPT`c2#q{RwVj)df04znC?*i zPUUe`T6ui6k>MMz?Q^fU9vv1RG#={*H) z1}z<3Y~)R5D>c(WqJhtf_)O2ZywbUQ_RNUW{Sg2-jJKd%As?f3;x^^0(|?;V+|suW z5_?dFVAPB#EPc-uFs?k-gRqjCKoz0ns@8>4xjJTAy0m zbgmUiNQ~D$a5Sa76dR1$kP-B7U8TmY-ZA^VYco}1z6%K)J zT~TuNYxs0@4k_3gTfmAjQcgIyyGl%QvvV`H0yKNsJAz3r1Nci)^l zh?uC^!pmTJ#R>{xmL-02LDB)XNu|`!i$nlO<9sJku_yL}68iZyw*$*~tF; zcB080g^q~g0M%d``G+!Xqw10~*nL7bz{!c|>8fTy_mVg?bB|_AXj!_+5yE0G1;Pb} z;{^JMXF&OpGJ#L$r%50*&JsoL3U{4owc729bh}Z@4QT3pjv2REhu0MC=MS^6^Nbk6 zYADJS+4O73^!2;;vm?oNP-oC@Ps^B$qq(iq^Y(n$MwVG@0m`;k0pTXXBWgCE)h) zjTn!U5a?WALs3q5UWbDC(dzzTYTP~`8RvRYI+jYFfL5sd?op-7+@nr)j;O`+2|bLK z=xdP2J(ebyhO@&R`G1*B5sL@y7lcns$Horn_oZ6{+u4r(CC2u936*{LVu44k$wiFB z`k7AlxO=GuIU!t(UtlN;><}a>e-{kOzcU4d%>!C{=6uieD6M=t&6L#_T8lb9J9YTL~WT`!VD;X1|OM=0N z)knb9EAS4yshfXzthj^~?Cd}_kG_eQaa|$}?-W={w4uQB#mje1)_h=YGxqpMW(0%y zVuDu~&-#fMp(;sF`_=q@{z2zfh+IC8sgjUq@i#IvKEME;2M2Se5G<#P{> zVmL0-t_21|G%Zu!N-){uI{zTjn56C#LWt_eYTX3}4eh`0LC4!gM|*dBZzVjEWML(d z&E{=xA3w0n{4%1{0woIf97pz4LJ%7I7|=KV&L~NY5byI!p)YTtWbJO37*N3*%o`Nc zs9NZLiEoPzb^d;C>=%+x=?+1*yDcAJz@}4b~0jiz}1`9~k1Fo|R9;eTmo-xL)6?&!P z=ItdJ=K)A|Vxu4uATlj*_Dg>CYz);3VUBq$z2a0KH`7ag_rO!!=2~iVZhR>{YKS0c zbW>leAm^SkVK(jQdvwz0TH&w|l*Cb@iPqzU94V=!Q9WPO>xPV>ZUd$B2VJ-yn?Pug zo4v9LU4SqR^$+^hA+*bwqkFX^6lV$SP1|6Lv;mT=;7Pvx`-&83;Y#m)+dJ&G#kHyq zn|1siHodrIBiMynyo=HZ^j4Cqi-x8=fZ_06w|iEXj$ZW3g}zWTLfQdIXY z@os2}7|e-!B00@k5A)3|j4gJT)Z6&g*vpWxGp@IBp*zO5`B_(z!*a3smV51Mz~WRd zOW=)~>KEB-1c9>|E@N?eM?u?!e(%p~eP!u_@bQg$oQ0p+;r=0K1J;}}DBHL%*=&h- zUhHHFAL9M|O z3!18rFSIP(*Y%1j_j;Dq%H>XL&_u){7vA^k9`ZB6{)Ew%GeZbt$SIpfKOf!$+*1Otz+5YhVgo>0L~G$y zLe>b@sK?sQY+4t&k$aIiAh3x>XjaCP5Z)l)u5kBO>|>Nx;gPz^e68Ty-Sdi=Rko^0 z>XBFO(KVcDO_rZ%Z?M{(FERQ+ml!jsr(XxTF0qbkm!n1@pvblm3wYSu?z|(_G#J^6 z+PVqfi~3BIMc)USRbxd}mSvyz$z!M#K*x5Zj;$3yW8aEWKU zKaVUzu^qN(TXfiO6l=W~f3N*?m77VA^F#lb-Cw1vFHN=Iy2VrlDn>S~d5lcUW$TU(x$wU({6cscTNphK z8S~*SRAsnrDXCH|`p6y19O${M(itB(U2%=Qxir2?#N}jjT|6cqI1+uiCf>U4h)$f3 zV-26<#xoOy+F(_*t=M{4S ztAL0UefVrb31>HuOn*}LEVn~$*{2?(VaLKrC_`1oc!FFvlacSGCy@zWhb? zrJsyjn5q>w!yYe42}&}8dWnH&{>-HIe!P3fK3%~(|Dfv=`W@9J#~JKhwo_d2QVgT5 zkH6&v@2_viKI>NRC+QJXVBF2wvgZo-m$4c2c-%C09Aiw4S6&rLo9Y~7sxXYb#(o@b zqwau59fAX%p#)(+amInOv-l9cWV{&$%V$4nzpkPr0PA?h%jCE#$O`a8A8Pr?#T*;W zU_SJ$CVCyP3Mya4SWb(^m}!!ai+aIfL{(AlnQU)WWn5UWy@BojjHSy~But9*D z42yZeefj*jG7bBw`u@<`7=h@+SPbEM%05owefHh2LJkq4-RU-qL_}mqvi?`s-@Vp9 z$I=eI_GSw%bF?>J=d+=hW*;z?pVi%fkd;}?pdK@xpiRSQk_6CwC}y!7?Asspv%#M{ z5ax0*n0rnZD=0Q0A$2ZY@VHcCN7wa!z=UY2#U;rMD!~r+43V2wW(TORp|`I{D>u}CkxMLI(* z$XUO1mp%}AeDDTt^F;@xZ+!RjA|3nvjHTXt`E|@y?AiHc*Enz@!Gpt??r;=*MeGnP z3yPIUGBW##HRV%RU-0PM=fI$(AhM9iQ5zDV97JX3JRRHWsYLDK7Qa_Iz>g?y%-B`O zekcbwww?F#DTiQ%WXLBr70UuDJhp<3ned^ESa@Npsf`&#DEICD_m__NV%BFe2+Xef zN~HbLKdm9R`%9xk&|3uIE1Hf&82WubbTNm*_W}w0{t=%wWIKlGwy^k%e+=r(f1C*1 zKi2e-Pj*3jJVP#)%<4;)2XYoB9Z=d&dW*GA9;$_9N+q3NT(CTFkbct`HX%1}5xefHm zIvip`yYzcM0J}n}mr13Q0ylEo1e>^}b~|d9m~Aj^QF89ztZ?Gcp%yV1gc>71T{HdQ z1bGJB4B-0ZlZXK*x>pXVh|&l~DJ+t%>AFp7;&2A2g~YdxP$sF4|>%f@KR@iQ=k?aErB|Y{sxgb?2K&+aFgkrO)%h+0cGz4vLLL zWV_D{U_9=%N0lzhHwJ14;QTJ@L+{T#JNGGlQW8=Ce%G%;=Mv*Gc+x6e*6cQBQPu+8 z58Io9I@pIPFVPtc53>=n$*-5R4NbaG$-{2i5ny!Wg zx(4q^En&TNrz5>m=Qc7yKL{353sJ=G4kL46dG^SRyGTBEAYj1)HDO9~k00SNZ5H0^ z-oYZUySCu7AKOPRiOv6!=n*z&mUi(namUyhk~(qzDbkFiZ`Az4MG_DZ#q_7BEJX<^=AfE6ajyh)e=Y*m{JZ+dG z%8<(acd;UvY8Rv{GUl;v^tvHi56IAI0DAvyXsrkuPeY?E6s zXKVo(I0#7~Pt#`srVVMy$Stf*G5u_gHw852XKIhTx z=h4$uu3)myv@v~e1umj30v~GE&k8g4fPXy?IDPJn5*tlD_<5Z&ZSym4`}L7NjN6q7 z!O~%t>>g6H6zLy#7*o4_)_&(#Z&_oj3t_rsyX3x9h|`emTV-@V=)|JmLH!aeMnYbh z1@OmX4Q2}QO_~n`i@r?o~jT%LqPg;u|``P~VKB3O0FBe<4S2$;_bXC8POq%#yx6y!~GY@O&o>>fdr zS_41zwc>O{b#+QJe4{4pM5rENg=gF_HcrHp08BcDC`Rbxm_#KN&(%bgK6&9!XwUFh z409{t27(=e1`+aW_dl3_CfVk$j4&0!$0WoQBEX}^GulqV;#w#C0l{^;kC3uDqDO=# z6if3nLU@E4Y)$YA>q`LLCcGeCMnW4Td8oVdE0CcJVvd^D5LBxn%^YZ8)@vR#$!B^F z=Q9l8EVonfvw?$op<#Y+Zs|$%~ zvT~zs_t*=0c&GD{fqpShmVy2T?FKVH>stn|f&t~Bd5cc#@k?vNToY58+yx|MJ-vAB z)3+7A8~7yM_k{N32B=+p;mR-p^5l<$({0+lHImP98WMonn2;AVag~!++_z%H#We(< z)r5doY-Vj?j~}8?T2e%#ML6SFqqt*JrS+`hC;7)bk9i331JA-+A@|^|Cu%Vkwhti^ zkw(hg6Bx+0BmuH*tIuCz|KVK`?vR@;oBVX0s zFb60Lt~EOF>l5sg+Ic|(WHVV25D<=wiX?M9axxl>NkBYm)XEg=@o>#ERywLIZ$lQv zTv%2M=oweRBPdzezS@8oHSN{M!tW4;@n*9aP`Y~D3;bRViNa0FN;UG$_|GyFaEogS zcZ9p2neLeSyT14y0Kc1{t6_H{&rhV(*_3U3J_Be71_gE}Sgi~>1uo~q6PR#LmMp&= z(E|KOqdWYopW|ue7KBxz<^wC`nCSg*{BU7CnDXohRjY~z?}i_x81)?Kzqw25;3VwW zG)qBuZn0ejN>BK|8`R%F7-Yd^y1x>ZLU;jXj6oUNRwv%Q_xNFE$7UW$58(*4Ts;vw zX_6}VnI@ex9-V6{u!wT~ZLx}oPlrMlWfid<0@dWUGuj1Gw(KFksnC^4-X|X&D_!mb|~!f-F7>fjhPvMAvMZ_S5 zKE~~L-o^e~ByU@TrMvZ0vmVlfG^F$i?ojs$9*ucSJ0|BFW}TH<)bpKwb$oZ3e+%S4 zptr_FaY=p}^of^vM}-pW`o3X)nfnyFcaO6^G_ZV745W>?UMN%9&h{>D%}p{eRmfii zetoo+7hkOSszw7p6&1CZBW}_Zz(;I}=`70SJF31*nPQniJ7!S(`snisZtI5X7WMcX zWp%WaN_qE#XO(yVPmha#n=b$50|^>rePr|{Z7)URzq)?^W}cOk3s2nrU;m^B0bIgl z3-?-l|A*~3x4U2TKV8fJd^*l7z#$BnQjGs<&GO&8E5IYc&0bOe^Q(VZ#Q$<1lW7C) z;83x9d?tT)p8l5$Zt{P`|G%G&DQ~(r_XPpSsyO@Ycm7g?;q?Aspmh2D!HDEObZ>7@ zp2q96ZSc#2@A2@#YLZS^Z!|@~&ev#ip8I@&XbxN0i1j*N2a2aW8CwA8$0~dFPak;h z8yuq(I03Q%Gai$QSCE#L76B;B(eFxpzsA*8*0YIyS&?l%ZEo{=p^s@>{Lbobqf%Af8MxX zQ-3#v0Z2!%Lc8j1=s*4JG3KZc38xoJfq-p5r3)#oix7A(0~rPAYbp=d+_xORUq29l zD~15sH-VcvdeZ;qr^(s1G!2F_SNVm zzWUpX&zT_SfMHyh#{Jvjgu&XeM?<3|w%}=z4mhp!KfY2ct*pcc^b<2ca>i@kZFs2< zG#|iydii<5o2%-L#)n-tt#_nx;_K=BZrhWOo;(rawtT4faU68$x}ISj1}Mm$#!6(G z2DUKX2TGfDqcYS~R8(Cr>?+A}K9rYtUH+`5e+@W2Hu+}&trBg@^xx%A$O-ga3WbSi z4yb)Y0MjA{lwYKg76vqy;I|q|=eOvKshQB048NWx2LA1iFjcgB4Wo4!F z8aQv4Y!8*6l4S+;}X#u3nHNNsO#o6E%g`0;~MvGS1iRyhXn28FM$uRjnZ zq=V?4+`={-D-Tv-W3oF_-VS<%baj3A-WpviHG z1zb%sCHqhMFmTCx?3T}yQNk(r5rOLY57u->A)$MK**4<#gY$Vnhb?yBt>@JDf$$`aog}_bAV;KRk@F z$Re-(XyYfXc+l|*_wspReIPz<*U!RMh|HQ6jGfOr;^*=U4h;#+sa61=Bow`3!lR4E zv25 zRuLw@!wtUzIw50)p+(U=4to}%6;R4YON|}=>SV=~vGrtJ0|Ar0b)E)H=mjegaC9iy zw0+%h7qB9zko^HR4dqvBns(wO_ad~ZdoY;7$SeUf}PAi z<6jUOVpC=DJ~#1mbCqNCmjP6~zLoVvBB9KW*{|sz=znJ z7Ez*fN=>8}r02*mT=X267TK{7oFg-wbDlW{vL)ysGMeLr{kJS!4 zN90P5FY^g^Abm6qK(1OU4wu@*w7&(ADQsTodN^2&hV^XT!^BkrLDQ+$h1!A)Lwkly z*Re`e^n|~r~)m+qNJ`he(x>H3HO#!VMHUu z&oU3i@@lEpYX#W@K@9`#A2rdmk z7*D*%vCNyzi#M+u0ngovbHoc|2ccem52PPOP9rxhf=X@@ZBL}OOEMMPB35xN zbMe}0?l5Fs-mdmTNVNjqE+W^pl;?d3Z#dp5mAP~T9_9alhh6k2h`G|ug!w``mNeQHqL<*QHcVLt z$s)V!gY=_Qzqhn(U`tbXL@{m$!hcXOVMhlH5d{C;J*W=dB8OHS{0kc1ozZPaygv$9 zM#EmnF)&mSibb$1kky2Q(aJ%dIF%D#20|k`C__Uczj57) z2<+78!owm$4c4Ekpwa-y zByMafsERt|^*J>R1d}JNeG_}|#w$_IU+F%G2A#?)DM1~!EFISyvd`rCc3OV&KzxLy zOX5-LEiqHreuC!+ea;y~MluItte!={IQ_iKIjkUdpZVjFC4#1h3ML-5dHrM5pByB0 z&~Y#dc;*mDpDbxGw2XnVd?};x#7$fF@BIS<(J{5}IRxU5j^sHs{9uzvfb-5xB+n}j z09@bwwb5jnhc={fz&v&T;k)m%c$Ho(|2fwS!Zs6l-TGu)YV-d8SJ}0PGuiiXhhmD^ zEG#)Tha5XtPa7hmCFRh;+N9#qgQ)kJ>ZL+qH-^k1ikxC;5jj=zB&|`4M1@R8+laCU zLoJEt{f+m%uDreP)%EPJUDvK%_jmX1`}_TTKEL1RbL*PM%5sg0s;R5eEAM6Y*!BNe zaq>2i!Nu z8G-X1I~Dcii;Gxn*NB;Kdg!BaFhrPqwKp6YZLKh1vRtdqFk+`_qvyl{>u7NJX5BH2W0A&k&pg5Pqc>AeGs&@?G z(>9fb3rY92EizF}GsmcXz7w@%WgV!%F~rcs)5h%clr-YteMh0*k#K?+Q}_1gq(75ErqxS27K*(qokr0<8mh zTY}AwcQ0MtfY20w`V_^Zx&`5R11LK`rM=qQ*6{6Jhc+udAnhy{1y}g&FK|Sl^K#a)w>k=p&(9=t4=L<6<(>L!~fxF z%WK}4f}k~z5?}NC1o2h4Foa8#5y?a?DP~$9tZBp63DTPuixcRa((|p`MU(w#!sV6r z$iqJVX9kC_S)4XrFy}IjqvKD4&BB7HC~|Y0Q1O*suAZ&V-E*p2Q&6zng6T8?%;30A zn5`e{nKf>s6py1{!t*(IV|aKPQ<0@ySUnNorS_H9kSPR|_bePUi9VGiR3Tx;qDBgx zyI=$c>f(w(4GE585O;dmoDaj;$R&TK%9T8TUe{^pG)X9t2`_}&4GC|z!L6~X?Z%B; zvj1qz3yjKY>X5?zvtd%heS}Ixlys7{LNao(>VD`k`Z8@>!`Xm3grCNV%MeOY7DAi? zinuVzMtd{ZGkn13w}df)uDw+Iyp6w;V_{1}C^e)q7n!JX-{+YU!}nwLv&j|Q`j#!T z9cN}I_)<`~?C=NlT(XUGiFz>9>GaK@3#Tg=*L?(*ZAdy%IDSr*i|Q%hjJ^SO*BzPs zMFKLe46$pf#z^5q&bko*yw{H4#gC_~1dNae{CHmixkf(mC?*YXlQ)>wj7!aj>VSN~ zYWg})zvoyC_<`k)a=e)#nH%tA!A@N8qyk)jBm|l%RD&DkxroeGc4-y83TV$;uHzIL z-^c?F#s~)YR~xLy68^)G_Qdy_UR9Z!7UC2n&IF970!OF={GoiQfpm{g=;ki0^E!4q zgh>`=Jaz57;x^n{LL1RP@Z7y0m>*Jy1sA7sr!|Y5o|;H8jt`Hxf4pz|u=x)68=6a% zDX;|j#1JRz>Zfp(mRLfb$B>o==Shw=i`5onR!s3EUhkX-OeIp=_LB$Ua~^)*Y$ab2 zhQI(TXhZ$%#Pnu3M**3B;GgkMi;!$rTT-eK3(kq_aG}dOYr+jiMrry&>$fI)u?eGk zK=xScz;#LVKh{aIH{nC~O=&~-m1wJqw|BBUQU~9!PYn6;qOP&~*fFoh@NZ%+x0`0_ z{!331>cgvdvtd3)2~U|8^o`fE(jqe#E8tF1N#gu0qebMT-jpu_9^Jut1=rEDdU{FB zNw%z_b7-8kCB7a6zsV=6M>PEYgAyS+t?|3xJt5Cm5vjydytsH;XC_Upvc3?RuJ$N3 zfL(+I0?5@ZkGPU5@iPn**d*qNGjuoA`3^EqX3P{aIv9qn=SD)3X8mM!+m~aN2AljI zaO|+Vy@L{(2F_Zy(+AubMbd5OVsc(?^UdPYcDsqTf_=M!N`c5@61uybC%1Pk(k5;QOqNY7r-HZ~=M?DleS>(S5bYc-rGFOY$>SLh{Uyj#X~S`~^Z48q z1=dF{CXdcJ(AjHHE(BVHsN8j>14kyE1~bc|*7RNWGkM;sR$S6Gh|4^2xw-mLa_7}b zXpJopx*}@yJz)tKJTi3#HO=3SC-Q`JZ3rz$A6o*u3I;+%NzWq*UXf7&_QqR1TkS~*oEyoq(H_-hNYb9wUswDq%g^D-PwBXfKQL@>N}w) zCf#u!`p>1N6^U5ju2l4UPhbphUd-(I~# zny7aDyg<_;+VhnTt@o{%Krver^?IVMo_sK&HaT*hZLM+xdgURjM)P^a%a|R!p7g4s zZO87uVhIHc)aPu(9;kE^>_|=kERXdcl}<3`a*9=^GqIjeRtUdMCw-N~ITFlCEa^O886 zRl9lE&QI{iJ@4={K65AriQdO_FX=AmkroG%So-?^y@Yp2*E6G``~o7J}WMFZ}0R>erLzr)Y6J66U`zXoBvsScmD!*`acxv7zY%3 zp7N7m4|Whpr3k{lDC=2BzP=O8|NWLki-xnJOwP#2 z$bzTkIqqh5NB>6pNy%%pKx@)Ux%u(GBdWoFx1i{lZhcV3dfyzcIptx&XA6eqS)z0p z`B$8B`%m}&vC_G3`>qEd-&=(o@UzehW;&o|&&bz6XyN}qtn?s%z>% literal 0 HcmV?d00001 diff --git a/paper/oregon_ldte_costs_comparison.png b/paper/oregon_ldte_costs_comparison.png new file mode 100644 index 0000000000000000000000000000000000000000..82fb79320865f03f96ad253e29143c1ffdd09a90 GIT binary patch literal 69226 zcmb@uWk6Knw>}EeIdpfcASvA-AyNk2At2q|H6RTVN~3_%-Q6HDf;0>%-3$W^aW{I- z|NPFq_rv{gKZwfgnZ5SA*ZZtzJ?q&MrJ<&XhfRfzgoK2rtR(*$2?=cs3F(mITuYieq~ERCLfw{N%{EjIZcWr%Gpi@-&Wyq9rv#5?dG+Bq7EJ7||9As|tP%mq7}kHjCxL+@^M8M=jD|4#|9VeI zu9MpTIxnisYUck$0sIRqq+8@aKa_w7^=}}`JviL043w0VqVyoCn^lnIa;rC=u(0r| z`7QMRHizyh-Tq?JoA}#3<9of=S!ZKn>V>Lb*rKGLGI{B=`rd6uLsBP%}uJvQ$Yan=KoB$-}8ftr=gSweV z3+NT&V`HgFNkqEU7EuNsbC%W4>-|`?GkdUM6p(nXkpDhAJGU1N1FGEy;c-^epGKtQNKmf2k1C-+)H5)rcIjQFH5?bxI2adMp*g|as^nnKO24A?t_mYy2UOn_e9oB+( zCnz8p-pllO@l){42D?;Zb#-;;^}7-+8jBnP+H-D*c|MuxH?LjYdrN6%%_*Ph`(9m`l1-6uyV<5kiJcw*m2@{wQPR5t4u9Fn1T7Y&3^ zDMpqi6NCs~<*m5)Y=Bz!MGbP8ty}l{wa}N*o@KeV0R##P3L*F8Q2<=*`;Q+y4vp*_mV&HkBhs`+Ma9VBZtAvmnIH8S`-eFdXXV!*5?%l zO|R)!PS4Izj`Eh7s&Vj|<&os14WFe6)-vNUg-TvS;_(;Y$?B4i_hu^6BI$gT#w6g2 zrV9mm@zoRz&G^L1Gz!<0hQ#J~{ihII<-9{L!ku?v3}4Sjv; z--1#KZ@%h%fv!4%N8eA5Uks@SyhgEPm|!n?uE``*LY6e$nZF8Ar2kDneiV3*Fn4J2 zJTkiK;||D66k7F3sHv{r_yWE8Q)x|!>E^{F6HhN6O{^~NJWHU%{D@3QKweB~)a_2Cx!pp&YEft(#fM1$RYt6V z8(fiimmY5d`d-0@H6NB#!BS&e6p^AqKDpUEb}oGu?&yE^Orx` zd!sJF+TwA5?Ps*U+H=e-7e+Kj9q?8Y|3;XBXn4nM7yd?5GtaV*z@!w1yea7HeVyL( zU97tf--~_9dc0v|G0-_Sec;W>wOrVG7knIPQ{;4z*`0DgbRxk(UU}vA{$@*xd0FnZ z+gG8024p30QpEQf6%MD8) zR+qk+MZSkkRt7mZiKoST#Wlj{?JDA(^3f;LXzNsZ^>!IX+ z78HlbjaDm-jpQrsZ4OJ#D9gs(-lkM=33<6}=YGoO7(|e>aldC^Hx_#_QV7^xd=YID z0?F&ZYZO{{)Ev+l59NB(?@BL|#eb*^&g|;xZTCm?`^TdTgD&G@q2#id8Mt;*>bUhe za{+1>z?N=TtcA-ih;opZGy1hFot=(w?|Dlb_SC)~rE04PdxS&oCxuxp-7Be(8d2kp zWe<)21ZA6gb7L}}jx<9eigQvL)S_^nf8n4{-A{w z&z(0p*He|gpk2yo}|Efru9+iyf2L)(meSIvIWe}W5=Okc^>)=4xhx1THnM6(`kv@ zt+eJt!0UO)(dHB@j7QCT<&)bgnegdpty|xl?cCO`9_A`;0eqfbiiPz^;&xMO@!Ukw zz&k(MN8z`|Zjcw(S@F+V$Uzd9oL){cZ8(3b2IwZ~l?xzI#duT9vr#LN^msIWTLi}n zAI)Ud`6H6oYH}UuF$E4G6mqwU>sUQ&cUR+qAJ%E}4XjCKRk+(&0&mvo>okKR7HK^f z^ETx(oJLp0?dPg3dwDDcs&S9A*3Z_Fk&dWiKc5(IucIY$uYj53DC+6D*`z_@5g*hF zG4A@vjo(&L#ZnYJlht_91exe+l4I!=%ur8{ zm&uHI=VdL^7OC;E`D#aUGG)wSxH=NBoFwPsgq!$ z=D~e~e1i#_BCUB1ii(Y`8ygxLY}@+U2w|q;@HD4^K3LoyHuNbSK~*3PF(MQxq&Xg+ zK!+-qzpqkyw%xx)lCvN&#(i{jRIkDLR_?8I=%qBAaD5p&5gCSu=#;}fEX#wJLJlFF z+xuNmT04KnFcu*@shZoL7iR%L&1ku$!DZ@j&g#s^gktRqio!L83QQ+M1j$9CC(ye+ zV{9?DhPJJF+Ci9w!sFH$`2hf*419eeo<#GCn)s2xE^fVImLesgU+}^4QbCz@&>l|K zTj7~Tw&SIBDlkbUhYW3(1X2##jA*-@o49l?UOOm1U_Bsck8HCW9&5$8$R4AqLNMfv zw|15%2LrIEWZK>D{kpeJey*#4!54aa$Hkt0*Rp~)L`SP)ot`KYnmlBUo~upxSpwEA z>oum7FoG<$Vw6iGC+Ld-W>NguxO$;wFS!avI}k5@2Q2`DwzXH9p3d=IZFF!X%%x(C z&h&Mqtz>?vp=KFpnb7RA>`b{he&I&dIF|0#b%fC4v&IUU0hQ7p1Y2u88#}uN82R~g zA))x$0X(5ap0#HFL+-Ae1M-f`@N&hr?VTO0(vCl$Z51*x3e=W77VLX=r&c5B%iX(+ zCojKDh`O&uv0sxtGZ5YvrKF)bFVm|V76NKK@A!b%Vd#1p4!z%Ta6se(S@k!n@SLsZ z7?-bbU)1h*4dfd$Y{Wu5=iua|IdhG355)3Zjzuhw3PEXfBKIag)a`nEpFWwumg}&RJp+Z(>)bbV+^RNo06QM!6_5XZEK`K}FgYyY zJov!sp$O7pEsUEtc4kRJ4&)Z{2 z+~2kbtUn#3e!vqxZ(D&jeSem7PowGBofWBV?C;)tdmP1{jp-{<50vIQ_tPPEkgrGM zUU`cf;-L*;9*(}|r<$DsAXog)72|rS78lD-7tgD6PIPs!Lk6aTI!}agEIN~@bYq_H zYCf^4ZjjOhCI5TWY4b}09}dW*-989Hx&8QdB^1Q=_Wio!0RKK+TUO%F`IzN#lL@(4 zQ)lq8%y=blUJ>yY8H55|0aWGOUPa5qO0a`W9eol~JhnooE&n#xpO~uZBWxQ>;Ye7z z7Q?55@RrCxizTzm3A$4KC`D@47%#C7Y$r>yVBZ zKSTR5YU&LVND((#Vce64!YBHS(uXS#POV9g9!cwwuF&*q!;j%LrY2Q#lTkm8&kErQ z0hLYoS@fVa^Gh^B67bi~DuP4;2OSpD1!Xd58jGD{km_55>wMCbVk|44e0Ol>#t4`QF@erkA*cAG%N#j}>?IK`K}8~TskX4(X`z(6to6?$F5=LbI5dR_ znuz!Nz=Op425mx+%1R*H6lNK`|-R#(d33@h$JXL)Q!er&^FR-fB(I zL^J7-MS(DS`O?Z+G~!WQR3d5wE0IENzCFk{Lqoy#jJhC0!&tN9#D{N%j--6lF~z%j z<~mq+*0Y&1&uEn1?`b9g7ygfIp?f7m*tWb216a`)ED5ra+DD&}_Yu#~aP4 zN>K>Tvs7viTeY$Vn+QLpdW-Q$7CU0!6pX)5ZIT#-&UHb^HNk;(k%#JK-4)N|H`*Pn zP>xI`QaKoeNim=PWe{Y$&GS;+3(2c5ynNuQznzza$nr%fS?rAys?;kaziACI&<$o6 z!mfG+A=AWDJ4JyWV~KkvFToRl$?Hdf=Y-p$S7jECPu+t8q7qFkE<#$is*J`gL-afG zw$i_9TrCl%C$_1iAR_~kiV1m4VqcJF^~}`zS%j*&wZL-APT?0OEx7b{Q`Z%~D&~8? zBQNOzPg?AV*ZIq4V1YtNIsYOC+lbfRBXK=QI!=HWj^fvHn2FP}!r78QCp3Q=V6xR;~l;1ji@^-HqWnD$QP+U$A z?V%@&MT)j(1))gpoagN!Q0`VCD!Vg9d_eI7Tk}!gC`Ae?$)aH+aq|gGFroRC9~Xpu zl>$>W-rpg9@8KFXVGL8mMFgnfP?;qj-S;iWrK;GudI z=D{gr--sXQ${@*LOH>|9^ak|fv{yux3I#Me;obzT7=$kf)^@q)-l|^EFo~ir`w~p9 zXvd&OLN58G9nTacPYLZ4*2E%AqUp`W1!Diai!ibm&!ilB!mP?Nrs(A8SKBE^m;dz` zNnjd9BHx_A60O5ZKuIGK#FD9Ae@KhlUl)w}0n9^x(4tPl`t2Nv%8L>v=R zsZdKJpETse0~IN^HE1ceQrd`!1TD%r`zx-ChqWkTgreVBcVjSyy+Q?cI%r|M2QKuW zIAmV%x0asxOUw6LkFG#GiP5^A9=qW?d|^&NrIkM(j&Ss8^uS?ST5Soe>|U`Bq<@xi z`|l>M-w#OLInTo7!sjfI;ihgDt48tT5#z|b@qpX~!;oH`w`hQs|7 z31&1q2)gKE&zy379|efT_+f3V0D9^#y7LrsVIl_y^XCM|=3r_zpLS5n3MQW6c^Gqs z90!G3RD|nsJaZTcZcSGhou@c118*4Qn7%|Nh@S1n#H1UKQd2+)JCr_72g}B~)7+Yn zh&;|6WkBNzuCDC*qZ|m5W_q4{i6@C=ABcwTG-f*`)RND#FX6R#CgVIM1cM`)_oqN@ z>$dVhm=c<}+cLmG^gunqK9P)p);tWu8`7~n0%|8q;erXVrX9(Y9wLx4UdW!MI0b08 zSbsfAQf>+~_FO*5I&?LQf{BjCJ?^WtN$uBn3M@EG&G;nF61x5R;!PV7aui_}nbF_%W2esmK#Q4tTXaBynCEu7>ImleN0(n|>gQ~BG z=Mcwhg*7Z8H27dEH1dIXB-h=>Tt}SsDmVKJN99DA&Lqw~yLc7-Xx>FlV9-HFKf2{Hz*%TIKj)I%6(X}ra=$N7w$ zaAWQ|okPUVyX&qPm3sJGYKvXn9CFV#xC5_7DicZbE>t0F(lb7#3pd-Zx=-*lBIWj` zz2=7T|U}OkA z-6Bik3u+K1#YGfHx{0H_yh5%gjnr{}g0w(bXji-h)qa=JNJJ-@&m`Wh{>?FEZ01VG z0EoNLI(XSLQ%?R(1o4L$7ix*RE{U4PJmwE9Bz}K}=0#%As;3KRz;>9ir=5EfuA+1b zh=Be_QhwG!Cg8_xxEc6yv)w!m6F<+bDRP|y80Z~vKlNC3p?o-=P(-=3rY3wxVFo_J z7i8^XkVTT(O^%w#+P+F3gz*_2GdNz#0b9scxI#!EjuBPSN6AXqNp)pa{F18ERvd|v zazai5-y=kH z=f_t&Ig~PMAe^WJIH!-{6L;Ycps%MX1)u$% zn)*7EfILP&Mn3jbp^@V1i<2s9JSK(u7y2Cf#2JW1E6@+dM=hE&E&Uk~!J}>CnXikx zj#o4EgLffhj`7BJ@hF~-*uIZ5|X-Au3(TtN1x9a6l|d|F$9=EK7KUyBsX+R6zpefR%f{lstosE9U#Dx>!)gt zISY+UP^9;NOZ}3fkOFu2X}gd>wT7ZbpoXMCL@04;7|Ir<<0F%PUqD_oKpcy| z5YTYyeoBK9>nFe1jRS&VD;ov?Nw8Bz@SV(lqO-qJEz?onJ}-N{xe31PtLPPm9(qKw zh0!C04Hs%PeVj6ecETBOY8DkWXGfy7`i# zhVe1HaYnYZ8y<&vqQGGgaJ9Mq3};s=e=-3aN078!`5<4&12clQ1t6uZ`UU6`ICu>Z z#52!^X&v0ym{R44EvRyrqHL*Vu;djo*)b0RPEZ#>e|hL!LFvd&bcl6-bbl+A#cYDi z)XYmYwPLc6UaUS?78Hl%CzwJW^5ig3Cw|a@h8oUT3u9O;xG7Ep7Yd0l58&Lfz12Z) z^l)y=&&~uo+9#Fy5>W_trc1^N>H-dp0ZoTY^Tjl$4n?c2kKFCWN2>uJ5vjZ(7;yd@ z!CY6QEbP?~|9%`%N&qFDB~K{(eaqGn(vjG^^x38wM)eINbP3_UU&dm`|NMau`)PnH7jZ`;k|MEbRLbhi03 zUbO=<#Pyiey&v=%wijvN;;~u86A!c4o-|C2izl|2sOzo3*#3r!pH3d5rwZZ z))2$m?Q4LLK;wa%0-BU>1hhq%u$ad_W+`HgykmQ0bd(n6c-RIvf+n%5oV#vpZ6Sa* ze!FG&M~NRE9$v7$QNWcYG@A6O+n3Lusr>JOQ9u#h-qltP$Zlq4X6tWz+`1xBa5IZu zk>j-%RS8=KWDJV>y9jEB30PTKSV!|B{NaDSM?c0|>*1 zzBeZW{_B&e@vkA3uKdXV3P^KYEz#jqkzG>-P>0V1_^@2Oqp2&n|=BudJ+8 z-sY8Zyx|Zg7iRDBJMx8|{hprQg^?M=D=Z9nRgkYEK;=&hPS)w~b>N#$!jJaq9ifLU zfS^!^zt-FZ58vB8pHznf(;`>Xy|LgiWJ&7j^~~_6woZo6cc%aby`$gc@@?F#YWu1M z=(M{J`D^Ly7wSjh;oN>*+V9?dgOZM&P+Dc?!9!4Zr*Da%GxEES9>#p%;E>Ernlk-M zyW&aWw&@;2_0`V1fot5(9k=UtSefG$8JFHyID>&0FxWh7i9E2aMW{5MNzVH_)*?_Y z!mllOcz9M?PKTfR|Ni~k=UJTSE(a9Qvwj4|+qA+H2k3k`o!V7izVyqv-M8}hg&)#h zz%iN?*Xd51#{C_Bp1s0Uwx10E3Km6ujv_Fn9lSUT9_G<%>$0Lb2;g`9^Rp7sbWp!= zeruKE?=fY5dp*-JZLbk2?d#ie1c(l;6}h?69iE))%S3t=-E~JNy4;N`k8Z^bLUZoU zi~Ip8rJ1%qUY?qGUJ5(IFH^fB2cgaUr1A zHovBZ-$@NbhWLzypT$aGo|!0MBuLuN#h_2*2RA-Ub6E(b-Dpb9%vQgtDe-{chk@9W zhF)2;VTOTo9FL$+BW?N z>-TQ#ER>xD7PtVTdT=x(xYCMkzrTO$N149{4We`j-wP!}Lqos`@vGACpKF8Z_p>4=I%EQfS>}02NQf7Cv4lE1jFR{4;#W|1eZbe(^Ve z|Nim+i^U{KSNMKzJb-PoP?h*&AJa%0ciO|C%G}|;msowU)%(m15spJX{Gk4t9_0YF z>$CE6FVW4EH~BY@|Ccp|<%u$Drqso(xl13Z_8=g>GVD#4CBOJcHUt!eZ{NPL?q2|v zAlmN_e@-IKvqtIZ3S(}oZ0|xfz-q8udBBh&Bq)eps7U9{qL2grpL+%vXYvzQ&#wHE zSQxr*2AShiS`M34j6wMLF9G|9fzQr27uav@_}RPtnsKt0rDuVN1?Z!_2VV-{9BBk{ zQX&|9R$l(h8_*8f4a>8$=^k@(rW_tRJm%m?KM(kO2}-+N^=A7kO>BS7H;EsfvZ?kq zjcg{>+0G0rvuBa$q-A7$uQ2w9{jLlQg>@~U$QMUC+;`(}D*+YiXO@`uv-V3f8%M|Z zUquZ6>EB7(wopl$n4V@cs?dJ&^l4<-z@F+C!SLHJfX$R{Jq-ZZ zAwc-G7lBQ^d1o*_JdE99Tr~OpjguqVf7*KzrWkC(&a9osM7B^<6UEOfd+?Jp_^m#W zTmv#G#f76o(L;>54b(%Iz?G9`u09O z)sDEOE(I;qiJ$%CFK_@|94?j6Z2|aQ;E>XF5jvIid zWJW&Qqs8!nNBG}A_*)c0sUQ48r$C(=v2R!^l-$0D2i~(gwqF$*xD2sg2N)aY`0eRK zlV*&3RXq+DzcV2QBrkX~p4+Cf5wiYg+(`i}p9Zexta7-38vZbsv<(^W0ZAEzJ2NA1 zA1<}9^1B8acm8-+vH{4@skRkuTFC)p*%JEMey7S0PN8K;qcbHplG9-J;na~{4>_r~+u)$ay0LvC{7yk+tcq2|Q@?7wIw|eA|2%4+T zSIvidPdY`WG*p`8MV|I|Hzw5PEkT}%iFyw9*T??);7AJ3)t|}Y-NJobOVZ!s^*1u- z;BJv9;m~Xu%5Ay?fLZJu?Q18tcC;BY|7>%pu&}TIKb~RvBVIE*Y@d{I;EB7tyu^j) zD4-Nz-=1Dt=oN_jFKJ(sjpr&}Tvc=1U|~+^ze{sNdUQ2bo>g7tULbHGxZ8ZX=<0&)1(VYzMSrd5-pvQp8=er^u&2ij2@)IJ#83eig0I zTb_kSq|CpxRNKi+Py6!sW@Q8CHGgz}GIo%)w(X>J)NZtMbgW+gjeCLD}AXjd&-@P8Nbmhlg^$WUl75Yef;zZgAu%~vD; zaw|pQK=@xqer2+ctCy}G7xew2@JV|^3p{G^<+%OBsja|k|Iza2F7x}7y&tL-zWw}r zcocfLD<9jKLy=E<+Rw*P85N8~q3=KQTSC7ezUG!oqxt>|d2qY2?a*{c^59tu8zNIz z{7p36Nymi#EQTS-xCN(MtN$mBZJ-+>iihxU(8VMdfkJnJ)nBLmKg$b zR`xRXQ}V7TEEwUeKR;*L3jF$a3qR#KsVVf$t&@0m=>Kej*ayY4Yg!LX$2ETX?^sRj z*Y&OzQags)5B*!;YnR|2yVtvQ*7uF( z?jgbH+~T#pEr6WP!jZyZYeb*LlzI~RmvZ069t>8RHgH& zp8h&a`#K6CJ6ijDi{vveOY&J5Pw=k4Ry1)t01;G?>Gw?@B}UY}B6<1uVt&OtST%T| zOMJ!^&@3LU_!M`CuUlgq3xMc<5yJ$q>}qBhTEN<&k!LsNR<6m*;95tUnZJk%@Y+X@ z+!Yl3xJcw=ybDRylI>|1HtHlQhH~SPpFJ#rGA)2qmBKAqYc7pMFNMd)hNe%g^Cu+P zu>X~<9XWHyg4yczBS$)Hrji2Le z5%UYmq?}~>nNz@rIycjzzWl*Y<@3Jk=byD~9SSe)zcm|JJbua1aS3tq$(*+d!Sawe zxIN%&v8ey~EQ>3HGKzFV9$4$IR3u}{IzH6C#KVQG6e@`?XzcNA3>G-NKY3%3ldAu{ z8%iuYblYS=rTMA*UfH6MOzl%ymj3h8o3E7Y${z3i=!6(~I+o7s^#&T-8~~hF`q=iu z!{cBI&NW9i^nCNjtZ|lcnXFG$ole1@^&RH;YAPmgtH2_G^XJ=EJFc*(k-el>LsPET zG6;<7T;E{M5<|A|)b7Sl4;4?@9k^43jfkaPXNPLGLMMFeU7WACp&pNy4LC_7wR-<6 zA;aWj$AZ;{5_+les9tKlPmSI>s%tGWdmgr=6HPbdAD#)Wm2e7jsCv{i&|a_BZ)bicL~$ruYF;P;XeM=B#mi4Ld`U9E0Po?Jp%48uAT`&0?jv>UEkp(klo4xv3v$h|2+t`vgP7mfGa* zZJGzNtYq;J^vb^TaIUXL^-p43f_^HmtOiG$v_hCFAFT2F$usjM{C!r$Ld9GGou@GK5dQ!Lz0-@^Utj_ze|`zmLnT{vTZszq$qtU^B>uT4y{T z$#sgQNuU$eGRstcD%E_|Z%ZSC;(E)nW#X}C!9d54Zc5K280IejLwi%}g~60UB(^Ay zlVtYYicPF!WDve)_5;LWU;yzw@=Z3Z<{Hv`ZUR}eb}zP6aj-16$e=L%*Q&n8xHMtD z7MWcarIaRl-fPEF9d_yWGsT=~ttU1vtlx4>Hkg49BFgAWXblv2~AZ$w~L<2vV z21^`%z7H9<8ma%jiM{{9jPxqDni0Dlx>P}Tq;P)KTJ7zuGVuVTUmx9&vi2M|x2;)w zf*%Vx6Ng5qkjz(=$;}HjsPsMhZ?O9=ryV%+wN#u4Sw2W^)@Y_uZMf9Z#kwyK!&0}V z%CeyK_f2Z8*!;IA6)y|s*RkYxWL5IiqL4~02>h4r98&`wZmk%u7mr;YmVL1IA&*$l zVZ!2OT?ZTbP z(9}gkx-H%;FacnpeL12GoZTJN$O^5N;#K2I(B4renQv^@1Q<#zp3MX95+t2>=A%J= z3%Z`NpyOeLE!t--SUj+!{!IQU@*C;SH#8UQ1V0-*?PjcZ=U8hpE?qJ^-_Sv}xGu=9 zX6gfn4$GIQ{Ef2ppY22}>`i;KNE{Qt+wTtpWI}Iux47Bu#?}=(&!s@w!Ub+lDTmKXPNF3oF`;oGr=@*<-yNr%Yp`dG5aZQkw$WvY!lh+ z!WnQGDMq5E8AIckXNgy@JsW9vzr}Bk-KhmGhcZ+xh;73@ZcD10P`a{SR?&@LB{2fp z^ygWvqMw|k?amVKtt?dDdlZ)U_|AgD+-Zo1tYYyo_fcRNGZt!EaeAbt0?NTnCSjFv z@MKkehOh*ahbx}d-~ZsB6a!RU!N`4I*PM-aTfCYaf3H^g14Dw0c<5!yzI4nNJ>hrD z_TJhr%f40DdhCGheX`n4jYfP=)kegcaRbj!g&tB7#kf@ziC4l>v*+I8}7}DZH0{+wfk$t{}hnHIui9H*@}P8+v(6WjC)hCS<=*Nz$?8$Gp}Y z$HGVR%dGZC4a%FQ-)9I|{#Jw?QfGX|C-0a98D&!!hBsmU6)I6>^6yE;Ctw6n2AA$s zv?wk@Q{y+HyxeXsCMJ_1+bd-U`nQm+-zt`-o82kLl?npq<%fpV-`$h<_m{uBW4qg@ z)3>{CKy%HpiF=_3Sqtw-1OX>OJHhm72}pAZtRWQU-s+iaOMJo6)h(Rrn%XnI z+)0PLruzihxpe$DJNIFLW`QpuA#RtSI-G7uXscu*Ly5`s2XdmH$BvziYwggPST;yL zN3Qji0uP_5;tc*AfYV|q1on99{__0vnBetvhUGXiO*~Css~NSq#dZ;f8!Am;?5M)^ zZ650@WM@vRLubFbL^y8H`6uI=xfB5eZu?A=Y=`q~h>~z1hs!pTJ}~6_DL`;Z?QBnN zr@F&?=`fRPUcOguDKN~2UbXHYE_03SrP}s7GCkrYrQ0MJF)dkh5)pXf{m?JRzV37; zcueQOf|>n|vi7-%(=smfEP4^!FcR#Bx$}9Y4(2Zb-)89Yj2jdB$@rE056NjOv1~rj zlM>~26d3bWzh1`}$h4Z4&-*J)xCi`V6aO!A9(7IMg3+6lqo>7%4i+xsE{*r(1%t5@ zz|qS~HTs{0ghmK|D-P!Wc!6b3jB4yO`>iw1(s*1F_Nb!rosz$d*>C#jhQvPbwc@RS zs2ao9f*w3tb|lEw@qwbD^xKZ>ulgS|+)@aLc)f%egO^X*`z3kjURz!20P={N!%<3V zEZ1v)FZZ8BrTA`m*5~bIQ*0t=QBqTh4~0LB<$bS6E`_4m??m1jufqB4(z!Z?GB?~D z8=Vgn#HFETpIQkMj~7I5A^nwHvI2ZevB_O(5%P54RX|(G!d;+o`*DtcJ{;(K|Ma13 z_@G@pqZoRfrA=GuX_^JNc*wA~)tq>MP6Ff(@_iEHTv7#VwjTc5mR_1u@l-`V&z!o4SwiP+e3&p6FNx-+{(QsWbX?hI1LA?uZ?8&GeUoT9 zoBN7=6?hnuDw&?#wxz;1$_g8LQ+CwvqOIMgwB6csv%3p1;o4cCHbFKvcD687KwaaA zn6>3n%lgI}eSUJrS1pU=#Yi;=8&Jwa-J#JRYP7w;vNKG0E1> zDoWh)HJU3h+GDLz`D$mBu`#p3|KXoH9fQS*_4#1*$teL*{Px+8)*poKMEMpSEyfpa zM-Z@E&m?s>v%EXzApyt7(ic6b?o;b1QHjSj5ws(KepX#%8&3x^ngLV&IU}Y76|;kd z2i|7yjpGH&m$#K0!HzsZ&3$6qv=>6#F~N;-)*i2l5yd0LH4003WHUWc2PE@Y%J#tH zqiEAiq02I{ztrmJ)phDS%fETOuOd3t@M8A`1+J{E_*X=$UY9`%d&791GJ6mmYVZNK{oRhGpcJ$br1m-Bpb_I znMK;a+uyC$_wnu%Q5%LI=BHi~zg7Qb-KSCtT(mkmySe8{1D#RU{ z=)lUx=#a|B`mCU(kUAFYE-VP>L1rl7uK5h2qVIf9pKDt0Y2;6Q_8S_*GCm;WUEOv4 zet_M3dML4U?WwZ`^I~1dER(&`L10FHo5)WdIZTQJiT-WeO1V>_(MVqllv&Fn{vpmz zA_Yle+xO2DDPQpkj$$~VW|VYa!>o;&U;6tFcz8n*cXxv;j0@!8RG_PVo&JV{KiZ1t zJ9E@yf*RhS6$lG&@g78)ff9}Dfe-TMzcmL(Fz_UmEZb1 zPxcCE(6KkYKwBu&Pto3BGZ95Ojy?bAU8s9tMkZ1dEQF@HAD-dq(~;cM%n*p` zBubNqNU)La#f}ULEk>Z{i~Lb=Yvqk)t)~vO#e|1_efqs5CCM3ukvzot2`f4tE5F#A z+l5tkpqR}x_yA3zt&!c_K*!8}#UUuzzxr*)JcWEO3oTzXt~F|FC>0rMHB zI`thFut^ciU)7W>eZxz~5(7e#$DLJ?K-&_FrR&I7 zH;3;VS{tVWe?>BuQE7k!pNYTu&@hU{7=?xU8&HN6*v;4YZBB!aeiAgWGu2avH1jh4 zq77*o*}|4@r%omWqVF>I4n&U0LT|%woVh%BFMGQnXK#}`hw5mHd2ne!0E^uYY{mB) z2Vq?nT=fk+GEeFB+L{mMoq`Yw7xfp1%G_VkI=5r^i|Tb!h?wY{gvC+hs;WP9$y4RM zd*_9%GlC9{-&~2fyNKj{!5d0^(Mp&8_e<H~vtj_+Jr<-W)GMTOKtbHP z9obe-@-lc%N^wn_Q?8d~=zE&9N$iDnk%ri36Gq8Qu2{zH`1hv74hYi$u>`0eNzh9C zx|xID`%IvR-9cP)Gw)Zg+B(`%FGhur#{#zOihyIuI%Y=ejiA-2cmtA%*|?dmHJ{?k zIJofXsQQPe-z*FODV?`1D&g2VDbuV(ZOo{s*1_&m{iywG@lozvay70P@Qe?SQG<54 zbBr*LybH|F(Q~`ns+0K;xgouj?Ym^Y-Vbu^67WAcFds#}EI_?uD*eXW8y7J8Ok#Q@a;*jA^xrJo%oCrZ>>S3P>2k8P&Y|ZzUOm946jS)7DN-} zxc6xL=y>4**Igbp%qvH#=w;#3sm9{50VKeZ&jAWT7*;o&cQ2X`)v~$`oPs0$0=N!7!Qf z9Wj7veCemsuF!p?SEF)Gh4|`VnYHnb@A~fyn1yO(74T}&XsKdiq05qo9sOzlQyC)s zhhS`gIhtJm&mZdL)Br;PYqMMw~`z+Si{E8=--Ix%JNoCRE;kDsIGIaYjpABIxIhTK)~hL1ctZa6hq~AUmH3tRUWdvdbLW2t!HYw^>%Z`3sNVbzjGecupVwM;CbEglU4K?67>`sIiN6Tsi^#Ga0>6>m84n`Jal zK*T+AsHd;>cN&@G2n|0B)hK;?8>jCT>LvNPvb-WR%&QfH*Fk-pF3V^ZCM*4_VyGff zm$l}t#(91IGb$fgE?x%;1!3+m0zluN-&Bc=^ad6+MfWG5i*+?ZV zwbbb4V+|>a67n1Sp?#vNhIz|@4$p@H(|2Rt@#Y8C_VsbR#6`M+b04Y8xCDIgbaQ0$ z!LtcG;4|`P^vk#We6!yF8Sd*lG`$!TUdQ#mT#eiWo^4BU-oyE`a58 z&@teQIz!UGIeLF~t3Q^y&#X#GR6a!z_FR*JiAA-tWxu+`s$qZk6la5Am{-OEA0wN%&SrZpMKpdXdA+$lBl+R$Uw23^dO=DQfijt-YlE_t$lTcXO7za94x2vqVB4$ zhiQ_raO<;c!DSR?)U$275Bs*bK(9tUsgq?vkjcx`t27ubQctfwFyEDsk+^fDF;ZPz zTs|U~+&+e_Cpc|&m48EbNKk~us){MXVF7aAH` z(kr)v{iN8y#o|YH;zSxrL&2D?;LS>n;9>`z5^0u!J!>n;NF``niT(>$oraWAF7w-m z1b$svCQL>!0(=95QWIdcI)|;S&tNrYY1!oexv4dAMD=!+(G z0amr-GHxIVu(34(Ua3vsDf+_pcvmf5-N(Q~!xfvEu4BqquCdlX|1a9kGAhb9YWLC& zLrd2HQWDY~LrDqJor0t^NC^@{!+@01C`fmQbV^8fcXz}4%>P~Mtn>AJINw}LafT=E zd+%%iuIrXx@`23FzFTtBL%(#dS;%VpCXhweiRZW;fJRe zig%#0{_?a>0Q7aF^RZ#V;V}GfF+yE#OCpLn+$&`0kSGeERi{gfO>%uf?+)nM6QmPG z>G(|S(os|l7`p!D3v>sBAeVoN%>4i%MtEyc>9Op05%B6Cht2?dz$?p~F!MG8;>vi~knt z-mZI#7KN@s+|%^1CP?33F$~4U`8lM|^giMja@Xm{ka110(sW)bRM$kK3X;E~+CAEL zK_VZ?d%?eRv35j?G_Sp>^%2sDpBm#`8L=BGxOaK`n6KXUE*8`fDV)kqJys!VA}CRg z-tKZDnuI7M+>fZfvFKA%8iRVo7Ru!aJe)GTG=urRmBw2M?QeEYBX?xX@4g7w2-yVn zQS0JXacn^&rl*acJ!S;lvZ>_GW37_7dv^g<7`3#UgX3wZOUiF7S%*C>zAXeFI%0b5Q|>5++B83*-`C znC#Cm6E9Co_fn3BPKxo|xrV91&8yr_B*LRI;gLS_iq~F#_1E(E6zgK>3!u8MA&y`@ zUINb0RyiY_rp3|a*r{3bFJox!?qlVH=!cY;(|Wv9KDuEeNYA#=?2}?Sa3yz$f#;u( zqd)P712?lTZ-x*8eH1ounf=k`VK3*AjeFlwX0ijqT4%7c1=-cx@C0v{PM{s#*~c^p@x2>WD_S9>+_ae?z>bW;4$r}^VdLUL zZllInR!3WPe%nse_V;bP2`ymrutj|-s|__YY|+g(GylBJ50Ds(H0etfD)5^4U z%3DzkfPlbkY)Wb0zr%5)!om}N23?7%s~*bz!lI%u2PB${Pk&gMW}Z;+w@>uc=DM<4 zRLIE`6$;SLw9gIBE~fG)2dpoLCMzo)8AbGhY-;KToR*%Vi)UNpObn_@&VI{?Gt#8r zk3pe4_2P8;A%t_$R*-#ZptT={#L}vKTgUTp5H2g9xteJYhrMxWcip)M8e~Ox6hTMW^9~@@c z_D5`pYH}xI?ZC~AWx@m!^)}G%TpS$5a|A5XPbwAt+&f|IA6_j9g0N^>iaFx2t%ciT{h<}*gI4;%~@OM!2M+kl^7fG!wr`*i^r&9FslRn-{P z-!G^md=^_^P6M%?>Iy+G+BX`gM*82=h%(-PzDAj9TYRXU*qR<--U^Tw=RK5R$z*q7 zzgv7f>s1NRzR zX0M$W>c*^>GraQ1LP9+Z1oZdAnew#aG&D2COM>46Ww3#Tfs~uHgvY#NPWG{`ta7{l zp*58_2<}8aARqi&p5<(sI{4R>eOa`}SjJjI&IW&p8ow$WbNZh~R=VwK07Y5qg8%)Y zdENXbU^<_^ZaWF#RB&)8nfv`^@O6`NjL4B7oX*_DpHCOzq47) z?}pu}XM|d_918p+gfcqux3Z`?c)xYf!{!QpyqL`&JROfA{oXI3aWmv?UFa~F@I5mT zXz~`Cj*PwSJ1rWYCiL=AYn8P4YaKOdT)lLWAf_1sH@CJXGI-9?3Op|&e*hEh%gtUt zEFlm;jJ<1OUKo!9C4*7PTF0>M`|8_4|2jIPfdHzcprEe;53$4)88ftxRF5!|&ebe% zCwLC7LU6%wT11eGJ*TufdIV;u_&O$@p!k5@0C=Z7;%g?#hf)4>$ zu@7&ye$dgRp2tJMxk}(q{wo$w@GV&VM&TiRByFdE6P<1sCr=0bJH$k2YfHH*apo;= zVRLPkcDKt+Wn7mX8_$}kit}|-_v|52!nY*_@ZCnzcHy)4{4f5Bx*0y_SC9)$iqkJN zoljT(yK{?-e$jsB92F5i`*Jd=5`?CIz%so@#a;op8*y70hY#xIWw}+d&ap%WLabN1 zu`^!1yBNgPYh9XGU^Au5&~yqPE2WZSb}u2;8>qRT`I1zk`WS$+Mz}2>#}9{vhDG_R zs>QWwuHwOZIsmc%=IrmSZ9`DO=|D4*!IdPtxWm;)m$RV+D_~-Hc7GjE#?J{A4{#Yz za2Y6m&x_eIq0icXbomr^{hXEAFSaukgW|j8N}9peoz!=mLMW|}$MJJiD5B40VSB7X z?*m%`F(carbu6S~_eUzB;wYrxanybg9l(P4N$+6u1=h_FLXR zyXooGN_jrtk|+`k_;-=8QG9FtTgS07T=m2|;3OSQ%L@^UjbtF1G(hhcsamY(DIg*s zJC+&b4w$utA{T`Sg`|0!>5uWYpfP;pejG@M?9#APP$v8;ofwn_Q7Hc6m1>e+W^O+| zuM)U(zLwgp$ZX$XoL5(&(+DVc0HGJn8n{CgFmYqar*zKM z7p+*uAl}Jbx~hJ#yKmD>dETJQ)WlnaY{TXFg8S5oh|9ojPj|!dS=-h6r>n!tio;0s9ZEs(G}1 z@8ZeS=Y?ijx8_~!F|5LJa{RP^B{_*^?1rjfADI4VQ9V7nDtmKQi>ynNCabjN|4O4cbTdvW zxy=Ei!wj!y487@kRX`yKQf1rI?#su$O&kKc_;9)RCJ7IBhbDt;KE`9y<($Izqg9(x zv(`|SNCdo!7uPv}q&2Lb45EsBLCh{g)J6HcdzXEUL571q1MLX?OR!bW&Yi@qb8-`c zFi7^~W6kt5IJ6)!G9!_VyC?ca$%V)DqSbVO_qRLo)}8PmSV&K(5vR6SRQr;5d5W@3f7atig{5)Z#W|T4>OK&faz%T zaQpot8vO%gL~ZH=O0;sPHgcwbV4;5Zk$;`r(T-*yHEo(v?)bSYxQ<6B#2DHP+Qc!V z3=#}J4a`7j3#%V2LTrdfBA+MksApLQPw^BT2~6gAysOhRq{%Hol%k5i6vKHsCeB(% zr9wriL9zX9F(Df>1T=&M4s-YOEaE@WK_SKqmpPbQ0C|ZzHNuw8oIfIDq&-;@@%x1@ zK$8eN9L}(|E#LUmM0h(K+X>Ur6CmCJTYgWA7!;pW&E8k~&}y{*%f6J~5b2MivG3rs z6|(kS_qV_-XGEgQ1^NTj9iIoH-^jv7%?U7TDrlDwe?CWf8o!hoA!pz!)Eu2}4{D<8 zS{D?ghP6mPM0py4*O92$>a}s`47X*>>&Mig!f~sEqdpm!^Lk7bDeM{_UXTC4O`+w= z9e=r2I~*2#w9;_uzvJ_;2(F;*6xcJ0fv_>yH3z0#4a*$ws9=vJ1$R+EyIbx*YjK6v zW2w*S+kShP^Cf=GP1nz~6!~##pCkb6b-(2ijfu9WW}HHJj+)Li$!o85x46e#iW47& z9bJxEt`KNyfAe{t=b?p6%1@H_?eeNNHx=O?c$g;VAO2;g$s0{lrDA;v3_9?)`G|gH zo(#KcA=;N6!?{dtOaOhDkT;Gw98Qe zGx8N-khZx88}{gOpj@P5y?v0j}poGPR*2njMhf_rB2|&A`SE+ z#3sDJ|Avb}k4_HqM>~f$Uy{K?zG`69{VT(CNX;broNBjrrhJH&MNmM?;>YPTslKp! zTy-MpM!JGr_Fbleb>G#8l^?4=$Olt$$21wyZn7}~ao%`K+`~?J)cIdWbA{(h$v`P+J z_S5aomR2%rj+E(Gxe#>Mlx)gflsI-uj(LhPE32@F>RHwD2suoBB8T`Z|Sl+8K^p}^0>OhWgf<%nq~4=mq!W9&5snbV>L2$GQIe|L$Sp|u#Pz? zXbDmV0&yR=8QIExuw;hyKte<_W(|h)Vi$5eVbF)>Kj+vGI1$dI*h4w3QHyz&g!6X6(hx z6!b|z4tb{{%)`7{pxWieM{3OnXu;l4iu5Kjpll%$cX6X&t+uZTmLw`v0(3Dh*Z_&D z7WsR|0vH&!&d}0f#PSQ?iLvI{fy;eHIZe0jP?RBv449PC!p1ax&%%@n?6dQuMN_Zf zv$M7;6AsPtWy=#I<_ZM$9XW_FL4s|5%-`kTxmT~hJ*iSbcN1z(03nm9e@a{V1MJjL z;DWzzoq{@<1?BxK(XQXv0DzrTLu?KopqZXPU$Dt+i6t(d#ukGlD)nMwdr zKZX0!tEV%JI{kIZiGD)b2kXLbxLX#jxOyMNz`=17`94_&C@(QX$IRd`%y^Y`8y7VJ zAN9$a*b7htEhI@-$<5f74X0Ty*xTE&Va^at@#^_DER!=cwMFa_m zmk-9rCLfdP67{3$WFa_uNDX;(EBFgg+_(p09w^h=u07C2P%f`FpLuJTvtU5tEHW^I z^69wN?=TN}N73DN$}smAhaGE_f)?>g?Vx*tPN7+jGs7?WdE~-CkmHp;JDHm)Urib5 zI&~8qCkHq2l&ZghP*3QfZV!Lxi39pD0=KBbP0zlr4YwQt5S&0CtQ@h?mH@**TK?cg zO9(5LdN0q|_ae*Xx9`8+U}(0 zD&FxFzt5-l)cvPE^JwZ8yDCvOuhjENLYnu(n>P>i-oH%Pr>AB1xx!9xBWcrz?(>za z0$L*WdusJMQLaZXEjaeOhBc+p!?8f#UU)IMu?}9!)Ss`jksQ$itf9-S!SkVj{#AL~ zeQFswk^Ty#PiLAbbgb-cwv9zTYlj}V%EWNScXO#bPmvHm5gUaIF_QtyLxjF?MG6bH zcr|Kn>)fNbRGHAqPq;K9w38E-j+m}=IzaN#klp2rJl;va5Ga=A5;%~fwF#SAt(amg zAyV$S#KI;*NpDpS}6KSsp(j_1+~1p7 zc~~LuCBn`bV(}uBWMXNE7>!f#EsM#_v4cmW!O#opQtCzgh^KITBcb+Bok~)y$ZEih3|Vtopxu4 z%Kj?yxIFvRNwk3{`XnBY;INZ}qAZMm>QWffv~P)tHex-q0=t(&W`U-%6T_>=8jsP7 zNA<8gzYhnf>D1W_sBwKs>8Qb!L<*MM)NVz5NJ3su zRp$~e(*ei-He>%jrucfSTCW>{VL>*qeyK91nF02%;{1qiL!5X8N&yzl~4=HyFCU#>XqL6KNAXH-e_X-0*pV0HRPe*9Ig z1I73mg;mcp?@Yk+HpVo*nHx{kRy@rd20myc8Xrj}7j>v%>+wpIfBciP7;}^up2!gf znVfY1GfFl3=|G0g26NdnGNLaSj?IzjFQBn6~z96V>L`3=pp7HUiw5daf{=s$R^E%D(}7<8viXhKat(z0fj zY^NJ?yYY3egj=4F6O-861ZpRbW24<0#hH~H#JB{eeh1VtDk+mBJRgTnnW;Y8^WY<@ zPu-Z9+ul2OE0`BCAVuLLy)K)f@}LsR62Dzq#cYvlMGHSj%M!9~O~KTK7k=oi<*fl24<%GMKUmY#tv6?A8m4@S zE0Okris8-CJsLeCAczY_;Mr(4^=ArdC%aaHnFZvPy=BastP0NA6%6r=f0;_b!VyNk zxZ%3JEe&r7!ws$bzLaEO+$hTs5({ilhAtb7Q74R;rDaL2B&q5=GFyqmWLhWuU7s+Q zfm;DH{$HbY-O}eGUOvA59NEZ?VUgoRtHGq`#z!Fgnr_<2Gaa7F59B;>NZhpJZB&n| zVl~I2)lxXtO98o?;=6l$u-q8oxBgG(nftXYsV^svNbFd00W&Efn)**r8;;oms4E)% z?5Xw%Y5u(>bm^fAonBq0w|S`iw5`E+0zHth`-{Ro)nQ3E926VBNS*%K ztP$6+ZJZG&+cDE9I+<2g=3B>lWKoJ;+UOb9=sIWu6Y98dn5G)e|Mw|%AtS2n1!8a!-Q&#%b%Y&BIRBx~tfg9L+Fbzz}a=fdLrRqn>q?>&H zw3vLNIVpfHRn^H0*SYT@JC}Q!l&LnEEKtp$4H)v|LLxp8DfSN$5Koja9=L(~(I7S( z0%S~4b+;kT7DhJP8oHu69&bSv4E*$9T}?dZKWJE&+4~N&PmltIj~zgmaai)4n=CjPohg{epe% zChK_3Q}P5nl8t1~8-s-P^x8Cj5vxQ3`wy_}ag()KYz|EFMs!33nXw5d=o1>h;J!7{05t41mrwA@UWy|t-$sS*?oorvV zSXGPJ-?Zkgyy}#L0ELQa-anB&)6nw1vDQt0ATXb z#KPd=uJv?GJ{G;E0$`2Ga`SA%fP>3}FAAxPZ|Ue{cWx?i2_gN#bJk!}Qdr>)3@Ipsv9 zM(=wXzYweM`1I^$Ut5qQa1D2wf)xa%M0H>FUXx+H^Tcl`&AI#IA8&->(d+LlrVFI- zwW39y%IstNFfEL8MKiNL)$hkfw&yci+LParr@V9{faBrQZJ$~q;ALQcO|PNSnCM^U z&p#@weDV`@l{O|m=G$5m`5ghn8Awo`y88UwT-W}kEw2}sSIfuoU*!Qv>Y@u*x=zhb z#S{+TKm9ix_hBQT5;c^*2s~z{vp_(|adybHf4=sBJ9WF?K);N~WHNJ_1iAvMz9c39> zop|53*{VR@l=zG7M4?poU3LOOv9YG|a;u5*@zP{Tj{W)kUtzKpA4nKd*W0KlO!S2% zy7qZp>UA+0N}4h*a!(#`T-F$99EosB#s*K%8vSTbCi?pC?3%MTB7J?tqF<>Qy_e~d z0Qd_{-+Kv_FPnW_X3ayj&+dAq8$?^+AqCJSW1ZEW-!N-DKK5#WI6e6i)Jp&;wA-oN z3))&Tr<8#DrxmcnR-`(dzJ9AQHabr`Ef$b|A?RbN-Ot{zIz!QAyLS0wxLRl~ucggM zL|uEbOyd+zt|^MFipGDj{%@k*pY)Q*T*jDBR*isK*Ztt;8doJvaI|K5kX%}bKFL$G zNyekK{Qa*?K_lG9i)?Z~C!gODIK8XIbh! z<3g)wIZ8PlbJ}o1cIi4W>K&KmE^|ylrHimCbRlNJcSZJa-9w>{nu(Ea8jrf%z_dmA zmfg{p39?0zsgirqY3BR1;>uesI^~V+MiTnNz=-(4cqI={MR*V${Nh_q*5TJ{feE1~ zNkpa>OyMg|xJu!L3#^K~Pqmv)m9@?#Kg93d^fX$)W^!qK*L;IDBPML=Na=e9e(TNx zZ%b1T$K1N#?8F-_Hg&-q;t7AH(_*oaMzBOmnFHm=KTSbSHnxP~V)l=YOEi;cCyj-y zzF!IVd*_x?ER5VQpUm#OaA+>Zz8EQ&qhQkyuhwMjh_}9Mr7X&x0SifXF%i(yH#Ym7 z`)A{g0prz_QhFiCRIQ2wFj+lT9KU0leJVSwJa0(ndCc2&0p#qlogTMehmZW;+n&ep zyoEdRoWDO}hvEx6iyrV}a6M@eHIl=ANnTj=*$v@1qp9zw0LzzGm~1Q_#^zpUI_=o6 z-^j2Zg<^w4Yi8yQ!rZ2Zh@Y0mgtBvNr$#M{2M68_R;*kH)lFW9*3w*M04U0fJ)gr1 zW^=x-3$uh%<33>-w`qlytZ9(AinmmbXWUK6Alp1Il!=UdlVQX|zjX@__B< zgvTqP&SI}nV+IohVMFwA=0&7Ma`y}y5{47pwt&aS&ni-)@t+nMWXDbo3@_S zc6mZMk?|L&O*D6<8{4CDTt@9sq$qJoG|ZeKJ1kqE5dkI%EN}l66nv=COJdif2iULT zAKqK3jktKGUViKb+-At7N#LNAc2CM|jEbMN)fwNcF9`aitX1%9J;q2DAiH2RJV`XGu=gVq4*MZKc= z!^gi|>Imw!G>xY+tORZMCr!3W6^FYHjj^fBBZ;?}X`HqHF34@Mi*O++6#mnKCRM8G zjA6cKAYv=rsb^2#cwlMn^+Z!IGL4W(qrAcnr!V-$2CBmu{OVtU`r_%=D5(0=l%tn&E&H7MGxQw$P`GZfDz3qIS`cjw{ zB~zxzfh%iXFF7P77hC0u>Gwbp?3n6VjHQyOyO|RFl;wDIILo#qRaRVJK&NDv^sf+z zT2jzdBwwYZ9k`D)dB|ZmYSc{PFkq@%_HP1;6rtyb^UFt0-&HIP1A`)M)|JY^*9q$} zBC3u}hy3s}Vm5)ZYrY9A`@V;V%K6dY1%W1=fC#%(F`rt@U3jHFF)@MMb*qWsfeV#W zflNJ_m&|>&ADVd;$*IfkDj_VeB>PtJpu{7ZX!+^Dqe5U77)8d5uIG{QT*|!f)))2Y ztg`n{rF?E#*YXNDUcKfWxck}`Ks%o?%T%SB%6KlpEwSlmfv@;*By((=|CR0!?H%>g zaXhpY@9G&y!g3=TkRkgJPRM{oDV>7@`KYT5CKx&PH4bq}-0#6#Ea)fBoVf{paGjUf zOi%LMC=4+9DZOb;>aYEQWF$(@^7L}VgcD#$;3A2m@DmeDO;XRs47lP;4L(7q_O{O8 zK;t)_TratfM`L(}B1Z6Dkt(EWMBG5%z+e+Ni(+~s9ZCQ{S^%kE%=37uL}ngGyF{N6 zIBKjMo+);A=oDbkeaf8at3kr93!2#~EGpGFeawk`4TBT6cRRW6nIuZ)e3p2q*vTH% z!V{!IV>9Oi)Wm;@GyCrDr;GaV5Ym?h(eUVD@x=t%>?%%tC^?7kNPkg_XrY_~lW9Un zH*>c?7|Rk;fEuZd(Bf`UWPiI)GJ7@RK<+N~Tzrd0?4AzIGH$)z@u}PoG4jU>#Exg_ zRgMqYex3M0j!VoG3@(3nbmEg33mQ;zFdtSE&HfRm*?Z`&^u0yJMJ3$Ab9#F(!KCfd z&PW`X!2KNB8rePl#%K~BJ91wyTD!{a?a^Cx_M!8;a69$-0?PJ4Vh4*l$M|#0Nx+_k zkvYql0$@3N3~y4r8=aU2w+tKCVCw@FHKq&Z+JEo)N9j;bu9$DiBr{D!BqVP=J?n^w zz)P`S0rG9l*E2rk7(W!$i(gGkynYP(EI9JK;T!GZb3g60@547}YdI;P!)TxQpqoxa z5zwZtFHpUAJ6tlP6^|I-ryKeXAIn5*1}e$Ee3@yqwl{wqZ&Xn(otwL_GbulGY)oh& zEg7&Qb}5Z^LYd)rXPTS;Q))Aovi0R=Mt&nP2jNbTOGiYbEYEAlfj2D*>xKyY+ zqn(fbDep4AGfZg|5vl%sN9{z=Qj+o(+}RMSc5QJi_6vU^-29F6a^w9*nZXk)fZn?F zvUmooRHCcx!a-DkCL~lK@?Dr%-*|&1o~!;RJ`c#M5NQ7N)!ud)49wD9&M^d({LIX^ z6Y1l7H{JVztQywDvH}yj^T918Z6;TE2u`?>0e6F@BF}DD)`|RZ9avS9y=zL~=(CLM zIY1#*uDrqG9ixV9P5Ymv@jlc^WjtP0xg5CnQjdy*)jw4WZG59?B>=^CA{gK)g?DrQ z3QGn0Vllh~1apjF7xXP7k*oU;((HjwLcf{|MFB0rw2(v#Fv@--pTtFI@l$HDzZ++& zaZYvZp(GyVt1u)8JimRSSxUysxT9S`GHEE#z`b8E(-2OU3C2(qi0VpSWx5=_x=W4^ zU=-B*Ffq7!i9_$p^oR&OpV$)4eILF6|YaC*=A zc6&zZdE6KA>!Ai%LgoyjFmi#4nj`FQz*2FzVDv@-mPFe-R?<6%Y;?MQl%r_se0joS z^O)(*;N$j3dI9=Zd@us@s=Z}7ImOfY_EnP)zn&knt@<%P<9TBarGBP~2{Dac-@TSJ5(9rdQud>FXNc7@J23}pkNrsJv z^pXDGRjOKMEi?L}A1S_nH)+&^yTI+4zAtz!MDw(G-Q``$!U(}n3*%Mt0XnhipNrW` zjoG)@ncRT*I071*Nf@e(G+kk>y?y-JZi3w(xlXpbT_;fvfC{4Ye_ADvrhMyUQC56= zM83N?T_k;>K27;k{ckvGupKzU_xbvGe;;5)D)$PtU?yTbjWlc6InPuaG5CPdNNPd| zR4yI}b>yge3k-W}2)O^8t{RAM&agO>7C75|^vab&xd7s!5S-k>0qOS@4Yq$16eN{p z+=PPt-M+kt3Y84@zWlcr>{ZVg9Od8E-UpI-*?Szhq(%pgRO>z{{=t6e#k-ky;fX?3 z3rcs#&87(Q)c$2@X%&@{yY9VkY;`w{4^wvN)}$vJzI<NJXCuevYfXjSeJTB`Z z?o5WsFvAU$^nS7wx%CPV0pA*#rBiAJXljaGz(XXN#R%=$Gz8U7QH8taJZZdja&3T< zhPcVHZyTH>nBi9T+w4u32bJDfa-3*|+hj+7MM5>{IKeX*gH=!X8?gEp(?>vWRF`n& z+OO8DyitmrT=wBv(de?KR?-!q&-yCmwpxv+GQM*L*`>sIG=zo*9LHZHlJI~^X|qOd zcPP2HdvqkP8(V}C%{$jAEVEoCr->(s`+{6a3rIy#%Dd=JT#4-i@omHK&x}`&gObKD zf*vp)91{d>eHC8zUfR7-H|%y+s@28hNV7(~mWOVnd@56K_k~^CNNd!Tq}YP*8bSrJ z0tmd@g^rQ_e7whtyz7vWbJcT^vy~>ifut*rMqK=Ln(z1A z>E*t(+&8s4P6^3Z&EQ|k35Ed+mz4!BV|jxK5vH0ie76oAn_A_FJmP;UU=E=ua^~6n zQE~dJ|B|CnA3SKAtXL}K4BG~xgW#b=y@3$NEE#CXC)`uoCH%e@W6e&_TMp5p;+rw= z!DTM?+%!3T`x3wzMrE@lWq7w2bZLp)OJi*+``$!H|=TL^Zv)I&#gm~BdX1gC~2g1GA^xf7De9iKc(6$Nx z(4+a^??9m;Eh@ra@4hTx?M%pofr8(4(#yymeM-I=6}HcRj=GoZd3C%CWChMVp*}KT zW8;er!Wfx~6j5{xM|K|;(R`a*#9QPLU1w5})0h9;;M8oLWz|sWJ^jkU(!U-PxsrJ=c6kM5;XkpO>y+B9<-pKPSCL1Md;qJIUl@i zOO$YWs`{L~>^+d4`}w=th3CfKP=QU)<0sXYaj8}zli{?0B%nvTC~rLi1qa;Sscqh- zk0^P3sP*06Iqrj{F`^kIY3B6KG60p=u4hg4hu>koNO22nsU>z4hA8jYYK5H26W_h}C1H4ap|xf>ImC#77}WIu~k z{fH+d+RP&?ysION|I|R)A`fK%F9_nadr<=6lB^zRCl{FRz3Z1U^#^J zvyx~pkzFi;UKXh@PuUX-#s48mCrLt#+!8}4)N8x<=<=yY%O=Dm6mFv|!G-4l z($NH!&`lJ^iGzPxWc0ar0u}%iCnLF>w?ko~EnkAvjwO0ZCs9!XbSxGD01e01^PPT_t6lE>3Na5 z!OW%S1OTO?nwE@PTx{B;Pz#We-KU6Yl1kE)*DmPsH<>+;V8Rc%MEz25569I3KnoYz zbVGG>`&}YAzfP>geD|69vW+zZ_Q*iAaviD{lrP!x;C$>-XaE$h81jd*J;swi&2zSpg@kTW9bUQXrr1JDRc;-1(QCJg7kn z4{Ad?V;Rfq?O%Oxf~}*$b@#G1>rFv37-r3Gs=sI6+QU`-6ixP6ZiSqeLtNY~)4vli z_oyD%j@QO*ULZryP|kn zQYy*vdlg^fQaSG!?kjr%-rv)nqx+*QKMwyE)C#R8$yW?xETL{%O$Y*L(JnPKzh-6)IGK_gVh}$T@P8MRNLSdlIZk@KOaN% z&KLQ}e?AP=WyD*wx<)1bWs6L+3C)X#hW5CuPDs~$kUCQUsIfg25&tyR^@tu>oMif41Il}s?r9R(Qshu@q%u}gnoI}*uoCLKX%Eam=5%SX#k8Hx%2I-Dc~ zBShmJRG1Kh?bqL+e0BYTi==bxFGV%B3!1oekA3f;QCWF~HLo|TcYi2X?2ll`VM^=2 z{+JbyAIep_|9-=e0Pr;+nb<;*I>i}`C5Ij+4ch)~strF|Dnneq*sK95U-Zn%5>gpU zch#a5rx?U8?K;W6-*Rr>%_FL>ftvB9VqUp=rKzDJvq~29@V5E3%J~4-6GL!SY-0K{{YBWKk~Ui z?kmd_YO9o1UWd$OyIj6ARe;?`WXF$0g$;(cd}pTTI!?a3FvT)ra^GBBZyqvHR+y|1 zwBs@}$lEqqT&DiSXksL-6f1~WK|edY0#-=?Oc9GX6J9&@U?tA zsU9Lg?DH}i5P5_DScjs=un@U%Ns5NG3~;G0B+A~yl>!Tp5l%M#SnDZbxT46Nh~@;O zV_5eq`g-i}W5Ey6b4z&J{P{JY48z>eGM1s(VST=97lfOm5i$(0;>j2I%j18?tv;W+ z=^57<>K~TOg;%v$G*t`g=T(%cI~dY_%EivEWnqP(dBbEj1gz5B+6t6H6-on~Sbls8 z7KNJx3*I3pn}JSuT%fsef={Liwdbzx?+A;yRf4_ z)~&zG6j63%2Jqq;>|zsn4BX`YkxLTELj-@xev7O&mWIebO44gU@^V#d{??+3_S3=U zhxmU3$k-iL6ERRWzJdW>K1D0wmV24Mo6+x5I+(le5r^xpdyh>Im7DmnzF}M;$l3eH z2Anhxk4WD8fpvL6tVvA$4ph`w$q>H0D{l>t{Lo#+~@LZt<2xxkjC0c|r zH;_~_w(a#Rq7c{TS7ut zjTV#ttVH4CbI#?v%8R<&5(DR)`hmRt{(<`dy!{}5n$Ic&+#Iqo4|DPO*XGQ- z;nsh0$5zB+)tae~{Sg2F0?n_{DAclGnWgI!bVW~3Pgs6JuoznDTGegI9nLbH$Djb> z#+@g2Q!wigh9p!i!Lw7tsXiWj63G*Gvi#1QcCbIo9I*6YZS-`u#;K(0p{g9K2AQKe z{I_TG4Tv7L@;-F=4Uc|;PsTw!s=y2R$^0Fk6%Bt@_ zJT+LdqqibF?}|^LH2x@g%^n2GX+Ucs#+*)TAHqM(n*^-YMw z`y@+C>jF@Q!{8M9$)ANrlNa(-ytR0K>GA!-ionO_LCoKo*b{yw$B7;Jfart zAt$?Z3F*H$-l* z2Ma7Edh{HV02k6c#M|C4-ON(_n_gfJ^E=|2J)7n@yK-8ZDlmZ_$Uh#dbH_$8|3;%$ zJ>>9cxf*rWbcI9%lXRj#K&uYQ0b*&#@s*6{uhZZu%DpWoP%JW9f=+(3sun#FrLCml zZ@|izSN7jJod=*h>ZYlP2qM~U$Gh6xsH^O`tfY0u<%^Y{8ksEraSEca_>ems4J3mt)^IH)VG|Kg8&B@ zQ0=18+iA<$yWl<6t+t(gSpUwtV(iu>iy=h7n|i2g8t`=tsoYHH!yY$0?*vHN^~wM) zQ5=y~N5n`PpIz7bX_RiPyH{)%jeHXWlQ}gk@akw^3m<=MD>#pJ2fF95GkIqW$mmT3^nzy{U7l+mJL z%l7TPbvmmx&F5P9gle;46V7b{U*JOr{&p`-Kr=5t|6AGaPlJ)b$rOG-pr>4=uO&}E zCdDDlA@ayH2*|=cYhM;0wYo{$-WV5&&vQ)l1}u2}93zwJ#6Ml%vH%W0bj6o5{A%`b zL_{5Sr7U+Hw=_%7U#n*f@`m((b-Yd$t+oIj3TgDdj3_ROQk|BT79$te++|v@#d>X1 z_v|*2YpIZJujJa;^s$#wvUNJYQekXC%v=R}IB7ZUp2NYUFM?KN;l!Y*HijqstDkn% zdYg|Q9*~61({0wC6;S5`^-YND$i$NTLjkHr zh%Y@K3ZIMpZ3Rz>jlMFw2mFHphObC2!$c}1~1&-Uo+wq;N z{&-g61Rw;R6`olFDrcUhw&H-lHt>J<0234RNr$^x0rBytk7pf_ zruaXMy>~d)Z}>l~sUmx%P_mL0B4kANjOx=_OePRM#4s>}x7 zM#M`3#Oss&MSF8KyR6dNH}|D`H$abf&R_^dMYF?JS`?CUl(#mQA9}hO3<|~Q3To10 z%#Yvv6xYPhdZ0s_`H(8b6z>@_33P$nU87;=_T%WY9ve6qG^7rU@Ac_cR?p)Fq&@c^ ze;A%d8|?U&Ux!+izx07cNNyx#vrfXb#uEvlSz6@_TF>V8@m?rPm%Ke_S^^dJwq;;{DOr^_T)XJ-s zVAxad&vZN8%-^`1-{Ng3p84^nY;uv5v0WAzA7F-RGS%`FCDBt?lVP(NSS{n2Ib;N> z@*0wtF88+pvqR6JhiQxEKophFsBjf;r5nsSmzoA0M`u#6pw%S^G($5{y4GPp7bH#<>5_sd@Y3IQa?ksH` zdgn;eD}^|oHD(rJN~LWUjboOYInlv+G~L6UtX}zITr)AZNoZcXvWbuih6)|h!kEY* zmRJ2$Xu6%Zm3rm5O3;cLAka^}xsDv^E^jMLLwaF3*G;3m;iV|o{?VK^XiO)L(Yeh# z*(7u37iY-UIFN{7wLTHam5Bpf(H&2|&bf{R)+TDZz5XCMzYegB~-S=Y&!&4D3md9|#zqgeU+J#1?-CC)r+A4Ov^NM9$WBfaRi(YK zDbVs~R;XH?IHkqHp3tzn
*=^nW791~k${e2tA1DU&L{(r9EOa%7=LNRy>ivXsaHkX z+&G~DSL^xbVnOt2-O|wMQ;yI;DA=H36b@=FHEh;*7Sg!lGK-(foF`%KRVet-z$781 zM|SpY)YZ#fhZz@M=8Kdzc}mJB{N-+vE^r;u5_-)LVC2xw?L|}o0Ks)`6AIue_jWZ z$=qIC-Xh#IZE#+>8e?d3^{TJZ@Y5F;h)p_29#AQ^yrNkl)o$fPCT6XbtM!M zj{5l>1;GRE56{@&{8inVg7sBU!QghroL_ z=}Qr;aEdNJ6z;Ly50G=on1`sym|o_9l6_RL~VLDZKoVgfM8uW{n}d}n8T zJhfH4AuzrXvt5qyHS;%{`kveD9KX#=up0-WDn8}lnhx^*YcLn4hN*GpN5#L6ccKhH zq?GG5^2y+!=4Ld{Mw-uKhD&Y@2?a{ZuNnwcFFuda77w7nK39z2V=m!^$(Pq>(7^c? z#x}m&@nTwcd@qUsZ)IsI{62wz#QOTBT_VoC;eAGWo+1@?sRm#JY%uD)`ymx?#IYQ_ zRhV)DW36Q*F|YSd5-1|0qq}hse8Vj$Vr*Bp*0`hFu|Qm7b@3_>f;Fi9@u<_ z7(vFoU!h7lC^!w+tgNgm@e3NZwj8ve!TTnQQbXmqYh0UydZez+?8iDpdK?LsJl*GBgu|_;7V<>V1*@!5?yB zVxDB=6ZmP=8Y$JAK}yGf(9fy2uHs<5Z6+%-5e+n{;=r>ugO=wAEA9}K2Zq;~$HT(l zS=+XaJ@rlrUaR zQ}bDGQ+K*NBP%QGa@laXO^7{J%ZY<_eS&DR2noK{lyX#ac6w~R^X*F;_}TXe-cys( zIu*AM8g=l*As~@amSUxbFJT-cXd!9Ce z?cvVb^cEa3N|ohT8tk})B;`gXuX28HMtser?wW*p^I%Q34@A0Iiu5g9rn*e=#R&He zow-Ey@N7ZtUbg@68#4kW!3Bm*J#2L;kjn1()*(~)q%w@}dO2feTOwE<+IxJd7g*M- zJ(%Ilx!A5*x2GQTlC{ovCgc1XegE*Q#Z z$h6TZk@*dsM`m4<*4SV zhzfd^%Jm9k88WX)%RPzm%^R5wVQSiK=k1Ea4ofAyK4yi<$Y3h;E%|mq`WX%Phx&W(($Jiq_ynHiG|;+6NZ?SguLN&3 z@5gQLrnbxHr4v)mgmd9`CMF6ajUC_FN4-S%=Y@mA#Q;GYYirt`A8xzrGwU_0 zFO^rcY;C2OMj9%gFHMP+7weuCI)x68B#rdN*i&}wj~;eI0VNDRkuNOeKPIn|x%_!L zS@yRL!F#^)>vtwWAdhBAm3IbVDrR=Hnmkn6{b%PQ%u3ta+S(ihiC8uB)lN|Jamh!q zbveN7(fA8HM3rAXR}`4Sh4z<=b=*HXW1bAwxR(F6xjp8x#LmLOp>zp#aF2NIS~s7Y zXG5|%Tm07)(-%LiA=tS$8`&Q40tH$I6yhxbGm@!glz-+py|zG@JpN_NzLRmp+(+Ns zVA3p0qBh$m>k=bRz8wa9l z&uzL4`jnB0=cj)jPBV+1x5QhDM8>#|z8I%*x=Oqt3ie@oy^XU!*;~8#NlS>wfid|! zDNT7lJCMB;##j=Ma&~yp#e=UUCwOu{j`nGYP6-edP92lxrV*2N+}s6dBXEzlnqZ<- z#>fk-llE-0l-{*0#_NM(2IP4<-%h#B;*NT8&!n-Sb!oMSvGyZW!ZMED;s8iEX*Um{ z2`6j*3Vup6U7+#hJf6?F@nZ(RCH+1NF@ds9#in zd{Z0gAoNeS=c9jwtTV|K2%&{0+&&9m`)T=1V*8my9MEAkMnt+>VPGa1na%}M2uE8w z0+;Owk!_J_@Wyz0dOlX6fA%vs?C9v|@vyg4Watf2pPtd-ThlfMig8h=xFEBVpP;V_ zAfiQ7noSW9@PM0xc-aAB!`0dnMf}AL{`BENcBDWnhP@~ZOFKn3MVCT_ zPe1yAp|D8?KUNdDV^>YF?uAS;xx;LimhVrvA7HH~1w_)KK`)O~PWSCD+;MpXf{%a` z1T3NqF!>aD*;w(?5bsip$3~rn3(powALh%3d(KYJ{l{m6A=ZL{1H3wedS^w0J+0DD)f@|#=0nqY z<-xfT`tRiP{Qm9g#LyAL!`yidN*u!*11RO6WEi?;zBRHqKUZRL9z29DJGHtZgVx!u zXIMsyHBT$d#vjEyxjNk~Q*N6596)p?4fE)HU#MVD@NA$izsE06+!S3Bk{5a}IZ&IG z_A&LlGC+5;$;%mM$OKV6Ri7mMZ|J8lWzLH|zs>sRmvRHyAg3B?pWNQXR3T6S!>KCX z6)F5vJl*jqyg9>zNy=1R67(%UF?L1@F73ew7t_Iwo>re}=}pa)ixOb{@1r2r+#4uS zn^Bk5c(Pb7+$H;}f~3u*)f8k&y>9?H-mBU-A_;SSof(Y1_1^|(B3Psq!|GXl{=Muu z*-TJx*W&5BN1haANh18A|7Z^3+Lm)UQ3sc$;;+Eey#>gosR!hw8Hu?#$saQX?u2vH zR{&5^cEkq_MH1PVoPQ@c6bG*L^~D@mT$uYjhmq|M=3t4@zV5)f$Uq<7au3mk;xd6g znCf1N4j=;yqiWq6c)begiGZhv&u|Ler1!rHnz!*TkSt0<)$3Q^zN<&lHE{qZfT z$kk|Hkd;^`+?wv*{)lzrIq--r+D`^V#V>=ej^)Rm{^gCiVpFZMw|5ZM@+J98@V1QY zB+&Wf^ryE4@>7Bdqg?Y-5ODGLpM#fy+}*k1*(_OL{;L)#@NZ2oQ#SQmW|BiyNl?^=wl$3bSnU&XJUxaQxH= zGk~(16xHLfK|!d5j}@l3%si_jC+d6P_lQ4StEf#Gpb^!hjdnaoHcf_<_7#1wP^k`( zMkJFY3dNe%*w{F9qv`h3j`+tH?F;&I!Z-{M7d4xz?xG)U@ZmF_Rm7NK#8#8-R~Mt8 zB`R(2ITJH>@~jcS>J~kn(IWtgM{HNR9Ew-;Git|Z-;DUd5|i3B19O0a(%!zpj<`o} z)bFm8kLMG!BQUA`z|+T%AO8|A9%O?Nzv0j+B$X4-AVoa4%e%lDf2HMK05H)Ws%)ZT z5)QhAkp+=e1Vv{Bqk`mbPK-e37z};)pE*o)Z|V51@t#^%6zC^UidQfWldPs-#YO^J z*e5CLK42DlGTVSNxxvJ!s39VdUm5U;1YCh#fmr?G8EaNiyf@*?J%+SG*H*xE}-piX& z!0txgiue7DMb*hdYH>aywGpcqb7ngXD zLxsLj*8VUgqn3$0KW|V2>1#UJ;oq9wYF9O5QS)?@{X&vJpF(|%*3vE1d;4GrMRxAF z|J@&N%hfJUd3c?~ZtwCz*DNr|n&>j=F`{D=M}-%cL=ZY?(^mg~-@$Ymz9Zq%ZCV;S z=vae^Ibk8~`|tc`mIJW|yAmzXC8|~HR$)0>$t##nx3+nwK(@dPQ#&it(%OPHZ&C`! z#?;tzYElm;+2!^Y@3+2;bmQdv85D|rH|+dpvHS6qiQ2#o0E$+`w^USAEP6g6&K0BZ zbxd^m#W>dYK^T2w9eDS0(0?p4Ei$O4ar1B#3vGC5v5FC` z!#`eV{INs{*6G&?ftiShCkq1n zjgq{X>2h@@-(4azy^)}w{gcJ|tuAvx8!Pk@;8)_Bc1|U^86^33^?Nqv22RxPooH|b z@@}LNej#2pOW%N%Du&t};bRP)W;C#}BY6q&d!68H({C8_`1K}XxSpN$_Yg%yWg(iR zH|rte&uU1rU}@Ju1JlVzbM!d?`*~)(eWLT3kxYJiT@w*>h^TmTJ3_(RVuynLRLlICov+10D=~ZHm$m%&NBh}UkFG7(Uh&wLzsd{fQXXsHKz|ROK zx?ULaQsMf=;1j4}qV_Mc%n4~x>;-j4m2L-9SFf(!dDfrb*KBY3^p?$-Pz<-J|8Gt$ z{=a+ubsUCA>(Qowsjc@FxT~iQHpzLP8m3e@D=aLG&INK3i*S-3zIMQ?1amc{&uvRJ zd}`qv$+2U*hJH&r5Olt4X(#?C$ulyMV>7wz=P(>E5%SFd3SQmpYQe|i`;h)DS>8Xd z@qfOd{!N$Y-`yWG{;c_(W~=%6`LArQdt2S^Og(|7QCCyL*Nsj{U}t1x+#56VFy@Ps z9?1VZF0^A)3zoKZDnvA&;?^JEgQ5P~jIL#4R2zHwhs^-4zW#psBIdKDQ96uwneSRp zBK0UO+@>@z__58XD7Hb>Oogj9x$K^QCZ#-*4AIkK>62D#>T5gZjz%}6S}CPS-#0uU z@p8nsZw^eMGUDa>jI24GiE6!o#LWKhD&<;C!m5nMPxtqvUB6a=O2d20VY5&3naQ>m z7@xS5Jw<&;2iOAO+A`S~Fyrxe+hYcr0MWWOH4fkFDrRG!^12l=(yja!^7zv zt6qxlS0Y3R)tHVec2@>6KYLWCr<)GAgf!99(9op1gd5ki&2%@%{rZ3bV2kVUdo-sZ z$N2u(cCY`LBo=Bd(IM9sscKR$!69l@EzKKcwz_)3SUM&2cZnh^rAcMur19)@!2-cq zSuuGw#(!HUaA|_tz|W>;^QW(bgv^*T12rQ-!uOa^#;UbZbpoNc2HAf3yIp z!-dg_&yGytvWYLZv`ZhIh`VDG1=F9Gj-*Z{<$e0ZZ|c3arc>JRwYK5TJn8ByO`ui} zUSK)qbV?cLV_th|SRguCU3Lvfv+v>x&NL6vH=?cl7EoA0Q@xbCQ@;S zxznvhrft2h2N1V*uGiz$z%&kynhjO*_MnDjM78psq$yyh2-u4scf+(}@!xJAC!JET z$1+MP!DPLZ4F>~QDapDgG&IYw*CO_8-H38`Wvh~rhyGy(S$75&Ql3us#sJ*ahNOvx zo)wV4{Yp2N=$h}}FM7D3tlk_OVu&#KXhG>wf@~!Ov9_a?uda+z9`o?7klU zx$L8Xc8yEv$jj|f&u!wcqvE7P1&+D{+n1g1c%MZb>@f17l%i)7L#d z#u@omT~5i}MPeIkMI>UKO50_pp+rX(fQ#rHOpftoD+41|zY1Y|tRZYYKlcl6o2z6N zzrDFBD=FzCaYZ6X6y}+v){0sU-6ntSB}OLTjZ&DnM|L;EzOiqKq5!QU^b7zn=9Qd zl*b?H8mCMDsXA1G{Lx!Nod55^oG7PmxpiMd9AIYXurRJJ{m|3z3%rgQr70dS(Iu1C(=m5!5;wNDl{bFKoRp4l zQsv*ixYsbc8dX$biVkb>y@AymUJy3+Lg=zWiR}K99#w1BXkh1iM;(Iw!W}cxRw_m#QDkMF_ze4yRl47{a%%{QxL&u4eWLmzX|NE{!NGtN1&HLrmIlFpPMa=yZ z8WT3A+;_DwN59DZ&?V;Xi{BD|BN&pO)5^zDNW|U$@S`Beidz-CZfZ!7eAO@AZ`lOFOw_`(aklpl)QSfK96m}27CDMIo(mzS4DDS0NU zp|Kan3ajcqxUy-C&1o?3kZRoG1XY4fBO%!oHoZ>|-o80&H}LqaWRV*~E37<8$LOIo z;$_xn`EUu)V#P%evTzoi1H&IMu~005#EPn6y6;JkUn}{!`_|yor2~1xJ=I4v)nqZR zZdA-$1?H&-q_)%T_U)oIsqQ|zbo^}~Pt&;L>i2Dyb9El|6jfUbSi#gR*uCrQ@v8%c z#wi}+Bp*>@{X}@w?^5rY|o)@wFefix$NyJ=~^d>*{RGY_8Z|qEkLDn;(gJmmHTM1JjCm#~!Y86H z$+@~#P0!9M*}q^2_7lTcq+GQUnWu_?NHpR*o8Mj`c2Tv0N$Gl|!?`tfR*#R{Z*EFV~X|GV88 z5cRiOjAyN2pzUI{1xsaW+L~9fN7KC_OvI6W3k@lcOsB)Z?W-lp|5u zyZn~KOGDa_|LXju(8@>$uzLB1^vA*hyK>>eoHMVAGQMDry~Y6X1?-xcFidIiTm?5e zK{<|35t%t3^GMl|3xegERKBZ!&9rtcbGKTA+rl)aI#kJ zR=vSku5`480zdxFn=977O0qHv_taFcO$xPYu!F9e)|`>naYSTO|HGY}XpyZiJ2Omj4p`{_BZjD|23N-_sW!uw9c?hI0`39xteva!Zi2M`DzL+?G`EyaOt2LSk!#l!Xsso}Gwq$aJdACK=-?m^?<4!aZb}8s&N|ENmDKEy zhd-J_NH1yn7b~Gz3R@C?f!YO#b}J7y3q17032VTEUGEBUSF2r*ERaYIWDvi$c!LOK zK$_rd7t%uyL8&T#kSvUFgd3y)7||jREV}r6XOW(3x!kaF{)SrM`-46>s#ey!-yX*f7=y-UprX-W3_Gjmlt8HJZH`ALEF$vOlM0YhZ)-9$s>&+K2qhY zQ{XOURxib>k>=B2lLR2yjyDH%V=99lV3-tG;;)|(_2lOoTA~k5XnD#r$*+dLw_0AiZl%)9JT=aU76Dh&*lTNpZ_+q zlM#FAsRf0ngQ8e*r6NEU7HL;`wtn3v9VsJ{6p&QaC303t#5xc9?>;`Sg@ls;)SpS` za{D~+He2r9yODEHbBm(b= zKPryi^+?2h5%>K3;IieDroo7(7SPG7XCH^wOTNl}xtnClX&@|NcBX!Mj^L-cC#!4M7(4!@9*-P0k1 z8hz1OR}8yhwVkF@G#RdcdT!^QoFhCqsv&%lSs_7CeRAy+J{7lwg@wgN-h1~tLue6a zQfU5ujF#`o3*A1XNMzy6D5*ZwC&H4B~lZPnY#s5wLEnw`Qeh z%GGj?cQ2j9!6~`T!q{P4_%2tkLN&kc$fW{u@|#xg`}+E9C;gb^#7-Py;sI_ia7esl z@O52M3^O!qnuex<+rE{y{H;gcXUL9kfl~iu(RDpTX70gL`2su~oKBMT2>v_x_hFi+ z&|ryqI2k47QZjW%Xp4{7SJ!&gpLO5Yy@lWw!GQxpxD;ng`B6tRN1W;W*k#7ib&22p zJ*q2m@aMjKEq|;i%3xq@$Vcp1&NBx?%!AC(^!_tAIG?+S5 zkH1HPg}DRV*rcUIKcrkjO--#iQP_3Ss`3=CAM9cf+Q6=))IFEgq0%<=lAwxYeylswlcY~r! z_L}^MIqmn4LdkY=SuF<1#~2UW_T)d4eq4@)JF&+wM^5K(JN@~k5R6NeOfu4SsDi4{ zM~`!i7cKqXH%f}wVkbBB{897_zVcd?F$NUblK+;3tuB?I{Rb#sfh1~>9zcBUj-$`{ zbIVUNmjsV!SRK|UhiOTan=eo=S`PBUnbN6@aEb2;{yJ!-vpzKKH^IwR@q~?p6y;a4 zDCbHGFLHN)JkDmkR)46>O4lZ-X(|kjCV81FP;l3E%ksRBcuSsfQ!sj^?DwFB^Q2y> zy8zSbrJ)kIZa@WJvG=@@N(sCSBL84l~$iu?%@AE4U+kh z5tZ-4k}XHTFR$L{>5r-BUP6ydj6P-&F6}M40LK|8?c3sjTWbUhw?*%Vvp*KV;faWf zY7M5jySNe}l`1G(5X`IoFn?)#e{+^WH-|ti{;IBhE@OXAP@n$ugKDewE4L-ez}! zp=gZ_R4kB$G@1VT?5TLP7gK3n#zCN?@klWzs`X`Lw_rhOI{7l(LFRpvz+roWAvg+w{Pzq6W~Dq+eE6Cw0`jlUo-Aix8x+7kxwe3$|d3E-yIUHm7PM=`e=RI^A@G^^9O6vs+=ws zF!rB<&+rxm(gt2+#LQpQyKaEoC5D9^sJA#kn>OA*ffhr$B z^T)?ur*n%t(7Tv9<`C)Y#Dmq&#W0VIVH@fYmG`z*a!99I2cIV8C2<;5f9mRYsI*ca z-k9S0V$AUCeA>&J=eOBtVFoRk!^cPV@O5UYITQ{74DnC@M(;CLie`t8?5=bd5RKQl z)}8i0MY4FTr>X{M;-VuWnqh8nuv0Za!*3b3GnJF_%kA^yc&v0TU3_5ts_fW zJW1W5R=V;zFWK2@olU#h&TvM=ye}_uCm_D`92{qyP~^Mlhq&D4*A#%WfU67%==0s&(Oo4aCD;lEtYOV6Y_RI|4JkQ9agqYPWoY}pPG7L zUK%ckSO0WY?S0?;It+vvhh09Use;_k4iXcJVpq(g$2+jU|9ZsGVW)=`)7M3>OA5%x zNTZCE_@(d-_t*{LvR6?On{LWX9eAy|<~`kuPFy1mv+#3FXWA2|)w;j6xmtzdj>`k6 z=6VTainF@0*O}~c?T0AZEI;3-)v(EqMv5SOxIp6?--|c;`!8B*=S5IeBj%j56`V!N z2XB(F+G#KXo3@??>?;xqIpxjGvp(tNPjhW@eQ|bFFV#{<859>wT^OR2aUXUIkiv!c zvIp%ioibFdXmDli@>t1mNeQbNvlV7o=e8g`E-ir{AY>1Y;wSCsIO|H5K4nRsVZ#bT zqltH)M)FHY#%`XtI!YP&Q&?x|8LZB=yF_t!n|@~JMQz^Ew9+~2ceB1%w@tQ@ISS*ayRE^lSv9Ckrr(BZA%Z8=A@g$qc2EQx=5-L^wwZ8gfQ$^+Y zrc_i-vs%uSjyC_l+l>g^%&yJm=ak_@)s{OJ#CNShKmboN(VcWVI@ncx@ZcJt59qN) zRXif%_@{=A6*|4eSr*$r{p&h8&!jJ~b6kM`UV=rwKN6_hot<7;rnBybb#|It8+KuA zAU6+*iD{#AE*}$Zopkk=y*_!k;N#W83LZBVj8E@3i^k_vmF;qZRysqJCDA{f^Vn&w z!haA%{ycn{VZL~AP`ZZ0uTiW=F@FDo+oWw38#Urn3$Yb62b))tTW+g0dOPNlMQX?;p04M}KM{RoJxxtN7~4JL%!!1A z_#;>KVb;17|Ws%*z2pM#4B{RV{g8z;F2 z4QfWw{&sc4%8ddbxjeE0^n%9U@Uc)7xFJegDte&Fp=N9OquAYL74vPoJd9*PX_$25F@w}YQV9sWWgynr;=C!Im-I_(`v#Np7^Mv1uXJV%UuiS?*X zqWhkCO;!j1Lafnxwz5y8{370=Fm=Km!*^;Xg(!mknpWRWQo~7|+L3sm$lrO0C-2 zK;MA?7G_z(GI6v_J+toC2NFMN2C{1?_%&4B4+BO?Mny$sed-NEhu0za7sE44v@?q7 z=+!eT7aC@4wfV6+#A{$gyG+BXlmTTOn=$Q-VY^HYAI2sdUvANdpqpt{!)2;^dTs*} z*%zjJkF?l+zLUS+MtN3|q2tv1SCXm7PLun#Ck$sZ>a*d)c?vc4!_2fr%gf689XTNm z8!c%!n1mc3vEe%}A3taejG1KJ(7t^Wo&p+SB!+yhiGp@``dPyy7uU#})4~iLpp%Mj zqC~*?nr?zTeI&RHw5o$C**YSJ8xNPHeIpp*@x_yRYiDaH0?)yHG!U&M!*>1dGsle4 zY)N+fMIJY`c(5P0QzNcC%K3GG`5`K5;$do{y(i~~zF1EMv(|*D2dUd8 z@ta?ag|MO(@lkdJczJnM4Gi+jp`1|Lanw>n{jp+Wg+<>aVc`xDV#A7A`I@>`LVibc zZoLe1gg-T<{wH>VsQf*~pXuoKD}0urVC&`cD;ICJc!Pi0f@YLA8Lf2auJiDyvH3kx zfdoP^JWqooB56d368<-=u;$Ad+X7v(zj3oe{nhWbzcvL5@L?ub@51bQCCNTd0LyCY zONMeKrc|I0>UgVNCl8|Q4rU&2D<_Ev8?!5E6d2?Ue_%ilw|2h2-?-w~2F=P)cgg{j zZ*RUnu3=mrES4_vq`H&tvhUfSoXl3^Z-;1JlgAGbUHL@XGWgq|j_GK9^8e`1Rg^n} zrXI=%>ES@1N1(wntKpx+b8Aj%X=xHypZ#9yNEFr;fB(RCX`nFG#4DE5!rsOfRL$*Ol z@%t61FO9TJ-Ajka@ovRS7#xdM1fOw;+;`Y)PyR|s!+*lgulCT$PAey+vc9OHLM}FE z8gw>Jta58i7a&d>p#}pMsCJ;qRkW^{iHgYQHEkuM;oaYPv>|PuJx@J6 zAmH}UuTs<0D!yfQn_Bn*iZ&Um-s7}kd3XZ6qR7+6r>glMSay9tR!xiLsT3uf*0uIL z%>o0EP!8j7fniWdFUvp@wl>w&=1WYUmb-lK8njYrP;s013V{#6{pd4Jjq)ao=J5Z- z6;NWF{xDOiGQ`<9e~CeT7$PUiDDn(_6s_eIvgl#cu4Z^x=y}J6S^mn!A&Wtpu{=?0 z!vT@JXE8={wdKcWHQ78?Pc_&2?+D-)M>`Fq$_g5mPh0DK@ElZi8JpA@+y6{_oq_be zJguheDnu9ZP|IVRB=_dWJ1+X&10Xh5I`RxNBepA9668V|bc!5mupd3xDoud}y$4S= z(<^n(xyK6bNcsz|hgz4&z|BAJhe7g+w!1UYZ9s;d{%PkkyI1yT3)RzjI)uQ(e!i2A zbgGwpGJt|zaB_`9zj8h?*g-K#MBipXub<`nCzih99QG`J%>b{n;%rQyw?kK!{oA`f zLXg9@U_*OnYSOlXVV({#fSn^FS?=|K*xuJ4&AX}7dp*|T;nkelUyGNPRWOo4N={Y6 z)-BJ%4b%hV^(N1Ek0VBTm^g0ks6N^KE7oCnby2k=xE9G*LT;yK{=w@#Cph%$fZC2O zuF*B}vG1;|H(I3~tkS*RP}u-WgpB}V#4QsOZf3Kx$;a>GU&Z<(d@=%GhYF3Q_4{l; zYO&2WpW(HWbZ7(zd?<4;gYj~|--}~QbG6XfS~LD_>f;1{LMO9Suxp_J`hdQZW>R>U zOnFW20kmK=iAD+7&1wmYq*dNB)7}GT#U1Oc;NXG<6R{XgZ+IDrzz~fJn4cP;2nfqGp}k1p4EBbst{x zIWQFYe46Trsc9BznC&p@K?3a^;iHm#^8a;DO8qc7`IR>@^kU&LsE66*x@stcE8(vl z&unJDLxfk@x4l@mNKroau$?#h)e86Xl0WA;$Ve z0mCxRHT^E%-?7w+Cmmb3K6r!SvPG*XhK|&=4+dlr$1b5#Jf~r;9@13O#Hx|oyz#EH z*GH_a`&kb<*#&VpuH7Y$D-YCQw|M-ZZX)wo8IaC-;S&A~=`vF@a9LnPN?_9Nza@yF zy>dCrVptQ;t4!JEai@eSnt9+uIS4h5D)q@1;i&tJNP;Xe!_pc z(AiP8%v`p(&Y2tao3jX}lLGGGAHkD`j8b>6h6FXCKt?v5QK44&+1D9^nia~EU8mSJ z01Qia$OAw3h12EzH3U%*C!$=}w#g}Ar3ap7{}J5y1m7AlABHP)P1eV`;4&zyagU<+v& z5`Q16J|#}yxBQ_HOm`zA2gOlU3G}{KKSWS?tq9cH6id7l#p36g&Q9Y+RIr1r#2INJ$%&k2=P=tBLJ^QEOOz zN-SH|=cYoxW~MhnTRvUdA-|+kIP~t42)L!MvEiLG$UOO+2E>Yu&dtg>NR$k@a8anE zn!)@aRC{;>O7|jV!{>?uLYEw3es$}q+hLMI`1PI=d+3*1_{b%+vAuKgKN6bosRg!+eo@5SYY-Yd~F`!mW6}$Rk}(&(s3^la?|RT|iL_bH9|llz zDvdGue-DtJD%vitiBgY`%Qz0-3GQ1azVx$W9jcvg5WvcOnOsGK)|nY7Ly17bQb-w~ z_rYUhb#`WZ)T8I++5}Sn|HO(C{gn0x_tie25re#WR;r7BBepQ^Qj^BDW!0_ zq-!L}29x*+y|vuH_xNlMu9b8!H48a8d6E5ss-MJlTd4C~IwPxYbO7bR7HLrQXal)| zyOYj5Pdg&GjK47?QBpj4vAOALSevRovL#QM94GWEY0wMFH>B;cA zQsUK7=Y7H!0IL(SP?Oijoe%C7O}xLT+RJj7O~LcTzo7GJ%O_ZeP9b=YJ61ys&M~9oW6?Sv2fQ6fM#0#iuIPN$yhBGO4Bb#94)vxKd-rcQ=@(&-1XSkDyGPd>X zw3nD9hER*-&x2k#*^lnLFjhW*WcGC2_(4YP%XJ%%{WyYm0L1-JeDgV7?>_E=tX!@mfvpK-8&lxD{` zFPDVgeKhsS1) zVaq~h5=UDUdro=X(;taQjrjbdT?h@+^Z7jO7_!5KzC=`Q!Mc2Aa{M;?|A^w`n;QH` zXMjGir(dVkBGP&xMW=oQsyE&W8*YjwaT&jvz7Wa3ec1r(tMF${>W!+#e5zT0v>5fk zL1C2R!c#x#Nh7C=Yo}!pXNF&r$;hr0Cv$J)O&N?*(Bw8>pKh`35SVL5AW856!jYCK z!QFzJZG_T{KLMXh7(pN!l*^MJDjYnLFF8d!E8we3sLw?7A_JYDU$K z7nJM0#|Ch}AHyTAMK(A)IFwG$&{Xs%3TixjXx#UyQvpEmOCnwx2ueqQ_WS=x5jms` z%R$87S<^#keK9*oYj zMi_${7g=ZLia;tJDMaUn$w+()xylJitI(n%?w1{a^_w<_(jld}&96?J+7RK@D9!Rw zj0E6D4w)LU*eknzTQeXyyHT=kEy!m6aB&yPT}dk@{aMq%s&tdQ<9N7Uv+GC*TY{6t z)S><-PD`PPqi^-Wie|`jgB*%T#2CLS9-xZjgz`G(w!TYhk~{w!zqM>Y{MMi2qA9yo ziRTQJ$b)+*m~&7m8E8?l4sA9q8w))S${0Xo78@!SUB`^xrMM#7|5?lr!>(Iq;>hq( zvYLP=a+-q)qQsUuYAy^y11T|wC#|2W(zO^-egy2qwV=SK)ex(^%B%3 ztL^4GVrb4=zCTJ{n=M4#dBhaXC4!m$7j4~(dOj?SGZ2FH*B%uNY31$w3!k)KH#p-_ zgfwohB#&Lgn)J^~6J|U2{sr{8_a>{)1c21;fYhq$&?8R3y|Xtv@D6<@9oLvIBay?k zo6o)&m2z>py>WVtRYo{fq)7OWm7ZU$kbU}S#-B|J-H31yah46KZ~sG4*)OYGjFj6L zj^b5u89*DneUtZ-Tz_#Se)jG^O@u5R=z z@-7#^h()kqk)vpfm zy!QTE2!@$lb2f@_q*Y-ZY6IJVW#@WnY3XJ#A(`bFBsu!zKklt;B#^jsU<+^~}C&sAM+a7ynPA8AAA%|G>TvI*Ct&^3D z2r5UQ7>|e+(xZa3O*jeBVy7`ceC;*jvXqA;Zk{}eXl|}gelwn|4y6&VydaoXYIMFq zL=xR*x{WMPXK%wUOtDD^glli=_3M-RgDGy9xbX-NXwAqnAmIDI!-h-nd`1L+{q?KC z`7HrWT1twzhua9MsRfYS0#So<>x7*{cYuNeeSbdN?pM{mx69_q@p-ir&JDiethp|xoQoyr9wQe_-Q24Rg$v&?S|sX571t( zc6KC0f;fk?8t!S(<@tB`1E_c#HGdVmUV@)L5*1cpB#f}uXyfz>Ej$&c+q{n(viZ*Q z@81#(7Ho|=^y|ZVIHLTVFvq}23y6^M;|EDi%*PQejx=)zq$S-K28-wD_LX5s15zLj zfyCLEjI67DlgV(?=Nv|u5t;b?QTJnAFN}imZjKN)9MbI%mzdLoS=R`{qas5PxXY90 z(b$v94L~+X9(uKnG9+#QaV5L_Fo3?d0$f7~uLiZ?+PoYlPKfgMy(5G5xS8Pu1o!6Kt&7_K7~{$u5qgWS6we^>t^Gb50dBf@KH_@jc2IPC!>p!{EC zz)H0ve#~hqXdTMWnqeiFLBd@OuMvepf>Rb6Orrd_>Wz-k=+S&+p*tjD?K9q?=h}}~ zSmqAuw{68Vj8Z`BT31N&f$Q+)(Su;oBd4|f!z_afd}S6YST({Ajv2u%Nr(tXQ{g_J z#bQTtQj?RfRS*$-R$xfxj;y3Ryunt*96 z^7!i>!~0Bw>D&1LXUwtnr75f^1}}k-p(6{sJMNPQ(3YQifYi-Fm~f+Fbvj%*$~Z(= zgHul8dE)3~;ium|3+HcF$P>c%6rBc1Bb{^q%6Woti4aZ$f33PaWI#5S2Qi_C7D<9{ zmMM_UzcP7I^?x=(84399AHxC0p-$u%g?6;*J9WG5bo+h(r?W4Qhq8V9p6XV2dnww4 zqI6r6U4u#`30cOLtrDRyBpnq=>5D|=mZw;fF$^^y05?CHU12@JZMtvDgUhE*H6 zg~3q8BIe8sTZ(A038?J~K zb7)xD?iWo>h~B^B@b!$kIHg|l^F2BMSoLDIzHBcnHjAj_yfYC?VX0$6-~1^s+H28c z$t>VHlqZ{7dKY=2Bciady&kw8$KUbAupeX1mww5y44ZA+q|abQoB7E?m?^5YfAeK>Mb^{-W|y-0cO`U}@JLU}Po1 z_vNcc9z5!(h2Oa?_Y0?eFgjUWnMO{t5umSr6kcgK-|Cb1MHLX3xeM==XJeOsrQ?+< znluQ+$dCo*+z1I)aoG8aJ2f$KPLnrJ`AS%~@H3}jqM)=Z^Ua%Mpb!)m{tTYJ%&(_k z^X*>!d6V?;v%E{&{61n!L8} z6$MByGgoupYc;^*o?kl?Bdeux_`4)26XoKstXr=YBIfec-8cy_ua6-5l~sQwE!g9u z&!U~=qia~NbEal$s8a#aH!Lm&(bIHNxL+KkmUlw4NOZIKnOMA`7_a+8jX*8%&jz-Y zPT#g$p_7|1Z!VZ$;xRQk@Dpsjzm8GPw6d2%HhPrbl9qtLPaWUf|HXb9RVo3pmit^{ zb1D$Eje4RzftWPg!9$vS02Q+SXH;-$#d7xnbd$61f=@aE%z=vqk^8=`?@$2z98RzA zZ=%yMTi}OmGclk#wa$e@U)#8=u|PrQtMe5ryALxze>j+5b}y<=j34c6+{ZuIgYNeJ zK;$+Lh`n&tZ!6s{GX7>Y9fEF1$nOq$&AaXpzwjsQqG6}kKxQkK-U`NhGRGa?^Ib$^M7R(zt4T@ z=*M#asVNjO76+#PMQTd7n1vWR2?TudHTQjjCbje}qkmDF43zZ0>v4O06W@V%eHL@A zMfH&F2jR!E*&f}-=cAIn0EK5c-^3o2- zw<6Ms+9u2oV!;p2Fj&@p*A*eVh>J7?MjLVM&EY%-^v~(_82;uXdD{t*?1!$Ix8ybJCdz_wP?7z|hK2odZi`(Re=?9$gYHm-x7d+;Ef0nV~( zYeen7*!Xv&PtZAt7|C*m`)qDvA@rt*dFAa?VUMA|IE$kwVr#jLGMj32>~vq3)zTqV zWc6{M?n5-b>2D)mW(pY8U-4EVFR|yCU<32E7coD7H(ImLc44(!3ng}{$UdU!KrR*q z`MjF6fqEtX*L~P|en;myehN@mtVHeec<=MO(@SV4mk)g5QE$wqnF6Zqc?aWz;8vlp za=po2=nLP1dv!VIxzTt&%TF$6r^<nfpalOAv;r$E_9>q5gm@3u zIPN~mLtfd8Z=0RiQot=9qtM`W{bb zPBA;AoUyOP%^_+L+A`Q3`urO@Xw%%c+Wi}rlr|;oW2d?;rSZ#v4QCle4oLW;C!WH3 z$}KI(^JmMziJ@CBCyYQ_DQvaiUhoo4w;pJ5`4c=S@7x2~#_`)n*O#&`SD5hsc4Jgc za53(aU-v*LCGSk*-y)z93Vp1U!($(2Tnz`t(l80N zp)%y28=MiSSA%MAJDD&$t8eO-f} z;2z+F{ws$!T#zZl-8mHPTJr7AC1_oC_(DymhR@!VI+*OyU5zN5yYaZ@xTJG+Z@k_PMRFZxDumnVrbA>( zzD@bVKwwi0zh_h;hj&_<3?$v&AK3o-%sVODcs@C@Z#|Yolr^jW3(N^{83jH;k5_vN zf-`w+PZ0=Vp%D>kfP5g44EHaMzcKOwe#~@Pu!1%mc-L;E+S;v2EVmE=DL$%vq3**mH&uX9)Lv9bO01Q zqMzDLQ_QAopp_;BsECzV0y-@F*l2s3{WQy9ZwCuiw`;OP0UbGF=6 zuI#73Q;^wm?v4W}^anBl*Vdr7jD&cW)Bd^U-5c!zA@?p`0a4plGIl6pfCDlef9xg3CX%))91X&5IPPhZKm(IoNdZA2TyWNb;_5nQ zBL24Xzq~mHwaTuAdt$u3?K|<8-$UN~1?F)tBd53@mUmDu1~TEyC9vnxaH0?YuJVnz zc+pC%P~%hIXKB2~lZ|Cv)Q_N}d&NzzW<8|B>EW&nu(Oie1MqPl$yG}xu51+C`S8P~ zyRZd(O1YlMjXoB>g)@2u)Vw7g63!Q{(Duw`dqpfRz+H#O_A5h{-3sS_(LW6XA92q# zuV(yQLiuni@K?%pd1>2tCw^04#dyKtv9e-6IPQLtX5HB+2owT=`8V#d?yNfw?Wh&~x2k?d-#C9j_; ztta%0U~&uY)4t@pVoRWBii5DQX@wf{|32uDcr5SIJtE)dPLYRw2P-$nS}WkD z`z_N#U)Tt^Q)>3@+FK9&4YM5wYk3?9%YejTAzL*7sLxATH-dLPI*7_!wyfaiqp;1% zq$ab=*y#nYg!A3YDR-8bB@Lg9`w9CO1GEtEYz?kQIn!wQiDk2B1LC9)sp}}M#4p3n z!bcKT{u7`8U0|V}=$Yg9npOxM)oj8d64n>vIk$Q5VPUwVw~sRKnEkR2C&4<*9&ktD z95O1eF8?Fa^W`JP@{zW-FBTom?p8*Jjb3$J&NSVwn}b;GK2C72mO=E`Wh#R$;Uk=& zD#vz=wSui-;j94ccYGSpvJ=w(u!h5sXlG|+gg)R?IjyDj0NS@|zcwc(6%-Vd(*>aJ zrr<$ zlqn|gjM}M{7sMY+ib3*?nq^1K+y=uyp??sBF*c>#J#nE=><0vOtkvK2V#y9W`tY$+ zfw=~3#qeq%l0EfaTyXD&(#55Z&(6z>o9gLZucl5y-&w5ZM9lylVkZsD`lx+~8JKlo z$`wBWT-jPA>JIA9H(%LOG=Qh7i#oUllcVh(g8o-PBXvs(1iHRU*~{f-2)**EM3$h* z6sP0q;iwtvr|?pBj9A$H0>0v;nF=i2OjF$!Jh5AmH2d|b+Z)6+@c8-d0fIVW+Q|FP zay|Q%AKX4l;%4!uY_*l z6qJfv)_U(xZ3Ob*)nd95h?WjK=X)f5H3~3=eA5d(Sp)eyWEjhj$Hw#7A%YFgQ4dnH zhvt5HCV^k?u7;Jp@?+H1tN*`9=ksPI&3&&3jgF!!?c9FziQpe`m)TxI z<+gj32_c2YuOPPL#ZhQ23uXRd^8c5mlzyr6xwx>sA?c|ZLyqC4fY^lfT z1%bVNshKd%|n_#gV z0z!R0#Gen}@jbYErKiXcwSm$pX5JXeU5U5;q`>Ij-vUfi$^O@l(PC@hq@GVrzbFc7k8b$RH7O9ot?VD3YA@V zx02CPs5V^27oU#RKRo#xb~p4--Ie*8gKdo?0o&I6`HdS{VBur%RqNvR#Nt@4yW4+7 z6)PF#+yhq6&&i&fb>K1HE@)tGf|VSx{;Ko|$_;Tp$9H+HcE_p=y!yeiE_p|^j=;&Y zf)69foCBAe)Q}Uw4$68z?;YKP{!AoRBZjrBBu>Q~j;wX?D_ue#WS~gl)`Mr6@c{V5 zgYBJNTP8nZaaVj5y^zWN<2RPGL?=={5m*yqE@Za~1wY?H$7&;5-e^M?0A}ZXP8`s` z{!GK_`U?*rB;bw&=o_6SqthPPzzW;oB`)$anqA4IQx z)-i3T*~%?Q(T zwR1L7m!urN4viB0N~4{*+26Vg^+)rmv3&a*Ak`9)u&g`fftd7V<*bJo{D)3*b*fh4 zPJ9YT6%{N_W38q7r1v1>9@G(CL?)jWn>REtIH0M(2z4#)S=q=V>kyxHMxQX~Rk|45 z|Etke?LE*9o|mIeUfg}wTJZt&_@upx#N{U5PdT;`8D^!X!2eE}5&`ljG^cA^;b4aIPU!Q9+Uk^V%7og`d?h_l>S3l7_iRvD~rbx z7I>#8TQ#BEJ=L(pQmw4WawR=#M_oE2vbE0KRN|JKb3+l?;T#(koLk@Z)<484eU7&m zx#2(#w-db4*79KXqw*r7Ai4VPJm?%)o9Z{(^yHR^*|}okbWn@?Ki{x!NJ=77->Ob1 zY$%^6LTG*}@}FN+ziz1Az0P{HYB012T{QYRy9$eevDY+S;^Um23;Ri3+N5l+MjDun#k_xidWW3jQy3+oN+do}5y`aq-ou90 zAe`R`s+@eVdNX93p?PJ{)@&|nD<|h^Fe-ykXF`KiD2J_p+~o*-yjt;hTzqV7Y$z0% zfnEdJWBNbB?pb-69@oUl9T2fvwIWSEQu1wj&QcssIemh+9TsTHMvub=Iv{y9)?ed3|dXTxc#2 zXHCYTPj_Ds{jiWTjY<45N4QE;Aph_>kvv|oxQ&?4xLYf6E3e3HQ}nr8AMp}*YcnbP zw)H#a$R3Cpgn=*;B*nNHAhO+OszuSN<6m!!6f{`2WE+f%!259bye9dMpPWS9F*ha0szJqsz^kkf>(}6*MneB-d3GfqG*-)E(ay%?w4W zXH#tLy7HQ@MFG0mYy8ao#)7HLm5=5dc1hQL1IK4*V)7Vo)s;uIgaEhxQf>I7?BbtM zZqUCIZ52a2IpHd~QwqvoS3v-x%@Z)+$&og2AWksXuEq%|T1beAHOAoGUc|lxsvb?w z_=Sb4MoI5BG1DsxFiGNIug92z6Bwf*P48J9Cbt$n<6z1#8ZPk(><}Hw+R}&>A}}S9 zomTk$b!Q#WS!*&ty21`@t{Xhb4oQf&Edisd*GfG32%%w4KS1{75s&AOqZ3?SR!Md! zF7HW4LZ;(o{v1M+#GdSGe$%O0cu*?Fu{$aUAE{ii+kL@lWOIbo+d!^+U~CRI0XwS> z;5Ee=JZHcoFbzYnQtQJ>Zt}sDQqdZV z8+OdEgGiZ1MS|FM;0qpoagcRufO$j~Kfg9Z-;77^T&x=m8~m`V5&DPfd6-{Sl3UYw zyleuYH6Uq#i_9~ii61{c5d#fp!%nD(8kf%FV3*t3RYLNZi0K81M@#7GbmU&_U7Yil z$EA+Es1%cLf^#Dct6+OTI=?{ zgFwVUgi%t!shgeOKQuxM)76oWFPfoA8e!0kA)&2nhC3zM=}Fw}B2vt`Db3i#27P(~ zq<#Z(fb71W_AD8YLn>7z>Y%1ndL~w9YJ|U`<~HRLc%j+n(1fN$o-*fWPn%+rJN==` zplfg1E^_jFM^3s7&r+6HsO1!1J2qt?ydO0Tp#Y(G=lhssmG03bJF5J&hm~oxM)#dm zyPPplnXNeee5!V)uJEGN0jq(@M9LF0$KxM(xek> z3$*=^5G7R(e%LG8xDKsJ+|RNURTJ374EW@ZcZ7fweHIYT^BdwZBVW=qy5rm7suF4=ne+a*l-dd|P^I~8WAeP$5ZDP~p zpT6W%s7y2IMEI28Rnh<<(4z2>S#H>^%mo66AuiY!_T>q<=NIU8ywbVcM7JS@NVwq~ z)D5j*SH#!#2NSH?>A>sld>fX@?$pxcnf7k?;p=UOv-G|Z()lQV-}Ml)2-Hi5vO=Kz z2ZMp)pGqH7?OsS&U_5)`c9j;B#*s$A1qY0|FvK!!+Noa5*$*|e z)h_lEENuH+#36<;XOXuet}i}il%g8bUr9NUNqu_L$+<8gYOPd|bJ|oqWh4`}6t@X2;!L+2 zAy?@G^F+dsGV`F?{7w0en4CYS3TM6Ed^zaUaU2T%%}xpA+w9R~uCt^?b{@Fm^sKCK z4%~0%415>E006Zk>w8yyQKi_KKB* z!+g^A+?7Mae4>pRZ24OwQi@^@T$zF#YBZK_7WuPfWr zL`Xo|paB#%eUUT|OsO_Hy>xM|bO@H87`?N1;WdQh`cBXRQ)3H-q-fYgQj`qCizPrj zacBDhrUmb2oVm~ml{z{awm1VZ%rJ!)9Ma$JaMBe&gs3j>%WMX1b1wV<%uPK7ZJv4v zFM$peU{D$+6%WtQrb`fZE-5QC!J>13GKi2Kp>8W30qC?9zjDhILQSCao|1Bp(R6Ro2YRr z^)n2>RFAC~%SiUCx-hHY)~YVE+wmT@_KFpe6^pQ1Npz#{k=6v^3KX~qgkQvrSrn)_ zc)gtQRCFrySvCZ?q|Iwx)F*Rero0o%M@d2N`iXQHx7n+QmQK)(}kISqdo^iX<@r^%W|3{Q&F?h!77C3u}2xP}cnMOG+O0Puo2iSL4QW$V8kG2^*+}N>hB}42wW33XqAJ@d} zV|QAbdjo2W9kU(iTX|&@4!8cj>hYtx9dpbm+bnVDEXgR{$fz(Y-nOMY|4oeLepxG1 z9gxC|c=Ww9m(uGy>y{FW=GxZuL!`%qOo~z-KWowgjW$l86v7CkA5OPM7L@5h`<$;0 zYq{06dodud%N=b0QR5y@v}vzkal^QAVAY@~!#j`S15GyfwQB|=~ zNf-uF+?UUN{aOzx;3I{3en1_Ja_vN53{kSM!LuWKADrsAi26V#M{8Jf>jY%ll=JGMj;Ewtmc+`oyS4aP_r7vl5_F8zWFc zp}bLnAC=}aF9uXyke|LL$Pt4wWvmPemF=t*bl|MHK`+TA=-y}nb~UP}f9Kv0xk||9 zh;zPqc&EFy%br&LhVw0;k<;s7|8%a8P4udR6rmB>)_U`UmjSJ9H3q9UvoJRn4egcE z^ofZ&U>sr*U~!`d1giBwc#ZNZ?fnJ;gdA!1ngqLFTjA)K*VJJ;%5+NJ0u?)Xs@H4o zyCpy_b-+k=0vgs33f5ww!-wR_a2@1Fbi~v&)YiT*+FM%MGh-7AiG}}4>M)?f@)nCA#A40K z$w{*nn5V}rN;b-m2up1GJ#x4FwS!Q1FxS)97YK40?fL{JDp~GAXW(qDQ+^ zm{nq4=b?-D9FIOF4_Pe0oCn5%5*PX8Y&DkYyaa!ge17{ZL$5!hFw@}qUO6jJ-8OsJ zPlI#G2QsA-#8j50Ud+1rdC!+cMO|GmyWmpx?b(i1K?387vL8Xg{Zc`<9UbOADh(K1 zy?WJgp=~62NT=AhEfOioFctVyDPQJ#7fU{Yb7qnN5$N^=_VUB*^WEyjO6>iZFg?|% zei&N2&wGK0iHaxczk2GQ7lZxdsy+c(>42w?*SF{IXrI;Du0RL!b1e3UC|)W;*ziIL z%nTobR4D6K?w{c6J_!p79zsro4l!`<3=}Cw8nuqPcaH#FL~oS}Ee71hM^E@B=ccDW z)3T*N=!NvSU|E5D3b+UPuXE7N!$@3{+D;+w+DeWehe~-f1tvU}B;M+&*dWEKB0gWtS^ITGT9BqJ@p9@-qF8bDGucO}m#+b3e)8?LP;&Q9eY(q%&+2Hbtn+!yPcoC8fde%KLI>*Q9OA&`FXvjS0+jbB;bo)<52D zfKGlk5OmeL5w~Xhh#e?SosLah9v*F9A~)j+ObPd1n8*;>H3dShhO0iTMy_HkuRqY; zYy>jAkPSmLQ%nduTpzNSUk zMKeCZ_RM+lvzBP#w-@dgQPb4g-Ue+btTT`+tfqbrw2jOQK`iz9-Kapg(l75&j_&COSB_(lFh6>^4nwItw zSg_-%7Xj0}EZ8?u!6zyps>v(5D#nkSisSmet9SYiQc23r%?5Fe9y*Yu?vnN_<+t!lVZ|}mFt2* zQD-Cw5+xy|@{bf%4q4Vk4#+`M`LQQWo8O`QkeLUHx6l)iVrqeMozczI<0f+mXMv$E zgg}L#2E}JJE|e$h?!Mc-QCRzFYCbmq5M8Q^2A(?tT76DIAUjzCX0*L9(0w7@j43!X zkEq&Aze{e+)nt2g8oIqej87Sc6tpO3as2)pjr3hNbwqF@ z{)dlgSBhxG!9e+;%ITZs=C{mLL@p=l-Sq}&_1dN>q(ceXy&J9lkDGslp2vZyA85GA zevKH{>U=2k+Dr=4d*_E_oS&sF{H}d;0NO+nZe7YT(A;}Ozt>98TNkc5Yi{ScaU;nO zAo(7@xLuV3!&G@u)Q_eTqC5x*cOXFBeNW;Y&Is*n3RGe-8R0G%o-peKWcbNf@8DX4 zj~II$fZa7w0N##DgzhI?wKQr@&C=Gou~;76NgCjZM!6R>J+^o1Pvw&{4hRW_0L1kp zcx3K%h5%EN`_`3uI`p?3 zL6Qm0ErE6k7{l^q$ca3;b)2Kf4j9y0{aR+qgdIt>WJ<#G{u#7x)7?-HhH_w<4JX0c z2~g(Yku);7;9Z-?S0oiOiuhN?&g}#Ypr%+s1Xc6=fbXuL*+d9`t&w8JaCc#@u+gj3 z=3YK1yO`XqMZr|m_@;;zr=akN;}YT8Pz%7~$alIc9%>IG`_1SOr?-b~BWw2`?wtwT z@V-4W7^ZB5&jXgF)JH`e_g>RfI}SqbaGKz9=Q#@yUVKPR_1{6b6mNG}SWJx4j}3Q% z_&*f-n`CS}1)H>U<;k6z@1UT*;w6WW-q+6KTY7WX{BEE3Wu-~}6G7vh7eT7aU9>$C zl+qOP0W2E?kMkkU%-A@iF@HEY`r26q;6yW3e8{6(%L>V%(=huuq$qsNAw-WO)h5XQ zguB6?ZmYW~V|gR$NzE-`h>%8$+~_-{-U7y24YaS6MIZU0$cr2DNfI1-MVvNx^68Vz z@~31!U4ll!p%j?Hw|{udjF8X;frukOGpaz0TX2U8N(6rT^DRd(+ccXvlu78%rB1~~ zDLY4JkY`nkWabzn0s@b(J>(vPR5xV4A_dr&1v^g>o`HL>=(>#IDxYNSwJoVU7gyOkZ;O(1B|feYX+vV{ z#bXGk8P!QIOAqUB?-RFi{<&<$%18+Gl=_^vK$X=RSmGVR2M!#}&L7%Y{NtXHYKdr0 z^*y5--BkK;*e5`Huh3EWw)tJk+oYdq#Qmu!F11b1dzBg*8()cqppfx48yIp*P3g^J zy`_t(PKGWEZ8y)@+~Aff9ie0^4H)N~o?o1AT>LtwB{KU|FFCvLM~Km%m!ZFC$fb17 z1fh(=A|g_V9ZD&D3u$TkHfL>m`r8HQwQTcPANVS-5sE?&%?^tT+$X7TbWs;F4M2QS zw$QePPJrC+I%Sxy%Gsn1R(X;uopMrSKLa9#XrSB0C2~4O9Xx;2YhxNkZ?`_(#O1+L z=<^(Q(}~OGw3vl=g}g&r^DhylUn>=ow`kxf$e*Ks8*%xkMdp!8JCgQ`V%nw)t|2TV0@NTiLLszslOHNyg^&rF7e!Y= z-~@fF=9`^IVheJ#GDEXCU(7(c7>UNkQz46r~o5}lWfCemAIWKiD`NW8xxmtgjFzv|G?24 zC?r%6LqA8zMh?{{s!N B6#4)F literal 0 HcmV?d00001 diff --git a/paper.bib b/paper/paper.bib similarity index 88% rename from paper.bib rename to paper/paper.bib index f6897d4..6634a0a 100644 --- a/paper.bib +++ b/paper/paper.bib @@ -116,6 +116,24 @@ @ARTICLE{2020SciPy-NMeth doi = {10.1038/s41592-019-0686-2} } +@misc{hillstrom2008, + title={The MineThatData E-Mail Analytics And Data Mining Challenge}, + author={Hillstrom, Kevin}, + year={2008}, + url={https://blog.minethatdata.com/2008/03/minethatdata-e-mail-analytics-and-data.html} +} + +@article{finkelstein2012, + title={The Oregon Health Insurance Experiment: Evidence from the First Year}, + author={Finkelstein, Amy and Taubman, Sarah and Wright, Bill and Bernstein, Mira and Gruber, Jonathan and Newhouse, Joseph P. and Allen, Heidi and Baicker, Katherine and {Oregon Health Study Group}}, + journal={The Quarterly Journal of Economics}, + volume={127}, + number={3}, + pages={1057--1106}, + year={2012}, + doi={10.1093/qje/qjs020} +} + @article{kobrosly2020causalcurve, title={causal-curve: A Python Causal Inference Package to Estimate Causal Dose-Response Curves}, author={Kobrosly, Roni W.}, diff --git a/paper.md b/paper/paper.md similarity index 86% rename from paper.md rename to paper/paper.md index 6940ee4..b324ab1 100644 --- a/paper.md +++ b/paper/paper.md @@ -60,9 +60,13 @@ In the R ecosystem, packages like `qte` provide quantile treatment effect estima All estimators implement a consistent API with three primary methods: `predict_dte()` for distributional treatment effects, `predict_pte()` for probability treatment effects over intervals, and `predict_qte()` for quantile treatment effects. The adjusted estimators use K-fold cross-fitting to prevent overfitting and support both single-task and multi-task learning modes [@hirata2025efficientscalableestimationdistributional] for computational efficiency. Bootstrap methods provide confidence intervals with multiple variance estimation approaches. +![Distributional treatment effects for the Hillstrom email marketing dataset [@hillstrom2008], comparing Women's vs Men's email campaigns. The simple estimator (left, purple) and ML-adjusted estimator (right, green) show that adjustment substantially tightens confidence bands, demonstrating the variance reduction benefit of regression adjustment.\label{fig:hillstrom}](hillstrom_dte.png) + +![Local distributional treatment effects for emergency department costs in the Oregon Health Insurance Experiment [@finkelstein2012], estimated using `SimpleLocalDistributionEstimator` (left) and `AdjustedLocalDistributionEstimator` (right). Health insurance coverage shifts the distribution of ED costs, with ML adjustment again yielding narrower confidence intervals.\label{fig:oregon}](oregon_ldte_costs_comparison.png) + # Research Impact Statement -The methods implemented in `dte_adj` have been published at top machine learning venues: ICML 2024 [@byambadalai2024estimatingdistributionaltreatmenteffects] and ICML 2025 [@byambadalai2025efficientestimationdistributionaltreatment]. The package has been used internally at CyberAgent, Inc. for analyzing A/B tests where distributional impacts are critical, such as evaluating interventions on user engagement metrics where tail behavior matters more than averages. The documentation includes tutorials demonstrating applications to the Hillstrom email marketing dataset and the Oregon Health Insurance Experiment, facilitating adoption by researchers in economics, marketing, and healthcare. +The methods implemented in `dte_adj` have been published at top machine learning venues: ICML 2024 [@byambadalai2024estimatingdistributionaltreatmenteffects] and ICML 2025 [@byambadalai2025efficientestimationdistributionaltreatment]. The package has been used internally at CyberAgent, Inc. for analyzing A/B tests where distributional impacts are critical, such as evaluating interventions on user engagement metrics where tail behavior matters more than averages. The documentation includes tutorials demonstrating applications to the Hillstrom email marketing dataset (\autoref{fig:hillstrom}) and the Oregon Health Insurance Experiment (\autoref{fig:oregon}), facilitating adoption by researchers in economics, marketing, and healthcare. # AI Usage Disclosure From f05daa23e33a9bf801f8a192f5067d0e24f0bfc3 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 21:52:07 +0900 Subject: [PATCH 07/14] Pin GitHub Actions to commit SHAs for supply chain security Co-Authored-By: Claude Sonnet 4.6 --- .github/workflows/draft-pdf.yml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/.github/workflows/draft-pdf.yml b/.github/workflows/draft-pdf.yml index 998b3ac..044c679 100644 --- a/.github/workflows/draft-pdf.yml +++ b/.github/workflows/draft-pdf.yml @@ -6,14 +6,14 @@ jobs: name: Paper Draft steps: - name: Checkout - uses: actions/checkout@v4 + uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4 - name: Build draft PDF - uses: openjournals/openjournals-draft-action@master + uses: openjournals/openjournals-draft-action@85a18372e48f551d8af9ddb7a747de685fbbb01c # master with: journal: joss paper-path: paper/paper.md - name: Upload - uses: actions/upload-artifact@v4 + uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4 with: name: paper path: paper/paper.pdf From 23fc3a5fba2cec5d195c95131a61c3fa12fa8634 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 21:52:54 +0900 Subject: [PATCH 08/14] Update Databricks affiliation to Databricks, Inc., United States Co-Authored-By: Claude Sonnet 4.6 --- paper/paper.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/paper/paper.md b/paper/paper.md index b324ab1..337b72f 100644 --- a/paper/paper.md +++ b/paper/paper.md @@ -23,7 +23,7 @@ affiliations: index: 1 - name: Keio University, Japan index: 2 - - name: Databricks Japan, Japan + - name: Databricks, Inc., United States index: 3 date: 24 August 2025 bibliography: paper.bib From ef09dcfc78532d418e078593dd96a84a953b9e43 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 21:55:08 +0900 Subject: [PATCH 09/14] Add explicit permissions to draft-pdf workflow Addresses CodeQL security alert: restrict GITHUB_TOKEN to read-only contents access following least-privilege principle. Co-Authored-By: Claude Sonnet 4.6 --- .github/workflows/draft-pdf.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.github/workflows/draft-pdf.yml b/.github/workflows/draft-pdf.yml index 044c679..0edd22d 100644 --- a/.github/workflows/draft-pdf.yml +++ b/.github/workflows/draft-pdf.yml @@ -1,5 +1,8 @@ on: [push] +permissions: + contents: read + jobs: paper: runs-on: ubuntu-latest From e29baf5caabf7d3f6402e5369c798a8a591a9ab3 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 22:00:12 +0900 Subject: [PATCH 10/14] Fix figure paths for inara PDF build inara/pandoc runs from the repo root, so relative image paths in paper/paper.md resolve against root, not paper/. Copy PNGs to root before the build step so pandoc finds them. Co-Authored-By: Claude Sonnet 4.6 --- .github/workflows/draft-pdf.yml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/.github/workflows/draft-pdf.yml b/.github/workflows/draft-pdf.yml index 0edd22d..2814c2e 100644 --- a/.github/workflows/draft-pdf.yml +++ b/.github/workflows/draft-pdf.yml @@ -10,6 +10,8 @@ jobs: steps: - name: Checkout uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4 + - name: Copy figures to repo root for inara + run: cp paper/*.png . - name: Build draft PDF uses: openjournals/openjournals-draft-action@85a18372e48f551d8af9ddb7a747de685fbbb01c # master with: From 00146497b0a8124a73455b4f813d93b3218501b4 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 22:01:25 +0900 Subject: [PATCH 11/14] Fix multiply-defined LaTeX labels for figures Pandoc extracts \label{} from caption text and also auto-generates a figure label, causing the label to be defined twice. Remove explicit \label{} from captions and replace \autoref{} with plain text references. Co-Authored-By: Claude Sonnet 4.6 --- paper/paper.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/paper/paper.md b/paper/paper.md index 337b72f..929f1d2 100644 --- a/paper/paper.md +++ b/paper/paper.md @@ -60,13 +60,13 @@ In the R ecosystem, packages like `qte` provide quantile treatment effect estima All estimators implement a consistent API with three primary methods: `predict_dte()` for distributional treatment effects, `predict_pte()` for probability treatment effects over intervals, and `predict_qte()` for quantile treatment effects. The adjusted estimators use K-fold cross-fitting to prevent overfitting and support both single-task and multi-task learning modes [@hirata2025efficientscalableestimationdistributional] for computational efficiency. Bootstrap methods provide confidence intervals with multiple variance estimation approaches. -![Distributional treatment effects for the Hillstrom email marketing dataset [@hillstrom2008], comparing Women's vs Men's email campaigns. The simple estimator (left, purple) and ML-adjusted estimator (right, green) show that adjustment substantially tightens confidence bands, demonstrating the variance reduction benefit of regression adjustment.\label{fig:hillstrom}](hillstrom_dte.png) +![Distributional treatment effects for the Hillstrom email marketing dataset [@hillstrom2008], comparing Women's vs Men's email campaigns. The simple estimator (left, purple) and ML-adjusted estimator (right, green) show that adjustment substantially tightens confidence bands, demonstrating the variance reduction benefit of regression adjustment.](hillstrom_dte.png) -![Local distributional treatment effects for emergency department costs in the Oregon Health Insurance Experiment [@finkelstein2012], estimated using `SimpleLocalDistributionEstimator` (left) and `AdjustedLocalDistributionEstimator` (right). Health insurance coverage shifts the distribution of ED costs, with ML adjustment again yielding narrower confidence intervals.\label{fig:oregon}](oregon_ldte_costs_comparison.png) +![Local distributional treatment effects for emergency department costs in the Oregon Health Insurance Experiment [@finkelstein2012], estimated using `SimpleLocalDistributionEstimator` (left) and `AdjustedLocalDistributionEstimator` (right). Health insurance coverage shifts the distribution of ED costs, with ML adjustment again yielding narrower confidence intervals.](oregon_ldte_costs_comparison.png) # Research Impact Statement -The methods implemented in `dte_adj` have been published at top machine learning venues: ICML 2024 [@byambadalai2024estimatingdistributionaltreatmenteffects] and ICML 2025 [@byambadalai2025efficientestimationdistributionaltreatment]. The package has been used internally at CyberAgent, Inc. for analyzing A/B tests where distributional impacts are critical, such as evaluating interventions on user engagement metrics where tail behavior matters more than averages. The documentation includes tutorials demonstrating applications to the Hillstrom email marketing dataset (\autoref{fig:hillstrom}) and the Oregon Health Insurance Experiment (\autoref{fig:oregon}), facilitating adoption by researchers in economics, marketing, and healthcare. +The methods implemented in `dte_adj` have been published at top machine learning venues: ICML 2024 [@byambadalai2024estimatingdistributionaltreatmenteffects] and ICML 2025 [@byambadalai2025efficientestimationdistributionaltreatment]. The package has been used internally at CyberAgent, Inc. for analyzing A/B tests where distributional impacts are critical, such as evaluating interventions on user engagement metrics where tail behavior matters more than averages. The documentation includes tutorials demonstrating applications to the Hillstrom email marketing dataset (Figure 1) and the Oregon Health Insurance Experiment (Figure 2), facilitating adoption by researchers in economics, marketing, and healthcare. # AI Usage Disclosure From 654ac352089c01d3828cf032ec8af3c5de8b2695 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 22:05:34 +0900 Subject: [PATCH 12/14] Fix figure 1 filename: hillstorm_dte.png (not hillstrom) Co-Authored-By: Claude Sonnet 4.6 --- paper/paper.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/paper/paper.md b/paper/paper.md index 929f1d2..e8b79f4 100644 --- a/paper/paper.md +++ b/paper/paper.md @@ -60,7 +60,7 @@ In the R ecosystem, packages like `qte` provide quantile treatment effect estima All estimators implement a consistent API with three primary methods: `predict_dte()` for distributional treatment effects, `predict_pte()` for probability treatment effects over intervals, and `predict_qte()` for quantile treatment effects. The adjusted estimators use K-fold cross-fitting to prevent overfitting and support both single-task and multi-task learning modes [@hirata2025efficientscalableestimationdistributional] for computational efficiency. Bootstrap methods provide confidence intervals with multiple variance estimation approaches. -![Distributional treatment effects for the Hillstrom email marketing dataset [@hillstrom2008], comparing Women's vs Men's email campaigns. The simple estimator (left, purple) and ML-adjusted estimator (right, green) show that adjustment substantially tightens confidence bands, demonstrating the variance reduction benefit of regression adjustment.](hillstrom_dte.png) +![Distributional treatment effects for the Hillstrom email marketing dataset [@hillstrom2008], comparing Women's vs Men's email campaigns. The simple estimator (left, purple) and ML-adjusted estimator (right, green) show that adjustment substantially tightens confidence bands, demonstrating the variance reduction benefit of regression adjustment.](hillstorm_dte.png) ![Local distributional treatment effects for emergency department costs in the Oregon Health Insurance Experiment [@finkelstein2012], estimated using `SimpleLocalDistributionEstimator` (left) and `AdjustedLocalDistributionEstimator` (right). Health insurance coverage shifts the distribution of ED costs, with ML adjustment again yielding narrower confidence intervals.](oregon_ldte_costs_comparison.png) From df0817a04bae4e6ca3b8343ecaa5ce8e35ca6463 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Wed, 8 Apr 2026 22:05:47 +0900 Subject: [PATCH 13/14] Remove unnecessary cp step in draft-pdf workflow inara CDs into the paper/ directory before running pandoc, so images in paper/ are already on the resource path. No copying needed. Co-Authored-By: Claude Sonnet 4.6 --- .github/workflows/draft-pdf.yml | 2 -- 1 file changed, 2 deletions(-) diff --git a/.github/workflows/draft-pdf.yml b/.github/workflows/draft-pdf.yml index 2814c2e..0edd22d 100644 --- a/.github/workflows/draft-pdf.yml +++ b/.github/workflows/draft-pdf.yml @@ -10,8 +10,6 @@ jobs: steps: - name: Checkout uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4 - - name: Copy figures to repo root for inara - run: cp paper/*.png . - name: Build draft PDF uses: openjournals/openjournals-draft-action@85a18372e48f551d8af9ddb7a747de685fbbb01c # master with: From 6a4df3dbfd07f28b3941b2e55be5842bc7a6c442 Mon Sep 17 00:00:00 2001 From: TomeHirata Date: Sun, 19 Apr 2026 23:35:49 +0900 Subject: [PATCH 14/14] Add citations per reviewer feedback - Add Econometric Reviews publication (Oka et al. 2025) - Add NeurIPS 2025 reference for imperfect compliance paper - Add ABEMA field experiment paper (Yasui et al. 2026, Japanese Economic Review) as industry application evidence Co-Authored-By: Claude Sonnet 4.6 --- paper/paper.bib | 16 ++++++++++++++++ paper/paper.md | 2 +- 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/paper/paper.bib b/paper/paper.bib index 6634a0a..f6a8001 100644 --- a/paper/paper.bib +++ b/paper/paper.bib @@ -77,6 +77,22 @@ @misc{byambadalai2025imperfectcompliance url={https://arxiv.org/abs/2509.15594} } +@article{oka2025regression, + title={Regression Adjustment for Estimating Distributional Treatment Effects in Randomized Controlled Trials}, + author={Oka, Tatsushi and Yasui, Shota and Hayakawa, Yuta and Byambadalai, Undral}, + journal={Econometric Reviews}, + year={2025}, + doi={10.1080/07474938.2025.2538843} +} + +@article{yasui2026abema, + title={Distributional Treatment Effects of Content Promotion: Evidence from an {ABEMA} Field Experiment}, + author={Yasui, Shota and Oka, Tatsushi and Byambadalai, Undral and Oishi, Yuki}, + journal={The Japanese Economic Review}, + year={2026}, + doi={10.1007/s42973-026-00234-y} +} + @article{dowhy, title={DoWhy: An End-to-End Library for Causal Inference}, author={Sharma, Amit and Kiciman, Emre}, diff --git a/paper/paper.md b/paper/paper.md index e8b79f4..e363807 100644 --- a/paper/paper.md +++ b/paper/paper.md @@ -66,7 +66,7 @@ All estimators implement a consistent API with three primary methods: `predict_d # Research Impact Statement -The methods implemented in `dte_adj` have been published at top machine learning venues: ICML 2024 [@byambadalai2024estimatingdistributionaltreatmenteffects] and ICML 2025 [@byambadalai2025efficientestimationdistributionaltreatment]. The package has been used internally at CyberAgent, Inc. for analyzing A/B tests where distributional impacts are critical, such as evaluating interventions on user engagement metrics where tail behavior matters more than averages. The documentation includes tutorials demonstrating applications to the Hillstrom email marketing dataset (Figure 1) and the Oregon Health Insurance Experiment (Figure 2), facilitating adoption by researchers in economics, marketing, and healthcare. +The methods implemented in `dte_adj` have been published across machine learning and econometrics venues: ICML 2024 [@byambadalai2024estimatingdistributionaltreatmenteffects], Econometric Reviews [@oka2025regression], ICML 2025 [@byambadalai2025efficientestimationdistributionaltreatment], and NeurIPS 2025 [@byambadalai2025imperfectcompliance]. The package has been applied in industry settings, including analyzing the distributional impact of content promotion on user engagement at ABEMA, a major video streaming platform [@yasui2026abema]. The documentation includes tutorials demonstrating applications to the Hillstrom email marketing dataset (Figure 1) and the Oregon Health Insurance Experiment (Figure 2), facilitating adoption by researchers in economics, marketing, and healthcare. # AI Usage Disclosure