{ "cells": [ { "cell_type": "markdown", "id": "5fbc2d16-59f9-4be3-b93e-1a5440c7efd0", "metadata": {}, "source": [ "# Tutorial 5 - DIY KANs" ] }, { "cell_type": "markdown", "id": "1afe6a1e-3ab4-4f3f-ad47-f6cd66419504", "metadata": {}, "source": [ "In all previous tutorials we've been using the `KAN` class from `jaxkan.models.KAN` in order to define a network. However, if one requires more fine-grained access over the network's constituents, they may define their own custom KAN class by using one (or more) KAN Layers from the `jaxkan.layers` module." ] }, { "cell_type": "code", "execution_count": 1, "id": "0a2ef2a6-f681-427f-8252-ade2111ce0e6", "metadata": {}, "outputs": [], "source": [ "from jaxkan.layers.RBF import RBFLayer\n", "from jaxkan.layers.Sine import SineLayer\n", "\n", "import jax\n", "import jax.numpy as jnp\n", "\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import mean_squared_error\n", "\n", "from flax import nnx\n", "import optax\n", "\n", "from typing import List\n", "\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "\n", "import os\n", "os.environ[\"TF_CPP_MIN_LOG_LEVEL\"] = \"2\"" ] }, { "cell_type": "code", "execution_count": null, "id": "5e74e6e0-a74e-4ddc-95fe-bc3195072571", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "60e70a5d-0340-4bdc-af4e-150d0098a87e", "metadata": {}, "source": [ "## Custom KAN Class" ] }, { "cell_type": "markdown", "id": "d239ed7c-6245-4f43-8a92-c4767f0dd654", "metadata": {}, "source": [ "In general, the custom KAN class should inherit from `nnx.Module` and contain an `__init__` and `__call__` method. Apart from these, there are no other pre-requisites. For example, if one does not intend to perform grid updates, there is no need to define a `update_grids` method.\n", "\n", "As an example, we will create a custom KAN class to perform the same function fitting task as in the first tutorial, this time using a mix of RBF Layers, Sine Layers and additional transformations in between." ] }, { "cell_type": "code", "execution_count": 2, "id": "f821497a-cd3d-458c-8a13-9dab2b85964b", "metadata": {}, "outputs": [], "source": [ "class CustomKAN(nnx.Module):\n", " def __init__(self, rbf_layers, sine_layers, add_bias = True, seed = 42):\n", "\n", " self.RLayers = nnx.List([\n", " RBFLayer(\n", " n_in = rbf_layers[i],\n", " n_out = rbf_layers[i + 1],\n", " D = 5,\n", " kernel = {\"type\":\"gaussian\"},\n", " add_bias = add_bias,\n", " seed = seed\n", " )\n", " for i in range(len(rbf_layers)-1)\n", " ])\n", "\n", " self.SLayers = nnx.List([\n", " SineLayer(\n", " n_in = sine_layers[i],\n", " n_out = sine_layers[i + 1],\n", " D = 8,\n", " add_bias = add_bias,\n", " seed = seed\n", " )\n", " for i in range(len(sine_layers)-1)\n", " ])\n", "\n", " def __call__(self, x):\n", " for layer in self.RLayers:\n", " x = layer(x)\n", "\n", " # Apply a GELU transformation\n", " x = nnx.gelu(x, approximate=True)\n", "\n", " for layer in self.SLayers:\n", " x = layer(x)\n", "\n", " return x" ] }, { "cell_type": "code", "execution_count": null, "id": "b5f07c8e-cb8a-4bc0-a2db-1b1bce2c57aa", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "ef6157a1-d691-4ef1-90aa-115774b610b0", "metadata": {}, "source": [ "## Data Generation & Preprocessing" ] }, { "cell_type": "markdown", "id": "166ad90d-430e-45ca-86a8-e6e4dbc1c943", "metadata": {}, "source": [ "Again, we create data for the function $f(x, y) = x^2 + 2\\exp(y)$." ] }, { "cell_type": "code", "execution_count": 3, "id": "b986e75a-6d4a-402f-bea7-36d13f4a7866", "metadata": {}, "outputs": [], "source": [ "def f(x,y):\n", " return x**2 + 2*jnp.exp(y)\n", "\n", "def generate_data(minval=-1, maxval=1, num_samples=1000, seed=42):\n", " key = jax.random.PRNGKey(seed)\n", " x_key, y_key = jax.random.split(key)\n", "\n", " x1 = jax.random.uniform(x_key, shape=(num_samples,), minval=minval, maxval=maxval)\n", " x2 = jax.random.uniform(y_key, shape=(num_samples,), minval=minval, maxval=maxval)\n", "\n", " y = f(x1, x2).reshape(-1, 1)\n", " X = jnp.stack([x1, x2], axis=1)\n", " \n", " return X, y" ] }, { "cell_type": "code", "execution_count": 4, "id": "db8bf4bc-49fe-4014-94aa-4e5dbf632369", "metadata": {}, "outputs": [], "source": [ "seed = 42\n", "\n", "X, y = generate_data(minval=-1, maxval=1, num_samples=1000, seed=seed)" ] }, { "cell_type": "code", "execution_count": 5, "id": "ad91a42a-ffe8-4b8c-b1a7-cc9b60dfe5c0", "metadata": {}, "outputs": [], "source": [ "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=seed)" ] }, { "cell_type": "code", "execution_count": null, "id": "b6300ac1-525d-4990-b6d0-6c3809bd5c12", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "f681d135-c885-4e59-87d7-1ea16a157c50", "metadata": {}, "source": [ "## Training" ] }, { "cell_type": "markdown", "id": "529ae82e-b422-4ec7-b88a-f3be7cd99367", "metadata": {}, "source": [ "Let's now define an instance of the CustomKAN class, with 2 RBF Layers and 2 Sine Layers." ] }, { "cell_type": "code", "execution_count": 6, "id": "471d7532-5434-498a-9b99-40896001776d", "metadata": {}, "outputs": [], "source": [ "n_hidden = 5\n", "rbf_layers = [X_train.shape[1], n_hidden, n_hidden]\n", "sine_layers = [n_hidden, n_hidden, y_train.shape[1]]\n", "\n", "# Sanity check for dimensions matching\n", "assert rbf_layers[-1] == sine_layers[0]\n", "\n", "model = CustomKAN(rbf_layers, sine_layers, True, 42)" ] }, { "cell_type": "code", "execution_count": 7, "id": "6723a412-c313-453c-aa88-9274687ee54e", "metadata": {}, "outputs": [], "source": [ "opt_type = optax.adam(learning_rate=0.001)\n", "\n", "optimizer = nnx.Optimizer(model, opt_type, wrt=nnx.Param)" ] }, { "cell_type": "code", "execution_count": 8, "id": "f18ab849-3c05-418a-a30b-3928f77c332d", "metadata": {}, "outputs": [], "source": [ "# Define train loop\n", "@nnx.jit\n", "def train_step(model, optimizer, X_train, y_train):\n", "\n", " def loss_fn(model):\n", " residual = model(X_train) - y_train\n", " loss = jnp.mean((residual)**2)\n", "\n", " return loss\n", " \n", " loss, grads = nnx.value_and_grad(loss_fn)(model)\n", " optimizer.update(model, grads)\n", " \n", " return loss" ] }, { "cell_type": "code", "execution_count": 9, "id": "b06282a5-8899-4801-9c9e-d8b73b55d74a", "metadata": {}, "outputs": [], "source": [ "# Initialize train_losses\n", "num_epochs = 2000\n", "train_losses = jnp.zeros((num_epochs,))\n", "\n", "for epoch in range(num_epochs):\n", " # Calculate the loss\n", " loss = train_step(model, optimizer, X_train, y_train)\n", " \n", " # Append the loss\n", " train_losses = train_losses.at[epoch].set(loss)" ] }, { "cell_type": "code", "execution_count": null, "id": "d1efa9ea-a3a7-4fb9-bd3d-dd9185e8d437", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "faab949e-dadb-4cec-bc13-63b10cb609c5", "metadata": {}, "source": [ "## Evaluation" ] }, { "cell_type": "code", "execution_count": 10, "id": "86630d29-9b9f-452a-b2f1-399b02768829", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnEAAAGJCAYAAADlpGXRAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAhGdJREFUeJztnXl4VNX5+D+Tfd/IRiAhIQQStrAGERXRVKSuqBWBKqDVumBtsbbaVgVbi4q1/hSqtiJorbijrTuo1AUIBAjIkrCFsCUQyL4vc35/5JtbhiQkhJm5Mznv53nyPJk7d+49n3tvbt655z3vsSilFIIgCIIgCIJb4WF2AwRBEARBEISzR4I4QRAEQRAEN0SCOEEQBEEQBDdEgjhBEARBEAQ3RII4QRAEQRAEN0SCOEEQBEEQBDdEgjhBEARBEAQ3RII4QRAEQRAEN0SCOEEQBEEQBDdEgjhBEGyYPXs2iYmJ3frs/PnzsVgs9m2QIHRC63V34sQJs5siCE5FgjhBcBMsFkuXftasWWN2U01h9uzZBAUFmd2MLqGU4p///CcXXXQRYWFhBAQEMGzYMB577DGqq6vNbl4bWoOkjn6KiorMbqIgaImX2Q0QBKFr/POf/7R5/dprr7Fq1ao2y9PS0s5pP//4xz+wWq3d+uwf/vAHHnzwwXPaf0+nubmZGTNm8Pbbb3PhhRcyf/58AgIC+Pbbb1mwYAHvvPMOq1evJiYmxuymtuGFF15oN1AOCwtzfmMEQZAgThDchZ/+9Kc2r9evX8+qVavaLD+dmpoaAgICurwfb2/vbrUPwMvLCy8vua2ciaeeeoq3336bX//61yxatMhYfscdd3DjjTdy7bXXMnv2bD799FOntqsr18kNN9xAZGSkk1okCEJnSHeqIPQgLr74YoYOHcqmTZu46KKLCAgI4He/+x0AH374IVdccQVxcXH4+vqSnJzMH//4R5qbm222cXpO3IEDB7BYLDz99NP8/e9/Jzk5GV9fX8aOHcvGjRttPtteTpzFYmHu3Ll88MEHDB06FF9fX4YMGcJnn33Wpv1r1qxhzJgx+Pn5kZyczEsvvWT3PLt33nmH0aNH4+/vT2RkJD/96U85cuSIzTpFRUXMmTOHvn374uvrS+/evbnmmms4cOCAsU52djaTJ08mMjISf39/kpKSuPXWW8+479raWhYtWsTAgQNZuHBhm/evuuoqZs2axWeffcb69esBuPLKK+nfv3+72xs/fjxjxoyxWfb6668bfhEREdx0000cOnTIZp0zXSfnwpo1a7BYLLz11lv87ne/IzY2lsDAQK6++uo2bYCunQuA3NxcbrzxRqKiovD392fQoEH8/ve/b7NeWVkZs2fPJiwsjNDQUObMmUNNTY3NOqtWreKCCy4gLCyMoKAgBg0aZBd3QTAD+cosCD2MkydPMmXKFG666SZ++tOfGt1yy5cvJygoiHnz5hEUFMRXX33FI488QkVFhc0ToY544403qKys5Oc//zkWi4WnnnqK6667jv3793f69O67777j/fff5+677yY4OJjnnnuO66+/noMHD9KrVy8AtmzZwuWXX07v3r1ZsGABzc3NPPbYY0RFRZ37Qfk/li9fzpw5cxg7diwLFy7k2LFj/L//9//4/vvv2bJli9EteP3117Njxw7uvfdeEhMTOX78OKtWreLgwYPG68suu4yoqCgefPBBwsLCOHDgAO+//36nx6G0tJT77ruvwyeWt9xyC8uWLeOjjz7ivPPOY9q0adxyyy1s3LiRsWPHGusVFBSwfv16m3P3+OOP8/DDD3PjjTfys5/9jOLiYp5//nkuuugiGz/o+Do5EyUlJW2WeXl5telOffzxx7FYLPz2t7/l+PHjPPvss2RmZpKTk4O/vz/Q9XOxbds2LrzwQry9vbnjjjtITExk3759/Oc//+Hxxx+32e+NN95IUlISCxcuZPPmzbz88stER0fz5JNPArBjxw6uvPJKhg8fzmOPPYavry979+7l+++/79RdEFwSJQiCW3LPPfeo0/+EJ06cqAD14osvtlm/pqamzbKf//znKiAgQNXV1RnLZs2apfr162e8zs/PV4Dq1auXKikpMZZ/+OGHClD/+c9/jGWPPvpomzYBysfHR+3du9dYtnXrVgWo559/3lh21VVXqYCAAHXkyBFj2Z49e5SXl1ebbbbHrFmzVGBgYIfvNzQ0qOjoaDV06FBVW1trLP/oo48UoB555BGllFKlpaUKUIsWLepwWytXrlSA2rhxY6ftOpVnn31WAWrlypUdrlNSUqIAdd111ymllCovL1e+vr7q/vvvt1nvqaeeUhaLRRUUFCillDpw4IDy9PRUjz/+uM16P/zwg/Ly8rJZfqbrpD1az2t7P4MGDTLW+/rrrxWg+vTpoyoqKozlb7/9tgLU//t//08p1fVzoZRSF110kQoODjY8W7FarW3ad+utt9qsM3XqVNWrVy/j9V//+lcFqOLi4i55C4KrI92pgtDD8PX1Zc6cOW2Wtz4BAaisrOTEiRNceOGF1NTUkJub2+l2p02bRnh4uPH6wgsvBGD//v2dfjYzM5Pk5GTj9fDhwwkJCTE+29zczOrVq7n22muJi4sz1hswYABTpkzpdPtdITs7m+PHj3P33Xfj5+dnLL/iiitITU3l448/BlqOk4+PD2vWrKG0tLTdbbU+Jfroo49obGzschsqKysBCA4O7nCd1vcqKioACAkJYcqUKbz99tsopYz13nrrLc477zwSEhIAeP/997Fardx4442cOHHC+ImNjSUlJYWvv/7aZj8dXSdn4r333mPVqlU2P8uWLWuz3i233GLjeMMNN9C7d28++eQToOvnori4mG+++YZbb73V8GylvS72O++80+b1hRdeyMmTJ41j2XrePvzww24P3hEEV0KCOEHoYfTp0wcfH582y3fs2MHUqVMJDQ0lJCSEqKgoY1BEeXl5p9s9/Z9oa0DXUaBzps+2fr71s8ePH6e2tpYBAwa0Wa+9Zd2hoKAAgEGDBrV5LzU11Xjf19eXJ598kk8//ZSYmBguuuginnrqKZsyGhMnTuT6669nwYIFREZGcs0117Bs2TLq6+vP2IbWwKY1mGuP9gK9adOmcejQIdatWwfAvn372LRpE9OmTTPW2bNnD0opUlJSiIqKsvnZtWsXx48ft9lPR9fJmbjooovIzMy0+Rk/fnyb9VJSUmxeWywWBgwYYOQUdvVctAb5Q4cO7VL7OrtGp02bxoQJE/jZz35GTEwMN910E2+//bYEdILbIkGcIPQwTn3i1kpZWRkTJ05k69atPPbYY/znP/9h1apVRq5QV/6JeXp6trv81KdDjvisGfzyl79k9+7dLFy4ED8/Px5++GHS0tLYsmUL0BKUvPvuu6xbt465c+dy5MgRbr31VkaPHk1VVVWH220t/7Jt27YO12l9b/Dgwcayq666ioCAAN5++20A3n77bTw8PPjJT35irGO1WrFYLHz22WdtnpatWrWKl156yWY/7V0n7k5n15m/vz/ffPMNq1ev5uabb2bbtm1MmzaNH/3oR20G+AiCOyBBnCBowJo1azh58iTLly/nvvvu48orryQzM9Ome9RMoqOj8fPzY+/evW3ea29Zd+jXrx8AeXl5bd7Ly8sz3m8lOTmZ+++/ny+++ILt27fT0NDAX/7yF5t1zjvvPB5//HGys7P517/+xY4dO3jzzTc7bEPrqMg33nijw6DhtddeA1pGpbYSGBjIlVdeyTvvvIPVauWtt97iwgsvtOl6Tk5ORilFUlJSm6dlmZmZnHfeeZ0cIfuxZ88em9dKKfbu3WuMeu7quWgdlbt9+3a7tc3Dw4NLL72UZ555hp07d/L444/z1VdfteluFgR3QII4QdCA1icUpz75amho4G9/+5tZTbLB09OTzMxMPvjgA44ePWos37t3r93qpY0ZM4bo6GhefPFFm27PTz/9lF27dnHFFVcALfXS6urqbD6bnJxMcHCw8bnS0tI2TxFHjBgBcMYu1YCAAH7961+Tl5fXbomMjz/+mOXLlzN58uQ2Qde0adM4evQoL7/8Mlu3brXpSgW47rrr8PT0ZMGCBW3appTi5MmTHbbL3rz22ms2XcbvvvsuhYWFRn5jV89FVFQUF110Ea+88goHDx602Ud3nuK2N7q2K+dNEFwVKTEiCBpw/vnnEx4ezqxZs/jFL36BxWLhn//8p0t1Z86fP58vvviCCRMmcNddd9Hc3MzixYsZOnQoOTk5XdpGY2Mjf/rTn9osj4iI4O677+bJJ59kzpw5TJw4kenTpxtlLRITE/nVr34FwO7du7n00ku58cYbGTx4MF5eXqxcuZJjx45x0003AfDqq6/yt7/9jalTp5KcnExlZSX/+Mc/CAkJ4cc//vEZ2/jggw+yZcsWnnzySdatW8f111+Pv78/3333Ha+//jppaWm8+uqrbT734x//mODgYH7961/j6enJ9ddfb/N+cnIyf/rTn3jooYc4cOAA1157LcHBweTn57Ny5UruuOMOfv3rX3fpOHbEu+++2+6MDT/60Y9sSpRERERwwQUXMGfOHI4dO8azzz7LgAEDuP3224GWgtJdORcAzz33HBdccAGjRo3ijjvuICkpiQMHDvDxxx93+bpo5bHHHuObb77hiiuuoF+/fhw/fpy//e1v9O3blwsuuKB7B0UQzMSUMbGCIJwzHZUYGTJkSLvrf//99+q8885T/v7+Ki4uTv3mN79Rn3/+uQLU119/bazXUYmR9kpuAOrRRx81XndUYuSee+5p89l+/fqpWbNm2Sz78ssv1ciRI5WPj49KTk5WL7/8srr//vuVn59fB0fhf8yaNavDMhjJycnGem+99ZYaOXKk8vX1VREREWrmzJnq8OHDxvsnTpxQ99xzj0pNTVWBgYEqNDRUjRs3Tr399tvGOps3b1bTp09XCQkJytfXV0VHR6srr7xSZWdnd9pOpZRqbm5Wy5YtUxMmTFAhISHKz89PDRkyRC1YsEBVVVV1+LmZM2cqQGVmZna4znvvvacuuOACFRgYqAIDA1Vqaqq65557VF5enrHOma6T9jhTiZFTr5/WEiMrVqxQDz30kIqOjlb+/v7qiiuuaFMiRKnOz0Ur27dvV1OnTlVhYWHKz89PDRo0SD388MNt2nd66ZBly5YpQOXn5yulWq6va665RsXFxSkfHx8VFxenpk+frnbv3t3lYyEIroRFKRf6Ki4IgnAa1157LTt27GiTZyW4HmvWrGHSpEm888473HDDDWY3RxB6PJITJwiCy1BbW2vzes+ePXzyySdcfPHF5jRIEATBhZGcOEEQXIb+/fsze/Zs+vfvT0FBAS+88AI+Pj785je/MbtpgiAILocEcYIguAyXX345K1asoKioCF9fX8aPH8+f//znNsVjBUEQBJCcOEEQBEEQBDdEcuIEQRAEQRDcEAniBEEQBEEQ3BDJiesEq9XK0aNHCQ4OxmKxmN0cQRAEQRB6MEopKisriYuLw8PjzM/aJIjrhKNHjxIfH292MwRBEARB0IhDhw7Rt2/fM64jQVwnBAcHAy0HMyQkxCH7yM7OZsyYMQ7Ztisj3vqgozOIt07o6Azi7QgqKiqIj4834o8zIUFcJ7R2oYaEhDgsiAsMDHTYtl0Z8dYHHZ1BvHVCR2cQb0fSlRQuKTHSCRUVFYSGhlJeXu6wE6aU0jLfTrz1QUdnEG+d0NEZxNsRnE3cIaNTXYCcnByzm2AK4q0POjqDeOuEjs4g3mYjQZwL0NDQYHYTTEG89UFHZxBvndDRGcTbbLTIiZs6dSpr1qzh0ksv5d133zW7OW0ICwszuwmmIN76oKMziLdOONpZKUVTUxPNzc0O3c/ZEhISQl1dndnNcDrn4u3p6YmXl5ddumO1yIlbs2YNlZWVvPrqq2cdxDkjJ666uprAwECHbNuVEW990NEZxFsnHOnc0NBAYWEhNTU1Dtn+uWC1WjutZdYTOVfvgIAAevfujY+PT5v3zibu0OJJ3MUXX8yaNWvMbkaHbN++nXHjxpndDKcj3vqgozOIt044ytlqtZKfn4+npydxcXH4+Pi41EACHQN26L63UoqGhgaKi4vJz88nJSXlnIJBlw/ivvnmGxYtWsSmTZsoLCxk5cqVXHvttTbrLFmyhEWLFlFUVER6ejrPP/88GRkZ5jRYEARBEOxEQ0MDVquV+Ph4AgICzG5OG5qamvDz8zO7GU7nXLz9/f3x9vamoKCAhoaGczp+Lv8MtLq6mvT0dJYsWdLu+2+99Rbz5s3j0UcfZfPmzaSnpzN58mSOHz/erf3V19dTUVFh8+No+vfv7/B9uCLirQ86OoN464SjnV21y9LX19fsJpjCuXrb63y6/JO4KVOmMGXKlA7ff+aZZ7j99tuZM2cOAC+++CIff/wxr7zyCg8++OBZ72/hwoUsWLCgzfLs7GwCAwMZNWoUu3btora2luDgYJKSkti2bRsA/fr1w2q1cujQIQBGjBjB3r17qaqqIjAwkIEDB7JlyxYA+vbti6enJwUFBdTW1jJu3DgOHDhARUUFfn5+DBkyhE2bNgEQFxeHn58f+/fvB2Do0KEcPnyYsrIyfHx8GDFiBBs2bAAgNjaWoKAg9u7dC0BaWhrHjh2jpKQELy8vRo8ezYYNG1BKERUVRXh4OLt37wZg0KBBlJSUUFxcjIeHB2PHjiU7O5vm5mZ69epFdHQ0u3btAiAlJYWKigqOHTsGwLhx49i8eTONjY2Eh4cTFxfHjh07AEhOTqampobCwkIAxowZw/bt2yktLSU2NpaEhAR++OEHABITE2lqauLw4cMAjBo1itzcXGpqaggKCiI5OZmtW7cCkJCQAMDBgwcBSE9PZ9++fVRVVREQEEBqaiqbN282jreXlxcHDhwAYNiwYRw8eJDy8nL8/PwYOnQo2dnZAPTu3ZuAgAD27dsHwJAhQzh69CilpaV4e3szatQosrKyAIiJiSEkJIQ9e/YYx/v48eOcPHkST09PxowZw8aNG7FarURFRREREcGOHTvw9/dn4MCBlJaWUlxcjMViISMjg02bNtHU1ERERAQxMTHG8R4wYABVVVUUFRUBkJGRQU5ODg0NDYSFhdG3b1+2b98OtPwzqaur4+jRowCMHj2aHTt2UFdXR0hICImJiTbXbHNzs3G8R44cye7du6muriYoKIgBAwYYQ+nj4+Px8PCgoKAAgOHDh5Ofn09lZSX+/v6kpaUZx7tPnz74+PiQn58PQK9evSgpKaGsrAxfX1+GDx/Oxo0bjWs2MDDQON6DBw+mqKiIkpKSNsc7Ojqa0NBQ43inpqZy4sQJTpw4YVyzrcc7MjKSyMhIcnNzjWu2vLzc+IJ36jUbERFBbGwsO3fuNK7Z6upq43iPHTuWbdu2UV9fT1hYGPHx8cY1m5SURENDA0eOHDGu2dZ7hFKKoKCgc75HtB5vd7lH1NbWMnz48HO+R9TV1REaGuoW94gDBw6wf/9+u9wj8vLyABg4cCAnTpygvr6empoa/Pz8qK6uRimFl5cX3t7e1NbWAuDn50dzczONjY1ASzHa2tparFZrm3V9fX2Nbr3T1/X09MTX19fIvzt93YCAAOrr62lubsbT0xMPDw+qqqoAjK7e+vr6Nut6eHjg5+dnbPf0df39/WlsbKSpqQkPDw/8/f2prq4GwNvbGw8Pj3bXtVgsBAYGGm3w9vbG09PTGHTg5+dHU1OTzbqnHkMvLy+bdU89hkFBQR0eb09PT6xWa7eOd0BAALW1tdTX17N3716Sk5Nt7hHl5eV0Fbca2GCxWGy6UxsaGggICODdd9+16WKdNWsWZWVlfPjhh8ayNWvWsHjx4k4HNtTX1xsXCvxv+gtHDWzYtq+Yz/+7idtvnERYkF7faLKysrTLmwE9vXV0BvHWCUc519XVkZ+fT1JSkkt2W1ZVVREUFGR2M5zOuXqf6bxqU+z3xIkTNDc3ExMTY7M8JibG+AYNkJmZyU9+8hM++eQT+vbty7p16zrcpq+vrzHFliOn2mrljkVf8I/VR1j83maH7kcQBEEQ3JnExESeffZZs5vhUrh8d6o9WL169Vl/ZsmSJSxZssSoyeOo7tTyqpbHuG9/lcf1GWFu0VUC9ulOtVqt5ObmukVXiT27U61WK1lZWVp1p6alpZGXl6ddd2prF4tu3alWq5WSkhKtulNDQ0PJysrSrjvV39+/TXdqe2UzTuWhhx5i/vz5Z92d+vXXXxMYGGh4dqc79YorrmDIkCE8+eST59Sd6uvrS319vXSnng3n0p3aXRxdJ27YrOXUNzbj4+XB9tfm2H37rszWrVtJT083uxlOR0dvHZ1BvHXCUc6u3p1aU1PTZtTsqT1hb731Fo888ogRmEJLcNTaFamUorm5GS8v5zxTuvjiixkxYsQ5P9Frz/tskO5UWqL+0aNH8+WXXxrLrFYrX375JePHjzexZV3nx+clAdBsVaxYvcvk1jgXHat8g57eOjqDeOuEM52VUtTUNTr9p71nPlartc2y2NhY4yc0NBSLxWK8zs3NJTg4mE8//ZTRo0fj6+vLd999x759+7jmmmuIiYkhKCiIsWPHtulFO7071WKx8PLLLzN16lQCAgJISUnh3//+9zkd2/fee48hQ4bg6+tLYmIif/nLX2ze/9vf/kZKSorRW3LDDTcY77377rsMGzYMf39/evXqRWZmpvE00VG4fHdqVVWV8dgfID8/n5ycHCIiIkhISGDevHnMmjWLMWPGkJGRwbPPPkt1dbUxWtXVmXnZYFZ+u5dmq+KZtzYxPTPN7CY5DUfnG7oqOnrr6AzirRPOdK6tb2LEra85bX+t5LxyCwF+3jbLPD09u7WtBx98kKeffpr+/fsTHh7OoUOH+PGPf8zjjz+Or68vr732GldddRV5eXlGt3h7LFiwgKeeeopFixbx/PPPM3PmTAoKCoiIiDjrNm3atIkbb7yR+fPnM23aNNauXcvdd99Nr169mD17NtnZ2fziF7/gn//8J6NGjaKmpoZvv/0WgMLCQqZPn85TTz3F1KlTqays5Ntvv2038LUnLh/EZWdnM2nSJOP1vHnzgJYu0+XLlzNt2jSKi4t55JFHKCoqYsSIEXz22WdtBjucLc7KiSs5ut/YZ1NzE1lZWS6f7wL2yYmrrq7WMieutLRUu5y4lJQULXPi/P39tcyJa25u1i4nrjXX1Rk5cTV1LXlYzqa6uho/nxCbnDhvb+8zlhhpzQNrbX9rntjDDz/MxRdfbKw7dOhQBg0aZOTEPfbYY7z33nu88847zJ07Fw8PD5RSxr5bc9FmzJjB9OnTqaqq4ne/+x3PPfcc69atY+LEiUDbnDiAxsZGqqqq2uTEPf3000yaNIlf/epXAMyePZutW7fy5JNPctNNN3HgwAECAwO5+OKL6dWrF0opUlJSqKqq4ujRozQ1NTF58mRiY2OJj48nKSnJ2J+jcuJQwhkpLy9XgCovL3fYPgbfvFSlTH9ZzXr8E4ftwxVZv3692U0wBR29dXRWSrx1wlHOtbW1aufOnaq2ttZYZrVaVXVtw1n/LP/kB3XR3BVq+Sc/dOvzVqu1TfsqKyvP2P5ly5ap0NBQ4/XXX3+tAHX48OE227n//vtVamqqCg0NVYGBgcrDw0M98MADxjr9+vVTf/3rX43XgHr77bdtthMSEqJeffXVDtszceJEdd9997X73siRI9X8+fNtln3wwQfK29tbNTU1qYqKCjVs2DAVGRmpbrrpJvX666+r6upqpZRSTU1N6tJLL1XBwcHqhhtuUH//+99VSUlJh+1o77y2cjZxh8s/idOC/3vauvPASXPbIQiCILg8FoulTbdmV5g1ZSizpgx1QIvOntPnHf31r3/NqlWrePrppxkwYAD+/v7ccMMNxtOrjvD2tj0OFoul3Tw9exAcHMzmzZtZs2YNH330EY888gjz589n48aNhIWFsWrVKtauXcsXX3zB888/z+9//3uysrKMJ3KOwK0HNvQUokL9AYgOd7158RxJv379zG6CKejoraMziLdO6OgM9pt26/vvv2f27NlMnTqVYcOGERsba3RtO4u0tDS+//77Nu0aOHCgkfvn5eVFZmYmixYtYtu2bRw4cICvvvoKaAkgJ0yYwIIFC9iyZQs+Pj6sXLnSoW2WJ3Ed4KycuIKCAkr/r1bc7kOl/Pkfn3JFRpxL57uA/abdKisrc4t8F3vmxOXm5lJQUKBVTlxUVJSWOXEAoaGh2uXE1dbW4uPjo1VO3OHDhykoKNCuTlxn0251lBPX0NBAY2Ojse6AAQN47733uPTSS/Hw8ODPf/6zMa1VfX19hzlxrflsrW0AaG5uNl63lxNXWFjI2rVr8fT0xNPTk4aGBmJjY7nvvvsYP348Dz/8MNdddx3btm1j8eLFPPPMM9TV1fH555+ze/duJkyYQK9evfjiiy+wWq3Ex8ezfv16Pv/8cyZNmkRsbCxbtmyhuLiYxMREyYkzE2fkxN32x3dVyvSXVcr0l9XEe9902H5cDR3zZpTS01tHZ6XEWyecmRPnSnQ3J660tNRmvfz8fDVp0iTl7++v4uPj1eLFi9vkr7WXE7dy5Uqb7YSGhqply5Z12J6JEycqWpKYbH7++Mc/KqWUevfdd9XgwYOVt7e3SkhIUIsWLTI+++2336qJEyeq8PBw5e/vr4YPH67eeustpZRSO3fuVJMnT1ZRUVHK19dXDRw4UD3//PMdtkNy4noQk4ZF8M3OMgBGDYw2tzGCIAiCYCdmz57N7NmzjdcXX3xxu2U3EhMTjW7JVu655x6b16d3r7a3nbKysjO2Z82aNWd8//rrr+f6669v970LLrjA+Pzpc6empaXx2WefnXHbjkBy4lyASyeMNn7flHfMxJY4l5EjR5rdBFPQ0VtHZxBvndDRGTinWQvcGVfxlidxHeDMnLjy8nI8PCxYrYqYYAtbt2516XwXsE9O3PHjx4mPj3eLfBd75sRt3LiRkJAQrXLifHx8CAgI0C4nrq6ujoyMDO1y4ioqKhg9erRWOXGtiey65cQppYzRoO3lxLWu6+HhgZ+fn7Hd09ftytyp7a17tnOnBgYG2hzD7s6darFY8PLykrlTXR1Hz50KkJWVxZzFO2hqVvQK8WPdizMdsh9XIysri3HjxpndDKejo7eOziDeOuEoZ1efO/X0bkVdOFdvmTu1BxEUFERibCgAE4b1Mbk1zkPHP3zQ01tHZxBvndDRGbo/7Za74yreEsS5AAMGDCA4wAeAypozFzbsSQwYMMDsJpiCjt46OoN464SjnV2108xedeLcjXP1ttf5lCDOBcjJyWHv4VIAsnYWmtwa59Gaa6UbOnrr6AzirROOcm6dkaA1l8zVcNV2OZpz9W79/OkzTpwtMrChA5xa7Le0lAuH9+GTrAPUNzaxaPkqfvXTSS6btAz2K/abm5vrFknL9hzYUFpaSlZWllYDG5qamrQs9ltZWUltba12AxtKS0spKSnRamBDVVUVWVlZdh/YUFpaSmVlZZvpp05Pyvf19TWK40JL8nxdXZ2RaH/quj4+Piil2l3X09MTHx8fIyn/9HX9/f1paGgwBjY0NzcbAw68vb2xWCxGW09d18PDA19fX2O7p6976gCE0wdBnGldi8VCQEBAh4MgfH19aW5utlm3pqbGGKzg6elps+6pxzAwMNBm3VOPIdDt4221WiksLKSuro68vLw2cYQMbLAjzhjYcPToUXYfa+aORV8AEBcZxJrnpjlkX67E0aNHiYuLM7sZTkdHbx2dQbx1wpHOSimKioo6rYFmBq3BnG6cq3dYWBixsbHGLBKncjZxhzyJcwE8PDyIifhf/7ouBX89PPTszdfRW0dnEG+dcKSzxWKhd+/eREdHG09zXIXi4mKioqLMbobTORfv1jIo9kCCOBegoKCAlLR047UuBX8LCgqIjY01uxlOR0dvHZ1BvHXCGc6tc326EkePHiU+Pt7sZjgdV/HW7+uSixIe7IunR8tj1WmXDDK5NYIgCIIguDoSxLkAw4cPx2KxEBLYUmakolqPMiPDhw83uwmmoKO3js4g3jqhozOIt9lIEOcCtI7gq65tyXVY+e0eM5vjNFq9dUNHbx2dQbx1QkdnEG+zkZy4DnB2iZGkpCRCArw4UdFAWIAnTU1NLls+AOxTYqSwsBAPDw+3KB9gzxIjBw8epLKyUkqMaFJixB73iNbj7U4lRmJjY7UqMVJYWEhlZaVDSoy48j2ioqLCcD2Xe8SwYcM4dOiQ29wjGhoaKCgoOOd7RHtxhJQYsSPOKDGybds2hg8fzshbX6O6rpEgf282L73FIftyJVq9dUNHbx2dQbx1QkdnEG9HIHOnuhlpaWkAnD+0pcZQ/7gwE1vjPFq9dUNHbx2dQbx1QkdnEG+zkSDOBWh9vPyjsf0AyD1YworVu8xsklNo9dYNHb11dAbx1gkdnUG8zUaCOBciOjwQgIbGZl769zaTWyMIgiAIgisjQZwL0KdPHwCiwwMAsFjg51f3/ByDVm/d0NFbR2cQb53Q0RnE22wkiHMBfHxa6sNFh/kDoBQ0NlnNbJJTaPXWDR29dXQG8dYJHZ1BvM1GgjgXoHWIdXDA/y6Kv/+n53enukqdHWejo7eOziDeOqGjM4i32UiduA5wdp242tpaDhw4gK+XhfomRf/eIUbdG1erAQX2qRNXWlpKbm6uW9SAsmeduNLSUrKystyiBpTUiTv3OnG1tbVa1okrKSnRqk5cVVUVWVlZ2tWJs1qtUicOqRPnsjijTlxNTQ0BAS35cENvWUZDk5XwYF+yXvqpQ/bnKpzqrRM6euvoDOKtEzo6g3g7AqkT52a0fisHSIkPB2Dc4N5mNcdpnOqtEzp66+gM4q0TOjqDeJuNBHEuQFlZmfF7sH9LXlxlTaNJrXEep3rrhI7eOjqDeOuEjs4g3mYjQZwL4Ovra/y+88BJADbvPmZWc5zGqd46oaO3js4g3jqhozOIt9lITlwnOCMnzmq14uHREk8/+OI3vP/NHpJ6h/L5X25wyP5chVO9dUJHbx2dQbx1QkdnEG9HIDlxbkbrCByAqy8YAMCh4xU9fuqtU711QkdvHZ1BvHVCR2cQb7ORIM7FaJ21oalZydRbgiAIgiB0iNSJcwFiY2ON32PC/zdk2c/Hk9SZS7EquPL8/jwzd5IZzXMYp3rrhI7eOjqDeOuEjs4g3mYjT+JcgMDAQOP3IH9v4/f9R8ux/l/G4kdr9zu7WQ7nVG+d0NFbR2cQb53Q0RnE22wkiHMBWqtQA1gsFjws7a/3q+e/dlKLnMOp3jqho7eOziDeOqGjM4i32UgQ54LERQa1u/zjdT3vaZwgCIIgCN1DcuI6wJlzpzY1NRlzp1ZUVFBeVddhuzJ/8Trv/vFqjhUVuv3cqU1NTVrOndrU1KTd3KlJSUlazp3q6+ur5dypTU1N2s2dGhAQoOXcqQMGDNBy7tTevXvL3KnugDPqxO3Zs4eUlBTj9RurdjF/2VoAvDw9mDQynlXZBW0+52GBR2afz4wfpTmkXY7mdG9d0NFbR2cQb53Q0RnE2xFInTg3o6SkxOb1jB+lseDW84mLDOLhWeexZF4mnu0kylkVzF+2ltSZS3ljlfvVlDvdWxd09NbRGcRbJ3R0BvE2GwniXABvb+82y6ZnprHmuWlMz2x5yvbI7PF4WNof8dAazA2csZSpv//AkU21K+1564CO3jo6g3jrhI7OIN5mI92pneCM7tTucP0fPuSH/Sfafc8CzL/1fCMAFARBEATBPZDuVDejNSHzbHjvT9ew+43bGNY/ss17Cnj0lbVc+7uVdmid4+iOd09AR28dnUG8dUJHZxBvs5Egzs1pDeYW3Ho+p3e27jxQwoS73zClXYIgCIIgOBYJ4lyA6Ojoc97G9Mw08tp5MldcVkvqzKWsWO16Ax/s4e2O6OitozOIt07o6AzibTYSxLkAoaGhdtvWe3+6hgW3nm8z64NVtXSvulogZ09vd0JHbx2dQbx1QkdnEG+zkSDOBWgtSmgvpmemkfuv2+gdYTu32x+Xr7Prfs4Ve3u7Czp66+gM4q0TOjqDeJuNBHE9mP8uvonBiRHG6yarcvnBDoIgCIIgdA0J4lyA1NRUh237gz9PJTXhf4HczgMlzFv8tcP2dzY40tuV0dFbR2cQb53Q0RnE22wkiHMBTpxov96bvfj3E1NtZnz4aO1+l8iPc7S3q6Kjt47OIN46oaMziLfZSBDnAjjjYnhk9nib1wtfN7/Gjav8ETgbHb11dAbx1gkdnUG8zUaLIO6jjz5i0KBBpKSk8PLLL5vdnDZ4eDj+NEzPTOOysYnG67qGZtOfxjnD2xXR0VtHZxBvndDRGcTbbHr8tFtNTU0MHjyYr7/+mtDQUEaPHs3atWvp1atXlz7vqtNudZfBN79CU3PLKffz8WTb8tnmNkgQBEEQBAOZdusUNmzYwJAhQ+jTpw9BQUFMmTKFL774wuxm2bBx40an7es3MzKM381+GudMb1dCR28dnUG8dUJHZxBvs3H5IO6bb77hqquuIi4uDovFwgcffNBmnSVLlpCYmIifnx/jxo1jw4YNxntHjx6lT58+xus+ffpw5MgRZzS9y1itVqfta/aUoXh5/m+QwzNvbXLavk/Hmd6uhI7eOjqDeOuEjs4g3mbj8kFcdXU16enpLFmypN3333rrLebNm8ejjz7K5s2bSU9PZ/LkyRw/ftzJLe0+kZFtJ7F3JL/8yWjj9+raBqfu+1Sc7e0q6OitozOIt07o6AzibTYuH8RNmTKFP/3pT0ydOrXd95955hluv/125syZw+DBg3nxxRcJCAjglVdeASAuLs7myduRI0eIi4vrcH/19fVUVFTY/DgaZ18Md1ydbvzeZFWm1Y1zlT8CZ6Ojt47OIN46oaMziLfZeJndgHOhoaGBTZs28dBDDxnLPDw8yMzMZN26limmMjIy2L59O0eOHCE0NJRPP/2Uhx9+uMNtLly4kAULFrRZnp2dTWBgIKNGjWLXrl3U1tYSHBxMUlIS27ZtA6Bfv35YrVYOHToEwIgRI9i7dy9VVVUEBgYycOBAtmzZAkDfvn3x9PSkoKCA0tJSJk6cyIEDB6ioqMDPz48hQ4awaVNLV2dcXBx+fn7s378fgKFDh3L48GHKysrw8fFhxIgRRhdybGwsQUFB7N27F4C0tDSOHTtGSUkJXl5ejB49mg0bNpDaJ4DcIzUAfLx2P9PGBjBo0CBKSkooLi7Gw8ODsWPHkp2dTXNzM7169SI6Oppdu1py6FJSUqioqODYsWMAjBs3js2bN9PY2Eh4eDhxcXHs2LEDgOTkZGpqaigsLARgzJgxbN++ncLCQhITE0lISOCHH34AIDExkaamJg4fPgzAqFGjyM3NpaamhqCgIJKTk9m6dSsACQkJABw8eBCA9PR09u3bR1VVFQEBAaSmprJ582bjeHt5eXHgwAEAhg0bxsGDBykvL8fPz4+hQ4eSnZ0NQO/evQkICGDfvn0ADBkyhKNHj1JaWoq3tzejRo0iK6ulREtMTAwhISHGFCxpaWkcP36ckydP4unpyZgxY9i4cSNWq5WoqCgiIiJYv3494eHhDBw4kNLSUoqLi7FYLGRkZLBp0yaampqIiIggJibGON4DBgygqqqKoqIi47rOycmhoaGBsLAw+vbty/bt2wHo378/dXV1HD16FIDRo0ezY8cO6urqCAkJITEx0eaabW5uNo73yJEj2b17N9XV1QQFBTFgwABycnIAiI+Px8PDg4KCAgCGDx9Ofn4+lZWV+Pv7k5aWZhzvPn364OPjQ35+PtAywCgyMpKysjJ8fX0ZPny4kVMSGxtLYGCgcbwHDx5MUVERJSUlbY53dHQ0oaGhxvFOTU3lxIkTnDhxwrhmW493ZGQkkZGR5ObmGtdseXm58ZT+1Gs2IiKC2NhYdu7caVyz1dXVxvEeO3Ys27Zto76+nrCwMOLj441rNikpiYaGBuPL4qn3iMrKSiZMmHDO94jW4+3Me4RSiqioKMLDw9m9ezdAl+8RpaWlZGRknPM9oq6ujtDQULe4R2RnZxMUFGSXe0ReXh6AW9wjKioqjJGa53KPGDZsGIcOHXKbe0RDQwMJCQnnfI9oL44oLy+nyyg3AlArV640Xh85ckQBau3atTbrPfDAAyojI8N4/eGHH6qUlBSVnJysXnrppTPuo66uTpWXlxs/hw4dUoAqLy+3q8uprF+/3mHb7oiGxmaVMv1llTL9ZZU6c6nT96+UOd6ugI7eOjorJd46oaOzUuLtCMrLy7scd7j1k7iucvXVV3P11Vd3aV1fX198fX0d3CJbUlJSnLo/AG8vD3y8PGhostL8f12qz8yd5NQ2mOHtCujoraMziLdO6OgM4m02bh3ERUZG4unpaTyub+XYsWPExsae07aXLFnCkiVLaG5uBhzbnVpdXc348eOd3lUyZUxvPlzf8nj347X7eeSnI53anXry5Eni4uLcoqvEnt2pOTk5xvXg6l0l9upODQsLo7i42G26SuzVndrc3MzIkSO1606trq5mxIgRWnWntl6TunWn+vj4GG46dacGBARQWVlpeneqWxX7tVgsrFy5kmuvvdZYNm7cODIyMnj++eeBlmG/CQkJzJ07lwcffPCc9+mMYr9ZWVmMGzfOIds+Ew1NzQy9ZTkAXh4Wdr5+q1P3b5a32ejoraMziLdO6OgM4u0IzibucPkncVVVVcY3RoD8/HxycnKIiIggISGBefPmMWvWLMaMGUNGRgbPPvss1dXVzJkzx8RWuwc+Xp54eVpoalY0WRUrVu9iemaa2c0SBEEQBKELuHwQl52dzaRJ/8vVmjdvHgCzZs1i+fLlTJs2jeLiYh555BGKiooYMWIEn332GTExMee0X2d2pwLU1taa0lVycXocqze3POL9y5sbGdff12ndqQC5ublu0VViz+5UaPkW5w5dJfbqTh02bBh5eXlu01Vir+7U4OBgamtrtetOBSgpKdGqOzU8PJysrCztulNHjBhhuOrUnZqcnExBQYF0p7o6zuhO3bx5M6NGjXLItjvjQGE5l93/LuD8uVTN9DYTHb11dAbx1gkdnUG8HYHMnepmNDY2mrbvfrH/u0CsVufG82Z6m4mO3jo6g3jrhI7OIN5mI0GcC9DaxWYGFosFb8+Wy6ChycqK1buctm8zvc1ER28dnUG8dUJHZxBvs3H5nDizcGZOXFNTE3379jUt32VkUhAb9rZML/b0ig30D65wSk5cVVUVzc3NbpHvYs+cuOPHj1NSUuIW+S72yolLSkrSMifO19dXy5y4pqYmevXqpVVOXF1dnZY5cbGxsVrmxPXu3Vty4tyBnlxipJVt+4q54eF/A+Dn48W25bOcsl+zvc1CR28dnUG8dUJHZxBvRyA5ccJZkdrvf4+FlbKa2BJBEARBELqKBHEuQHJysqn79/HyxNPDAmBMZOwMzPY2Cx29dXQG8dYJHZ1BvM1GcuI6wJk5cTU1NZx33nmm5rsMjAtg1+Fqmpqb+fM/PuVHIyIdnhNXUlJC79693SLfxZ45cT/88AMBAQFuke9ir5y48PBwTp486Tb5LvbKibNarQQGBmqXE1dTU0N6erpWOXH79+9n37592uXE+fn5GcdBp5y4oKAgu9WSlJw4B6JDThzAe2t289DfvwUgNNCXjf/4qcP36QreZqCjt47OIN46oaMziLcjkJw44aw5NS+uvrHZxJYIgiAIgtAV5ElcJzjjSZzVanVqLlp71Dc0MWz2qwCEBPiQ/fLNDt+nK3ibgY7eOjqDeOuEjs4g3o5AnsS5Ga194Wbi6+NFcIA3AIMSwp2yT1fwNgMdvXV0BvHWCR2dQbzNRgY2dIAzBzaUlpZSW1tretJyY1NLeZHs3GMsfPkzHvrZ5Q4d2FBUVERubq5bJC3bc2BDUVERWVlZbpG0bK+BDU1NTVoW+62srNSy2G9paSklJSVaDWw4efKklsV+a2trtSz229DQIMV+3QFndKfm5eUxaNAgh2z7bLjnmdWsym7544uLDGLNc9Mcuj9X8XY2Onrr6AzirRM6OoN4OwLpTnUz4uPjzW4CANdNTDF+HzUw2uH7cxVvZ6Ojt47OIN46oaMziLfZSBDnArQ+cjWbpN6hxu/fbj3s8P25irez0dFbR2cQb53Q0RnE22wkiBMM4qP/99jWKp3sgiAIguDSSBDnAiQlJZndBAC8vTwI9GsZ6zI4MaKTtc8dV/F2Njp66+gM4q0TOjqDeJuNBHEuQENDg9lNMGj+v0dwG3YVsWL1Lofuy5W8nYmO3jo6g3jrhI7OIN5mIyVGOsDZJUZ69erlEuUDBieEsHlvKUrBS//eRkpYtcNKjBQWFlJVVeUW5QPsWWJk+/btHDlyxC3KB9izxEh1dbXblA+wZ4mRXr16aVliJCAgQKsSI3l5eRw5ckS7EiMVFRXGta9biZHm5mYpMeLq6DJ3aiunzqF65fn9eWbuJIfty5W8nYmO3jo6g3jrhI7OIN6O4GziDgniOsEZQVxjYyPe3t4O2fbZkp1bxIzHPgYcXyvOlbydiY7eOjqDeOuEjs4g3o5A6sS5Ga2Pw12B+Jhg4/cRA6Icui9X8nYmOnrr6AzirRM6OoN4m40EcS5AbW2t2U0wiAoNMH7Pzjvm0H25krcz0dFbR2cQb53Q0RnE22wkiHMBgoODO1/JSXh4WAj0b3lEnNTbMd3HrbiStzPR0VtHZxBvndDRGcTbbCSIcwFcpd5MK/X1TQBs3XfCoftxNW9noaO3js4g3jqhozOIt9lIEOcCtA4tdhU8PC0AKAdP2+Bq3s5CR28dnUG8dUJHZxBvs5E6cR3g7DpxtbW1LlMDKr1fEBv3VmC1Kp5Y+hmThoY7pE5caWkpubm5blEDyp514kpLS8nKynKLGlD2rBOXl5fnNjWg7Fknrra2Vss6cSUlJVrViauqqiIrK0u7OnFWq9Vw1a1OXEFBgdSJc3WcUWKkqKiI2NhYh2y7O3y5qYC7/rIacGyZEVfzdhY6euvoDOKtEzo6g3g7Aikx4mZYrVazm2BDfPT/EjZHDYx22H5czdtZ6OitozOIt07o6AzibTYSxLkArV0rrkLfqP8Fcdm5jisz4mrezkJHbx2dQbx1QkdnEG+zkSBOaEOAnzc+Xi2XRkrfMHMbIwiCIAhCu0hOXCc4Iyeuvr4eX19fh2y7uwy5ZRmNTVbCg/3IemmmQ/bhit7OQEdvHZ1BvHVCR2cQb0cgOXFuRusoMVciMtQfgNiIgE7W7D6u6O0MdPTW0RnEWyd0dAbxNhsJ4lyAqqoqs5vQhsqaBgAOHqt02D5c0dsZ6OitozOIt07o6AzibTYSxLkAgYGBZjehDZNGxgNQ39jEitWOmejXFb2dgY7eOjqDeOuEjs4g3mYjOXGd4IycuIaGBnx8fByy7e7y3bbD3PrE54DjasW5orcz0NFbR2cQb53Q0RnE2xFITpyb0Vqh3ZU4tczIyBTH1IpzRW9noKO3js4g3jqhozOIt9lIECe0S1xkkPH7prwiE1siCIIgCEJ7yNypHeDMuVNra2tdau7U1nkRQwJ8qKhpYHRSAHv37rX73Km1tbVazp1aW1ur3dypUVFRWs6dCmg5d2ptba12c6d6enpqOXdqTEyMlnOnhoaGytyp7oCOc6e2kvmrtzl4rJIbLk7hz3dcZPftu6q3o9HRW0dnEG+d0NEZxNsRSE6cm9H6rcXVKC5reZLw+QbHtM9VvR2Njt46OoN464SOziDeZiNBnNAhrYMbwoP1q8YtCIIgCK6OBHEuwPDhw81uQrsUl9UAcKykxiHbd1VvR6Ojt47OIN46oaMziLfZSBDnArQm0roa1144AIDGpmaHFPx1VW9Ho6O3js4g3jqhozOIt9lIEOcCVFRUmN2Edpl1+RAArApe+vdWu2/fVb0djY7eOjqDeOuEjs4g3mYjQZwL4OfnZ3YT2iW2VyCW//t9cL9edt++q3o7Gh29dXQG8dYJHZ1BvM1GSox0gjNKjDQ1NeHl5Zol+9J++grNVkVkqD9rX5hh1227srcj0dFbR2cQb53Q0RnE2xFIiRE3o7VgpysSFtQyMrVvVFAna549ruztSHT01tEZxFsndHQG8TYbCeKEM1JR0wDAroKTJrdEEARBEIRTkSDOBYiLizO7CR3i4dGSFWd1QKe7K3s7Eh29dXQG8dYJHZ1BvM1GgjgXwFUSJNvjyvH9AbBald3LjLiytyPR0VtHZxBvndDRGcTbbCSIcwFaJ612RX4yaRAAzVbFS//eZtdtu7K3I9HRW0dnEG+d0NEZxNtstAjipk6dSnh4ODfccIPZTXE7WqfeAhgxIMrElgiCIAiCcCpaBHH33Xcfr732mtnN6JChQ4ea3YQOiQrzN37/dtthu27blb0diY7eOjqDeOuEjs4g3majRRB38cUXExwc3PmKJnH4sH2DI3tisVj4v7ENWK323bYrezsSHb11dAbx1gkdnUG8zcb0IO6bb77hqquuIi4uDovFwgcffNBmnSVLlpCYmIifnx/jxo1jw4YNzm+oAykrKzO7CWckNiIQgL7R9q0V5+rejkJHbx2dQbx1QkdnEG+z6VYQd+jQIZsodMOGDfzyl7/k73//+1lvq7q6mvT0dJYsWdLu+2+99Rbz5s3j0UcfZfPmzaSnpzN58mSOHz9urDNixAiGDh3a5ufo0aNnL2cCPj4+ZjfhjFTVNgKQd7DUriNUXd3bUejoraMziLdO6OgM4m023Zp268ILL+SOO+7g5ptvpqioiEGDBjFkyBD27NnDvffeyyOPPNK9xlgsrFy5kmuvvdZYNm7cOMaOHcvixYsBsFqtxMfHc++99/Lggw92edtr1qxh8eLFvPvuu2dcr76+nvr6euN1RUUF8fHxDp12SymFxWLpfEWT+NOr63jt850AhAb6svEfP7XLdl3d21Ho6K2jM4i3TujoDOLtCM5m2q1uTfy1fft2MjIyAHj77bcZOnQo33//PV988QV33nlnt4O402loaGDTpk089NBDxjIPDw8yMzNZt26dXfZxOgsXLmTBggVtlmdnZxMYGMioUaPYtWsXtbW1BAcHk5SUxLZtLaU3+vXrh9Vq5dChQ0DLE8K9e/dSVVVFYGAgAwcOZMuWLQD07dsXT09PCgoKKC0tZeLEiRw4cICKigr8/PwYMmSIMa1HXFwcfn5+xpDmoUOHcvjwYcrKyvDx8WHEiBFGF3NsbCxBQUHs3bsXgLS0NI4dO0ZJSQleXl6MHj2aDRs2oJQiKiqK8PBwdu/eDcCgQYMoKSmhuLgYDw8Pxo4dS3Z2Nuf1s9I6LKSmvoGsrCxSUlKoqKjg2LFjQEuwvXnzZhobGwkPDycuLo4dO3YAkJycTE1NDYWFhQCMGTOG7du3U1hYSGJiIgkJCfzwww8AJCYm0tTUZDzpHTVqFLm5udTU1BAUFERycjJbt24FICEhAYCDBw8CkJ6ezr59+6iqqiIgIIDU1FQ2b95sHG8vLy8OHDgAwLBhwzh48CDl5eX4+fkxdOhQsrOzAejduzcBAQHs27cPgCFDhnD06FFKS0vx9vZm1KhRZGVlARATE0NISAh79uwxjvfx48c5efIknp6ejBkzho0bN2K1WomKiiIiIoL169cTHh7OwIEDKS0tpbi4GIvFQkZGBps2baKpqYmIiAhiYmLYtavlyeeAAQOoqqqiqKgIgIyMDHJycmhoaCAsLIy+ffuyfft2APr3709dXZ3xJHr06NHs2LGDuro6QkJCSExMtLlmm5ubjeM9cuRIdu/eTXV1NUFBQQwYMICcnBwA4uPj8fDwoKCgAIDhw4eTn59PZWUl/v7+pKWlGce7T58++Pj4kJ+fD7TMMxgZGUlZWRm+vr4MHz6cjRs3GtdsYGCgcbwHDx5MUVERJSUlbY53dHQ0oaGhxvFOTU3lxIkTnDhxwrhmW493ZGQkkZGR5ObmApCSkkJ5ebnxFP/UazYiIoLY2Fh27txpXLPV1dXG8R47dizbtm2jvr6esLAw4uPjjWs2KSmJhoYGjhw5YlyzrfeIyspKJkyYcM73iNbj7ar3iObmZnr16kV0dDS7du2itLSUjIyMc75H1NXVERoa6hb3iKysLIKCguxyj8jLywNwi3tERUUFHh4tnXrnco8YNmwYhw4dcpt7RENDAwkJCed8j2gvjigvL6fLqG4QGBio8vPzlVJKXXXVVeqJJ55QSilVUFCg/Pz8urNJ9X9PBNXKlSuN10eOHFGAWrt2rc16DzzwgMrIyOjydi+99FIVGRmp/P39VZ8+fdps71Tq6upUeXm58XPo0CEFqPLy8rP26Srr16932LbtxcAZL6uU6S+rwTe/YrdtuoO3I9DRW0dnpcRbJ3R0Vkq8HUF5eXmX445u5cQNGTKEF198kW+//ZZVq1Zx+eWXA3D06FF69erVnU06lNWrV1NcXExNTQ2HDx9m/PjxHa7r6+tLSEiIzY+jiY2Ndfg+zhUfL08AGpuszFv8tV226Q7ejkBHbx2dQbx1QkdnEG+z6VZ36pNPPsnUqVNZtGgRs2bNIj09HYB///vfRjerPYiMjMTT09N4HN/KsWPHHH4AlyxZwpIlS2hubgYc253a0NBAdHS0S3eVTD2/D2/+t6VL4qO1+/nDjPRz7iqpqKigtrbWLbpK7NmdevDgQYqKityiq8Re3al9+/YlLy/PbbpK7NWd6uXlRXR0tHbdqQ0NDQQHB2vVnVpaWkpRUZF23am9evUyXHXqTo2KiqKgoMD07tRuDWwAaG5upqKigvDwcGPZgQMHCAgIIDo6ujub7HBgQ0ZGBs8//zzQMrAhISGBuXPnntXAhu5yNgmG3SUrK4tx48Y5ZNv2ZOCMpcbvQ5J6sfLxa89pe+7ibW909NbRGcRbJ3R0BvF2BGcTd3SrO7W2tpb6+nojgCsoKODZZ58lLy/vrAO4qqoqcnJyjG/7+fn55OTkGN+c5s2bxz/+8Q9effVVdu3axV133UV1dTVz5szpTtOFc2DSyHjj9x35J7nu9x+Y1xhBEARB0JxuPYm77LLLuO6667jzzjspKysjNTUVb29vTpw4wTPPPMNdd93V5W2tWbOGSZMmtVk+a9Ysli9fDsDixYtZtGgRRUVFjBgxgueee87hkf+p3am7d+/myy+/dFh3amNjI6NHj3aLrpKMu97BesoVkxTtx2PTB3Srq6SyspLIyEi36CqxZ3fq9u3b8fb2douuEnt1pyYmJlJWVuY2XSX26k718fEhNTVVu+7UxsZGBg8erFV3an5+PjU1Ndp1p8bExBjHQafu1N69e6OUclh3ampqapeexHUriIuMjOS///0vQ4YM4eWXX+b5559ny5YtvPfeezzyyCPGRdUTcEZ36p49e0hJSXHItu3NitW7ePSVtTbLFtx6PtMz0856W+7kbU909NbRGcRbJ3R0BvF2BA7vTq2pqTHmIv3iiy+47rrr8PDw4LzzzjMicKHrlJSUmN2ELjM9M40Ft55vs+yPy7tXs8+dvO2Jjt46OoN464SOziDeZtOtIG7AgAF88MEHHDp0iM8//5zLLrsMgOPHjzulJEdPw8urW4OETWN6ZhqPzvlfmZYmq+pW2RF387YXOnrr6AzirRM6OoN4m023WvHII48wY8YMfvWrX3HJJZcYdde++OILRo4cadcGmoUzS4xAy2ARd8l3AZgyOoVlHwVwsLgGaCk78tPzQ88q36WpqYnc3Fy3yHexZ05cU1MTWVlZbpHvYq+cuGHDhmlZYiQ4OJja2lrtcuKg5UmFTjlxwcHBZGVlaZcTN3ToUC1LjCQnJ7t3iZGioiIKCwtJT083ptzYsGEDISEhpKamdmeTLokzcuI2bNhg1/p6zuRcyo64s/e5oKO3js4g3jqhozOItyNweE4ctETGI0eO5OjRo0aEnpGR0aMCOGfRzTjaJbhsbD/j9x35J1mxuuuDWtzZ+1zQ0VtHZxBvndDRGcTbbLoVxFmtVh577DFCQ0Pp168f/fr1IywsjD/+8Y9YrVZ7t7HHExUVZXYTus3iX2Xi6WExXi/85/ouf9advc8FHb11dAbx1gkdnUG8zaZbOXG///3vWbp0KU888QQTJkwA4LvvvmP+/PnU1dXx+OOP27WRZuDsabd69+7tVvkuKSkpRr7LTy+K5dU1LbksdY1Wfvbnf/PsLy7p0rRbDQ0NbpHvYs+cuCNHjlBcXOwW+S72yolLSEjQMifO29tby5y41utSp5y4yspKLXPioqKitMyJi46Odt+cuLi4OF588UWuvvpqm+Uffvghd999t9HQnoBMu9U1hs1aTn1js/F69xu3dfqZnuDdHXT01tEZxFsndHQG8XYEDs+JKykpaTf3LTU11WVqpwjO5Xc3217MU2VKLkEQBEFwKN0K4tLT01m8eHGb5YsXL2b48OHn3CjdGDRokNlNOGemZ6aROSbBeL0j/2SnteN6gnd30NFbR2cQb53Q0RnE22y6FcQ99dRTvPLKKwwePJjbbruN2267jcGDB7N8+XKefvppe7exx9NTnl7+bd6PbAY5fLR2/xlHq/YU77NFR28dnUG8dUJHZxBvs+lWEDdx4kR2797N1KlTjUmtr7vuOnbs2ME///lPe7exx1NcXGx2E+zGI7PH27w+05RcPcn7bNDRW0dnEG+d0NEZxNtsul3stz22bt3KqFGjjBGd7sypo1N3797Nl19+6bDRqaWlpUycONFtRp6B7ehUsB3F8/fVRXy744RxLFPjQ1l8T0abkWeFhYUkJia6xcgze45OXb9+PeHh4W4x8sxeo1OtVisRERFuM/LMXqNTq6qqOP/887UbnVpaWkpGRoZWo1M3bNhAYGCgdqNTKysrsVhaemB0Gp3a2NhIfHy8w0anpqamdmlggwRxneCM0ak9kSt/+z67D5X+7/X5/Xlm7iQTWyQIgiAIro9TZmwQ7Efrt7uexEdPXtcmP+50eqJ3V9DRW0dnEG+d0NEZxNtsJIhzAXrSk8tTOT0/7orfvGfzuqd6d4aO3jo6g3jrhI7OIN5mc1YzNlx33XVnfL+srOxc2qItvXr1MrsJDmF6Zhpvf53HjvyTAOw5XMY1D63kw4VTgZ7r3Rk6euvoDOKtEzo6g3ibzVk9iQsNDT3jT79+/bjlllsc1dYeS3R0tNlNcBgrH78WL8//XWa7Ckr45XNfAT3b+0zo6K2jM4i3TujoDOJtNnYd2NCTkNGp3R+devrIs22FHix6e5vN8b39sgSGxTbL6FQXH3lmr9GpTU1NREZGus3IM3uNTq2srGTChAkyOpWePzo1KyuLoKAg7UanVlRU4OHR8kVdp9GprfN+96jRqT0RmTvVPqxYvYtHX1lrvPbwsPDqvUN6vHd76HC+T0dHZxBvndDRGcTbEcjoVDcjJSXF7CY4nOmZaVx5fn/jtdWquO+VPSa2yDx0ON+no6MziLdO6OgM4m02EsS5ABUVFWY3wSk8M3cSQ5MijdcllfVc/eD7JrbIHHQ536eiozOIt07o6AzibTYSxLkArTkjOvD+49cQGxFgvM49WMrU339gXoNMQKfz3YqOziDeOqGjM4i32UgQJzidbxZPx8vzf4WAd+Sf5FfPf21iiwRBEATB/ZCBDZ0g0245htMHOoBMzSUIgiAIMrDBzWgdcq0T0zPTuC0z3mbZR2v3M29xz38ip+P51tEZxFsndHQG8Tabs5qxQSdOrRMHLfOkObJOXG1trdvUgIKzqxPXUQ2o9DgrFw2J5JsdJ4zj/tHa/dTU1DBnYssACFepAWXPOnHHjx8nKyvLLWpA2bNOXF5entvUgLJnnbja2lot68SVlJRoVSeutLSUrKws7erE1dfXG6661YkrKChwWJ24riLdqZ3gjO7U3bt3M3DgQIds25Vp9f7V81/z8br9Nu/15K5VHc+3js4g3jqhozOItyOQ7lQ3Iy4uzuwmmEKr91/vncQV4/vbvNeTu1Z1PN86OoN464SOziDeZiNBnAvQ2qWgG6d6dxTI9cRRqzqebx2dQbx1QkdnEG+zkSBOcBn+eu8km1kdAD5et1+7OnKCIAiC0BUkiHMBkpOTzW6CKbTn/czctoHcjvyTXPu7lc5qlsPR8Xzr6AzirRM6OoN4m40EcS5ATU2N2U0whY682wvkdh4o4cJ7VjijWQ5Hx/OtozOIt07o6AzibTYSxLkArcPqdeNM3s/MncSCW8+3WXastIaBM5a6/YAHHc+3js4g3jqhozOIt9lIECe4LNMz01hw6/l4WGyXf7R2P7/8f1+Z0yhBEARBcBGkTlwnOKNOXHNzM56eng7ZtitzNt4T732TwpPVNssiw/xZ+7cZjmiaQ9HxfOvoDOKtEzo6g3g7AqkT52a0VtHWjbPx/u/zNzGsf6TNshNltQycsZR7//qlvZvmUHQ83zo6g3jrhI7OIN5mI0GcC1BXV2d2E0zhbL3f+9M1bQY8AHy+8QCDZi7ln5+5Rt2eztDxfOvoDOKtEzo6g3ibjXSndsCpc6fu3r2bL7/80mFzp1ZWVjJhwgS3mRcR7DN3anFxMX379u3WvIi/fCWXk5VNbc5bsL8Xi28fRFBgoMvOnZqdnU1wcLBbzItor7lTAwIC8PHxcZt5Ee01d2pDQwNjxozRbu7UyspKRo0apdXcqVu3bsXLy0u7uVOVUlRXt6S66DR3qre3N7169XLY3Kmpqald6k6VIK4TnJETV1NTQ0BAgEO27cqcq/e8xV/z0dr97b7nYYFH55zP9My0bm/fUeh4vnV0BvHWCR2dQbwdgeTEuRmt0bpunKv3M3MnsfuN29rkygFYFTz6yloGzVzKay7Wzarj+dbRGcRbJ3R0BvE2GwniBLfnvT9dw+43biMuMrDNe0rBn15bz8AZS7niN++b0DpBEARBcAxeZjdAaMnx0BF7e6957iYArv/Dh/yw/0Sb9/ccLmXgjKUAXDwinr//5jK77r+r6Hi+dXQG8dYJHZ1BvM1GnsS5AE1NbRP0dcBR3q1P5trrZm1lTc4hBs5YysAZS1n4+nqamq0OaUt76Hi+dXQG8dYJHZ1BvM1GgjgXoHXUj2442rs1mGuZ9cHS4XrLPtnB4JuXMXDGUi69720qaxoc2i4dz7eOziDeOqGjM4i32Uh3qtDjmZ6ZZoxSPdOIVoBDxZWM/tk/jdf9YkJ4709XExLo6/B2CoIgCMLZICVGOsEZJUYaGxvx9vZ2yLZdGbO9O8qdOxMWC9x97QjmXj8ST4/uPcg229sMdHQG8dYJHZ1BvB2BlBhxM1oLDeqG2d6t3a1d6XJtRSlYsjKHtJ8uM3LqBs1cyl/e3EhdQ9dyJMz2NgMdnUG8dUJHZxBvs5HuVBegpqbG7CaYgit5n9rlCrBi9S4WLFuHtQsPqpWCl/69jZf+vc1YZgFuvGQQc348lH6xITZP7VzJ21no6AzirRM6OoN4m40EcS5AUFCQ2U0wBVf2Pj2og7PrflXAW1/l8dZXeTbLLRaYMjqGfinVxIQHYOnC07+egCufa0ci3vqgozOIt9lITlwnOCMnrq6uDj8/P4ds25XpCd7dyatrD4sFbr5sMD+7aniPDO56wrnuDuKtDzo6g3g7grOJOySI6wRnBHFZWVmMGzfOIdt2ZXqqt70CO2gJ7m6cNIhZlw8hITYEHy9Pu2zX2fTUc90Z4q0POjqDeDuCs4k7enx36qFDh7j55ps5fvw4Xl5ePPzww/zkJz8xu1lCD+a9P13T7vLuBHdKtd8tCxAa6MPf7v8RyXGhhAf79bind4IgCMKZ6fFP4goLCzl27BgjRoygqKiI0aNHs3v3bgID286z2R7OeBJXWFhI7969HbJtV0a8W7Dnk7tWRqZEM//W80nqHYqfj/nf1eRc64WO3jo6g3g7AnkSdwq9e/c2DnRsbCyRkZGUlJR0OYgTBEdjzyd3rWzZc5xrHvqgzfKQQB+evXcS/ePCiI0IxMNDnt4JgiC4K6YHcd988w2LFi1i06ZNFBYWsnLlSq699lqbdZYsWcKiRYsoKioiPT2d559/noyMjLPe16ZNm2hubiY+Pt5OrbcPBw8e1PKbjHifmY6CO4CLf/EmR09Un/W+K6obuPWJz9t9b2RKNI/OGU9CTAhB/j5nve0zIedaL3T01tEZxNtsTA/iqqurSU9P59Zbb+W6665r8/5bb73FvHnzePHFFxk3bhzPPvsskydPJi8vj+joaABGjBjR7mS0X3zxBXFxcQCUlJRwyy238I9//MOxQoLgBNY8d1O7yzubVuxMbNlznGt/92Gb5RYLzJo8hJ9dNZyoMH/JvRMEQXARXConzmKxtHkSN27cOMaOHcvixYsBsFqtxMfHc++99/Lggw92abv19fX86Ec/4vbbb+fmm2/udN36+nrjdUVFBfHx8VJixAGIt3Pp7tO7MzE2NYaFP7+IvlHBZ+yalXOtFzp66+gM4u0IekxOXENDA5s2beKhhx4ylnl4eJCZmcm6deu6tA2lFLNnz+aSSy7pNIADWLhwIQsWLGizPDs7m8DAQEaNGsWuXbuora0lODiYpKQktm1rqdTfr18/rFYrhw4dAlqeEO7du5eqqioCAwMZOHAgW7ZsAaBv3754enpSUFBARUUFF1xwAQcOHKCiogI/Pz+GDBnCpk2bAIiLi8PPz4/9+1uesAwdOpTDhw9TVlaGj48PI0aMYMOGDUBL3l9QUBB79+4FIC0tjWPHjlFSUoKXlxejR49mw4YNKKWIiooiPDyc3bt3AzBo0CBKSkooLi7Gw8ODsWPHkp2dTXNzM7169SI6Oppdu3YBkJKSQkVFBceOHQNagu3NmzfT2NhIeHg4cXFx7NixA4Dk5GRqamooLCwEYMyYMWzfvp3jx48THx9PQkICP/zwAwCJiYk0NTVx+PBhAEaNGkVubi41NTUEBQWRnJzM1q1bAUhISABaHmsDpKens2/fPqqqqggICCA1NZXNmzcbx9vLy4sDBw4AMGzYMA4ePEh5eTl+fn4MHTqU7OxsoCWPMiAggH379gEwZMgQjh49SmlpKd7e3owaNYqsrCwAYmJiCAkJYc+ePcbxPn78OCdPnsTT05MxY8awceNGrFYrUVFRREREsHHjRkJCQhg4cCClpaUUFxdjsVjIyMhg06ZNNDU1ERERQUxMjHG8BwwYQFVVFUVFRQBkZGSQk5NDQ0MDYWFh9O3bl+3btwPQv39/6urqOHr0KACjR49mx44dPDkziZCQEBITE22u2fmvbeKrnKIz/2F0wMbcY2T+6p02y4cmhvHQ9BE0VhXj6WHB19cXf39/ysrK8PX1Zfjw4WzcuBFouWYDAwON4z148GCKioooKSlpc7yjo6MJDQ01jndqaionTpzgxIkTxjXberwjIyOJjIw0psdJSUmhvLyc48ePA7bXbEREBLGxsezcuRNouWarq6uN4z127Fi2bdtGfX09YWFhxMfHG9dsUlISDQ0NHDlyBMDmHlFfX298tvV4d+ceATB8+HC3uUdUVFQwevToc75H1NXVERoa6hb3iC1btuDj42OXe0ReXstodGffI+rq6tq9RzQ3NxvHe+TIkezevZvq6mqCgoJobm6mtrYWgPj4eDw8PGyu2fz8fCorK/H39yctLc043n369MHHx4f8/HzjeB86dMht7hGenp5ERUWd8z2ivTiivLycruLST+KOHj1Knz59WLt2LePHjzfW+81vfsN///tf46Sdie+++46LLrqI4cOHG8v++c9/MmzYsHbXN+NJnNTZ0Qt38rb3yNkxg1qe3MVHn/nJXU/Bnc61PdHRW0dnEG9H0GOexNmDCy64AKvV2uX1fX198fX1dWCL2hIQEODU/bkK4u36dDS4oru5d9l5x/jRPNsnd605d3dNHUF4cM/qlnGnc21PdPTW0RnE22xcOoiLjIzE09PTeBzfyrFjx4iNjXXovpcsWcKSJUtobm4GHNudarVaqa2tdZuuErBPd2pNTQ25ublu0VViz+7UqqoqsrKy3KKrZMCAAeTk5AC2XSXTxgbw+G0z2nSV3Pb4h6zf3fWuAGgpaLz8sx0s/2yHzfKoEG8emJrMjy89jy2bWs6Nu3WnBgYGUltbq113qtVqpaSkRKvuVC8vL7KysrTrTu3fv7/hqlN3amJiIgUFBdKdeiodDWzIyMjg+eefB1oGNiQkJDB37twuD2w4F2TaLcch3vpwy/x3zzq4aw+LBX525TDmXjcKf1+X/g4K6HmuQU9vHZ1BvB2BW3WnVlVVGd8IAfLz88nJySEiIoKEhATmzZvHrFmzGDNmDBkZGTz77LNUV1czZ84cE1stCMLZcM+UeF6bf0Ob5Webc6cU/OM/P/CP//xgs3z8kN48/8tLCQl0biqEIAiCmZj+JG7NmjVMmjSpzfJZs2axfPlyABYvXmwU+x0xYgTPPfecwyP/U7tTd+/ezZdffumw7tTa2lrGjRvnNl0lYJ/u1NLSUmJjY92iq8Se3ak5OTn4+/u7RVdJR92pcHZdJVFRUTQ2Nnapq2RnsQ9PrtiM9RzvTOlJYTx770QOF7Rs14zu1NbjpFt3am1tLcOHD9eqOzU3N5fm5mbtulMDAwONc6xTd2poaCj+/v4O605NTU3t0pM404M4V8cZ3anHjh0jJibGIdt2ZcRbH+zhbI+RshcM78OLv/4RPl6e57SdrqLjuQY9vXV0BvF2BG7VnSrAgQMHtPwjEG99sIdzeyNlz3aU7HfbjjD0luXG67AgX1Y+fg1xkUEOmYlCx3MNenrr6AzibTYSxAmC4LY8M3cSz8y1Tcc4m8CurKqeSfe9bbNs/JDevPTAZfj5yO1REATXRrpTO8CZOXFNTU2MGjXKbfJdwD45cVVVVfTq1cst8l3smRO3Y8cOvLy83CLfxV45cQMGDODEiROm5bss+fRQt0fHxoR589FTN5K/N/es8138/PwYOHCgdjlxTU1NpKWlaZUTV1BQQFVVlXY5cX369DHaq1NOXN++fWlqapKcOFfHGTlxubm5pKamOmTbrox464MrOnd3LtnwYF8+evI6osI6L/bpit7OQEdvHZ1BvB2B5MS5GWdT2K8nId764IrOa567yeZ1V7thSyvrmXD3CuO1xQL3TxvDHVent1nXFb2dgY7eOjqDeJuNBHEugJ9fz5pqqKuItz64g/Pp+XUrVu9iwbJ1WDvprFAKnn4zm6ffbOlqs1jglzeM4s5rR7iFtyPQ0VtHZxBvs5Hu1E5wRndqc3Mznp7OKXngSoi3PvQU5+50wfaLCeGDhdcS6OftoFa5Hj3lfJ8NOjqDeDuCs4k7JIjrAGcObCgtLWXixIluk7QM9hnYUFhYSGJiolskLdtzYMP69esJDw93i6Rlew1saGpqIjIy0m2Slrta7PeRFXvJP17H2TAmpRf3Xh6Lh4elxw5sKC0tJSMjQ6uBDVlZWQQFBWk3sKGiogIPDw9Ar4ENDQ0NJCQkyMAGV0fmTnUc4q0Pujif7ZM6iwXm3TiGn1/TNp/OndHlfJ+Kjs4g3o5ABja4Gb179za7CaYg3vqgi/OpgyVacurWnnH6MKXgL29l85e3Wp7w9I8L5aMnr8PL08PRTXUoupzvU9HRGcTbbCSIcwECAjovVdATEW990NF5emYaPxoRRWRkJNC10a/7j5Yz+OZlQMtTugemj+VnVw53eFvtjY7nW0dnEG+zce+vez2E1n5+3RBvfdDRGWy9n5k7id1v3Gb8DOsfecbPKgVPvbGRgTOWMnDGUn6+6AtHN9du6Hi+dXQG8TYbyYnrABnYIAMbZGCDDGw416TlyspKJkyY0Ok94sO1B3ltzdEzdr2eSp8IX7746zTjeLvaPUIGNsjABpCBDTKwwQVwxsCGqqoqgoKCHLJtV0a89UFHZ+i+9/V/+JAf9p/o0roRIX58u3g63l6u07Gi4/nW0RnE2xGcTdzhOn/1GtP6bUg3xFsfdHSG7nu/96drjG7XK8/vf8Z1SyrqGHLLMgbOWMqgmUt57bMd3dqnPdHxfOvoDOJtNhLEuQClpaVmN8EUxFsfdHQG+3ifmku34Nbz8bBYOlxXKfjTa+uNgO5VkwI6Hc+3js4g3mYjo1NdAG9vfSq5n4p464OOzmB/7+mZaUzPTDNen6kunVLw+Gvrefy19Vgs8IdbzuPmyUPs2p6O0PF86+gM4m02khPXCc7IiRMEQThXuppHFxHix3dLprt9LTpB6KnItFt2QEanyuhUGZ0qo1OdNTrV3tNuLfzgCDsLOu/uiQ3z5oPHr+bkiWIZnSqjU2V0qoxO7XnItFuOQ7z1QUdncA3vrk4FNqBPGJ8sut4u+3QFb2ejozOItyOQabfcjJiYGLObYArirQ86OoNreJ86FdiZArq9R8oYOGMpAD8a048l8zK7vU9X8HY2OjqDeJuNJEW4ALrm2om3PujoDK7nvea5m7o0ynVVdoExwvWfX5z9CFdX83YGOjqDeJuNBHEuQGvfvW6Itz7o6Ayu6z09M43cf93aaUCnFPxxeUvJkvPveoOuZt+4qrcj0dEZxNtspDtVEARBY04tWzJv8dd8tHZ/u+udKK9l0MxXAEjtF8G/F051WhsFQWgfGdjQCc4Y2FBRUeEyj2adiXjrg47O4N7eXSlZEhHsy7dLZrSZ8sudvbuLjs4g3o7atky75Ua0DmfWDfHWBx2dwb29T536Ky4ysN11SirrjSm/rvrt+8Zyd/buLjo6g3ibjQRxLsDJkyfNboIpiLc+6OgMPce7dUDEmeZxzTtUagyGWPFlnhNb5xr0lHN9toi3uUh3agc4s9hvWVkZF110kXbFfouKiujXr59bFPK0Z7HfrKwswsLC3KKQp72K/SqlCA8Pd5tCnvYq9ltVVcX555/v9GK/zrhH/PKVPE5WNnImzkuN5J7JsW2Od08s9rtx40YCAgK0K/ZbVVVlnG+div02NTXRt29fKfbr6si0W4IgCB2zYvUuFixbh7WTfyU/Pi+JZ39xiZNaJQjui+TEuRmt3zZ0Q7z1QUdn0MP71HIlw/pHdrjeJ+vzGThjKVc/+H6H67gzOpzr9hBvc5EgzgWwWq1mN8EUxFsfdHQG/bxbB0OcNzC0w3VyD/4vd+6NVbuc2DrHotu5bkW8zUWCOBcgKirK7CaYgnjrg47OoK/3/FtGGSNbOxoMoRTMX7aWgTOWsuyT7U5uof3R9VyLt7lIEOcCREREmN0EUxBvfdDRGcQb4Jm5kzqdGWLh61kMnLGUW5/4zFlNtDtyrvXCVbwliHMBWkci6YZ464OOziDep9KV3Lnvth0xulr/tWqno5tpV+Rc64WreEsQJwiCIDiV1ty5joI5pWDBsnUtAyEeWunk1gmC+yAlRjrBGSVGSktLCQ8Pd8i2XRnx1gcdnUG8u8qZ5mxtJSLEj3UvzMDSQZes2ci51gtHekuJETejtLTU7CaYgnjrg47OIN5dpTVv7kxP50oq6hg08xUGzljKNS74dE7OtV64ircEcS5AcXGx2U0wBfHWBx2dQby7Q2tXa8tAiPbX2VVQwsAZS5lw9xvd3o+9kXOtF67i7WV2A1yVU6fdAsjOznbYtFulpaXU1ta61JQ6zph2q7S0lNzcXLeYUsee026VlpaSlZXlFlPq2GvarebmZvLy8txmSh17TbtVWVlJbW1tj5x260z3iNLSUkpKSs7pHjE4qoFXfzGU0NBQ7li8haKSGk6nuKyWgTOWEhbgxfO3p5p6j6iuriYrK0u7abeUUoarTtNuNTQ0UFBQ4LBpt7qK5MR1gky7JQiC4Bq0TPG1FusZ/ms9PPs8br5siPMaJQh2RnLi3IzWb9O6Id76oKMziLe9aSlT0pI7FxcZ2O46f1y+noEzlvLzRV84pA0dIedaL1zFW4I4F6CpqcnsJpiCeOuDjs4g3o5kzXM3nTGY+3rLIQbOWMqPfvU2NXWNDm+PnGu9cBVvyYlzAVyl8rOzEW990NEZxNsZrHnuJgCu/8OH/LD/RJv3C45VMuLW1wD48XlJPPuLSxzSDjnXeuEq3pIT1wnOyImrqKjQMt9OvPVBR2cQbzPoSs25tH4RfLhwql33K+daLxzpLTlxbkbr6CLdEG990NEZxNsMujJXa2uJktSZS3ljlX3aKudaL1zFW4I4QRAEocdx6lytV57fv911rArmL1vLwBlLue3Jz5zcQkE4d6Q7tROc0Z168uRJevXq5ZBtuzLirQ86OoN4uxotJUrWYT3Dvz0PCzwy+3xm/CjtrLbtqs6ORrztj3SnuhlVVVVmN8EUxFsfdHQG8XY1Tn0619H0Xqc+nRt6yzJWrO5at5mrOjsa8TYXCeJcgNaKz7oh3vqgozOItyvTOr1XR12tAA1NVh59ZS1X/Ob9TrfnDs6OQLzNRYI4QRAEQVtaB0Kc6encnsOlDJyxlHE/f53aeteoDyYIIDlxneKMnDilFJYORlH1ZMRbH3R0BvF2V7pSpiQuMpCv/980w9PdnbuLeNsfyYlzM1onGNcN8dYHHZ1BvN2VrpQpOXqimkEzX2HQzKUs+2S72zt3F/E2F5mxwQVoaGgwuwmmIN76oKMziLe7Mz0zjemZLaNUL/7Fmxw9Ud1mHaVg4etZAFjYzG1XDuM3MzKc2k4z6Snn+mxxFe8e/ySurKyMMWPGMGLECIYOHco//vEPs5vUhrCwMLObYArirQ86OoN49yRa52o909M5Bbz80Q8MnLGUq37b+WCInkBPPNddwVW8e3xOXHNzM/X19QQEBFBdXc3QoUPJzs7ucn0XZ+TEVVdXExjY/iTOPRnx1gcdnUG8ezodzdd6OvfdMIp7rhvphBY5H13O9ek40lty4k7B09OTgIAAAOrr61FK4Wpx6/bt281ugimItz7o6Azi3dNpLVOy+43bSIr263C9//fuZgbOWMqPfvUOxWU1Tmyh49HlXJ+Oq3ibHsR98803XHXVVcTFxWGxWPjggw/arLNkyRISExPx8/Nj3LhxbNiw4az2UVZWRnp6On379uWBBx4gMrL9YeSCIAiC0B0emz7glO7W9tcpOFbBhLtXMHDGUgbNXMprn+1wbiOFHofpQVx1dTXp6eksWbKk3fffeust5s2bx6OPPsrmzZtJT09n8uTJHD9+3FinNd/t9J+jR48CLX3XW7duJT8/nzfeeINjx4512J76+noqKipsfhxN//4dF5vsyYi3PujoDOKtE63OLbNCnLnuHLQMiPjTa+sZOGMpl9z3Fk3NVmc11a7oeK7BdbxdKifOYrGwcuVKrr32WmPZuHHjGDt2LIsXLwbAarUSHx/Pvffey4MPPnjW+7j77ru55JJLuOGGG9p9f/78+SxYsKDN8i+//JLAwEBGjRrFrl27qK2tJTg4mKSkJLZt2wZAv379sFqtHDp0CGgJLvfu3UtVVRWBgYEMHDiQLVu2ANC3b188PT0pKCigtraWcePGceDAASoqKvDz82PIkCFs2rQJgLi4OPz8/Ni/v6Vu0dChQzl8+DBlZWX4+PgwYsQI4+lkbGwsQUFB7N27F4C0tDSOHTtGSUkJXl5ejB49mg0bNqCUIioqivDwcHbv3g3AoEGDKCkpobi4GA8PD8aOHUt2djbNzc306tWL6Ohodu1qmYImJSWFiooKIyAeN24cmzdvprGxkfDwcOLi4tixo+VbZnJyMjU1NRQWFgIwZswYtm/fTmlpKbGxsSQkJPDDDz8AkJiYSFNTE4cPHwZg1KhR5ObmUlNTQ1BQEMnJyWzduhWAhIQEAA4ePAhAeno6+/bto6qqioCAAFJTU9m8ebNxvL28vDhw4AAAw4YN4+DBg5SXl+Pn52fkSgL07t2bgIAA9u3bB8CQIUM4evQopaWleHt7M2rUKLKyWkajxcTEEBISwp49e4zjffz4cU6ePImnpydjxoxh48aNWK1WoqKiiIiIICcnB39/fwYOHEhpaSnFxcVYLBYyMjLYtGkTTU1NREREEBMTYxzvAQMGUFVVZVQJz8jIICcnh4aGBsLCwujbt6/xeL9///7U1dUZX2JGjx7Njh07qKurIyQkhMTERJtrtrm52TjeI0eOZPfu3VRXVxMUFMSAAQOMofTx8fF4eHhQUFAAwPDhw8nPz6eyshJ/f3/S0tKM492nTx98fHzIz88HoFevXjQ3N1NWVoavry/Dhw9n48aNxjUbGBhoHO/BgwdTVFRESUlJm+MdHR1NaGiocbxTU1M5ceIEJ06cMK7Z1uMdGRlJZGQkubm5xjVbXl5ufAE89ZqNiIggNjaWnTt3GtdsdXW1cbzHjh3Ltm3bqK+vJywsjPj4eOOaTUpKoqGhgSNHjhjXbOs9QilFenr6Od8jWo+3u9wjamtrGT58+DnfI+rq6ggNDXWLe8SuXbuwWq3t3iNWbz3Jordz6Mp/27BAb/4yOwUfLw+3uEcEBAQYf1Pnco8YNmwYhw4dcpt7RHBwMIGBged8j2gvjigvLyc1NbVLOXEuHcQ1NDQQEBDAu+++axPYzZo1i7KyMj788MNOt3ns2DECAgIIDg6mvLycCRMmsGLFCoYNG9bu+vX19dTX1xuvKyoqiI+Pd+jAhqysLMaNG+eQbbsy4q0POjqDeOvE2Th3dUAEwMUj4vn7by47l6Y5FB3PNTjW+2wGNrh0nbgTJ07Q3NxMTEyMzfKYmBgjcu6MgoIC7rjjDmNAw7333tthAAfg6+uLr6/vObVbEARBEDrivT9dA8CK1btYsGwd1jM8S1mTc4iBM5YC4GGB+28ay+1XDXdKOwXXx6WDOHvQ+kj5bFmyZAlLliyhubkZgOzsbId1p1qtVmpra92mqwTs051qtVrJzc11i64Se3anWq1WsrKy3KKrxF7dqWlpaeTl5blNV4m9ulMDAwOpra3VrjvVarVSUlKiVXdqaGgoWVlZZ3WPSAnzJPdftxrX7LxX91Bc9r+eoNOxKli0YiOLVrT87Vx5XiLXjPTHz8fTtHvEsGHDDFedulMTExMpKChwWHdqV+nx3annijPqxG3dupX09HSHbNuVEW990NEZxFsn7O3clflbT6VfTDC/u+U8Ghqb6RMZTHl1PROG9bFbezpCx3MNjvXuMd2pPj4+jB49mi+//NII4qxWK19++SVz5841t3F2pK6uzuwmmIJ464OOziDeOmFv52fmTuKZuZOAruXQFRyr5OeLVtks87DAo3PON6YOcwQ6nmtwHW/Tg7iqqirjsT5Afn4+OTk5REREkJCQwLx585g1axZjxowhIyODZ599lurqaubMmWNiq+2Lo57wuTrirQ86OoN464QjnVtz6KAlj+6Pr67vUkkSq4JHX1nLo6+sxcvTwo2TBrHncCk7D5Rw7/UjufWKjvPDu4qO5xpcx9v07tQ1a9YwadKkNstnzZrF8uXLAVi8eDGLFi2iqKiIESNG8Nxzzzl8NMypOXG7d+92aImR5uZmRo4c6Tb5LmCfnLjq6mojWHf1fBd75sTt3LkTT09PrXLiUlJSKC4udpt8F3vlxPn7+5OSkqJdTlxzczOpqala5cQVFBRQVVVll3tEXl4eQJfuEYvf+p5X1xztUvmS02mtSfyzK4dx2RDfbt0j+vbta/yN6ZQT17dvX5qamqTEiKvjjJw4GaKtFzp66+gM4q0TruR8Nk/r2uO8Ib352ZXDKSqpZtLIeCJD/bFY2p+GwpW8nYmUGBEEQRAEwe5Mz0wz8uDOdoAEwPodhazfUdjuez8+L4lnf3HJObdRsA/yJK4DnNmdWldXR0ZGhtt0lYB9ulPLysqIiYlxi64Se3anbt26FT8/P626U2NjY6mrq3ObrhJ7dad6eHgwdOhQ7bpT6+rqGDZsmFbdqXl5eTQ1NTm9O7U794hXvy1l9eaW6/VcCAnw4i93jKW5rgx/Hw/6JSRo050aHh6Or6+vdKe6Os7oTj1y5Ah9+jh+KLirId76oKMziLdOuLvzxb94k6Mnqu2yLQ8LXH/xQC4d1Y9PsvZzrKSG399yHv4+XkSF+RPg522X/ZiJI8+3dKe6GYcPH3brP/7uIt76oKMziLdOuLvzmuduand5V2aVOB2rgne+3s07X+82ll394Mo263lY4Irx/Xnyrol4eXqcfaNPob6hibU7jnLe4Dj8fR0f2rjK+ZYgThAEQRCEdjk1v+5U7PHkzqrgP2v38592cvYCfL2YdmkqowfFEODrxZoth5h7/SiC/L3x8vRAKUVjkxUfb08AzrvzX1TXNeHt5cEfbjnPobXxXAnpTu0EZ3SnNjQ04OPj45BtuzLirQ86OoN464SOzmDrfa6jYu3J/dPGcOMlgwgN9KWhqRk/n86fWRWerCI4wIcg/87PoyPP99nEHRLEdYAzBzZUVFRwwQUXuE3SMthnYMPx48eJj493i6Rlew5s2LhxIyEhIVoNbPDx8SEgIMBtkpbtNbChddCSbgMbKioqGD16tFYDG7Zs2YKPj49bDGyw5z2iqanJmL2gq/eIR1bsJf+4OTMeDEnqRUaSH6AYObA3A/rFcuzIAYrK6nno9b14e1mYeWFvLh0eccZ7hIeHB9HR0TKwwdWROnGOQ7z1QUdnEG+d0NEZ7Ou9YvUunnlrExXV9bhiYBIV5s+J8lq8PT24fGQvnv7V1Q7ZjwxscDOCgoLMboIpiLc+6OgM4q0TOjqDfb07yr87nRWrd7Hw9Q3UNTTZbd9dobisFoCGJiv/yS7maafuvX3kSVwnOONJXH19Pb6+vg7Ztisj3vqgozOIt07o6Azu492dosedsfuN2+y6vVbkSZybkZOTo+VjePHWBx2dQbx1QkdncB/vZ+ZO4pm5bedp7wrtBYBJ0X72aNY5I0FcB5w6sAEgOzvbYQMbSktLqa2tdZukZbDPwIbS0lJyc3PdImnZngMbSktLycrKcoukZXsNbGhqaiIvL0+7gQ2VlZXU1tZqN7ChtLSUkpISrQY2VFVVkZWVpd3ABqvVarieyz3ClWdsmDY2gLsmX2Rzj2hoaKCgoMBhAxu6inSndoIzulOPHj1KXFycQ7btyoi3PujoDOKtEzo6g3g7grOJO86tRLJgFzw89DwN4q0POjqDeOuEjs4g3mbjGq3QnNZHz7oh3vqgozOIt07o6AzibTYSxAmCIAiCILghkhPXCc7IiautrcXf398h23ZlxFsfdHQG8dYJHZ1BvB2B5MS5Ga2jc3RDvPVBR2cQb53Q0RnE22ykxEgHOLvESFJSktuUDwD7lBgpLCzEw8PDLcoH2LPEyMGDB6msrHSL8gFSYuTcS4zY4x7Rerzd5R5RWlpKbGysViVGCgsLqays1K7ESEVFRY8vMdLePUJKjLgJzuhO3bZtG8OHD3fItl0Z8dYHHZ1BvHVCR2cQb0dwNnGHBHGd4IwgrrGxEW9vb4ds25URb33Q0RnEWyd0dAbxdgSSE+dmtD5e1g3x1gcdnUG8dUJHZxBvs5GcuE5ofVBZUVHhsH1UV1c7dPuuinjrg47OIN46oaMziLcjaN1uVzpKJYjrhMrKSqAlYVMQBEEQBMEZVFZWEhoaesZ1JCeuE6xWK0ePHiU4OBiLxWL37VdUVBAfH8+hQ4cclnPnioi3Pt46OoN46+StozOIt6O8lVJUVlYSFxfX6fRe8iSuEzw8POjbt6/D9xMSEqLVH0Er4q0POjqDeOuEjs4g3o6gsydwrcjABkEQBEEQBDdEgjhBEARBEAQ3RII4k/H19eXRRx/F19fX7KY4FfHWx1tHZxBvnbx1dAbxdgVvGdggCIIgCILghsiTOEEQBEEQBDdEgjhBEARBEAQ3RII4QRAEQRAEN0SCOEEQBEEQBDdEgjiTWbJkCYmJifj5+TFu3Dg2bNhgdpO6zcKFCxk7dizBwcFER0dz7bXXkpeXZ7POxRdfjMVisfm58847bdY5ePAgV1xxBQEBAURHR/PAAw/Q1NTkTJWzYv78+W2cUlNTjffr6uq455576NWrF0FBQVx//fUcO3bMZhvu5pyYmNjG2WKxcM899wA95zx/8803XHXVVcTFxWGxWPjggw9s3ldK8cgjj9C7d2/8/f3JzMxkz549NuuUlJQwc+ZMQkJCCAsL47bbbqOqqspmnW3btnHhhRfi5+dHfHw8Tz31lKPVzsiZvBsbG/ntb3/LsGHDCAwMJC4ujltuuYWjR4/abKO9a+SJJ56wWceVvDs717Nnz27jc/nll9us09PONdDu37nFYmHRokXGOu52rrvyv8pe9+01a9YwatQofH19GTBgAMuXL7evjBJM480331Q+Pj7qlVdeUTt27FC33367CgsLU8eOHTO7ad1i8uTJatmyZWr79u0qJydH/fjHP1YJCQmqqqrKWGfixInq9ttvV4WFhcZPeXm58X5TU5MaOnSoyszMVFu2bFGffPKJioyMVA899JAZSl3i0UcfVUOGDLFxKi4uNt6/8847VXx8vPryyy9Vdna2Ou+889T5559vvO+OzsePH7fxXbVqlQLU119/rZTqOef5k08+Ub///e/V+++/rwC1cuVKm/efeOIJFRoaqj744AO1detWdfXVV6ukpCRVW1trrHP55Zer9PR0tX79evXtt9+qAQMGqOnTpxvvl5eXq5iYGDVz5ky1fft2tWLFCuXv769eeuklZ2m24UzeZWVlKjMzU7311lsqNzdXrVu3TmVkZKjRo0fbbKNfv37qscces7kGTr0XuJp3Z+d61qxZ6vLLL7fxKSkpsVmnp51rpZSNb2FhoXrllVeUxWJR+/btM9Zxt3Pdlf9V9rhv79+/XwUEBKh58+apnTt3queff155enqqzz77zG4uEsSZSEZGhrrnnnuM183NzSouLk4tXLjQxFbZj+PHjytA/fe//zWWTZw4Ud13330dfuaTTz5RHh4eqqioyFj2wgsvqJCQEFVfX+/I5nabRx99VKWnp7f7XllZmfL29lbvvPOOsWzXrl0KUOvWrVNKuafz6dx3330qOTlZWa1WpVTPPM+n/4OzWq0qNjZWLVq0yFhWVlamfH191YoVK5RSSu3cuVMBauPGjcY6n376qbJYLOrIkSNKKaX+9re/qfDwcBvv3/72t2rQoEEONuoa7f1jP50NGzYoQBUUFBjL+vXrp/761792+BlX9u4oiLvmmms6/Iwu5/qaa65Rl1xyic0ydz7XSrX9X2Wv+/ZvfvMbNWTIEJt9TZs2TU2ePNlubZfuVJNoaGhg06ZNZGZmGss8PDzIzMxk3bp1JrbMfpSXlwMQERFhs/xf//oXkZGRDB06lIceeoiamhrjvXXr1jFs2DBiYmKMZZMnT6aiooIdO3Y4p+HdYM+ePcTFxdG/f39mzpzJwYMHAdi0aRONjY025zk1NZWEhATjPLurcysNDQ28/vrr3HrrrVgsFmN5TzzPp5Kfn09RUZHNuQ0NDWXcuHE25zYsLIwxY8YY62RmZuLh4UFWVpaxzkUXXYSPj4+xzuTJk8nLy6O0tNRJNudGeXk5FouFsLAwm+VPPPEEvXr1YuTIkSxatMimq8kdvdesWUN0dDSDBg3irrvu4uTJk8Z7OpzrY8eO8fHHH3Pbbbe1ec+dz/Xp/6vsdd9et26dzTZa17Hn/3gvu21JOCtOnDhBc3OzzQUAEBMTQ25urkmtsh9Wq5Vf/vKXTJgwgaFDhxrLZ8yYQb9+/YiLi2Pbtm389re/JS8vj/fffx+AoqKido9J63uuyLhx41i+fDmDBg2isLCQBQsWcOGFF7J9+3aKiorw8fFp888tJibG8HFH51P54IMPKCsrY/bs2caynnieT6e1ne15nHpuo6Ojbd738vIiIiLCZp2kpKQ222h9Lzw83CHttxd1dXX89re/Zfr06TaTgf/iF79g1KhRREREsHbtWh566CEKCwt55plnAPfzvvzyy7nuuutISkpi3759/O53v2PKlCmsW7cOT09PLc71q6++SnBwMNddd53Ncnc+1+39r7LXfbujdSoqKqitrcXf3/+c2y9BnOAQ7rnnHrZv3853331ns/yOO+4wfh82bBi9e/fm0ksvZd++fSQnJzu7mXZhypQpxu/Dhw9n3Lhx9OvXj7ffftsuf6SuztKlS5kyZQpxcXHGsp54noW2NDY2cuONN6KU4oUXXrB5b968ecbvw4cPx8fHh5///OcsXLjQJaYrOltuuukm4/dhw4YxfPhwkpOTWbNmDZdeeqmJLXMer7zyCjNnzsTPz89muTuf647+V7kL0p1qEpGRkXh6erYZ7XLs2DFiY2NNapV9mDt3Lh999BFff/01ffv2PeO648aNA2Dv3r0AxMbGtntMWt9zB8LCwhg4cCB79+4lNjaWhoYGysrKbNY59Ty7s3NBQQGrV6/mZz/72RnX64nnubWdZ/objo2N5fjx4zbvNzU1UVJS4vbnvzWAKygoYNWqVTZP4dpj3LhxNDU1ceDAAcB9vVvp378/kZGRNtd0Tz3XAN9++y15eXmd/q2D+5zrjv5X2eu+3dE6ISEhdvuCL0GcSfj4+DB69Gi+/PJLY5nVauXLL79k/PjxJras+yilmDt3LitXruSrr75q8/i8PXJycgDo3bs3AOPHj+eHH36wuRm2/oMYPHiwQ9ptb6qqqti3bx+9e/dm9OjReHt725znvLw8Dh48aJxnd3ZetmwZ0dHRXHHFFWdcryee56SkJGJjY23ObUVFBVlZWTbntqysjE2bNhnrfPXVV1itViOwHT9+PN988w2NjY3GOqtWrWLQoEEu273WGsDt2bOH1atX06tXr04/k5OTg4eHh9Hl6I7ep3L48GFOnjxpc033xHPdytKlSxk9ejTp6emdruvq57qz/1X2um+PHz/eZhut69j1f7zdhkgIZ82bb76pfH191fLly9XOnTvVHXfcocLCwmxGu7gTd911lwoNDVVr1qyxGWpeU1OjlFJq79696rHHHlPZ2dkqPz9fffjhh6p///7qoosuMrbROmz7sssuUzk5Oeqzzz5TUVFRLld64lTuv/9+tWbNGpWfn6++//57lZmZqSIjI9Xx48eVUi1D1RMSEtRXX32lsrOz1fjx49X48eONz7ujs1Ito6kTEhLUb3/7W5vlPek8V1ZWqi1btqgtW7YoQD3zzDNqy5YtxijMJ554QoWFhakPP/xQbdu2TV1zzTXtlhgZOXKkysrKUt99951KSUmxKTtRVlamYmJi1M0336y2b9+u3nzzTRUQEGBq2YkzeTc0NKirr75a9e3bV+Xk5Nj8rbeOylu7dq3661//qnJyctS+ffvU66+/rqKiotQtt9xi7MPVvM/kXFlZqX7961+rdevWqfz8fLV69Wo1atQolZKSourq6oxt9LRz3Up5ebkKCAhQL7zwQpvPu+O57ux/lVL2uW+3lhh54IEH1K5du9SSJUukxEhP4/nnn1cJCQnKx8dHZWRkqPXr15vdpG4DtPuzbNkypZRSBw8eVBdddJGKiIhQvr6+asCAAeqBBx6wqR+mlFIHDhxQU6ZMUf7+/ioyMlLdf//9qrGx0QSjrjFt2jTVu3dv5ePjo/r06aOmTZum9u7da7xfW1ur7r77bhUeHq4CAgLU1KlTVWFhoc023M1ZKaU+//xzBai8vDyb5T3pPH/99dftXtOzZs1SSrWUGXn44YdVTEyM8vX1VZdeemmb43Hy5Ek1ffp0FRQUpEJCQtScOXNUZWWlzTpbt25VF1xwgfL19VV9+vRRTzzxhLMU2+VM3vn5+R3+rbfWCdy0aZMaN26cCg0NVX5+fiotLU39+c9/tgl4lHIt7zM519TUqMsuu0xFRUUpb29v1a9fP3X77be3+cLd0851Ky+99JLy9/dXZWVlbT7vjue6s/9VStnvvv3111+rESNGKB8fH9W/f3+bfdgDy/8JCYIgCIIgCG6E5MQJgiAIgiC4IRLECYIgCIIguCESxAmCIAiCILghEsQJgiAIgiC4IRLECYIgCIIguCESxAmCIAiCILghEsQJgiAIgiC4IRLECYIgCIIguCESxAmCIJiIxWLhgw8+MLsZgiC4IRLECYKgLbNnz8ZisbT5ufzyy81umiAIQqd4md0AQRAEM7n88stZtmyZzTJfX1+TWiMIgtB15EmcIAha4+vrS2xsrM1PeHg40NLV+cILLzBlyhT8/f3p378/7777rs3nf/jhBy655BL8/f3p1asXd9xxB1VVVTbrvPLKKwwZMgRfX1969+7N3Llzbd4/ceIEU6dOJSAggJSUFP79738b75WWljJz5kyioqLw9/cnJSWlTdApCIKeSBAnCIJwBh5++GGuv/56tm7dysyZM7npppvYtWsXANXV1UyePJnw8HA2btzIO++8w+rVq22CtBdeeIF77rmHO+64gx9++IF///vfDBgwwGYfCxYs4MYbb2Tbtm38+Mc/ZubMmZSUlBj737lzJ59++im7du3ihRdeIDIy0nkHQBAE10UJgiBoyqxZs5Snp6cKDAy0+Xn88ceVUkoB6s4777T5zLhx49Rdd92llFLq73//uwoPD1dVVVXG+x9//LHy8PBQRUVFSiml4uLi1O9///sO2wCoP/zhD8brqqoqBahPP/1UKaXUVVddpebMmWMfYUEQehSSEycIgtZMmjSJF154wWZZRESE8fv48eNt3hs/fjw5OTkA7Nq1i/T0dAIDA433J0yYgNVqJS8vD4vFwtGjR7n00kvP2Ibhw4cbvwcGBhISEsLx48cBuOuuu7j++uvZvHkzl112Gddeey3nn39+t1wFQehZSBAnCILWBAYGtunetBf+/v5dWs/b29vmtcViwWq1AjBlyhQKCgr45JNPWLVqFZdeein33HMPTz/9tN3bKwiCeyE5cYIgCGdg/fr1bV6npaUBkJaWxtatW6murjbe//777/Hw8GDQoEEEBweTmJjIl19+eU5tiIqKYtasWbz++us8++yz/P3vfz+n7QmC0DOQJ3GCIGhNfX09RUVFNsu8vLyMwQPvvPMOY8aM4YILLuBf//oXGzZsYOnSpQDMnDmTRx99lFmzZjF//nyKi4u59957ufnmm4mJiQFg/vz53HnnnURHRzNlyhQqKyv5/vvvuffee7vUvkceeYTRo0czZMgQ6uvr+eijj4wgUhAEvZEgThAErfnss8/o3bu3zbJBgwaRm5sLtIwcffPNN7n77rvp3bs3K1asYPDgwQAEBATw+eefc9999zF27FgCAgK4/vrreeaZZ4xtzZo1i7q6Ov7617/y61//msjISG644YYut8/Hx4eHHnqIAwcO4O/vz4UXXsibb75pB3NBENwdi1JKmd0IQRAEV8RisbBy5UquvfZas5siCILQBsmJEwRBEARBcEMkiBMEQRAEQXBDJCdOEAShAyTbRBAEV0aexAmCIAiCILghEsQJgiAIgiC4IRLECYIgCIIguCESxAmCIAiCILghEsQJgiAIgiC4IRLECYIgCIIguCESxAmCIAiCILghEsQJgiAIgiC4If8feXMvkjZ7ElcAAAAASUVORK5CYII=", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.figure(figsize=(7, 4))\n", "\n", "plt.plot(np.array(train_losses), label='Train Loss', marker='o', color='#25599c', markersize=1)\n", "\n", "plt.xlabel('Epochs')\n", "plt.ylabel('Loss')\n", "plt.title('Training Loss Over Epochs')\n", "plt.yscale('log')\n", "\n", "plt.legend()\n", "plt.grid(True, which='both', linestyle='--', linewidth=0.5) \n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 11, "id": "7798eb68-d6b5-47ec-b845-cf3d60dccf23", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The MSE of the fit is 0.00063\n" ] } ], "source": [ "y_pred = model(X_test)\n", "mse = mean_squared_error(y_test, y_pred)\n", "\n", "print(f\"The MSE of the fit is {mse:.5f}\")" ] }, { "cell_type": "code", "execution_count": 12, "id": "f061b2a3-b14c-4c82-a74d-c077016ca7a2", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.figure(figsize=(7, 4))\n", "plt.scatter(y_test, y_pred, alpha=0.7, color='#a3630f', marker='x', label='Predicted vs Actual')\n", "plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], color='#25599c', linestyle='--', label='Perfect Fit')\n", "plt.xlabel('Actual Values')\n", "plt.ylabel('Predicted Values')\n", "plt.title('Model Predictions vs Ground Truth')\n", "plt.legend()\n", "plt.grid(alpha=0.3)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": null, "id": "88cc341a-8869-4dd9-be77-ef9bf2d8b5c1", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.12" } }, "nbformat": 4, "nbformat_minor": 5 }