diff --git a/papers/adam_richie-halford/adam_richie-halford.rst b/papers/adam_richie-halford/adam_richie-halford.rst index 6b6f5da29d..703bf28f29 100644 --- a/papers/adam_richie-halford/adam_richie-halford.rst +++ b/papers/adam_richie-halford/adam_richie-halford.rst @@ -98,7 +98,13 @@ resources. Here, we introduce a new Python library: Cloudknot :cite:`cloudknot-docs` :cite:`cloudknot-repo`, that launches Python functions as jobs on the AWS Batch service, thereby lifting these -limitations. Cloudknot supports Python versions 2.7 and 3.5+. +limitations. Cloudknot supports Python versions 2.7 and 3.5+. Rather +than introducing its own set of terms and abstractions, Cloudknot +attempts to mimic Pythons concurrent futures’ :code:`Executor` objects. +Users of Cloudknot have to familiarize themselves with one new object: +the :code:`Knot`, while some of its functionality will initially be new +to users of Cloudknot (e.g., the way that resources on AWS are managed), +its :code:`map` method should be familiar to most Python users. We propose that software designed to aid computational and data scientists should be concerned with minimizing the time from @@ -183,8 +189,11 @@ total, the resulting command line program downloads input data from S3, executes the UDF, and sends output back to S3. :code:`Knot` then packages the CLI, along with its dependencies, into a Docker container. The container is uploaded into the Amazon Elastic Container Registry (ECR). -Cloudknot's use of Docker allows it to handle non-trivial software and -data dependencies (see examples below). +Cloudknot's use of Docker allows it to handle non-trivial software +and data dependencies (see examples below). This is because Docker +provides a consistent and isolated environment, allowing complete +control over the software dependencies of a particular application, and +near-immediate deployment of these dependencies :cite:`Boettiger14`. Separately, :code:`Knot` uses an AWS CloudFormation template to create the AWS resources required by AWS Batch: @@ -217,13 +226,21 @@ the AWS resources required by AWS Batch: minimum, desired, and maximum number of virtual CPUs and let AWS Batch select and manage the EC2 instances. -:code:`Knot` uses sensible defaults for the job definition and compute -environment parameters so that the casual user may never need to concern -themselves with selecting an instance type or specifying an AMI. More advanced -users can control their jobs' memory requirements, instance types, or AMIs. -This might be necessary if the jobs require special hardware (e.g. GPGPU -computing) or if the user wants more fine-grained control over which resources -are launched. +:code:`Knot` uses job definition and compute environment defaults +conservative enough to run most simple jobs, with the goal of minimizing +errors due to insufficient resources. The casual user may never need to +concern themselves with selecting an instance type or specifying an AMI. +Users who want to minimize costs by specifying the minimum sufficient +resources or users who need additional resources for intensive jobs can +control their jobs' memory requirements, instance types, or AMIs. This +might be necessary if the jobs require special hardware (e.g. GPGPU +computing) or if the user wants more fine-grained control over which +resources are launched. + +One of the most complex aspects of AWS is its permissions model. Here, +we assume that the user has the permissions needed to run AWS Batch +in the console. We also provide users with the minimal necessary +permissions in the documentation. Finally, :code:`Knot` exposes AWS resource tags to the user so that they can assign metadata to each created resource. This facilitates @@ -369,11 +386,13 @@ summary of the status of all jobs submitted with this :code:`Knot` using Examples -------- -In this section, we will present a few use-cases of Cloudknot, including real -life uses of the software in data analysis. We will start with examples that -have minimal software and data dependencies, and increase the complexity by -adding first data dependencies and subsequently complex software and resource -dependencies. +In this section, we will present a few use-cases of Cloudknot, including +real life uses of the software in data analysis. We will start with +examples that have minimal software and data dependencies, and increase +the complexity by adding first data dependencies and subsequently +complex software and resource dependencies. These and other examples +are available in Jupyter Notebooks in the Cloudknot repository +:cite:`cloudknot-examples`. Solving differential equations @@ -397,30 +416,40 @@ In this unrealistic example, we wish to parallelize execution both over a range of different boundary conditions and over a range of grid sizes. First, we hold the grid size constant at 10 x 10 and parallelize over -different temperature constraints on one edge of the simulation grid. We -investigate the scaling of job execution time as a function of the size -of the argument array. In Figure :ref:`fig.nargsscaling` we show the -execution time as a function of the length of the argument array (with -a :math:`\log_2` scale on both axes). The default :code:`Knot` instance -has a maximum of 256 vCPUs in its compute environment and a desired -vCPUs setting of 8. We testing scaling using these default parameters -and also using a custom parameters with :code:`min_vcpus=512`, -:code:`desired_vcpus=2048`, and :code:`max_vcpus=4096`. These tests -were also limited by the EC2 service limits for our region and account, -which vary by instance type but never exceeded 200 instances. The user -interested in maximizing throughput could request limit increases. -Regardless of the :code:`Knot` parameters, Pywren outperformed Cloudknot -at all argument array sizes. Indeed, Pywren appears to achieve -:math:`\mathcal{O}(1)` scaling for much of the argument range, revealing -AWS Lambda's capabilities for massively parallel computation. +different temperature constraints on one edge of the simulation grid. +We investigate the scaling of job execution time as a function of the +size of the argument array. In Figure :ref:`fig.nargsscaling` we show +the execution time as a function of :math:`n_\mathrm{args}`, the length +of the argument array (with a :math:`\log_2` scale on the :math:`x`-axis +and a :math:`\log_{10}` scale on the :math:`y`-axis). We testing scaling +using Cloudknot's default parameters and also using custom parameters +[#]_. Regardless of the :code:`Knot` parameters, Pywren outperformed +Cloudknot at all argument array sizes. Indeed, Pywren appears to achieve +:math:`\mathcal{O}(1)` scaling between :math:`4 \le n_\mathrm{args} +\le 512`, revealing AWS Lambda's capabilities for massively parallel +computation. For :math:`n_\mathrm{args} > 512`, Pywren appears to +conform to :math:`\mathcal{O}(n)` scaling with a constant of roughly +0.25. By contract, Cloudknot exhibits noisy :math:`\mathcal{O}(n)` +scaling for :math:`n_\mathrm{args} \gtrapprox 32`, with a constant that +is comparable to Pywren's scaling constant for :math:`n_\mathrm{args} +> 512`. Precise determination of these scaling constants would require +more data for a larger range of argument sizes. + +.. [#] Default settings are :code:`min_vcpus=0`, + :code:`desired_vcpus=8`, and :code:`max_vcpus=256`. Custom settings + are :code:`desired_vcpus=2048`, :code:`max_vcpus=4096`, and + :min_vcpus=512`. Both default and custom Cloudknot cases were also + limited by the EC2 service limits for our region and account, which + vary by instance type but never exceeded 200 instances. .. figure:: figures/nargsscaling.png Execution time to find solutions of the 2D heat equation for many - different temperature constraints on a 10x10 grid. We show scaling - as a function of the number of constraints for Pywren, the default - Cloudknot configuration, and a Cloudknot configuration with more - available vCPUs. Pywren outperforms Cloudknot in all cases. We posit + different temperature constraints on a 10 x 10 grid. We show + scaling as a function of the number of constraints for Pywren, the + default Cloudknot configuration, and a Cloudknot configuration + with more available vCPUs. Note the :math:`log_2` scale for the + :math:`x`-axis. Pywren outperforms Cloudknot in all cases. We posit that the additional overhead associated with building the Docker image, along with EC2 service limits limited Cloudknot's throughput. :label:`fig.nargsscaling` @@ -598,31 +627,32 @@ neuroimaging and microscopy. And we've included scaling analyses that show that Cloudknot performs comparably to other distributed computing frameworks. On one hand, scaling charts like the ones in Figures :ref:`fig.nargsscaling`, :ref:`fig.syssizescaling`, and -:ref:`fig.mribenchmark` are important because they show that Cloudknot -does not introduce undue overhead burden and exploits the scaling -efficiency of the underlying AWS Batch infrastructure. +:ref:`fig.mribenchmark` are important because they show potential users +the relative cost in execution time of using Cloudknot in comparison to +other distributed computing platforms. On the other hand, the scaling results in this paper, indeed most scaling results in general, measure :math:`t_\mathrm{exec}` from Eq :ref:`eq.tcompute`, capturing only partial information about :math:`t_\mathrm{compute}`. Precisely measuring :math:`t_\mathrm{env}` -including the time for users to learn a new system is a human computer -interaction (HCI) problem that was beyond our expertise and resource -limitations to solve at this time. But we believe an extra 30-50% in -execution time may be acceptable when users do not need to learn a -new queueing system or batch processing language nor do they have to -select from a dizzying array of instance types. Beginning Cloudknot -users simply add an extra import statement, instantiate a :code:`Knot` -object, call the :code:`map()` method, and wait for results. But because -Cloudknot is built using Docker and the AWS Batch infrastructure, it can -accomodate the needs of more advanced users who want to augment their -Dockerfiles or specify instance types. - -Cloudknot's simplified API and ability to achieve rough parity with -other distributed computing frameworks makes it a viable tool for -researchers who want distributed execution of their computational -workflow, from within their Python environment, without the steep -learning curve of learning a new platform. +is beyond the scope of this paper so the reduction in +:math:`t_\mathrm{compute}` is admittedly speculative. But we believe an +extra 30-50% in execution time may be acceptable in some situations. For +example, if the amount of time that a user will spend learning a new +queueing system or batch processing language exceeds the time savings +due to reduced execution time, then it will be advantageous to accept +Cloudknot's suboptimal execution time in order to use its simplified +API. Beginning Cloudknot users simply add an extra import statement, +instantiate a :code:`Knot` object, call the :code:`map()` method, and +wait for results. And because Cloudknot is built using Docker and the +AWS Batch infrastructure, it can accomodate the needs of more advanced +users who want to augment their Dockerfiles or specify instance types. + +Cloudknot's simple API and its conditionally acceptable execution +time compared to other distributed computing frameworks makes it a +viable tool for researchers who want distributed execution of their +computational workflow, from within their Python environment, without +the steep learning curve of learning a new platform. Acknowledgements diff --git a/papers/adam_richie-halford/figures/compare_heateq_results.ipynb b/papers/adam_richie-halford/figures/compare_heateq_results.ipynb index 51323521da..ded4025f3f 100644 --- a/papers/adam_richie-halford/figures/compare_heateq_results.ipynb +++ b/papers/adam_richie-halford/figures/compare_heateq_results.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "code", - "execution_count": 2, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ @@ -14,7 +14,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "metadata": {}, "outputs": [], "source": [ @@ -35,7 +35,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 3, "metadata": {}, "outputs": [], "source": [ @@ -63,7 +63,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -353,7 +353,7 @@ "11 662.090 4096 cloudknot (custom)" ] }, - "execution_count": 5, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -364,7 +364,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -535,7 +535,7 @@ "6 400.223 175 cloudknot (custom)" ] }, - "execution_count": 6, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" } @@ -546,7 +546,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 8, "metadata": {}, "outputs": [ { @@ -760,7 +760,7 @@ }, "x": { "axis": { - "title": "Number of arguments" + "title": "Number of arguments (log scale)" }, "field": "npoints", "scale": { @@ -775,7 +775,7 @@ }, "field": "max_job_time", "scale": { - "base": 2, + "base": 10, "type": "log" }, "type": "quantitative" @@ -783,7 +783,7 @@ }, "mark": "line" }, - "image/png": "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", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdQAAAFfCAYAAAD3UD13AAAgAElEQVR4nOyde3wTVd7/P6iLt8UrNhPsum7dxbX+ZFddnvVZ9PmxrspCJuCtSVEf5ZGVnzfYB12aCeoaIZNWQV3xtoouikAmAVQuIooCK3gDUUEBuRQUCsqtUJpJkEK/vz9mkk5L0ullJnNazvv1Oi/adJK8mZnm0zNzzvcAHA6Hw+FwOBwOh8PhcDgcDofD4XA4HA6Hw+FwOBwOh8PhcDgcDofD4XA4HA6nHRwH4Iwmj50F4Dz9Z7mecx4AVwuel2tbDofD4XA6DacC+AuABwyPDQCwA8CTACoBnNbkOScBWArgWQCrAZQ287xc23I4HA6H02k4HsCLAAjAPfpjJ0ALwwL9+xEA7m/yvJsAjNG/PhlaiBbkeF62bU+x8j/B4XA4HA4r3ICG0CyE1pM8Vv++D4BHmmxfDuAP+tddALwP4JIcz8u2rdtafQ6Hw+Fw2OBmNATqBQA+AHCM4fsJTbafC6C3/nUXADEA1+R4XrZtiwCgvLw8VFFRQcb28ssvU2VlJW+88cYbb7y1ulkZjG3FGKgnA1iGhkFFg3HkJd87AFylf30CgHUAzs7xvGzbnp5LpKKigokdkoaVA2SENSfuYw5rTqz5AOw5cR9zWHNixccYqCdAu3R7GbQe5QQAN+o/+w2ArtAuEX8A7fLur6DdOz09x/OybXtiLhEeqOaw5sR9zGHNiTUfgD0n7mMOa06s+NyIxr3QS6ANVCIAL0G7jHuC/v0Z0AJzkmGbXs08L9e2WS/5VlRUOH7JgDfeeOONt47Z7ArJ9nISGkbsAg29TuN81QIAPzV5XnPbHgHvoZrDmhP3MYc1J9Z8APacuI85rDmx5tMcxwG40O434YFqDmtO3Mcc1pxY8wHYc+I+5rDmxJqP4/BANYc1J+5jDmtOrPkA7DlxH3NYc2LNJ6/we6i88cYbb7xZ2ZzONabgPVRzWHPiPuaw5sSaD8CeE/cxhzUn1nwchweqOaw5cR9zWHNizQdgz4n7mMOaE2s+jsMD1RzWnLiPOaw5seYDsOfEfcxhzYk1H8fhgWoOa07cxxzWnFjzAdhz4j7msObEmk9e4YOSeOONN954s7I5nWtMwXuo5rDmxH3MYc2JNR+APSfuYw5rTqz5OA4PVHNYc+I+5rDmxJoPwJ4T9zGHNSfWfByHB6o5rDlxH3NYc2LNB2DPifuYw5oTaz6OwwPVHNacuI85rDmx5gOw58R9zGHNiTWfvMIHJfHGG2+88WZlczrXmMKGHuqd3bt3X9m1a9dUSUnJIjQsI8cbI61r166p7t27rwRwZ3MHsqWw+EvFmhNrPgB7TtzHHNacWPNxHCsD9cwzz5zdp0+fvfPnz6dUKkUcNkmlUjR//nzq06fP3jPPPHN2e487i79UrDmx5gOw58R9zGHNiTUfx7EwUO/s06fPXqfDgtM6+vTpsxft7Kmy+EvFmhNrPgB7TtzHHNacWPNxHKsCtXv37ivnz5/vdD5wWsn8+fNJv/zbZlj8pWLNiTUfgD0n7mMOa06s+TiOVYHatWvXFL/M2/FIpVLUtWvXVHuOPYu/VKw5seYDsOfEfcxhzYk1n7xi1yjfVatWEYCcH9qqqtIPP/xA+/bty2NUcFoKAMdHCvLGG28dszmda0xh4T3UIz6o6+vr6cUXX2w0wvTee+8lVVXb9MG/YcMGmjt3Lv3444/tzRCOAf3YtBkWf6lYc2LNB2DPifuYw5oTaz6OY2egrl+/ngCQz+ej2bNn0/DhwwkAPfHEE2364J8xYwYVFBRQIpFob4ZwDIAHqu2w5gOw58R9zGHNiTUfx7EzUKdPn04AaN68eUREdPjwYQqFQvTUU0/R3Llzyev1UmVlJRERvfXWW+T1emnr1q20fv16uu666wgAXX311bR48WJaunQpFRQUEAAqLS2lRCJBK1asoD59+hAAGj58OO3evZsOHDhAI0aMIEmSaNCgQXTJJZfQnDlzaMyYMdSvXz+qqKigQ4cO5TWwWAc8UG2HNR+APSfuYw5rTqz5OI6dgbp58+bMpd6rr76aJkyYQCtXriQioq+++ooA0MSJE6m+vp6uvPJK6tmzJ6VSKerduzdddtlltGzZMho6dCgVFBTQF198QUOGDCEA9Pzzz9O6desIAN199900c+ZMKioqoiuuuIJqa2upuLiYunXrRs899xwVFRURABo1ahTdfffdBICWL1+e78xiGvBAtR3WfAD2nLiPOaw5sebjOHYGKhHR1q1b6cknn6TLLrus0X3Uuro66tOnD3m9Xtq+fTsBoMmTJ9PBgwepd+/emV7n+PHjMwE4a9YsKigoIFVVaerUqZlAVRSFiouLqaCggLZv307FxcX0zDPPEBHR8OHDqaCggFKpFH377bc8ULMAHqi2w5oPwJ4T9zGHNSfWfBzHzkBdsWIFTZw4MVM16cCBA3TLLbdQQUEB7dmzh6ZOnUrdunWjESNGULdu3ej7778nIm3w0RNPPEGDBg0iANStWzfasmULTZ8+nQoKCqimpiYTqJFIhKLRKL366qs0ceJE2rNnT6NAvfvuuzPP2bBhAw/ULIAHqu2w5gOw58R9zGHNiTWfvGJncXxkCdS33nqLAFBJSQktX76c3n333cyl2bq6ukzPFAANHTqU6uvrad++fVRQUECBQIB27txJ9913HwGgL7/8kqZPn07dunWjpUuX0ieffEIA6JlnnqHNmzfT8OHDyePxUHV1NQ/UVgI+bYY33nhrY3Mu0RjE7ku+EydObDRtpnfv3rRhw4bMz9Mjf99///2cz5EkiQ4dOkQLFizI9Fj37NlDkydPzmxTUFBAn376KaVSKerVqxdNnDiRiIgCgQAVFhaSqqqZQP3ss8/sTagOBngP1XZY8wHYc+I+5rDmxJqP49gdqEREP/74I+3YsYP27NnT6HFVVSkYDFJhYSHt37//iJ/t27ePampqGj1+6NAhOnDgQKPt9uzZQ3V1dRZEy9EJeKDaDms+AHtO3Mcc1pxY83GcfARqNurr66lfv37tmpfKsQbwQLUd1nwA9py4jzmsObHm4zhOBSoR0aZNm2jNmjVUX19vQ0xwWgp4oNoOaz4Ae07cxxzWnFjzcRwnA5XDBuCBajus+QDsOXEfc1hzYs3HcXigcsAD1XZY8wHYc+I+5rDmxJqP4/BA5YAHqu2w5gOw58R9zGHNiTUfx+GBygEPVNthzQdgz4n7mMOaE2s+eSXfhR2M7KhO0LT3vqbw5CUUnryEpr33Ne2o5qvGsAB4YQfeeOOtjc3pXGOKfPRQ31y6jkRJIVFSaPhT82n4U/Mz389auq5VH/7V1dW0Zs2adgXIJ598Qnv27KENGzbQzp072/Vaudi6dSutWLHCltduKfX19fTOO+/Q4cOHm90OvIdqO6z5AOw5cR9zWHNizcdx7A7UdJiOnbyEEqmDmccTqYM0dvISEiWF3mxFqH711Vf00EMPtXj7bHg8Hvrwww/phRdeoMWLF7f5dWbOnJkpcWjk8OHDdMUVV2RqE7eVgwcPUq9evdq8IDsR0dSpUykajTa7DXig2g5rPgB7TtzHHNacWPNxHDsDdUd1IhOmuUiHarbLv/X19fTMM880Klm4Zs0aqqioICLKrJHarVs3mjNnDhFpK9Kke4YbNmygSZMmERHRwoULqaCggEpKSgh6Pd9XXnmFPvroIzp8+DDdddddNGfOHJo3bx4FAgEqLi6mwsJC2rhxY9b32rp1a6bsYXpJujRvvvlmJmhVVaVhw4YR9HrFiUQiq2N9fT09/fTTmf/r5s2bSZZlAkC33347HT58OFOSsbCwMNNLLy8vp4cffpgA0OOPP07/+Mc/CAA9/PDDdPjw4czqPU0rUfFAzS+s+QDsOXEfc1hzYs3HcewM1KkLviJRUhr1TJuSSB0kUVJo6oKvjvjZu+++Sz179qS9e/eSoih05ZVX0urVq6miooK2b99O3bp1ow0bNtC2bduooKCA1q1bR6+88gp9+OGHRET09ddfU0VFBW3ZsoW6detGq1atyhTVX758Ob366qu0ePFiGj58OI0aNYrq6+spHo9TUVER7dixgyZOnEgPPfRQzvd64oknaNSoUUcsWD5s2DBatmwZERFJkkSjR48mVVVp6NChNHHixKyO69evp6KiIkomk7RgwQK68soradu2bdStWzdas2YNLViwgHr27En79u2jJUuWUGFhISWTSSotLaVwOEw//PAD9erViyKRCH3//ffUu3dvWrdO6/n7fL5mFwQAD1TbYc0HYM+J+5jDmhNrPo5jZ6COnbyEhj81P+cHeZrhT83P2ostLS2ljz76iIiI6urqaOPGjZke6rRp02j8+PGZbSdNmkSvvPJKo7BKbztv3jyKRCKNXnfZsmU0ffp0AkBFRUV08KAW+q+88krmMvCmTZuouLiYpkyZkvW9mjqk8fl8tGzZMkqlUtSzZ0+qrq4mIqL9+/fT1q1bszqme7yhUIhWrFhByWQy8/yampqMs/E9vv76ayotLaUdO3Zk/l/pHnVpaWnmPcaNG5f5OhvggWo7rPkA7DlxH3NYc2LNx3HsDtQR7QzUL774IvP9nj17GgXq008/nfnZpEmTaNasWfTKK6/Qxx9/TEQNvb9JkybRuHHjGr3usmXLaNq0aXTXXXdRr1696O233yYiahR269evp169etHUqVOzvlfToE3TNFDTBf5//PFH2rdvX1ZHIqItW7bQP//5TyooKKDS0lJSVTUTyMZ9UV9fTzfccANt3LiRhg4dSnv27KH6+noaOnRoJrzTDk3/T9kAD1TbYc0HYM+J+5jDilOtLPxRDQtvVW5Yx4QPM7B8yXfSpEn04IMPEhHRggULqG/fvplLvuvXr6eePXtSXV1dZvDOF198kek9EmlLw02YMIHWrl1LPXv2pIMHD1JNTQ0VFxdn7qF+/vnntHbt2syaqdkC9Ztvvsn5XtkC1dib9Hg8mbCWJIn++c9/ZnVctGhRZrDVvn37qKioiHbt2kWFhYVUXV1N06ZNy/x869atVFBQQPv27csEKhE1+toYqMbeKg9UZ2DNB2DPifuYw4JTQhaGqLJAqizQ9+8+4bgPU+RjUFLYZFCSPzQz66Ck6upq6t27N0FfA3XVqlW0Zs0aevTRR6m+vp4eeuihzMCg9CCcVatWZR6DvgB5fX09jRo1qtHjX375ZaOeYiAQIFmWGz22fv166tu3Lx06dCjre3311VeZ+7FGZs6cSdOmTSMi7ZJu+nlFRUW0d+/erI41NTXUq1cv6tatGwGgiRMnUn19PQ0ZMoS8Xi/t378/sy8A0Ouvv57plWYL1NLSUlq+fDnV19dT3759acuWLTxQHYQ1H4A9J+5jjtNOquwKp8NUDQuPOO3THG4AvwbQLcfPjwNwHgBXk8fP0h8/rgXbHkG+ps2Es0ybCbdw2kwikci53ml1dTXt27ev0WM//vgj1dbWZt22Peum5nqvpuzbt4+Ki4sz92UPHTpEiUTiiOflcmz6uNF5z549R7xWc3z++ed0xx13NLsNeKDaDms+AHtO3Mccp5zoBfwkIQtRPUwPJWVXqZM+ZvgA7ABQDu3D7dImPz8JwFIAzwJYDaBUf3yA/rwnAVQCOK2ZbbOS78IOI56aTyP0wg7+0MxWzUHtSEyfPp1mz57tqEO6B7t9+/ZmtwMPVNthzQdgz4n7mOOEU3XF6aeqsrBED9PaWln4o5M+ZpwM7QPtDP37qwC82GSbmwCMMWy/A0ABtBAt0B8fAeD+HNuekuvN81XLd0d1gqYu+IrGTl5CYycvoakLvuKlBxkBPFBthzUfgD0n7mNOvp2S4cKzE7KwTgtTV1VtRCh20qelpEOxC4AJ0ILRSDmAPxi2eR/AJdB6oMfqj/cB8EiObd253pgXx+eAB6rtsOYDsOfEfczJp1NCLuilysIPes90ZUIuOOIWIov7KM0vAKwEMBfAqU1+NhdAb/3rLgBiAK4B8AGAY/THL4AWxtm2LQKcLY7PYRfw4vi88caboVV9OI0SFeeSKgtU/a9BtGn96pzb2pKG7eSP0HoJnhw/vwPapWAAOAHAOgBnA1iGhsFIg6H1bLNte3quN+Y9VA54D9V2WPMB2HPiPubkwykhC7eqsnBY75m+SqFMp80Rn9ZyArR7ob+HFo6n6I8BwG8AdAVwA7Te6LEAfqVvfzq0S76XoeFS8Y05tj0x15vzQOWAB6rtsOYDsOfEfcyx2ykpCxV6kNYnIkKZ0z5t4WxoA4eMcxPvgRaq6cFKXQBMMvy8l/7cSwyPvQTt8m+ubbOSr0Ct37eFDn4wng5MH0IHpg+hgx+Mp/p9uedGcvIHeKDaDms+AHtO3Mccu5wodGFXVRZm6mH6YzLivt5JHztI9zrPMDxWAOCnTbY7CQ2DmmCy7RHkI1DrPn2R0pOBUxP/RKmJf8p8X7fsxVZ9+Hfk9VBramqarVhUV1dHM2fOpM2bN7fqvVasWEEHDhzIfK+qKi1cuLDFzwcPVNthzQdgz4n7mGOHkz4t5hP9M7k6GXH93kkfuzgOwIV2v4ndgZoO0wPTh1B9qibzeH2qhg5MH6KF6qctD9WOvB7qxo0bqbi4+IjVadIsXbqUevbsSbt3726xQ319PfXv3592796d8amvr6fbb789s9qMGeCBajus+QDsOXEfc6x2SlX0OKdhWoxQeaDcVeSkT4fHzkCt37clE6a5SIdqtsu/nWU91KFDh1KvXr1o0KBB5PV6MyUSi4qKqFu3bjR37lxKpVJUXFxMBQUFtHr1alqzZg0VFxcTAJIkiVKpVKP/TyqVokAgQIcOHaKhQ4dmyiCmfbZu3Uper5fq6+t5oDIAaz4Ae07cxxwrndRy96WqLOzSw/SzfeXn5By8mg+fToGdgXrwg3FaWBp6pkeEZqqGVFmggx+MO+JnnWE91BEjRtDo0aOpurqa7rvvPvJ6vbR79+7MOqc7d+6koqIiWrduHVVUVNDo0aOprq6O+vXrRx999BHV1dXRlVdeScuWLWu0Mk0ikaCrr76a6urqMjV8jT5NV7rhgeosrPkA7DlxH3OsckrI7n6qLKT0mrxzKHTuCebPss+n02BnoB6YPoRSE/9k+oGemvinrL3Yjr4e6sGDB6l3796ZgvWbNm2iXr160WeffUaFhYU0b948euedd6hPnz704YcfNnq9nTt30qJFi+ill16ioqIiWr58eaPeeSKRoP79+zcKVOPz6+vr6corr8y8d3OAB6rtsOYDsOfEfcyxwkmNuO9IT4tJRITHidDFSZ9Ohe2B+lL7ArUjr4eaSCSosLAwU1B/w4YN1KtXL3rrrbeoX79+NGPGDIpGoxSNRmnbtm2Z16urq6M+ffrQXXfdRQsWLKD777+fPv74Y1qzZg098cQT2j5LpY7ooRp9mq5E0xzggWo7rPkA7DlxH3Pa40SELomI8Lh+ifewGnHf4aRPp4TlS76dYT1Un8+XCWtFUeiKK66gjRs3UlFREdXV1WUWC//yyy8zgZhIJKioqIgOHDhAdXV11Lt3b1q2bBmtWbOGbrrpJqqvr6c33niDrrjiisw91D179jTySV/yTS84zgPVWVjzAdhz4j7mtNWJQhd2VcPCHD1MUwnZ3c9Jn04LC4OSkuN7Zh2U1BnWQ12/fn2j9y0tLaXDhw/T448/nnmsrKyMDh8+TNOnT6fx48dTfX093XnnnQSACgoKqLS0lK644grau3cv9erVK/M8r9fbKFCNPtXV1Zk/AnigOg9rPgB7TtzHnLY47Ss/53TDtJhdarm76WpmefXp1HSEaTMdfT3UH3/88YjnERHt378/q2c23/Q800OHDmV9raY+EyZMoFmzZpn8jzTAA9V2WPMB2HPiPua01ulAuatIlYVKVRYoIQvrUhU9znHSp9OT98IOL/2JUi9phR2S43u2ag5qR8LJ9VATiQT169evxX88gAeq7bDmA7DnxH3MaY2TPi2mWv/sXVJdcXrTRVfy6nNUkN/Sg+MMpQfH8dKDjAAeqLbDmg/AnhP3MaelTmrEJaanxSRkIUov4CdO+hw18OL4HPBAtR3WfAD2nLiPOS1xSsrukZnVYsKusU77HFXwQOWAB6rtsOYDsOfEfcxpzokIXdSw60X9Eu+hhCzc6qTPUQkPVA54oNoOaz4Ae07cx5xcThQ69wTDtJjaWln4o5M+Ry08UDnggWo7rPkA7DlxH3OyOe0P9eiuysJnWpi6qmojQrGTPixxFpBzdfTjAJwHwJXlOefpPzfb9gjyFaiHdu6k/dOn0+7x42j3+HG0f/p0OmTT0mmc1gEeqLbDmg/AnhP3Maepkz4t5ju9Z7oyIReYfubb6cMKXaEF4Do0Xv80zUkAlgJ4FsBqAKX64wOgLU7+JIBKAKc1s21W8hGotW+9RVV+H1X5fbSjbBTtKBuV+b523lt5jA5ONsAD1XZY8wHYc+I+5hidkhHX79PTYpKysJBCPU5y0oclrgLwMbRwzLaEzk0Axuhfn6xvVwAtRNOLi48AcH+ObU/J9cZ2B2o6THePe4wOJxKZxw8nErR73GNaqL7FQ9VJwAPVdljzAdhz4j7mpJ2SEff1qiz8qPdMX6U4jnXSh0V+Aq2Hmi1QywH8Qf+6C4D3AVwCrQea3pF9ADySY1t3rje1M1AP7dyZCdNcpEM12+XfRYsW0fDhw2nQoEFUUFBAn376Ka1Zs4ZGjhyZWeezoqKCFEXJrMIyb948Ki8vJyKixYsX02uvvZZZbaaoqIiqq6vp9ddfJ+hrrK5Zs4aIiMrLyykUChEAKikpoYQh/Ds74IFqO6z5AOw5cR9zKisrKSELo1RZqFdloT4RcUlO+zj5/s1xAnIH6lwAvfWvuwCIAbgGwAdouOd6AYAJObYtAoDy8vJQRUUFNW2VlZXtbsgSqDXT41Tl9zXqmTblcCJBVX4f1UyPH/Gz6dOnU0FBAW3ZsoU++eQTKigooKqqKiooKKDq6upMAfhVq1Zl6tYOGzYs8/XIkSPpww8/pHg8TgUFBfT555/Tp59+mlljdcWKFVRYWEjJZJJKS0vp/vvvp+rqarrhhhvo/ffftzS0WAaAJecAb7zxZmPbuIF2KcNIlQVKlP+Mtv37X847VXbMQL0D2mVh43ZnA1iGhsFIg6Fd8s22bc6V2O3soe4e9xjtKBtl+oG+o2xU1l5s0/VGfT4fLV++nIYNG0aLFy+mNWvW0KBBg+jw4cN0xRVX0Pr168nj8VBRURFt3LiR+vbtSzU1NTRp0iSaN28eERE9//zzNGTIEHrnnXfo7bffpsLCQtq9ezfdfvvttHv3biIimjFjBoXD4bbmU4cDvIdqO6z5AOw5cZ/c1I51X6DKwlx9JO/eVLjH5U47AWzto6ZkC7/fQBuwdAO03uixAH4F7d7p6dAu+V4GrSc6AcCNObY9Mdeb2h6ogTLTD/RcgTplypRG65CWlpbS8uXLaenSpXTffffRuHHjMgXgJ0yYQA8++CA9+uijpCgKBYNBuu222+jw4cONVpCRJIlGjRpFsViMotEoxWIxSiQSjdYOzbXOaWcFPFBthzUfgD0n7nMkiYhwm2GlGNr/ZC86EDm7p9NeaVjYR7k4AcBKNATqCdA+6M6AFpiT0LAMWC99m0sMj70E7fJvrm075CVfr9dLdXV19P3332fWLE0kElRcXJy59EtEmWXbFixYQJs3byYAFI1GiajxouEzZsygESNGaH41NdSrVy+qqak56gPV6ctGvPHGm9a+XfUh7Zx5H9WOPz+zqEjt+F/TzjfKaPOazx33a9rsDEUrSfc6jdNoCgD8tMl2J6FhpC9Mtj2CvAxKGn/k4uFpdo97jLb/z5Csg5KmT5+e/qOAANDEiRMzP3vkkUfojjvuyAxOSi/KvWXLFkqlUpnLvkTUqId64MAB8ng8mdecOXMm1dfXNwrUadOm0TPPPNP2hOpggPdQbYc1H4A9p6PZh0I4Ro24Bqph4W19wBHphe3/nZRdpeni9kfzPmovxwG40O43ydu0mfHjjpw2M35cs9NmpkyZQs888wzV1dUdMep22LBhmV5nW6iurs6sMXq0Ax6otsOaD8Ce09HoUxtyFaiy+wFDcQZSZaFGlYVnslU8Ohr3UYci74UdAmW0I1BGVX4fbf+fIc3OQZ06dSo99ljje6upVIqKi4vJ6/XSoUOHLAuVoxnwQLUd1nwA9pyOJp9U2P1fiYigqLJwsCFIXV+osnsYjXOd7IRTW2DNJ6/k+x6qkUM7d1LN9DjtHvcY7R73GNVMj7e59GBNTU3mUi+n/YDfQ+WNN9vbpnVf0ffzH6OaZ/4zc280UXEO7Z76P7Rl2WzH/dranM41puDF8TngPVTbYc0HYM+ps/okxrguUmXXC6osJAyXdTckIsLfakKF2UrN2u5kFaz5OI5VgVpSUrLI6WDgtI2SkpJF7Tn2LP5SsebEmg/AnlNn8qEJvzw+IbtvUWXhI0OIHlJl1xuJ8h7XEKFLvp3sgDUfx7EqULt27ZpKpVJOZwOnlaRSKeratWuqPceexV8q1pxY8wHYc+oMPqmw++dJWXhMlYVdhiDdroaFR9RHu/dwwslOWPNxHKsCtXv37ivnz5/vdD5wWsn8+fOpe/fuK9tz7Fn8pWLNiTUfgD2njupDIRyjhl3eplNekhHh/WREuJFCjZbWzItTvmDNJ6/YOShp7NixdPXVVzudD5xWcvXVV9PYsWMdH9jAG28drW1e/Rn9MOdh2v/kbxoKMIz7Je2c8Vf69svFjvvZ2VavXU9zF2tFJpzONaawcFASzjzzzNl9+vTZO3/+fOKXf9kllUrR/PnzqU+fPnvPPPPM2e097iz+UrHmxJoPwJ5TR/HJPuVFWK7KrtspdO4JTjjliwEPTP25KCn/ECWlVpSUH1euXsfUMXMcKwNV587u3buv7Nq1a0of7EK8sdW6du2a0i/z3tncgWwpTv+SZ4M1J9Z8AE8mnPYAACAASURBVPacWPbZ9Wj3bqos3KNGhK8NIZpUZeHlhHz2b51wyieeQPQPYiA2XZQUyrSgEl2+ci1Tx8xxbAjUdsHaLxXAnlM7fbpZJqLD2v4B2HNizQdgz4lFH33Kyz9VWajN3BuVhW+Ssuuve0PnnuaEU77eKxQKHeORlBJRUj42BGmNV1LG9y+bUphvnw4BD1RzWHPiPuaw5sSaD8CeE0s+yYhw474XriJDb5TUiBBPyK4/OemVj31UEor/1BNU/tcjKZvTQeqRlM1eKTqyJBRvVB+epWOWd+wclMQbb7zx1tHb1o/itO/5vpkQ3f/ERbRj9oO0+etljrvZ3VasWkuPT1tMN/59euay7j1PzKGZC5bTho0bcz7P6VxjCt5DNYc1J+5jDmtOrPkA7Dk56aOW97hYDbveMfRIt21//xmm9g9gzz4aKEUvFoNKVJSUOj1ID4uSMsMTiP7BCZ8ODQ9Uc1hz4j7msObEmg/AnpMTPvvH9jhflYWZhiDdmZRdf6XQhV1Z2z+AlfuIuohS9DpRii4x3B+tFaXYUwMemPrz/Pt0EnigmsOaE/cxhzUn1nwA9pzy6ZMMF56tysIkrRygQElZ2K9GhL9TqMdJTvi0lPY6eUNzThIl5V5vUNnYEKSxrWIwVnZVIH5qvn06HTxQzWHNifuYw5oTaz4Ae0758Nkf6tE9Ibv/ocrCAb1HmkpEhMezFalnbf8AbXcaWBbtIQaVclFSqjNBGlBWiJJyc9/QojZXcmJxHzkKD1RzWHPiPuaw5sSaD8Cek50+FDrrp6osjDFMfzmkht0vqXJ3txM+baW1TgMC8Yu8UvQ1r6Qc1IO0XpSU2d5g/P864dOp4KN8eeONt6OpbdrwDf3w1liqHX++fo/UTXsmD6ZvV/7bcTe72sbKSpq9aAWNnDAvM1r3+gfjVDF5IX36xRrL38/pXGMK3kM1hzUn7mMOa06s+QDsOVnpQyEcp8ruYarsqmoYcOR+Vy3vcbETPlbRnFPJyPiJohS7U5SUbwwDjb73SsoDHmnq6fn2OSrhgWoOa07cxxzWnFjzAdhzssKHCF2SYZc/IbvXZ4I0LHyeCve43Akfq8nm1D8YP0sMRseKUnR3Oki9krLKK0WHXDrshZ/k2+eohgeqOaw5cR9zWHNizQdgz6m9Pqos9Fdl1xeGKTCr1UiPa53ysQOj06BgvNgjKZNESfkxfX/UK8XeHlg27WonfNrDSQC8ACYD2AFgP4AYgJsAnGXFG+QLHqjmsObEfcxhzYk1H4A9p7b6JMOuy1RZ+NgQpN8mIsJtFMIxTvjYSWVlJQ2Qpv3ZE4y9a7ise8ATUF4aWBY/3wmf9r6GB41X71ihN+NjIwHY2tW2Ch6o5rDmxH3MYc2JNR+APafW+tRGhGJVFuYaau3uSIZdIyh0YVcnfOzGK0WHDK2YRYYg3ekJKo9c97fJBU45tXcfnQwtMO8BcC7QaCX2YwEUACjRt7FkWLLd8EA1hzUn7mMOa06s+QDsObXUJzXW9Yuk7J6qykK9HqY1qiw8ZCzKkE8fu/FKscGNCzEoa73B6FCnvYD276NjARirSfwMQBcAF0DruaZHUp0I4KdgDD5thjfeeOuobfPqz2hX7E5SywtJlQVKVJxDO2eOpM1rv3DczY725sIVNOyxhh7p0IpZNGfx5457NW1W5dM90O6f9kXDpd51AGxdsd1qeA/VHNacuI85rDmx5gOw55TLZ2/o3NOSYeFRfUFvrSiD7J7YXFEGO33sxlumXNZkDdINXik2GKAuHeWYtZYTAKwGMBpAGbRBSTdDC9jfWfEG+YIHqjmsOXEfc1hzYs0HYM+pqQ+Fepykyu4HVFnYpwdpvRoR4gfKz/qlEz52MygYLxYl5S1DkG4TpdidxtKArB+ztnIygEoAFwN4FsBsAF2hhWxvK94gX/BANYc1J+5jDmtOrPkA7DmlfegF/CQZFu5VZeGHhrmkrncSY1wXOeFjN94H47/wSNFpellAEiWl2iNFAyUj4yc65dRSrPI5BsAiNFzqHQNggv71L6x4g3zBA9Uc1py4jzmsObHmA7DnVLlxAyVk4VZVFjYbpsAsb0tRBkt8bN4/1/1tcoEoRZ83rEOqeoMxuX9oyilOObUWK30uArAUwMcAfgmtdzrSqhfPFzxQzWHNifuYw5oTaz4AW06Jcvefa56+jAxBukYNu7xOOtm1f64NvXGaGFAeFSUlqVc1OigGY8+0ZPoLS8cMaL/P8QAuzPEz45DtXwDo0Z43yhc8UM1x2inx2FlCUhZ8+rJTn9U8ezntD/Xo7qSTEaf3TzZYc2LNB2DDiULnnqDKrhcMQbpFjbg6xZSQppSMjJ/oCURHiwFlr94jPeyVoq95A/FznHJqL+31Sc9DfR3AnwAUQptGczK0Oaj/iYZLvx3iXioPVHPy7aTKPS5Jyq7hiYigqLLwneHDptFf8InHzhLy6ZULfszMYc0HcN4pIRf0SsrCN+lz+oc5DzO1j6zaP5cOe+EnYiB6jygp3xsGHM0eFIwXO+VkFVb4/ArAu2hcGcnYdqCDFHUAeKC2BDudKHTWTxNywdWqLDysyu53Des1Gtu2hCzEkmHXiFpZ+OO+f/5Jm4cnu9cnw4Vn2+XWUo62Y9YWWPMBnHVKhl0jDOf31mS44A+s7aP2+1AXTyB2ixhQNhmC9OMBZdE2zwTpfPuogZ4AhgKIACgH8P8A9IJW6IFJeGEH59u3X31E2xb+k3bF76F9z/0XqZEeTcLTTTXP/IF2xe6k7Qufp29XLT3iNTat+5r2vXA1qbJA+//xW/p21YeO/794460lbfPaL6l60vWZ833Pqz7a9M1Kx72sbrMWrqA7Hm0oynDnuNk0l8GiDFY0p3ONKXgP1Zy2OlEcx6rl7ksNl2+3Zul9phKy8EEy7C5XZcFTXXH6qWavW1lZSfRE4YnJiLBIew1X1YFyV1FbHK2gMx0zu2DNB8i/Uyrc43JVFrbr531Sld3DnPQxoy0+3jLlMm9QWW7oka4XpagfIEs6Wp1hH3VqeKCa01KnPaEzTknI7n6qLIxJRoT3VVlIZAnQ3aoszE5EhLJkuOAPbSnknfahCb88viFUhR+cCtWOfMzyBWs+QP6cKI5jVdkVVmXhsF7E/uv9Y3scsTIKa/uoNT4DAvGLxKAy3xCkVR5JGWYsypBvp3zAmo/j8EA1J5dTaqzrFwnZfYsqC8+rEWFV5gOjcatUZeFVNeK+o3as+wKrfWjCL4/X772SKgs/ZPugspuOdMycgjUfID9OyXCPnzVaWi0iPE0Tfnm8Uz6toSU+A6X4Lz3BaCxTlCGo7BElZVT/4fOy/h/z4ZRPrPb5FYBbAPwB2v1TW3ainfBANSftlAwL/5EMu+9TZWGGKgs7c4y+XZ6IuJ9Kht031IZctiyrlG0fqbL7Tf3993TWijKtgSWn6orTT/125QfM+KSxex8lI+7rDWUD96gRl+ikT2tpzqff6NfcYiD6oqEoQ8IjKeHmijLY7eQEVvqkl2kjAHcAmAStBOGxVr1BPuCBmptkuPBsVRZCe1/8M6mykMoSnsmkLLynykIoIRdcbfXyUbnItY9UWZip31Pdq8o9LsmHS3M+TuKkUzIs/EdSdv01GXFNUWVhY/p8ScruqU5cQciFXftIm1sqPJ/5f0eERS0pZM/aeZTNp9/I+BmeQGycKCkpPUh/9AaVCd77p+VlXnhH2EdtIV0c/3EAD0ML1GughWur5xY5CQ/UI0lUFPwmKbteyxKgO1VZmJmUXX9Vw27HFkFobh8lZCGqfXgL+5MR1++d9nGKfDmlwu6fJyPu65NhYZwqCx/luGrRqCVl99QDY8/6VT78msOOfZSQC3olZGGd4f/7kJM+7cHo4w3NOckjRR8SJaUmXZRBlJRXW1OUwWonFrDKJ10cvzeA/gCGQVv/lAD8xoo3yBc8UBtQw65BSVlY2OQD8LPt7z1DByJn93TKqylm+0iVhVd199p81EFl7ZccsMeJQj1OqpWFPyZkIaDKwmxVFnY1d9k/IQu31o4RLgSAb79cTIbjogVrWJicr5VTsmH1Pmoyt/TbZNh1mZM+7aWyspJKQvGu3mB0hChFd6QHHHkk5c2BZXFHrjSwuI+seJ10cfz90Gr5vqt/vwOArdfQreZoD1QKXdhVjbiGqrKw2vBhcFiVhZnJSI8+TjiZYeZDhC6GD+9k7Vihr5M+TtBeJyJ0qY0Ixarsul0vjbdSlYVDWcJzQzLimpIMu0YkI67f5xq1nfY5UH7WL/WrH+nXOqSGhVdSY115X1TDquO2r/yc01VZmJveJwlZiFHorJ865WMFoVDomOjbn5IoKd82jNyNLmlPUQYrYGkfAdb6FAJYicZVkq6x6sWzcByA8wH8vMnjZwE4T/+5cdvzALjMXvRoDdR95eecrkbcD6rGJaK0KkUTUmF3o33cEU9iInRRw64X9f/XgYRccLWTPvmmtU7a+eAS1bBrrBpxLzAMpjG2XWpYeEuVhYcT5e4/t2TOcC6fLMFap8rCy/kMViuOW5O5pQlVdt3upI8VeAPRm0QpusQwBebzAdK0PzvtBbCzj9JY7XMitJVmfg0tvE628sUNnA4tvJ8F8CaAl6ENfhoArVf8JLRL0KdBK9K/VN92NYDS5l74aAvUA+WuIlUWnlNlQW34oHRVJWQhkOsDsiOfxGpEeFr/fx5UZcHjtE++aM5JX2/zP5Jh14ik7J6qT206ouCGKgsfJSKuJ5Nh1+D2zvHN5bN/bI/z9cFLhxuC1T0xVdHD9ntz7TluFMexatg1NuMddn95YGzBeU75WIEnGLtNlJQN6SC9vfxN8khKiZNOTXF6HzXFSp+bcWQd3/3Qws9q7oC25iqgXW5eCW35uEpoRfkBYASA+wHcZNj2ZJhchj5aAlX7S9r1htp4ruiKRNh9M4XQ7OTrjn4S6wNmSJWFumTYfZ3TPvnA6JQqF85Nyq7SRMT1pKrNiTzQJDzrk7KwVg0Lr6hh4W613H2p2TnRHp9s7B/b43w93A83/AHkesHOYG3rcWsyt7Q+EXE9SS/gJ075tIe+oUXHicHYHU3q7X7tlWKDN25k+7xmAatH+a6EVsf3Ib3J0AYnWc0p0HqfgDb3dT+0ZeRWo2GaTh8Aj+g+f9Af6wLgfQA5h6x39kBNht03qLKwrMkH6Nzasa4rnXJqL23x0SvV6KNMBZ/TPnaSDAv/8cOch0mVhVmqLOzJeulWFmapsvuBxFjXVbse7d7NbqeW7qP9Y3ucnx6pbWjPJx89u9ApJyNN5pbuSsjufk76tJX+w+cdr68As8V4adcTVK5Plwlk7bwG2HOyMlBXArjNihdrIV0A/AVaT7gfgAsAfACtxwr9+wkA5qJh6bguAGIAioCjpzj+pvWr6Yd5Edr/5G8yH0qJinNoV+wu+nblvx33c6rtmP1QZn9sf/85x32sbt99voCq/3Vto/BMVJxLe1+8hna+fj9tWzyxwywk8N2Xi2jPlFsb/V92xe+mzV8vc8Rn0/q1tEv5fxmXvS95afPq5Y7vp9a2b9ZtoJff/IhuGjMzU7j+nsfn0OxFKxx366jNqoAbAq2naOyhjoE9PdTjod0XjQM4U3/sZADL0DAYaTC0S753ALhKf+wEAOvQzGXoztRDTTx2lpAMC4+qslDTuDfifmB/5OwzzV/Beic7aNc+koX7M/vGooWcnd4/qXLhXP0+ZLq83Y7v3y6nfBa3MKOt+2h/pMev9UvBZPj/Pd2SQglWOTVdtzQhC6Pa+97t8WkL1/xt8skeKRoQJWWncdSuV1JyDiR1+rzOBmtOVvn8BFrA5ese6g3QqjCdpLdT0XDZ+TJoPdEJAG7Ut/0A2qXgX0G7z3pirhfuDIGqTW8QXtUH3qQ/dFYlIsL/tKX4vBVOdtJen2RYuDd9/0uNuO9w2qet7I+cfaZ+X/RH/VL2flV2P0BPFJ7Y2Y6Z4VJw+h5rKiG7/9Ge8pYtcVJl4R61oUrYZrXcfWlb388Kn9ZyVSB+qjeg/F2vsasFaUB5T5SifZ3waS+sOVnlczK0MJOghevxhmYHEo5cxPx0AJcYHnsJ2uXfLtDKIKYf79XcC3fkQNVWdskUhk+3+VZPEemMJ7Eacd+hykK9qhUYuNdpn9ZATxSeqEZcow1XIg6qEeFp41WIznjMAO2PR30pwHSwJhMR4fG2BGtzTlbNLbXKp7VcG3rjNI+khL2Sss9wj3Rea+aRsnYOAew5WekThTY15WfQ5ob+HNrcz2Oae5INnISGkb5GCtCCy88dLVDpBfwkIQtDVFn4yhCiB9Sw+yW76qR21pM4IQu3pkM1ERH+5rSPGRTHsfofAtsMx356tukanfWYpamNCMUJWYgZglVNhoVx+0M9WlxTNpdTk7mltYmI+7+tM2+9T2voH4yfJUrKY6Kk1OohWu+RlDcHStGLnfCxGtacrPSJ4chLvgTgDKveIB90lECtrjj9VL1Xsl013h8NC4/URoSznHByCit9jKGqRoS/O+2TCzXS49qkLKw1HPtPmqun3JmPmRFDsNbr+yWRDAuPtiRYmzpp65YKYwwh/Vl755a2hvbso36jX3OLkvIPUVKS6Vq7XkmJDwrG21xbnbVzCGDPyUqfPwK4vknzQesxdhhYD9RU2P1zvThBZrFu7YPVPSzXuop2OzmN1T5J2VWq6hV7kmFhnNM+RpIR1+9VWfik0bEPuwY56dQW7PapjQjFakSIG4K1Nhl2l9eECnP+gW90ajq3NBkWxlk9F9eMtuyj/mVTCr1B5TlRUg7oQXrIK0Vfs6LWLmvnEMCekxU+xQAEAL+ANlXl14Z2IdD+Cc75hNVAVcPu36nauqOZQgzJiPC+alO1n5Y4sYIdPsmw+zpVK39HakR42mmfA5Gze+rzSNM90h/UsPtOirdsecSj4ZhlQx+gNz1zf1wW9quyK7w3dO5pTbfN/K5FelxrmFv6Q60s/DEfrrl8WoL3wfgvPAHlJa+kHBQlhbR/oy97H4xbVrqRtXMIYM+pvT4nQ7usW4bGA3/4JV8LoBCO2fbvf5HaeBmsg8mwMDnfi2Yb6WwncS5UWfBkRkqHXS864aM+2r2HGna/pDbUuE2oshCica5WlfU8Wo5ZLpoGqz6Aa4wxWDetX0t64X/Sj/k7rbkHazUt2UcDpfgvvVL0NVFSDuk90gOiFH2+f9kUJgpf2A1rTu31OQ7atJRe0KareKAVxE83D+wb6WsLLAWqXhowHaQ1ybC73Ir5du2ls53EzaFGhAGGD9gWhaoVPtUVp5+qHW/jiG3XPxNygekCD3Y5WYlTPokxrov0Kz3U0GMVHk6GC/5Q8/TvDUvJue9zws9Ic/toUDBe7AlGY/o6pKTdK4091W/0a7Z9PrB2DgHsObXX5wRo8zrvt0bHeVgJ1PQvfe3480mVhXso1IOZe9Gd7SQ2IzHWdZUqC0n9w3aS3T7JsPt/1UYlAl1vtHf92aPtmJnRNFgNbQMrBTCy7aMBgfhFYkB5Q5SUej1IE6KkPNY/GLd1IGIuH6dhzcmKQF0NrQRgp8DpQNVHFqbvle357ouFTJ0wQOc7iVtC7VihbzpUE7IQbe7eZVt8iNAlIbtvUWXhW8OH+/JkxPX79pm33clOWPHRg3WmKgv1u6fdTqz+4TqgLPo7UVLeMswhrfFISvja0BtH3AvOhw8rsOZkVaCuAFABbdk0Y3sO9pQetA2nA1Uf8k+qLFQnxrguYu2EATrfSdxSUhH3FQ2jq12vW+WjRlyiqi3YnQ7SymREuLH9xm13shvWfNRHu/dgzamyspIGBGL/JUrKgkyQBpU93oDy96sC8RavPWulT77f0wzWnKwI1KaLihubXaUHbcOpQKUQjtEHTZAqC/vSg45YO2EA9pzy6aOWC71VbeF1UmVhbrZlulpeE/bs3yZkYbEhSHcmw8K9dkzPOJqPWUthyUkcHf3TiH/MJUOPdKc3GP3bNX+bbNca06awtH/SsOZkVQ/1fmg90TOytA6FE4FKIRxjKGa+z3gPh7UTBmDPKd8+eqjqZf7c7zad/2vmkyoXztXL5aVHnKqq7ArbuWza0X7MWoLzTtTFG4xdKwajn6aD1CvFtnul6MiSkfGc9cfzhfP750hYc7Ji2kwlgOHW6DhPvgOVCF0MYVrbdEAEaycMwJ6TEz4J+ezfqrJQrc8HXkRPFGY+8HL51EaEs/SiHOlFCw6pYfdLbR252xr4MTPHSScxoNwqSso6Q4+06l+zPmJqH/FjZo4V02ZKAbS4wDLr5DNQidBFlYWXM5VcsgxAYe2EAdhzcsqndoxwoaotzk2q7P4wXSz9iBJ241wnqxHh7/oUDcOi7u4L8uXKj5k5+XbqG5p0gldS7vZIymZDkK71BGO39Q0tOo61fcSaD8CeE2s+jpPPQDWEqZprNCeLB4g1Jyd99o/tcb4qCz/ox3HZntAZp6R9KITj1IhwlxoWvjdMgfkiGenRJ9+e/JiZky+ngWWzunkDiiRKyg+ZS7tBZbknqFwPUJd8+7QU1nwA9pxY83GcfAWqKgvPpcM0Fe5xea7tWDxArDk57XOg3FXUEKquLzZ9s4qSEeHGhCysM/RINyXDrsFOOTq9j5rCmg9gv1P/YPwsbzAmG5dQ8waUhQPLpmVdXpG1fcSaD8CeE2s+reU4HDnw6Sxoy8Yd12S78wCY3qvKR6AmIu6n9A/ZZO1Y15XNbcviAWLNiQUfPVS3qrJAiYpzjYUC9iRl11+d9mNhHxlhzQewz0kvWD/BsPJLvSjFZpmtRcraPmLNB2DPyUqfCwBMhraMW7q9CcCuicenQiso8YDhsQHQFht/EtpgqdOgrXazFNparauh3fPNid2Bmoi4nkx/2CbGuq4y2561EwZgz4kVn1TY/XNVFjYZyto9lq0IuxOwso/SsOYDWO80sCx+vkdSJqUL1uv1dqe0dAk11vYRaz4Ae05W+RwHLbSazkPdAXvmoR4P4EX9Pe7RH0uXQUwvLj4C2nSemwCM0R87WXc6JdcL2xmoSVkYnwnTcvefW/Ic1k4YgD0nlnyS4cKzd02/l5LhHj9z2sUIS/sIYM8HsM5poBS9WJSUGYY6uwdEKfq8NxA/xwkfq2DNB2DPySqf9PSZW/TvjzM0O7kBDXWEC6H1QNMl4foAeARAOYA/6I91AfA+gJwFpO0KVEPPtK6lYQqwd8IA7DlxH3NYc2LNB2i/kyhF+3qCsXcNI3b3i5Ly2MDR0/iCBjbBmpNVPl2gXeJ9ANr6p/kK1JvREKgXAPgAwDGG7ycAmAugdxPPIgAoLy8PVVRUUNNWWVlpadv5+ijtflrkbKr64FXLX5833nhzrs1atILufaKhqpE/NIOem7GEvl673nE33vLfrAq3l5C/S75pjIF6MoBlaAjxwfrP7gCQvld5AoB1zTlZ3UNVw8IjmUn8EdfA1j7fygNkFaw5cR9zWHNizQdonVNJSfxYUVJuFiXlq4YeaWyrGFD+alVVI9b2EWs+AHtOVvrcDeAp5Lc4vjFQ02UQL4PWE50A4EZol4U/gHYp+FfQLk3nPOGtDNRExCWlw7Stxc5ZO2EA9py4jzmsObHmA7TMqf/weceLUuxOUVIqDZd213mk2O19Q4ssvSLH2j5izQdgz8lqn64ALgRwMYAzrXzhHNyIxmuxXoKG3vFL0C7/dgEwyfB4r/TGdl7y/WFuSL/M24O2LX7J8csQvPHGW9vb2nUb6MU3PqSbxszMXNq9c9xsmrlgOW3c6Lwfb863ze+9R5WV1gXqxTjykq/fqhdvBSehYaSvkQK0oLdsRQ9VXyCaVFk4nAi7b27Pa1l5gKyCNSfuYw5rTqz5ANmd+o2Mn+GVYmNESaluKFiv/HuANK3FAwut9HES1nwANpy2+P3nVfl9X1b5fQc3LVliic8xABZBu2c6BMC1hu8FK94gX7Q3UA1hWt/eMAXYOGGawpoT9zGHNSfWfIDGTv1Gv+b2BpQnRUlRDSu/zPWWKZc54cMCrPkAzjtV+Xy+Kr9PrfL7qKrU99mmzz6zdNqMserPudB6qXk7Aa2gPYGqRoS70mGqyq7brfBx+oTJBmtO3Mcc1pxY8wE0p4FS/JeiFH1ZlJQfM8UYgkp0QCB+kRM++X7P5mDNB3DOabvXe1JVqe+VKr+Pqvy++iq/bzz17WvZggbpQB0NZBZcFqEFaouqgjiBlfdQv39nfKbc3PYFTzl+PZ833nhreVuy/Gt68MV3yRvUeqPXPhCjyCvv07Iv1zjuxhtbbdNHH9HWu+6iKr+PqobcRpvnz2/0c6vyKYQj76Gug/1zUS2lLT1UVXbdnqndGhHustLHygNkFaw5cR9zWHNixWfg6Gm9REmZYxixq3oDypP9Rr+Ws/hLvmBlH6VhzQfIv9O2Ut89VX7fAf0S75Jvb7m+0XlipU8XANcDmA2tmMIINFPij1VaG6h6mNarskDJsPt/rfbhJ7E53Mcc1pyc9vEGpl2olwfMVDV6dvoS6jcy3nSxDcdweh81hTUfIH9OlSUlp1b5fHP1S7yHt/r9j1AodMwR21ngUwxt4NEvoFUn+rWhXYiGS8AdgtYEqt1hChzdJ3FL4T7msObklM/Asvj5nmA0pq34opAoKQlRikauDb1xGt9HzcOaD5Afpyq//7Iqv2+LHqY/bC8psW25zZOhXdotQ+O5nsbGzF98LaGlgZoIu29Oh2ki4pLs8jlaT+LWwH3MYc0p3z7aYCNliqFgfVIMKI9775/W3SknM7iPOXY6EdClqtQ3usrvq9PDdOH2wYO7N/ec9vocB6A/gPOhjeb1ALjG0DzQVoZhkrYOStq+8HlSIz1IlQXaMesBx2+S88Ybb9nbsi/X0Jh/vUcDgzESJYUGjY7Ro5MX0hdffeO4G2/stk0rV9KWoKQNPBpcSt/+619UuXFji55rRTZ1gVaZyDhF5hRopQDtrOVrOWY9JKE0ywAAIABJREFU1GREuFGVhcP6Zd5yu32sOkBWwpoT9zGHNSe7fbyB+DmegPKSKCl1eo+0zhNQXhIfiJ/tlFNr4T7m2OFU5fP9scrv26X3SrdsLym5NJ8+NyL7pV6C/cXxLaeiooIGBJWe2X6mh+kh/TLvk/nwOVpO4vbAfcxhzckuH/GB+NkeSXnWMNiIREl51ftg/BdOObUV7mOOlU7Ut+9xW/2+R/V5pVTl882tLCk5Nd8+fwIQhxag70Nb+PtFAFMB/B0daNpM/+Hzjq+oqEj/Em4QpdhTnkCsX0ko3lWNuAbmO0yBzn8SWwH3MYc1J6t9+o1+zS1KytPaYt4KiZJS75WU+MCy+PlOObUX7mOOVU7bSkt/VlXq+0zvlR7Y5i+510kfACiBVh2pw9K/bEqhIVAzbaA07ceHHvh7/cyHS2jN2Esm59OpM5/EVsF9zGHNySqf6/42uUAMKI83/p2NzWpLZaPOuo+sgjUfwBqn7T7ftVV+3z49TDdW+f2/cdInjRPLt7WLXIOS1q7bQLMXraCKyQvptkeiRwTsXx6dRU9MW0zzl3xB6ze07EY1b7zxZl1btXodPTFtMV3/YDzze/m3Z96mRZ+sctyNtw7S1q2j7x5/nPQgpS3hMFV+0/7BalblkxMLjFuOcVBSotz9Z1UW6taP/Q098/d7PhUDynuGGp+ZCeGiFHvdKyl/saO6ipUHyCpYc+I+5rDm1FaffiPjZ4hSNCJKSm3D72B0iRVF6zvLPrIL1nyAtjttKyk5f5u/5Gs9TNVtPh9ztde7Qpsiczy0+6YeaPV9O2RhBz1Mf9RLCr5MhC4AcM3fJp/sDcau9UrKC6IU29q09ypKypeiFI14A/HLS0rix7bXpzOdxHbBfcxhzam1PteG3jhNX0Ztv+F37WOPFLvKKSe74T7mtMVpW2nJ0MwKMX7fN9tKSlp8n90On1wcZ2hAw2Lfv7LqDfJBRUUF1Y51XZkO02TENSUdptkYEIhfJAZjZaIUXWwYop9u1aKkKGJAudU4gbw1dJaT2E64jzmsObXUp39oyinegPJ3MaDsNfxefe4JRD1OOeUL7mNOa5x2lpT8tKrUF0tf4t1W6ntx85AhJzjlY0anueSrykJK75lOpxCOqNeYi/6hKad4pdgNohR92SvFtjcJ13pRUpaJUjQklsX/A6CcIW2ko5/E+YD7mMOak5mPNzTnJFGKBcWgssfwO/S1V4rd0NLfHaud8g33MaelTttLSi6t8vs262Fau93nu9ZJn5ZwH4AoGqbNPA/gv6x68XyQCve4XA9ULUzjaNcl2wGj47/1BKKjPZKyVJSUQ00CdqdXir7mlWKDPdLUnH90dOSTOF9wH3NYc8rlUzIyfqIoKaNESdll+F1Z7w1EbwplKUaeDyen4D7mtMSpyl9yX5Xfd1AP0y+/Ly0910mfltIF2gLjPQB0g1Z6sN33EO0k1yjf6ldupMqNGywdUfbVmnU0Y8FyeuTl92jwIzMbT8sJxmj4k3Pp+RlL6N/LvnJ+9BtvvOW5rVu/kV5+86NGvxu3ym/QtLc/pQ0tLPnGG2/GtmnVKtry4AOZUbzfPfUUVa5fb/v7WpVPIWiXeXsDOE//OgbGQ7UpbVkPtfVQlwFl0d95A8rfvZLyiaFgN4mSQtrl4ujLXil2w+q16zvkX4X5hPuYw5pT2qckFO/qCSjDm9wiqfIElLsuHfZCXgc0srqPWIE1HyC30/aSksur/L4f9DCtriotsfyee2t8WssJAFYCmADgJGgDk26DFqrFVrxBvshPoDam38j4GZ5A7BYxqET1gUyZcL1pzEzSBjy1omnTe+aJUmyWKCkz9Nd9VattGn3eG1QmeCVlvBhUykUpGvJI0Yc8UjQgBpX7PAFluCjF7vRIsdu9weh/ewJKqVeK3SAGYt4B0rQ/i1K0L2shz9ovOms+AHtOlZWV5Akod4mSUmWY/rLDE1RsWQaxpU5OvXc2uI85TZ2opOTYrX7/I1V+32E9TD9uugh4Pn3aysnQpsgMNjz2a2iB+nsr3iBfOBGoTRGD8f/0BJVHREn5LMu0HMfb9Q/GySvFKgaWRXs4va8A9n7RWfMB2HHqH4yf5ZViN9wy9vWGcyqo7PEEoqO9oTknOenGyj5Kw33MMTp9e8v17iq/7+P0IuBVPp9MJSV5vUJq1T7qAmAGtACdB21AUocd5eu0g5H0ARpYNqtbv5HxMwaOnubqXzal0CNNLRpYFj/fM1r5PwOl6MUDgtHfewPxy0Up2tcrKdd4g4ooStHrvIGYT5SUm8Wg8j8eSRkmBqL3eKXoSDEYK/NKygOiFA2JUjTiCcTGiZLyD72w+ESvFH1FDCpTvZIS90jKm6KkvOUJxt4VJWVd48vT0VcGBeOOXoVg7RedNR/AWaf07Q1tlPsRRVEeLgnFmaimxtpx4z7mpJ2qSks8VX5ftR6mu6p8vj866WMFp6Px1JkdAPpY9eL5gtVAZYk3F64gUVI+aPzhGJvlGR3r8CexFbDmA+TX6Zq/TT5ZlKLXacVPojuyFT95dvoSujb0xmn5cmoJrB037mNO5fr1VOX3/yM98KjK71u4Y/Bgl2M+Fu+jrgD+D4Ce0NZD7XDwQDUn02uWoheLkjKjyaCqZR5JKbF7ikM2H1ZgzQew32nAA1N/rt1/V94RG1Z8STdVlGKzPJIyLF2e82jcR62F+zTPFr//vK3/+9d0kB6q8pc8QMhdhCcfWLmPLkZD73Q4gEkARln14naQa9qM08O9O1pb9uUaKn91IV1nKFR+m/wGvTrnY/pmnbXTj3hjo23YuJHeXfolPTltMQ2tmHXEffZb5Tfo0ckLae7iz/kCErxZ19aupW9nzKAtDz2YmQ6z9Y6/0OZFC51305sV2fQTAMv09jqAv0BbfWY/gJ9Z8Qb5gvdQzcnldF3w9TO1wVTR3YYP112iFA1dF3z9zHz7OAVrPoA1Th5p6un6/fhok1KAJErKIa+k/FsMxsq8gWkX5sPHalhz4j4aG/r3P35racl1VX5fvMrvSxou79KWBx+g7266iZlxOlbto/Qo398C6AfgHgAnQuuttnsViHzCA9UcMye92s29YkDZZPjATXok5dkBD0z9eb598g1rPkDbnQYE4hd5A4qUvdpXdLcoKVM8AaX0qkD81Hz42AlrTkezD4VCx2zz+a6q8vv+ZVirNN22VPl947f4/b076z5Kz0OdAeBpAOMBSNAGJp1lxRvkCx6o5rTUKRQKHSNKUX+T6T+HxUBs+kApenG+ffIFaz5AK4rRD593vBiMDdBHe3+XZUDRSlGKRgYGY33ac5+8I++jfHE0+lT5/Zdt8/meMhRlSLcftvr9T28vKbk8306twUqfPjiyOH4IDt8kbi08UM1pi5NXmnalV4q93aSHs9iKVUNY20es+QDNO/Ub/ZrbG1D+nygps7UBRI0CNOmVYnNFKXan+ED87Hz4OAVrTkeLz3a//4Iqn0+u8vsrm4Tonm2lvherBpdcSTn+eOvM++gsAN0BXAVtLdQLAAjQRv52GHigmtMep0HBeLEoKZO9knLQ8KH9tVeKDmlrqTnW9hFrPkBTJ+riLVMu80hKWAwoX2TphX7nDSrPeQJRT//h846334cNWHPqzD7bb7r+51v9JVKV37fKGKJb/b6aKr/v1a1+f/+WFGXorPuoC7S6vX8BMsudXQdtUNIZVrxBvuCBao4VTgPLoj3EgPK42HjB6CpRUkb1D01p1ZQr1vYRaz4AsHrtevJISokoKa+KkrKz6YAij6Qs9QYUaUAgflE+fFjcR6w5dTaf7YMHd99W6runyl/yYZXfV28IUrXK71e2+3zXri4paVUHrLPtIyMV0C7zzgcwVf96K7SVZzoMPFDNsdKpf2jKKdoC7cr3hg/4hBhQHm9paUPW9hErPt5g/P+KUiwoSsrkLL3QGk8wGhMl5ebmlg+0C1b2kRHWnDqDz66BA7tV+Xy3bfP75mtzRTMhemCr3/fGttIS/3avt80lJzvDPsrFMQAmo+H+aSU6WJgCPFBbgl1O3mB0qDeobGxa2nBgWfx8J3zaihM+A8tmdRODsQHeYEzWR+QeMTfUIymrxaBS7g3ELzd/RXth7ZgB7Dl1VJ8N/fsfv9Xvv77K75tR5felDCF6sMpfMq/K7791Z0mJJeUmO+o+agnpHuoOAB/rXy+FNn2GSXhhBzbbmwtX0Ign5zYKg8Bz82n+ki8cd2OlrVq9jl5/7zMaP3UR3Tl+dtZFDG6T36BHXn6PJs/9mJavXOu4M2+duG3YQJvnz6fvHn2Uqm7974aBRaV+2jI6SN8qClV+/bXznnloVmRT+h7qWADH698PA7+H2m6sOkBWki+nAWXR34lS7HWz0oas7SM7fLyB+DmeQOwWMRB9UZSUtVkCtF6UlK88kvKsJ6CUDhgVF+x2ag+s+QDsOXUEn6qSkv/c5vdNaDLNpb7K71u6zV9y7/bBg7vn28lJrPQpyvLYr6Gtj9ph4IFqTr6dPNLUIq3QupIyBEilGIjeUzIyfiJr+8gKH29w2q89kjLMK0VfyzYf1CspB/XF6R8TAzGvWWGFzriPrIY1J1Z9tvv9F1SV+iJb/b5NjUfoliyv8vnu31JamrdlHVndR23lOABPQpsmAwC9APxC/3ogtEFJzJSFagk8UM1xyql/MH6WNxiTxaCyJxMuAWXvK3M+JlGK9hWlaF8xoPzmur9NLnDCL01b9o9Hil8qBpX79B75rmyXcL0BZaEoxR4WR0f/1Dc06QS7neyENR+APSeWfHYMHuz69uWXqMpfsrLJXNEvqvwlD3x30w3ZOlS2w9I+AtrvczKA1dBKDQLAXDQUxL8RfD3UdsPaCQM47+QNzTnJG4yO0KfZNLcY+jZvUFmuFSZQJnqCyiNeSblblKLXecuUy66Voufa4deS/eMNxC/3BhRJ1FZnSWRxrxYlZY4YjJWJwfh/5sMpn7DmA7DnxILPD//93ydvLfWFtKktmRBdV+Xzjdk++IZfO+3Hwj4y0l6fE9A4UF8yfM0D1QJYO2EAtpw8gdgt46YsIn0KyAeipKwXJaXWJGgbN63Q+xpRir0vBpWpoqQ85pFi94uScrNXmnalp0y5oDVTS5ruH29ozkleSblGD/R/5wp/UVIUr6TcbcdcUJaOGcCeD8Cek9M+Vf6Su4z3Rr+rqKDtfv8lTjo1xel91BSrAvUv+vcvALhD/3ogeKC2G9ZOGIA9p2w+JSPjJ/YfpZw3MBjrIwajN3oCynC9MtC/REmZJ0rK503mvraoeSRlsyhFPxKl2OseSXlWDMQe9ErKX7T6t/FLB5ZFe6xavY7EQMzrCcTGiZKyLMdrrfEElX96peiQAUGlpxP7yElY8wHYc3LKZ1tpycAqv29Nw0hd3+KqkpLfsrZ/gM53zNKB+goAL4BF+td/hrYeKg/UdsLaCQOw59Qen1AodMzA0dNcYkD5zQBp2p+9UnSIdik29lSTXu/+1oavoR3WAjz2lBiM3tg/GM/7ghGd6ZjZBWtO+fbZUlr6uyq/b2k6SLf5S77e6vMNcMqnJbDmZMU91EocWRQ/3faDB2q7YO2EAdhzypePNzTnpP6jlPO8gfjl6V6vNxiTRSn6srHXe/2DcRKl6BJRika8AaX/wLJZjhc4OVqPWWtgzSlfPt+Xlp5b5fcrmXKAvv/f3tlH2VGU+f87eScxCREDRMKBQKKBFRQEQVxPwM2yskuQvFRVzyRA5CVBomLwhTdBYMEZs4KYJSiQCKsrnEBkWRYCu4KwHNbsoohxjT8Bk8lM151k3JX9Lft6OBx6/6hqu+/Nvbfm5d7uL/J8zqkzd/r2vf3pqrr9dFV3V+mB2JgLawekZ8sfgM+pFT7vhhsIf36d9Htwk48XySy/7fxBbByAIwEcFPqwBNQwZTtVougUG+m1VusHrNH9/kHyf7ZG/U2s9W2VSH/ORmpZvzEn7lHqLd8aBIpzGuzsPGigc+n8ShSdYo0500bqnNioSytaf6li9PqK0X9pjX60/4rLXVciS9L6yd1/cU8SK0Xz3Hy7y6yvq2uGNebW6jF11dWxUnUH43kr1+uhwuYzWjRcN3M3XAv5/XDPwT4LYANc93TU7AskoIYp0mmvUnMqWnf6H/5zNbfsDzX9l7smpLbaSH8jjvTlFWMiq9QHd69YMqvVzm/2Mtup1PS9Ss0ZMOb4itYLrdbaGrPaan1lbMy62JiN1qjvWaN/4B+j6LdG/+cIy4Yx3Wc71UfaWR5DKoc21aPelSsnVbT+gs1P3B3pbwx2djZtcLzZ63URsPmMhilwQTQ9w1wI4E4AXQBuyK0zCKDhbCYSUMO0y2lg0aLJVuvTrNZXWq3/2hr963oHvNjoHdbob9lIrYqj6NTe738/iaPo1DhSi60xn7HG3Bob/Vc+AP/b0A6i5qXYqL+tGH2XNerqWOsVFa0/VImiQ1nyZ6j8Wqm39Sp1cNy1dF4lio4biKIP9z7ySGK1Ps9Geq2N9J/GWt8WR/reitaPWaO3WWNespH+11EHo0j/qzXmJWv0torWj8WRvjfW+jar9Q2+V+G8SqTOysqMI1mtT+vbcFtitR7M7c/LcaQv33vO4lKea25HPYqNWe5PgNJ9fLiiVNOxstvpM1rYnNh8Rkta8TsArAfwWbjW6im55U/CdQvXRQJqmFY4JUDHQOfS+TZSK63R37RG/9RWz0bx20mGrVFbK0ZdW+lUf/ib5cv3ORkK+cRK7RcrdYw15syKUZ+0Rn/Vav1AbNSPrNH/MoRA8Zo1+ldW6yfjSG+qGHWtNebcWKkFA11LDqudt3E4+dO7cuWkPUrN3KvUnFipYypRdMqAMafHxixxwUeviSN9uTXmRneiYDbaSG+2Rj9aMfrvrdHPW2NesloPWKNftdXTYo00/Yc1us+XyQ+sUd+rGH1XbPRXYqOusJFaZY1RNor+oBJFx+2JosPrlUszWOt1cuqp46wxyhr1RC4vX7NaP1DpVH+YuGNIYT6t+q5YqQXW6H/OlfELFa0/VJZPq2BzYvNpBXMAbIcbZGK6/3uify8dc/gIQAbHLzT9/OdJ7+OPJX133JH0f/HqxJ53bt2Debz2M0nfLbckux94INn1j/9YmN+uH/4w6X38sWT3d7+b9N1+e9J/441J/+c/l8QXnD+kIBSvXpX0X3F50tfTnfR98xtJ3/r1Sd9X/yzpv+mmpP9L1yb9V16RxJetTeI1a5L4ogurBxBvdVqxPInP/3gSX7w6iS/9dNJ/+ReS/muvSfq6v5z03Xpr0nfnHcnu73w72b1lS9L76KNJ71M/SHZt25bs2r69/HpClHb96EdJ3+0bqupAfPHqZPfGu5JdP/1p6X5D2od/+Iek/4tfrPLvffDB0r1+l1MZQa9dnAbX7fsnuWUXwXX/Au4xnxfR5M5jaaGGGYqTVep9VuuLbaTvsfln2qrT3tioh6zWV8ZKLRjpvIhF5NFA59L5Fa0XWmNWW2NujCN9r430D63RlRZ1k+72LYht1ujvuy5r821rzO2xMetspK+xkV4ba32Rv6Z8ZqzUgv4oOqGi1LtjpQ7ZqVTD8XzZ6hGbD9DcqRIp41rr+bJT38s/VlKkT4j+KHqnNebunO+/WaMuK8unXbA5sfmMhklwj/CcBHdX7zS/bCmAZwCMBTDPr9NwSjkJqGFqnfaes/jAAa3PrhjTbY16yhr93w0Cxz9VtP661VrvVWpOo+8frU8ZxF1L59lO9ZHYmAt3b7wrsVp/1hqzOjZmeSVSZ1mtT4uV+sCAMUdVoujQvq6uQh8nY8ijPGw+wNCcYqXmxsass0a/kqvb/RWjru1V6uDQ51vtU8u/nHXWVPc7zPWeGLOuFfXtzVpmRcLmMxoOgbvhKP8c7Bq4bt67c8uObfYlElCbs0OpCb1PP53ERl1qjb7PGt3boOXVZyO92UZ6rVXqgy+fccbEdjmx5RGbD8DnxOYDDM/px6tWjc+1WtNrra9box+uaL2o9rp6u32SU08dFxvzKZvdE/BGHOl7R3JTXSt8ioLNic2nnRwIIDhL/O9aQP3tHZ9KzbVKvW9Aqd+PO9VHbaSWVYz5eGzMp/xdtTdUtP56HOlN7qYXtbXu83rG7KwTPP8nNuoZ/0jFklafqYdgq8RsPgCfE5sPMHKnuGvpPN9qzd/gZq3WN4wmoA3Vx0ZqmTX6V7mu6KcGlHr/SLc7Wp8iYXNi8ykdpoBqFy8+oHfr1sRqrStanx9H+tM20lfZSH/ZT+r7Lav1AxWtH/N3fL5gjX65YvQe6+7UbMtNL/EllyQVo/+yEuk17fjhDhe2SszmA/A5sfkArXFydwhXX2utGP34gNZnt9rHKvVBf83db8v8v0qkzhq5/eh8yoDNic2ndKgCqutSbUUQfMW6Z89+YY1+zmr9pNX6ryvGfNe6R1a+Gkf6uqrrfsZ8LL3ut0epo/PX/dgqjfiEYXNi8wFa/JiKa7X+uc0NnlAxeo+N9JeHev9AI5+4a+k8a/TDud/33kqkLmiV+3B9yoTNic2ndFgCaqzUB6x/BMIa9TfW6Psqkb7Tan1LbMz1bqQT9QkbqXPiSC22neoj1piT+415z54oOtwuXnxAu9zYKo34hGFzYvMB2ucUR6orNuqZmjuEn7Ba6+H47FFqpjV6Q/WlFnP93nPOmdIO75APA2xObD6lwxJQrVE/sUYnfd+4ncInD1ulEZ8wbE5sPkD7nWKl5lpjbq4akUrrwdjor/R1LT2ikU+s1H5W6y9aN3BHYo1+vWL0XaGhAlvNW7HMhgubT+kwBFQ/PFhijf71zl/+snSfWtgqjfiEYXNi8wGKc9qh1AR3WUX/fa61+YY1+gcVY6Ifr1o1HgB2/upXSSVSF9jqZ50fHTDmqCI8a3krl9lQYfMpnbIDau/KlZOs1rE1OqlofT5jAbE5iU8YNic2H6Acp4HOpfP9xA//Pxc0X6kY8914zZr8fRAvWK1PK9ovj5RZGDaf0ik7oNpIX+V/QD8FOAuIzUl8wrA5sfkA5TvFxizf91qr7ou1XlGmV0rZ+VMPNic2n9IpM6D2KnWwdVONJXEUnQpwFhCbk/iEYXNi8wF4nAY6l86vGL1+9z33UPiksORPHjYnNp/SKTOgWqO/5c9KH06XMRYQm5P4hGFzYvMB+JzEJwybE5tP6ZQVUCtRdJy/MeH1WKm56XLGAmJzEp8wbE5sPgCfk/iEYXNi8ymdsgJqNuKJuTW/nLGA2JzEJwybE5sPwOckPmHYnNh8SqeMgBpHanF6d1/tRM2MBcTmJD5h2JzYfAA+J/EJw+bE5lM6RQfUH69aNd7PhZnEkf507fuMBcTmJD5h2JzYfAA+J/EJw+bE5lM6RQdUN36uTqzRL9eb8omxgNicxCcMmxObD8DnJD5h2JzYfFrFTABjcv+PA3AkgOBQXUUG1L6urhmx0f/ur52eWW8dxgJicxKfMGxObD4An5P4hGFzYvMZLRPgAueLAN7ul00G8CyADQB2AIiafUGRATXW+jZrdGIj/XSjdRgLiM1JfMKwObH5AHxO4hOGzYnNZ7QsBLANwCCAGX5ZF4Ab/Osp/r1p+37UUVRAdQNl69et0W/sUeroRusxFhCbk/iEYXNi8wH4nMQnDJsTm08rGA/XQk0DajeAU/zrDgBPApjV6MNFBVRr1Fbf1XtHs/UYC4jNSXzCsDmx+QB8TuIThs2JzacVTEJ1QH0EwIn+dQeAzQCOAIDu7u7renp6ktq0c+fOtqbeJ55I0rlOd/3sZ23fniRJkiRJKiYVHvHaTG1AvQiuK7jee/vQ7hZqotTY2Ogd/jGZy0PrMxYQm5P4hGFzYvMB+JzEJwybE5tPK6gNmksBPANgLIB5AHYC2K/Rh9sdUK1Rn/BdvbvTeQ+bwVhAbE7iE4bNic0H4HMSnzBsTmw+rWASgO3IAmoHgLsBJD4dm65YeJfvL36R2JXnJdbopPehvyq9a0KSJEmSJLU2FR7xSuJAAG8LrdTOFmpszDo/iMO2oX6GsYDYnMQnDJsTmw/A5yQ+Ydic2HxKp10BdaBryWHW6Nes0Uklio4b6ucYC4jNSXzCsDmx+QB8TuIThs2Jzad02hVQrdFb/CAO3xnO5xgLiM1JfMKwObH5AHxO4hOGzYnNp1CKuoba+/RT7jGZ5V3JrhdeKL2PX5IkSZIktSeVHdeoaEcL1Rr1E2t0Ehtz/XA/y1hAbE7iE4bNic0H4HMSnzBsTmw+pdPqgGq1Ps/fiFQZWLRo8nA/z1hAbE7iE4bNic0H4HMSnzBsTmw+pdPKgDqwaNFka3TFGp1Yrc8byXcwFhCbk/iEYXNi8wH4nMQnDJsTm0/ptDKgxpG+zrVO1U9G+h2MBcTmJD5h2JzYfAA+J/EJw+bE5lMo7bwpadcLLyR2eZcbxOHpp0q/UC5JkiRJktqfyo5rVLSqhWqN+ba/drplNN/DWEBsTuIThs2JzQfgcxKfMGxObD6l04qAWomi43wwfW2ga8lho/kuxgJicxKfMGxObD4An5P4hGFzYvMpnVYEVGv0Nv+YzLrRfhdjAbE5iU8YNic2H4DPSXzCsDmx+ZTOaAOqNUb51ukrv1m+fNpofRgLiM1JfMKwObH5AHxO4hOGzYnNp1BaflPSSy8l8epViTU62X3vvaVfHJckSZIkScWmsuMaFaNpoVa0/oLr6tU7EqXGtsKHsYDYnMQnDJsTmw/A5yQ+Ydic2HxKZ6QBdY9SM2Oj/90ancRKLWiVD2MBsTmJTxg2JzYfgM9JfMKwObH5lM5IA6o1+pt+EIetrfRhLCA2J/EJw+bE5gPwOYlPGDYnNp/SGUlA3aPU0dboN6zRr8dKzW2lD2MBsTmJTxg2JzYfgM9JfMKwObH5lM5IAqqN9NPW6CTW+rZW+zAWEJuT+IRhc2LzAficxCcMmxObT6G04i7f3q2PurlOzz0n2fnzn5d+h5kkSZKEehuUAAATeUlEQVQkSSovlR3XqBhOCzVRaqw1+mV/7fSydvgwFhCbk/iEYXNi8wH4nMQnDJsTm0/pDCegVoz6pAumZvePV60a3w4fxgJicxKfMGxObD4An5P4hGFzYvMpnaEG1N8sXz7NGv2KNToZ0PrsdvkwFhCbk/iEYXNi8wH4nMQnDJsTm0/pDDWgWq1vsUYnNtJPt9OHsYDYnMQnDJsTmw/A5yQ+Ydic2HxKZygBdaBryWHW6Nes0W/sUerodvowFhCbk/iEYXNi8wH4nMQnDJsTm0/pDCWgxkY9ZI1O4khvarcPYwGxOYlPGDYnNh+Az0l8wrA5sfkUykgem+l94gn3mMyK5cmun/2s9Fu0JUmSJEkSTyo7rlHRrIWaAB2x0Tv8YzJXF+HDWEBsTuIThs2JzQfgcxKfMGxObD6l0yygViJ1gZ/rtPLyGWdMLMKHsYDYnMQnDJsTmw/A5yQ+Ydic2HxKp1FAHVi0aLI1+tfu2qnqKsqHsYDYnMQnDJsTmw/A5yQ+Ydic2HxKp1FAtcbc6Lt6f1KkD2MBsTmJTxg2JzYfgM9JfMKwObH5lE69gNofRe+0Rv+vHxXp5CJ9GAuIzUl8wrA5sfkAfE7iE4bNic2ndOoF1DjS9/prp/cV7cNYQGxO4hOGzYnNB+BzEp8wbE5sPqVTG1ArUXScD6b/2x9F7yzah7GA2JzEJwybE5sPwOckPmHYnNh8Sqc2oFqjt1mjE6v1TWX4MBYQm5P4hGFzYvMB+JzEJwybE5tPoYQGduh98MHEGp3E53882fnLX5b+wLAkSZIkSeJOZcc1KtIW6stnnDHRGl2xRieVSF1Qlg9jAbE5iU8YNic2H4DPSXzCsDmx+ZROGlBtpK+yRiex0TsSoKMsH8YCYnMSnzBsTmw+AJ+T+IRhc2LzKZ2enp5kj1IzrdH/ZY1OYqUWlOnDWEBsTuIThs2JzQfgcxKfMGxObD6l09PTk8TGbHStU/VQ2T6MBcTmJD5h2JzYfAA+J/EJw+bE5lM6PT09iTX6DWv0a7FSc8v2YSwgNifxCcPmxOYD8DmJTxg2Jzaf0vEBNbFa31K2C8BZQGxO4hOGzYnNB+BzEp8wbE5sPqVSMeZjPqC+8pvly6eV7QNwFhCbk/iEYXNi8wH4nMQnDJsTm0+pWKM39PT0JBWjPlm2SwpjAbE5iU8YNic2H4DPSXzCsDmx+bSLcQDeDeCw0IrN5kMtA8YCYnMSnzBsTmw+AJ+T+IRhc2LzaQczAGwHsAHAQwA2ARjbaGUJqGHYnMQnDJsTmw/A5yQ+Ydic2HzawUUAbvCvx8AF13c1WlkCahg2J/EJw+bE5gPwOYlPGDYnNp92MA3A/v71PACvwrVa6yIBNQybk/iEYXNi8wH4nMQnDJsTm0+76ABwIYAEwB+lC0OD40uSJEmSJEnDSaVFuYKYCOBZAPcDOCC0srRQw7A5iU8YNic2H4DPSXzCsDmx+bSDpQAeBjDZp+nNVpaAGobNSXzCsDmx+QB8TuIThs2JzacdXAHX1ZumQcg11FHB5iQ+Ydic2HwAPifxCcPmxOZTOhJQw7A5iU8YNic2H4DPSXzCsDmx+ZSOBNQwbE7iE4bNic0H4HMSnzBsTmw+pSMBNQybk/iEYXNi8wH4nMQnDJsTm0/pSEANw+YkPmHYnNh8AD4n8QnD5sTmUzoSUMOwOYlPGDYnNh+Az0l8wrA5sfmUjgTUMGxO4hOGzYnNB+BzEp8wbE5sPqUjATUMm5P4hGFzYvMB+JzEJwybE5tP6UhADcPmJD5h2JzYfAA+J/EJw+bE5lM6ElDDsDmJTxg2JzYfgM9JfMKwObH5lI4E1DBsTuIThs2JzQfgcxKfMGxObD6lIwE1DJuT+IRhc2LzAficxCcMmxObT+lIQA3D5iQ+Ydic2HwAPifxCcPmxOZTOhJQw7A5iU8YNic2H4DPSXzCsDmx+ZSOBNQwbE7iE4bNic0H4HMSnzBsTmw+pSMBNQybk/iEYXNi8wH4nMQnDJsTm087GQfg7aGVJKCGYXMSnzBsTmw+AJ+T+IRhc2LzaRfTAVwI4OrQihJQw7A5iU8YNic2H4DPSXzCsDmx+bSDiQDuBJAAWBNaWQJqGDYn8QnD5sTmA/A5iU8YNic2n3ayFMBnQytJQA3D5iQ+Ydic2HwAPifxCcPmxObTTpajJqB2d3df19PTk+TTzTffnNQukyRJkiRJkkJp06ZNb92AWo+eHq4WKpsPwOckPmHYnNh8AD4n8QnD5sTm004koLYINifxCcPmxOYD8DmJTxg2JzafdrIMElBbApuT+IRhc2LzAficxCcMmxObT+mwZQibD8DnJD5h2JzYfAA+J/EJw+bE5lM63d3d15XtkIfNB+BzEp8wbE5sPgCfk/iEYXNi8xEEQRAEQRAEQRAEoZZZAOYDmFq2CJzLfmVLeGYCeDeA8SV71Bub+R1wZbZ/8TqYDeBIOK+UcXB5dVgJPhPh8uLgOu9NBXBAgS4zAYzJ/d8oX4qoW1Ph8uWg3LLJcMOTpimlWR62imm57U6reW8q9v3dHwFgLqrzsxXMDLxXW7dDLocC6BihS70yyjPO+9R7fxyAQ2r+fxfaW4ZCAA1gEEA33JCF7y/R5WDvcHKJDinnA3gewNfg8ufQkjzqjc18PIBXAXwJLr9OKMilA8BVAJ5Dli+zAcwAsB3ABgAPAdgEYGxBTjMA7ATQA2AHqofcHAPgWQCfL8BjAtyB70VkJz+N8qWIunUMXN3ohsufC/zyjQCeBLDZO+2P5nnYKib5bdzvt90DVz5j4U54tgA4xa/bAZdnjwO43X/uHS1wmAzgg3D7WK9+/jFceXzNb3P/IbgsgPstzhiBT6Myyvs+67e/A0BU8/6VuX2Z6V+v93+vGIGPMEqmwBVoegBYCDcWcBmMAfAUig0QjZgEIEZ2Bn8N3KNIRVNvbOZJcAft9Cz7cAAnFeQzBe7Hmp693wR3QLkIwA1+2Ri4IPKugpzu8w6AOwCdgay1sBZDHNe6BSwEsA3ugJweXOvlyzEopm49jixfZsIdsKfCBdHalmCzPGwVs+ECUu33Hg7gQVT/7o+Cq+Npa7Abrcmjy+DK52Hs29JMA/6B/v9Pwz162MzlYP99+TIfDvXKKF82XcjqzxS/nbRlvwAuz9J96UJ20p2eIE0YgZMwStIK1AF3dhN8frVNXAv3Q14P4MSSHPJcDtcSuxauIs8q0SU/NvMkuKD2PNwP6m64H1uRzIBrZSUA5sD9yNOu53kY+Rn7SNgIF6gSuEA1xy8/FcCfAliM4ur0eLiDb7rvjfKliLo1E1l38tlweTTZby/x6Qq41k2jPGwl83LbTQD8Sc379yH73U9Adlya5t1adUw43H9fbUCdjeqW64cAXN/EJe39+D24shxJfa9XRnmvblS32p+Eqyuz4OrZsQCeQdZCTQBcB9cwuWoEPkKLmANXmI+g+rpKUSxA1jLO/7DKogMuUG2A62YZBHBaiT75ka/SM+lF/vV6DGGqvhYzCy5YDQI43S/rgOuaTgD8UYEuG+HO4vcDYOAOirPh8mg8XPkVFVDT3oP8wbU2X4qsW1PhfleDcAFtFoB1cF2saRfh+1A/D1vdZf8hAJfCBahjUd3aAlw3cO3v/jS4fLsSrbuOOh/1A+pRcMFpTO7/9U1croQrv7FwZV57TXio1JZRnkeQ5UkHXB7NgwvkR8Jdl0/35Ti4E7YVAL7iPztphE7CKEgrSu0ZY1F0wBV+guqz5+NL8gHcCcaLyLo2F8L9gMoiH1CnwLUiJvv/D0f9A0Q7mAl3Jp3yUbh8mQj3I78fxd4ABLi6c4x/nbYQ18DVoR3I6tOFBbjUBtR6+VJU3ZoNt9+XIev6G4vqLtfPw/V+1MvDVvcw5LsfO+BaUUfkltUG1CvhjgdHtdhjHur/XqbAtTTTculE9purdUm7X/Mt7hjDz7N6ZZTnIrj6AWR16yT/mbRHIYHLuzvhevjg920b9r2RUWgzaWvnJLiKNA3lnNXMgOtWOQCuW2MRWn8NZzjMhDvbm+3/vxjldYUD1QF1HNwPP7157FIU5zYb1deL1vptL4W7ljMZ2V2kRdGNLFieBHegmQp3V+T+3vEmFFOvawNqvXwpqm7dAWAVXH2ZAuBtcK3E7XCBdSJcPXof6udhq0/QrkF2c1hajxq1UOf49w+CCzTTse9dtyOltoV6MNxJaXop5WRkl7+WNXE5EC5gHQYXTEcS+OuVEQC8129rKbIu3Xlwx+rp3uXtAD6M7Ca4a5F1887AyAK8MEoOwb5nWkXcwNGMO+B+5GVzLqrPPmc3X72t1I7NfDwyt51ofMt9q0nv8s1v+2C4a3H5OjTSmzRGwjtQ3RKt7dlYhuLq9CS4g3W6743ypd11axxcyzi/7Wfggvp3csuuhztYh/KwFaTd8Ok2Fte8vxFZQF1Y456gdZeB5qK6a/c8uHwAqn9XG/06IZe0zIfb5duojPI3iqaXB9L3j62zL095z7QLn+U4Lgj7kLYqymwtNyI9Wy6DaeDrTmplK6YIyqxb6fOgtRSRhzPAdW1vPoAv5P6fjOwmpDJIW8f539eByFqvIaYPY11BEARBaBnpXeosjIO7c1gQBEEQBEEQBEEQBEEQBEEQBEEQBEEQBEEQBEEQBIGXiShu1hpmJoDrUZPhMhbFjzctCIIwYs5GNuB5yjK0bpCF9AH2Iobw64AbKzhBuSNUtZqT4Ua3GQ4z4MpwEdzziZtRzDCTQ2Uo9eJQuLJ8byFGgiAIo2QZslFSjs4ta2VA3YFiAup4v60tcJNw/65wH9xoOMMZzGEtsrF/16P+1GNlktaL0Mg8PXDu0uMgCAI9+YCatmLyAXUh3AE9HUVlLdwE5WMAXAJ3sL4dbpzZT8CNH/wc3FCQ70DWEnkcbvi6dGaLNDikE9W/Cje+8Fi4MVK3APgMqofkS5kGd6BNvS+B6+b9GrIBwP+4yWe2A1jp96Heto6HG97tebixZB+CG3N1tHmB3Hcnfv108PtLvP9X/efXw3V3pjPOJHAj80zP7edgbj/yzPTvpcEqH1Dr5V0arBrtd54OuHFk089/CVm37Fyfl4nPp/f45R9BNnzgRrgZT2oD6iwA9/h1tiIbpOAD4JjrWBAEIUgaPFfDHbhOgZt9Ig2ota3VjchaDBv8Z65CNtnz43DTwqXjgqYBNYELWn/nX38UwFn+9VoAF+Re5+fA3Aw3aH1KOr9jAje36jXIuniVd92MfbsJvw4XqM6EGxA8PUjXbmsusmC1GtmY1Se0IC/m+NcbACyBCzLPwLUi08/fABek0s98GC7IDfpyuc7vx+lwk1nX6xI9EdVBKA2oYxvk3Vq4sZUb7XeedIzaCwF8zL++CNnk1INwJ0yD/v/5fp0/h5uGLp2oezKygJrWkef8d97v15uZ+14ZY1YQBHqWwQ20Pg0uQGyDO3g1Cqj51k76GsiC47FwB+78wXIH3EEbcC21tPW0EdkBPW2JPYPsIPzROr7pATb9vjFwA30/jObX5ebABfSVcC2v2oCabus4/396zTIfnEabF13IAmo6f+gg3Pir+aA3HtWtt/XeuQNZ63ILgM/BXSNNJ5tO6UIWkPKeBzTJu/c32e88Z/jlz8NNeH6ez4/31Hx+LtzJ2WTv+ClkJxebUR1Q0/J+zudxOrD7ycjKdB04x8YWBEH4LWmQmAY3pVTtDCfp+5NRfQDOB5EOZF3HJ6A6INReQ52CrMWxEVn36Aq42VSWAHgXGs9eUtud2QEXXB72393outx6ZF2cn8C+ATXdVtoCO9n/fwz2DagjzYs0oKaTTJ/r82UyqoNzbXdo/rtnwHW53oGsrGpPPLrg8vXQms8f1CTvTm+y33kmwJXRzchasdcjOxFJZ3yaAdedngbLLXCB9Tns20JN19lUky+H5fJCAqogCPTUtrrSYJMuW+7/X4EsIKTXWocaULfDtYIXIOtuXZTb1ulw3ZZPwXVpzkP9gzmQBbJB/32dyLpaG93okjq8CNdKS1tK762zrVnIWkudyKa2OqEFeZFeD1wDF2zWw03SvR+aB9QNcD0HR8IFpm1wwSZC1v2aJw1utV2+45rkXbP9znOx//xJPv/S66WH+Nf3++Vpmaf5tATuRrG0Sz4fUNM7ku/2+/VJ//8RaH6SJAiCQMXZqA6ok+AOqulEw7OQXQN9Eq51mQ8im+GCSNrN+V64IJJ2veavoaZpHdzBfSKAW3LLt8G1ompbjbUciuo5HDfDHXibdflenFv/fv83arCtD8Bdt3wV7tprGlhGmxcAcE7OY9BvC6h+tKV2P9L5UXfAdanm83Ir3IToedLguKjOdzfKu2b7nWc2qstzENl8m6rO8pnIbkgahCvjnXBd//l9PAnVcyyv8MvTruS05SwIgvCmZixaM1dio7lYp2Jk86QeOMzPTUN4oIBDAFwG16Iaj6wVmN7404q8mAznPZw5RCfk1h8Ll4/N5rXtQfbYTD1q8y6037VMhzvhqu2Gneq/O7/dsRhaOaXr5cvoOrj9qL1OLAiCIJCTttDzrcANeHNNQg64lmECdxPRUGDc78O9x0klOgiCIAijYCzc9b4Tse9zmG8mZmDf7uBmsO33RLjWriAIgiAIgiAIgiAIgiAIgiAIgiAIgiAIgiAIgiAMn/8Doh+Q3pvjM9kAAAAASUVORK5CYII=", "text/plain": [ "\n", "\n", @@ -792,15 +792,15 @@ "https://altair-viz.github.io/user_guide/troubleshooting.html\n" ] }, - "execution_count": 22, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "lines = alt.Chart(nargs_scaling_results).mark_line().encode(\n", - " alt.X(\"npoints:Q\", scale=alt.Scale(base=2, type=\"log\"), axis=alt.Axis(title=\"Number of arguments\")),\n", - " alt.Y('max_job_time:Q', scale=alt.Scale(base=2, type=\"log\"), axis=alt.Axis(title='Execution Time (s)')),\n", + " alt.X(\"npoints:Q\", scale=alt.Scale(base=2, type=\"log\"), axis=alt.Axis(title=\"Number of arguments (log scale)\")),\n", + " alt.Y('max_job_time:Q', scale=alt.Scale(base=10, type=\"log\"), axis=alt.Axis(title='Execution Time (s)')),\n", " color=alt.Color(\"system\", legend=alt.Legend(orient=\"top-left\", title=\"System\"))\n", ")\n", "\n", @@ -812,7 +812,7 @@ }, { "cell_type": "code", - "execution_count": 28, + "execution_count": 7, "metadata": {}, "outputs": [ { @@ -996,7 +996,7 @@ }, "field": "max_job_time", "scale": { - "base": 2, + "base": 10, "type": "log" }, "type": "quantitative" @@ -1032,7 +1032,7 @@ } ] }, - "image/png": "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", + "image/png": "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", "text/plain": [ "\n", "\n", @@ -1041,7 +1041,7 @@ "https://altair-viz.github.io/user_guide/troubleshooting.html\n" ] }, - "execution_count": 28, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" } @@ -1049,7 +1049,7 @@ "source": [ "lines = alt.Chart(syssize_scaling_results).mark_line().encode(\n", " alt.X(\"side_len:Q\", axis=alt.Axis(title='Side Length')),\n", - " alt.Y('max_job_time:Q', scale=alt.Scale(base=2, type=\"log\"), axis=alt.Axis(title='Execution Time (s)')),\n", + " alt.Y('max_job_time:Q', scale=alt.Scale(base=10, type=\"log\"), axis=alt.Axis(title='Execution Time (s)')),\n", " color=alt.Color(\"system\", legend=alt.Legend(orient=\"bottom-right\", title=\"System\"))\n", ")\n", "\n", diff --git a/papers/adam_richie-halford/figures/compare_results_mri.ipynb b/papers/adam_richie-halford/figures/compare_results_mri.ipynb index e23757eb6b..1fba401745 100644 --- a/papers/adam_richie-halford/figures/compare_results_mri.ipynb +++ b/papers/adam_richie-halford/figures/compare_results_mri.ipynb @@ -430,7 +430,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.4" + "version": "3.5.3" } }, "nbformat": 4, diff --git a/papers/adam_richie-halford/figures/nargsscaling.png b/papers/adam_richie-halford/figures/nargsscaling.png index 288c120f6d..9a41dd56ec 100644 Binary files a/papers/adam_richie-halford/figures/nargsscaling.png and b/papers/adam_richie-halford/figures/nargsscaling.png differ diff --git a/papers/adam_richie-halford/figures/syssizescaling.png b/papers/adam_richie-halford/figures/syssizescaling.png index d5a52a490f..dfc202e207 100644 Binary files a/papers/adam_richie-halford/figures/syssizescaling.png and b/papers/adam_richie-halford/figures/syssizescaling.png differ diff --git a/papers/adam_richie-halford/mybib.bib b/papers/adam_richie-halford/mybib.bib index 9b0a6a717a..ebae1f205d 100644 --- a/papers/adam_richie-halford/mybib.bib +++ b/papers/adam_richie-halford/mybib.bib @@ -182,6 +182,12 @@ @misc{cloudknot-repo year = {2018} } +@misc{cloudknot-examples, + author = {Richie-Halford, Adam and Rokem, Ariel}, + title = {Cloudknot Examples}, + howpublished = {\url{https://github.com/richford/cloudknot/tree/master/examples}}, + year = {2018} +} @misc{cloudknot-docs, author = {Richie-Halford, Adam and Rokem, Ariel}, @@ -342,3 +348,19 @@ @misc{anwar_o_nunez_elizalde_2017_1034342 doi = {10.5281/zenodo.1034342}, url = {https://doi.org/10.5281/zenodo.1034342} } + +@article{Boettiger14, + author = {Carl Boettiger}, + title = {An introduction to Docker for reproducible research, with examples + from the {R} environment}, + journal = {CoRR}, + volume = {abs/1410.0846}, + year = {2014}, + url = {http://arxiv.org/abs/1410.0846}, + archivePrefix = {arXiv}, + eprint = {1410.0846}, + timestamp = {Wed, 07 Jun 2017 14:42:34 +0200}, + biburl = {https://dblp.org/rec/bib/journals/corr/Boettiger14}, + bibsource = {dblp computer science bibliography, https://dblp.org} +} +