{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "\"FMP\"\n", "\"AudioLabs\"\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "\"C4\"\n", "

Audio Thumbnailing

\n", "
\n", "\n", "
\n", "\n", "

\n", "Following Section 4.3 of [Müller, FMP, Springer 2015], we discuss in this notebook a music processing task referred to as audio thumbnailing. Our procedure is based on the following article.\n", "\n", "

\n", "

\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Introduction\n", "\n", "A prominent subproblem of music structure analysis is known as **audio thumbnailing**. Given a music recording, the objective is to automatically determine the most representative section, which may serve as a kind of \"preview\" giving a listener a first impression of the song or piece of music. Based on such previews, the user should be able to quickly decide if he or she would like to listen to the song or to move on to the next recording. Thus, audio thumbnails are an important browsing and navigation aid for finding interesting pieces in large music collections. In this notebook we describe a **repetition-based approach**. \n", "\n", "Often sections such as the chorus or the main theme of a song are good candidates for audio thumbnails. Such parts are typically repeated several times throughout the recording. Therefore, to determine a thumbnail automatically, most procedures try to identify a section that has on the one hand a certain minimal duration and on the other **many (approximate) repetitions**. The following figure shows some examples of musical structures as encountered in Western music. As for a rondo, a natural candidate of a thumbnail is the theme (A part). In a sonata, the beginning of the exposition (occurring at the parts E1, E2, and R) may be a suitable thumbnail. As for a pop song such as \"Yesterday\", the verse section (V1 part) is characteristic. Finally, in the [Brahms example](../C4/C4S1_MusicStructureGeneral.html) already considered in previous notebooks as our running example, either the A part or the B part may be chosen as thumbnail.\n", "\n", "\"FMP_C4_F04+.png\"\n", "\n", "One challenge is that repeating sections may show significant acoustic and musical differences in aspects that concern dynamics, instrumentation, articulation, and tempo. Therefore, a thumbnailing procedure for extracting repetitive segments from a given music recording need to be robust to certain variations. We now describe such a procedure that is based on enhanced [self-similarity matrices](../C4/C4S2_SSM.html) as well as [time warping techniques](../C3/C3S2_DTWbasic.html) for dealing with musical and temporal variabilities. As the main technical tool, we introduce a **fitness measure** that assigns a fitness value to each audio segment. This measure simultaneously captures two aspects. First, it indicates **how well** a given segment explains other related segments, and second, it indicates **how much** of the overall music recording is covered by all these related segments. The **audio thumbnail** is then defined to be the **segment of maximal fitness**. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Requirements on SSM\n", "\n", "The idea of our fitness measure is to simultaneously establish all relations between a given segment and its approximate repetitions. To this end, a [**self-similarity matrix**](../C4/C4S2_SSM.html) (SSM) is required. The following description of the fitness measure is generic in the sense that it works with general self-similarity matrices that only fulfill some basic **normalization properties**. From a technical point of view, only the properties\n", "\n", "\\begin{equation}\n", " \\mathbf{S}(n,m)\\leq 1 \\quad\\mbox{and}\\quad \\mathbf{S}(n,n)=1\n", "\\end{equation}\n", "\n", "for all $n,m\\in[1:N]$. Throughout this notebook, we again use [Hungarian Dance No. 5 by Johannes Brahms](../C4/C4S1_MusicStructureGeneral.html) as our running example. We start with an [enhanced](../C4/C4S2_SSM-PathEnhancement.html) SSM as introduced in Section 4.2.2.2 of [Müller, FMP, Springer 2015]. To fulfill the normalization properties introduced above, we apply a [thresholding procedure](../C4/C4S2_SSM-Thresholding.html) with normalization and penalty. As a result, all relevant elements of the SSM are between zero and one (with the main diagonal values being one), while irrelevant elements have negative score of $\\delta=-2$.\n", "\n", "\n", "\n", "In the following code cell, we provide a function for reading and computing all relevant data needed as input of our thumbnailing procedure. This data includes the audio signal, annotations, chroma features, the normalized SSM as well as other parameters and color maps. This function will also be used in subsequent notebooks using different music scenario beyond our Brahms example." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:46.931382Z", "iopub.status.busy": "2024-02-15T08:54:46.931139Z", "iopub.status.idle": "2024-02-15T08:54:55.308260Z", "shell.execute_reply": "2024-02-15T08:54:55.307647Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "import numpy as np\n", "import os, sys, librosa, math\n", "from scipy import signal\n", "from matplotlib import pyplot as plt\n", "import matplotlib\n", "import matplotlib.gridspec as gridspec\n", "import IPython.display as ipd\n", "import pandas as pd\n", "from numba import jit\n", "from matplotlib.colors import ListedColormap\n", "sys.path.append('..')\n", "import libfmp.b\n", "import libfmp.c2\n", "import libfmp.c3\n", "import libfmp.c4\n", "\n", "%matplotlib inline\n", "\n", "def colormap_penalty(penalty=-2, cmap=libfmp.b.compressed_gray_cmap(alpha=5)):\n", " \"\"\"Extend colormap with white color between the penalty value and zero\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " penalty (float): Negative number (Default value = -2.0)\n", " cmap (mpl.colors.Colormap): Original colormap (Default value = libfmp.b.compressed_gray_cmap(alpha=5))\n", "\n", " Returns:\n", " cmap_penalty (mpl.colors.Colormap): Extended colormap\n", " \"\"\"\n", " if isinstance(cmap, str):\n", " cmap = matplotlib.cm.get_cmap(cmap, 128)\n", " cmap_matrix = cmap(np.linspace(0, 1, 128))[:, :3]\n", " num_row = int(np.abs(penalty)*128)\n", " # cmap_penalty = np.flip(np.concatenate((cmap_matrix, np.ones((num_row, 3))), axis=0), axis=0)\n", " cmap_penalty = np.concatenate((np.ones((num_row, 3)), cmap_matrix), axis=0)\n", " cmap_penalty = ListedColormap(cmap_penalty)\n", "\n", " return cmap_penalty\n", "\n", "def normalization_properties_ssm(S):\n", " \"\"\"Normalizes self-similartiy matrix to fulfill S(n,n)=1.\n", " Yields a warning if max(S)<=1 is not fulfilled\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " S (np.ndarray): Self-similarity matrix (SSM)\n", "\n", " Returns:\n", " S_normalized (np.ndarray): Normalized self-similarity matrix\n", " \"\"\"\n", " S_normalized = S.copy()\n", " N = S_normalized.shape[0]\n", " for n in range(N):\n", " S_normalized[n, n] = 1\n", " max_S = np.max(S_normalized)\n", " if max_S > 1:\n", " print('Normalization condition for SSM not fulfill (max > 1)')\n", " return S_normalized\n", "\n", "def plot_ssm_ann(S, ann, Fs=1, cmap='gray_r', color_ann=[], ann_x=True, ann_y=True,\n", " fontsize=12, figsize=(5, 4.5), xlabel='', ylabel='', title=''):\n", " \"\"\"Plot SSM and annotations (horizontal and vertical as overlay)\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " S: Self-similarity matrix\n", " ann: Annotations\n", " Fs: Feature rate of path_family (Default value = 1)\n", " cmap: Color map for S (Default value = 'gray_r')\n", " color_ann: color scheme used for annotations (see :func:`libfmp.b.b_plot.plot_segments`)\n", " (Default value = [])\n", " ann_x: Plot annotations on x-axis (Default value = True)\n", " ann_y: Plot annotations on y-axis (Default value = True)\n", " fontsize: Font size used for annotation labels (Default value = 12)\n", " figsize: Size of figure (Default value = (5, 4.5))\n", " xlabel: Label for x-axis (Default value = '')\n", " ylabel: Label for y-axis (Default value = '')\n", " title: Figure size (Default value = '')\n", "\n", " Returns:\n", " fig: Handle for figure\n", " ax: Handle for axes\n", " im: Handle for imshow\n", " \"\"\"\n", " fig, ax = plt.subplots(2, 2, gridspec_kw={'width_ratios': [1, 0.05],\n", " 'height_ratios': [1, 0.1]}, figsize=figsize)\n", "\n", " fig_im, ax_im, im = libfmp.b.plot_matrix(S, Fs=Fs, Fs_F=Fs,\n", " ax=[ax[0, 0], ax[0, 1]], cmap=cmap,\n", " xlabel='', ylabel='', title='')\n", " ax[0, 0].set_ylabel(ylabel)\n", " ax[0, 0].set_xlabel(xlabel)\n", " ax[0, 0].set_title(title)\n", " if ann_y:\n", " libfmp.b.plot_segments_overlay(ann, ax=ax_im[0], direction='vertical',\n", " time_max=S.shape[0]/Fs, print_labels=False,\n", " colors=color_ann, alpha=0.05)\n", " if ann_x:\n", " libfmp.b.plot_segments(ann, ax=ax[1, 0], time_max=S.shape[0]/Fs, colors=color_ann,\n", " time_axis=False, fontsize=fontsize)\n", " else:\n", " ax[1, 0].axis('off')\n", " ax[1, 1].axis('off')\n", " plt.tight_layout()\n", " return fig, ax, im\n", "\n", "\n", "fn_wav = os.path.join('..', 'data', 'C4', 'FMP_C4_Audio_Brahms_HungarianDances-05_Ormandy.wav')\n", "tempo_rel_set = libfmp.c4.compute_tempo_rel_set(0.66, 1.5, 5)\n", "penalty = -2\n", "x, x_duration, X, Fs_feature, S, I = libfmp.c4.compute_sm_from_filename(fn_wav, L=21, H=5, \n", " L_smooth=12, tempo_rel_set=tempo_rel_set, penalty=penalty, thresh= 0.15)\n", "S = normalization_properties_ssm(S)\n", " \n", "fn_ann_color = 'FMP_C4_Audio_Brahms_HungarianDances-05_Ormandy.csv'\n", "fn_ann = os.path.join('..', 'data', 'C4', fn_ann_color)\n", "ann_frames, color_ann = libfmp.c4.read_structure_annotation(fn_ann, fn_ann_color, Fs=Fs_feature)\n", "\n", "cmap_penalty = colormap_penalty(penalty=penalty)\n", "fig, ax, im = plot_ssm_ann(S, ann_frames, Fs=1, color_ann=color_ann, cmap=cmap_penalty, \n", " xlabel='Time (frames)', ylabel='Time (frames)')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Path Family\n", "\n", "We start by extending some notions as defined in the FMP notebook on SSMs (see Section 4.2.1 of [Müller, FMP, Springer 2015]). Let $P=((n_1,m_1), (n_2,m_2), \\ldots,(n_L,m_L))$ be a path over a given **segment** $\\alpha=[s:t]\\subseteq [1:N]$, and let $\\Sigma$ denote the underlying **set of admissible step sizes**. Then, by definition, we have $m_1=s$ and $m_L=t$, and the path $P$ encodes a relation between $\\alpha=\\pi_2(P)$ and the induced segment $\\pi_1(P)$. Extending the notion of a path, we now introduce the concept of a path family, which allows us to capture relations between $\\alpha$ and several other segments in the music recording. To this end, we first define a **segment family** of size $K$ to be a set\n", "\n", "$$\n", "\\mathcal{A}:=\\{\\alpha_1,\\alpha_2,\\ldots,\\alpha_K\\}\n", "$$\n", "\n", "of pairwise disjoint segments, i.e., $\\alpha_k\\cap\\alpha_j=\\emptyset$ for all $k,j\\in[1:K]$ with $k\\not= j$. A **path family** over $\\alpha$ is defined to be a set\n", "\n", "$$\n", " \\mathcal{P}:=\\{P_1,P_2,\\ldots,P_K\\}\n", "$$\n", "\n", "of size $K$, consisting of paths $P_k$ over $\\alpha$ for $k\\in[1:K]$. Furthermore, as an additional condition, we require that the induced segments are pairwise disjoint. In other words, the set $\\{\\pi_1(P_1),\\ldots,\\pi_1(P_K)\\}$ is required to be a segment family. \n", "These definitions are illustrated by the following figure, which shows (from left to right), an SSM with a segment $\\alpha$, an SSM with a path family over $\\alpha$ consisting of three paths, a family of paths that is **not** a path family (since induced segments overlap), and a path family consisting of two paths.\n", "\n", "\"FMP_C4_F16.png\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In the next code cell, we provide some visualization functions for plotting the input SSM as well as a path family along with its projections. Furthermore, structure annotations are plotted horizontally below the SSM as well as a vertically as a transparent overlay on top of the SMM. As example, we synthetically generate a path family consisting of two components and use it for illustration purposes.\n", "\n", "
\n", "Note: Indexing in Python starts with index 0. In our implementation, the time axis is given by the set {0, 1, ..., N-1}. Furthermore, a segment is encoded by a tuple seg = [a, b], which encodes the index set {a, a+1, ..., b} . In particular, using this convention, index b belongs to the segment and the length of the segment is b - a + 1. \n", "
" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:55.339609Z", "iopub.status.busy": "2024-02-15T08:54:55.339306Z", "iopub.status.idle": "2024-02-15T08:54:55.601707Z", "shell.execute_reply": "2024-02-15T08:54:55.601134Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Segment: [40, 160]\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "def plot_path_family(ax, path_family, Fs=1, x_offset=0, y_offset=0, proj_x=True, w_x=7, proj_y=True, w_y=7):\n", " \"\"\"Plot path family into a given axis\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " ax: Axis of plot\n", " path_family: Path family\n", " Fs: Feature rate of path_family (Default value = 1)\n", " x_offset: Offset x-axis (Default value = 0)\n", " y_offset: Yffset x-axis (Default value = 0)\n", " proj_x: Display projection on x-axis (Default value = True)\n", " w_x: Width used for projection on x-axis (Default value = 7)\n", " proj_y: Display projection on y-axis (Default value = True)\n", " w_y: Width used for projection on y-axis (Default value = 7)\n", " \"\"\"\n", " for path in path_family:\n", " y = [(path[i][0] + y_offset)/Fs for i in range(len(path))]\n", " x = [(path[i][1] + x_offset)/Fs for i in range(len(path))]\n", " ax.plot(x, y, \"o\", color=[0, 0, 0], linewidth=3, markersize=5)\n", " ax.plot(x, y, '.', color=[0.7, 1, 1], linewidth=2, markersize=6)\n", " if proj_y:\n", " for path in path_family:\n", " y1 = path[0][0]/Fs\n", " y2 = path[-1][0]/Fs\n", " ax.add_patch(plt.Rectangle((0, y1), w_y, y2-y1, linewidth=1,\n", " facecolor=[0, 1, 0], edgecolor=[0, 0, 0]))\n", " # ax.plot([0, 0], [y1, y2], linewidth=8, color=[0, 1, 0])\n", " if proj_x:\n", " for path in path_family:\n", " x1 = (path[0][1] + x_offset)/Fs\n", " x2 = (path[-1][1] + x_offset)/Fs\n", " ax.add_patch(plt.Rectangle((x1, 0), x2-x1, w_x, linewidth=1,\n", " facecolor=[0, 0, 1], edgecolor=[0, 0, 0]))\n", " # ax.plot([x1, x2], [0, 0], linewidth=8, color=[0, 0, 1]) \n", "\n", "# Manually defined path family\n", "# For implementation reasons, the seconds components are of the paths \n", "# start with index 0 (corresponding to seg[0])\n", "seg_sec = [20, 80]\n", "seg = [int(seg_sec[0]*Fs_feature), int(seg_sec[1]*Fs_feature)]\n", "path_1 = [np.array([i+seg[0],i]) for i in range(0, seg[-1]-seg[0]+1)]\n", "path_2 = [np.array([int(i+130*Fs_feature),i]) for i in range(0, seg[-1]-seg[0]+1)]\n", "path_family = [path_1, path_2]\n", "print('Segment:', seg)\n", "fig, ax, im = plot_ssm_ann(S, ann_frames, Fs=1, color_ann=color_ann, cmap=cmap_penalty, \n", " xlabel='Time (frames)', ylabel='Time (frames)') \n", "plot_path_family(ax[0,0], path_family, Fs=1, x_offset=seg[0])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Coverage\n", "\n", "We now define some measures that describe specific properties of a path family $\\mathcal{P}$. Let $\\mathcal{A}:=\\{\\alpha_1,\\alpha_2,\\ldots,\\alpha_K\\}$ be a segment family. The **coverage** $\\gamma(\\mathcal{A})$ of $\\mathcal{A}$ is defined by\n", "\n", "\\begin{equation}\n", " \\gamma(\\mathcal{A}):=\\sum_{k=1}^K|\\alpha_k|.\n", "\\end{equation}\n", "\n", "Furthermore, the **coverage** $\\gamma(\\mathcal{P})$ of a path family $\\mathcal{P}=\\{P_1,P_2,\\ldots,P_K\\}$ is defined to be the coverage of its induced segment family $\\{\\pi_1(P_1),\\ldots,\\pi_1(P_K)\\}$. In the next code cell, we provide a function for deriving the induced segment family as well as the coverage for a given path family. This function is applied to our synthetically generate path family from above. " ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:55.604457Z", "iopub.status.busy": "2024-02-15T08:54:55.604250Z", "iopub.status.idle": "2024-02-15T08:54:55.609449Z", "shell.execute_reply": "2024-02-15T08:54:55.608868Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Segment (alpha): [40, 160]\n", "Induced segment family:\n", "[[ 40 160]\n", " [260 380]]\n", "Coverage: 242\n" ] } ], "source": [ "def compute_induced_segment_family_coverage(path_family):\n", " \"\"\"Compute induced segment family and coverage from path family\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " path_family (list): Path family\n", "\n", " Returns:\n", " segment_family (np.ndarray): Induced segment family\n", " coverage (float): Coverage of path family\n", " \"\"\"\n", " num_path = len(path_family)\n", " coverage = 0\n", " if num_path > 0:\n", " segment_family = np.zeros((num_path, 2), dtype=int)\n", " for n in range(num_path):\n", " segment_family[n, 0] = path_family[n][0][0]\n", " segment_family[n, 1] = path_family[n][-1][0]\n", " coverage = coverage + segment_family[n, 1] - segment_family[n, 0] + 1\n", " else:\n", " segment_family = np.empty\n", "\n", " return segment_family, coverage\n", "\n", "segment_family, coverage = compute_induced_segment_family_coverage(path_family)\n", "print('Segment (alpha):', seg)\n", "print('Induced segment family:')\n", "print(segment_family)\n", "print('Coverage: %d'%coverage) " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Score and Optimality\n", "\n", "Next, we transfer the notion of a score from paths to path families. Recall that, given an SSM $\\mathbf{S}$, the **score** $\\sigma(P)$ of a path $P$ is defined as \n", "\n", "$$\n", "\\sigma(P)=\\sum_{\\ell=1}^L \\mathbf{S}(n_\\ell,m_\\ell).\n", "$$\n", "\n", "This value can be thought of a quality measure for the similarity relation between the segments $\\pi_1(P)$ and $\\pi_2(P)$. For a path family $\\mathcal{P}$, the **score** $\\sigma(\\mathcal{P})$ is defined as \n", "\n", "$$\n", "\\sigma(\\mathcal{P}) := \\sum_{k=1}^{K} \\sigma(P_k).\n", "$$\n", "\n", "A large score $\\sigma(\\mathcal{P})$ indicates that the path components of $\\mathcal{P}$ express strong similarity relations among its segments. In general, there is large number of possible path families over $\\alpha$. Among these path families, let\n", "\n", "$$\n", " \\mathcal{P}^\\ast := \\underset{\\mathcal{P}}{\\mathrm{argmax}} \\,\\,\\, \\sigma(\\mathcal{P})\n", "$$\n", "\n", "denote an optimal path family of maximal score. In the following, the family consisting of the segments induced by the paths of $\\mathcal{P}^\\ast$ will be referred to as the **induced segment family** (of $\\mathcal{P}^\\ast$ or simply of $\\alpha$). Intuitively, the induced segment family contains the (nonoverlapping) repetitions of the segment $\\alpha$." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Computation of Optimal Path Families\n", "\n", "Similar to [dynamic time warping](../C3/C3S2_DTWbasic.html) for computing an optimal warping path, there is an efficient algorithm using dynamic programming to compute an optimal path family $\\mathcal{P}^\\ast$. For a description of this algorithm, we refer to Section 4.3.1.2 of [Müller, FMP, Springer 2015] and the article cited above. In the following code cells, one finds an implementation of this algorithm using the set $\\Sigma = \\{(2,1),(1,2),(1,1)\\}$ of admissible step sizes. Let $\\mathbf{S}$ be an $N$-square SSM that fulfills the required normalization properties (the matrix is also called **score matrix**) and a segment $\\alpha=[s:t]\\subseteq [1:N]$ with $M:=|\\alpha|$. The input of the algorithm is the $N\\times M$ submatrix $\\mathcal{S}^{\\alpha}$, which consists of the columns $s$ to $t$ of $\\mathbf{S}$. From this, we first compute an **accumulated score matrix** $D$ using dynamic programming. " ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:55.612017Z", "iopub.status.busy": "2024-02-15T08:54:55.611840Z", "iopub.status.idle": "2024-02-15T08:54:55.617570Z", "shell.execute_reply": "2024-02-15T08:54:55.617065Z" } }, "outputs": [], "source": [ "@jit(nopython=True)\n", "def compute_accumulated_score_matrix(S_seg):\n", " \"\"\"Compute the accumulated score matrix\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " S_seg (np.ndarray): Submatrix of an enhanced and normalized SSM ``S``.\n", " Note: ``S`` must satisfy ``S(n,m) <= 1 and S(n,n) = 1``\n", "\n", " Returns:\n", " D (np.ndarray): Accumulated score matrix\n", " score (float): Score of optimal path family\n", " \"\"\"\n", " inf = math.inf\n", " N = S_seg.shape[0]\n", " M = S_seg.shape[1]+1\n", "\n", " # Iinitializing score matrix\n", " D = -inf * np.ones((N, M), dtype=np.float64)\n", " D[0, 0] = 0.\n", " D[0, 1] = D[0, 0] + S_seg[0, 0]\n", "\n", " # Dynamic programming\n", " for n in range(1, N):\n", " D[n, 0] = max(D[n-1, 0], D[n-1, -1])\n", " D[n, 1] = D[n, 0] + S_seg[n, 0]\n", " for m in range(2, M):\n", " D[n, m] = S_seg[n, m-1] + max(D[n-1, m-1], D[n-1, m-2], D[n-2, m-1])\n", "\n", " # Score of optimal path family\n", " score = np.maximum(D[N-1, 0], D[N-1, M-1])\n", "\n", " return D, score" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Next, we derive an optimal path family $\\mathcal{P}^\\ast$ using backtracking." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:55.620091Z", "iopub.status.busy": "2024-02-15T08:54:55.619914Z", "iopub.status.idle": "2024-02-15T08:54:55.629288Z", "shell.execute_reply": "2024-02-15T08:54:55.628767Z" } }, "outputs": [], "source": [ "@jit(nopython=True)\n", "def compute_optimal_path_family(D):\n", " \"\"\"Compute an optimal path family given an accumulated score matrix\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " D (np.ndarray): Accumulated score matrix\n", "\n", " Returns:\n", " path_family (list): Optimal path family consisting of list of paths\n", " (each path being a list of index pairs)\n", " \"\"\"\n", " # Initialization\n", " inf = math.inf\n", " N = int(D.shape[0])\n", " M = int(D.shape[1])\n", "\n", " path_family = []\n", " path = []\n", "\n", " n = N - 1\n", " if(D[n, M-1] < D[n, 0]):\n", " m = 0\n", " else:\n", " m = M-1\n", " path_point = (N-1, M-2)\n", " path.append(path_point)\n", "\n", " # Backtracking\n", " while n > 0 or m > 0:\n", "\n", " # obtaining the set of possible predecesors given our current position\n", " if(n <= 2 and m <= 2):\n", " predecessors = [(n-1, m-1)]\n", " elif(n <= 2 and m > 2):\n", " predecessors = [(n-1, m-1), (n-1, m-2)]\n", " elif(n > 2 and m <= 2):\n", " predecessors = [(n-1, m-1), (n-2, m-1)]\n", " else:\n", " predecessors = [(n-1, m-1), (n-2, m-1), (n-1, m-2)]\n", "\n", " # case for the first row. Only horizontal movements allowed\n", " if n == 0:\n", " cell = (0, m-1)\n", " # case for the elevator column: we can keep going down the column or jumping to the end of the next row\n", " elif m == 0:\n", " if D[n-1, M-1] > D[n-1, 0]:\n", " cell = (n-1, M-1)\n", " path_point = (n-1, M-2)\n", " if(len(path) > 0):\n", " path.reverse()\n", " path_family.append(path)\n", " path = [path_point]\n", " else:\n", " cell = (n-1, 0)\n", " # case for m=1, only horizontal steps to the elevator column are allowed\n", " elif m == 1:\n", " cell = (n, 0)\n", " # regular case\n", " else:\n", "\n", " # obtaining the best of the possible predecesors\n", " max_val = -inf\n", " for i, cur_predecessor in enumerate(predecessors):\n", " if(max_val < D[cur_predecessor[0], cur_predecessor[1]]):\n", " max_val = D[cur_predecessor[0], cur_predecessor[1]]\n", " cell = cur_predecessor\n", "\n", " # saving the point in the current path\n", " path_point = (cell[0], cell[1]-1)\n", " path.append(path_point)\n", "\n", " (n, m) = cell\n", "\n", " # adding last path to the path family\n", " path.reverse()\n", " path_family.append(path)\n", " path_family.reverse()\n", "\n", " return path_family" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In the next figure, we show the submatrix $\\mathcal{S}^{\\alpha}$ and the resulting accumulated score matrix $D$ for the segment $\\alpha=[83:137]$ (`seg = [82, 136]`). Using a frame rate of $2~\\mathrm{Hz}$, this segment corresponds to the interval ranging from seconds $41$ to $68$, which is roughly the $B_1$-part of the Brahms recording. Note the following:\n", "\n", "* The number of columns of the submatrix $\\mathcal{S}^{\\alpha}$ is $M=55$. \n", "* In the Python implementation, the columns of $\\mathcal{S}^{\\alpha}$ are enumerated starting with index `0` and ending with index `M-1`.\n", "* The matrix $D$ has $M+1$ columns including a so-called elevator column (see Section 4.3.1.2 of [Müller, FMP, Springer 2015]).\n", "* Using the step size condition $\\Sigma = \\{(2,1),(1,2),(1,1)\\}$ allows the paths to adapt to local tempo differences in the alignment.\n", "* The optimal path family consists of four paths. The induced segments roughly correspond to the four $B$-part sections of the Brahms recording. " ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:55.631933Z", "iopub.status.busy": "2024-02-15T08:54:55.631755Z", "iopub.status.idle": "2024-02-15T08:54:57.233555Z", "shell.execute_reply": "2024-02-15T08:54:57.233023Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Segment: [82, 136]\n", "Length of segment: M = 55\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "def plot_matrix_seg(ax, M, seg, title='', xlabel='', cmap='gray_r'):\n", " libfmp.b.plot_matrix(M, Fs=1, ax=[ax], cmap=cmap, \n", " xlabel=xlabel, ylabel='Time (frames)', title=title)\n", " ax.set_xticks([0, M.shape[1]-1]) \n", " \n", "seg_sec = [41,68]\n", "seg = [int(seg_sec[0]*Fs_feature), int(seg_sec[1]*Fs_feature)]\n", "print('Segment:', seg)\n", "print('Length of segment: M = ', seg[1]-seg[0]+1)\n", "S_seg = S[:,seg[0]:seg[1]+1]\n", "D, score = compute_accumulated_score_matrix(S_seg)\n", "path_family = compute_optimal_path_family(D)\n", "\n", "plt.figure(figsize=(8, 8))\n", "ax = plt.subplot(2, 2, 1)\n", "title = r'$\\mathcal{S}^{\\alpha}$'\n", "plot_matrix_seg(ax, S_seg, seg, title=title, xlabel='', cmap=cmap_penalty)\n", "\n", "ax = plt.subplot(2, 2, 2)\n", "title = r'$D$'\n", "plot_matrix_seg(ax, D, seg, title=title)\n", "\n", "ax = plt.subplot(2, 2, 3)\n", "title = r'$\\mathcal{S}^{\\alpha}$ with optimal path family'\n", "plot_matrix_seg(ax, S_seg, seg, title=title, xlabel='', cmap=cmap_penalty)\n", "plot_path_family(ax, path_family, x_offset=0, w_y=1)\n", "\n", "ax = plt.subplot(2, 2, 4)\n", "title = r'$D$ with optimal path family'\n", "plot_matrix_seg(ax, D, seg, title=title)\n", "plot_path_family(ax, path_family, x_offset=1, w_y=1)\n", "\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Fitness Measure\n", "\n", "For a given segment $\\alpha$, let $\\mathcal{P}^\\ast=\\{P_1,\\ldots,P_K\\}$ be an optimal path family. We now explain how to derive a fitness measure from the score $\\sigma(\\mathcal{P}^\\ast)$ and the induced segment family of $\\mathcal{P}^\\ast$. One first idea is to simply use the total score $\\sigma(\\mathcal{P}^\\ast)$ as the fitness value for $\\alpha$. However, this measure does not yet have the desired properties, since it not only depends on the lengths of $\\alpha$ and the paths, but also captures trivial self-explanations. For example, the segment $\\alpha=[1:N]$ explains the entire sequence $X$ perfectly, which is a trivial fact. More generally, each segment $\\alpha$ explains itself perfectly, information that is encoded by the main diagonal of a self-similarity matrix. Therefore, one idea in defining the fitness measure is to disregard such trivial self-explanations. Assuming the normalization properties $\\mathbf{S}(n,m)\\leq 1$ and $\\mathbf{S}(n,n)=1$ of $\\mathbf{S}$, this step can be done by simply subtracting the length $|\\alpha|$ from the score $\\sigma(\\mathcal{P}^\\ast)$. Furthermore, we normalize the score with regard to the lengths $L_k:=|P_k|$ of the paths $P_k$ contained in the optimal path family $\\mathcal{P}^\\ast$. This yields the **normalized score** $\\bar{\\sigma}(\\alpha)$ defined by\n", "\n", "$$\n", " \\bar{\\sigma}(\\alpha) := \\frac{\\sigma(\\mathcal{P}^\\ast) - |\\alpha|}{\\sum_{k=1}^{K} L_k}.\n", "$$\n", "\n", "Intuitively, the value $\\bar{\\sigma}(\\alpha)$ expresses the **average score** of the optimal path family $\\mathcal{P}^\\ast$ (minus a proportion for the self-explanation). The normalized score indicates **how well** a given segment explains other segments, where the normalization eliminates the influence of segment lengths. This makes the normalized score a fair measure when comparing segments of different lengths. Besides repetitiveness, another issue is **how much** of the underlying music recording is covered by the thumbnail and its related segments. To capture this property, we define a **coverage** measure for a given $\\alpha$. To this end, let $\\mathcal{A}^\\ast:=\\{\\pi_1(P_1),\\ldots,\\pi_1(P_K)\\}$ be the segment family induced by the optimal path family $\\mathcal{P}^\\ast$, and let $\\gamma(\\mathcal{A}^\\ast)$ be its coverage. Then we define the **normalized coverage** $\\bar{\\gamma}(\\alpha)$ by\n", "\n", "$$\n", " \\bar{\\gamma}(\\alpha) := \\frac{\\gamma(\\mathcal{A}^\\ast) - |\\alpha|}{N}.\n", "$$\n", "\n", "As above, the length $|\\alpha|$ is subtracted to compensate for trivial coverage. The value $\\bar{\\gamma}(\\alpha)$ expresses the ratio between the union of the induced segments of $\\alpha$ and the total length of the original recording (minus a proportion for the self-explanation). Having a high average score and a high coverage are both desirable properties for defining a thumbnail segment. However, these two properties are sometimes hard to satisfy at the same time. Shorter segments often have a higher average score, but a lower coverage, whereas longer segments tend to have a lower average score, but a higher coverage. To balance out these two trends, we combine the score and coverage measure by taking a suitable average. In particular, we define the **fitness** $\\varphi(\\alpha)$ of the segment $\\alpha$ to be the **harmonic mean**\n", "\n", "$$\n", " \\varphi(\\alpha) :=\n", " 2\\cdot \\frac{\\bar{\\sigma}(\\alpha) \\cdot \\bar{\\gamma}(\\alpha)}\n", " {\\bar{\\sigma}(\\alpha)+\\bar{\\gamma}(\\alpha)}\n", "$$\n", "\n", "between the normalized score and normalized coverage. Based on the normalized values $\\bar{\\sigma}(\\alpha)$ and $\\bar{\\gamma}(\\alpha)$, one can show that $\\varphi(\\alpha)\\leq 1-|\\alpha|/N$.\n", "\n", "In the following code cell, we compute these different measures for the path family considered in the previous code cell." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:57.237008Z", "iopub.status.busy": "2024-02-15T08:54:57.236830Z", "iopub.status.idle": "2024-02-15T08:54:57.243718Z", "shell.execute_reply": "2024-02-15T08:54:57.242998Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Segment (alpha): [82, 136]\n", "Length of segment: 55\n", "Length of feature sequence: 409\n", "Induced segment path family:\n", " [[ 82 135]\n", " [136 176]\n", " [297 350]\n", " [351 391]]\n", "Fitness: 0.412\n", "Score: 159.199\n", "Normalized score: 0.548\n", "Coverage: 190\n", "Normalized coverage: 0.330\n", "Length of all paths of family: 190\n" ] } ], "source": [ "def compute_fitness(path_family, score, N):\n", " \"\"\"Compute fitness measure and other metrics from path family\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " path_family (list): Path family\n", " score (float): Score\n", " N (int): Length of feature sequence\n", "\n", " Returns:\n", " fitness (float): Fitness\n", " score (float): Score\n", " score_n (float): Normalized score\n", " coverage (float): Coverage\n", " coverage_n (float): Normalized coverage\n", " path_family_length (int): Length of path family (total number of cells)\n", " \"\"\"\n", " eps = 1e-16\n", " num_path = len(path_family)\n", " M = path_family[0][-1][1] + 1\n", "\n", " # Normalized score\n", " path_family_length = 0\n", " for n in range(num_path):\n", " path_family_length = path_family_length + len(path_family[n])\n", " score_n = (score - M) / (path_family_length + eps)\n", "\n", " # Normalized coverage\n", " segment_family, coverage = compute_induced_segment_family_coverage(path_family)\n", " coverage_n = (coverage - M) / (N + eps)\n", "\n", " # Fitness measure\n", " fitness = 2 * score_n * coverage_n / (score_n + coverage_n + eps)\n", "\n", " return fitness, score, score_n, coverage, coverage_n, path_family_length\n", "\n", "N = S.shape[0]\n", "\n", "segment_family, coverage = compute_induced_segment_family_coverage(path_family)\n", "fitness, score, score_n, coverage, coverage_n, path_family_length = compute_fitness(\n", " path_family, score, N)\n", "\n", "print('Segment (alpha):', seg)\n", "print('Length of segment:', seg[-1]-seg[0]+1)\n", "print('Length of feature sequence:', N)\n", "print('Induced segment path family:\\n', segment_family)\n", "print('Fitness: %0.3f'%fitness) \n", "print('Score: %0.3f'%score) \n", "print('Normalized score: %0.3f'%score_n) \n", "print('Coverage: %d'%coverage) \n", "print('Normalized coverage: %0.3f'%coverage_n) \n", "print('Length of all paths of family: %d'%path_family_length)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Thumbnail Selection\n", "\n", "Based on the fitness measure, we define the **audio thumbnail** to be the segment of **maximal fitness**:\n", "\n", "$$\n", " \\alpha^\\ast := \\underset{\\alpha}{\\mathrm{argmax}} \\,\\, \\varphi(\\alpha).\n", "$$\n", "\n", "By construction of the fitness measure, this segment has nonoverlapping repetitions that cover a possibly large portion of the audio recording. Furthermore, these repetitions are given by the induced segments obtained by the optimal path family of $\\alpha^\\ast$ yielding a segmentation of the audio recording into pairwise disjoint segments. To account for prior knowledge and to remove spurious estimates, one can impose additional requirements on the thumbnail solution.\n", "In particular, introducing a lower bound $\\theta$ for the minimal possible thumbnail length allows us to reduce the effect of noise scattered in the underlying self-similarity matrix. Extending the above definition, we define\n", "\n", "$$\n", " \\alpha^\\ast_\\theta := \\underset{\\alpha, |\\alpha|\\geq \\theta}{\\mathrm{argmax}} \\,\\, \\varphi(\\alpha).\n", "$$\n", "\n", "Computing an audio thumbnail involves computing a fitness measure for all possible segments $\\alpha$. Beyond selecting the fitness-maximizing segment, a visualization of the fitness values for all segments is very instructive and yields musical insights into structural properties of the music recording. We will consider such a visualization in the [FMP notebook on scape plots](../C4/C4S3_ScapePlot.html), where we also discuss further properties of the fitness measure. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Examples of Optimal Path Families\n", "\n", "We now compute the audio thumbnail $\\alpha^\\ast$ for our Brahms example. Note that there are $(N+1)N/2$ possible segments $\\alpha \\subseteq [1:N]$. Therefore, determining $\\alpha^\\ast$ may be a time-consuming task. \n", "\n", "* In the following, we apply the function `libfmp.c4.compute_fitness_scape_plot` (introduced in the [FMP notebook on scape plots](../C4/C4S3_ScapePlot.html)) to compute the fitness of all segments, which is stored in the data structure `SP=SP_all[0]`.\n", "\n", "* The function `libfmp.c4.seg_max_sp` allows for determining the fitness-maximizing segment from `SP`.\n", "\n", "* The function `libfmp.c4.check_segment` outputs the properties of the thumbnail segment $\\alpha^\\ast$.\n", "\n", "It turns out that the fitness-maximizing segment for our example is given by $\\alpha^\\ast=[353:396]$ (`seg = [352, 395]`) with $\\varphi(\\alpha^\\ast)=0.452$. This segment, which musically corresponds to the $B_4$-part (interval between seconds $176$ and $197.5$), is indeed the most repetitive part. The induced segment family consists of the four $B$-part segments." ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:54:57.247485Z", "iopub.status.busy": "2024-02-15T08:54:57.247164Z", "iopub.status.idle": "2024-02-15T08:55:21.885322Z", "shell.execute_reply": "2024-02-15T08:55:21.884711Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Segment (alpha): [352, 395]\n", "Length of segment: 44\n", "Length of feature sequence: 409\n", "Induced segment path family:\n", " [[ 84 137]\n", " [138 180]\n", " [300 353]\n", " [354 395]]\n", "Fitness: 0.4520185796\n", "Score: 144.0220022169\n", "Normalized score: 0.5953690608\n", "Coverage: 193, 193\n", "Normalized coverage: 0.3643031785\n", "Length of all paths of family: 168\n" ] } ], "source": [ "def plot_ssm_ann_optimal_path_family(S, ann, seg, Fs=1, cmap='gray_r', color_ann=[], fontsize=12,\n", " figsize=(5, 4.5), ylabel=''):\n", " \"\"\"Plot SSM, annotations, and computed optimal path family\n", "\n", " Notebook: C4/C4S3_AudioThumbnailing.ipynb\n", "\n", " Args:\n", " S: Self-similarity matrix\n", " ann: Annotations\n", " seg: Segment for computing the optimal path family\n", " Fs: Feature rate of path_family (Default value = 1)\n", " cmap: Color map for S (Default value = 'gray_r')\n", " color_ann: color scheme used for annotations (see :func:`libfmp.b.b_plot.plot_segments`)\n", " (Default value = [])\n", " fontsize: Font size used for annotation labels (Default value = 12)\n", " figsize: Size of figure (Default value = (5, 4.5))\n", " ylabel: Label for y-axis (Default value = '')\n", "\n", " Returns:\n", " fig: Handle for figure\n", " ax: Handle for axes\n", " im: Handle for imshow\n", " \"\"\"\n", " N = S.shape[0]\n", " S_seg = S[:, seg[0]:seg[1]+1]\n", " D, score = compute_accumulated_score_matrix(S_seg)\n", " path_family = compute_optimal_path_family(D)\n", " fitness, score, score_n, coverage, coverage_n, path_family_length = compute_fitness(\n", " path_family, score, N)\n", " title = r'$\\bar{\\sigma}(\\alpha)=%0.2f$, $\\bar{\\gamma}(\\alpha)=%0.2f$, $\\varphi(\\alpha)=%0.2f$' % \\\n", " (score_n, coverage_n, fitness)\n", " fig, ax, im = plot_ssm_ann(S, ann, color_ann=color_ann, Fs=Fs, cmap=cmap,\n", " figsize=figsize, fontsize=fontsize,\n", " xlabel=r'$\\alpha=[%d:%d]$' % (seg[0], seg[-1]), ylabel=ylabel, title=title)\n", " plot_path_family(ax[0, 0], path_family, Fs=Fs, x_offset=seg[0])\n", " return fig, ax, im\n", "\n", "SP_all = libfmp.c4.compute_fitness_scape_plot(S)\n", "SP = SP_all[0]\n", "seg = libfmp.c4.seg_max_sp(SP)\n", "fig, ax, im = plot_ssm_ann_optimal_path_family(S, ann_frames, seg,\n", " color_ann=color_ann, cmap=cmap_penalty, ylabel='Time (frames)')\n", "plt.show()\n", "path_family = libfmp.c4.check_segment(seg, S)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "One can check that all four $B$-part segments have almost the same fitness and lead to more or less the same segment family. For example, let us consider $\\alpha=[85:138]$ (`seg = [84, 137]`) (interval between seconds $42$ and $69$) corresponding to part $B_1$. The fitness is $\\varphi(\\alpha)=0.41$, which is only slightly below the maximal possible fitness. Again, the induced segment family consists of the four $B$-part segments, which reflects the fact that each of the $B$-part segments may serve equally well as the thumbnail. Since the fitness measure slightly favors shorter segment and since in this recording the $B_4$-part is played faster than the $B_1$-part, the fitness measure favors the $B_4$-part segment over the $B_1$-part segment. " ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:55:21.888231Z", "iopub.status.busy": "2024-02-15T08:55:21.887954Z", "iopub.status.idle": "2024-02-15T08:55:22.218009Z", "shell.execute_reply": "2024-02-15T08:55:22.217426Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "seg = [84, 137]\n", "fig, ax, im = plot_ssm_ann_optimal_path_family(S, ann_frames, seg, \n", " color_ann=color_ann, cmap=cmap_penalty, ylabel='Time (frames)')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Also other segments may have a high fitness value, thus indicating a high similarity to other repeating segments. For example, the segment $\\alpha=[261:307]$ (`seg = [260, 306]`, the interval between seconds $130$ and $153$), which corresponds to the $A_3$-part, has fitness of $\\varphi(\\alpha)=0.30$. The induced segment family reveals the three $A$-parts. " ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:55:22.220931Z", "iopub.status.busy": "2024-02-15T08:55:22.220755Z", "iopub.status.idle": "2024-02-15T08:55:22.547083Z", "shell.execute_reply": "2024-02-15T08:55:22.546073Z" } }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAE8CAYAAAD+LWvAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAACizUlEQVR4nOydd3hUVfrHP2dSJ3XSeyEklFATuqKABQRBFBtWFBsW1FVXkWVFfyLL6rouurKsKIIrLitrFI1BIlGCESKBAAEChBBInZRJMimTMsnk/v6YyWwCKRO4IQHm8zx5krn33HPO3Mx87znvec/7CkmSsGLFihUrFx9FX3fAihUrVq5UrAJsxYoVK32EVYCtWLFipY+wCrAVK1as9BFWAbZixYqVPsIqwFasWLHSR1gF2IoVK1b6CKsAW7FixUofcUULsBDiOyFErennp7PO/UkI8XwP6torhBgmeyf7AXLdp8v5HnWF9bNkpVMkSbL+nPUD+ACFgLIH19wFfCVT+57A14AOyAXutfC6KKAB+LzNsdqzfgzAB31xn+S8R5beJ8AB+MR0vgY4AMy09PyV9lkCPgfUQDWQBTx6oZ9L608X97uvO9Aff4DfA+t6eI0jUAEEyND+v4H/AC7AZKAKGGbBdYnAL20F+KzzziYRvrYv7pOc98jS+2R6z68D4RhnfLNNQhtuyfkr7bMEDAMcTH8PAYqBMRfyubT+dPH/6esO9NkbBzvgLeAM0ARIpp9DwE/A/WeVtwX+aCpfDtwLvAz8oU2ZH4EFF9gvZ0APDGpz7F/Aqm6umw98aRKTzgR4AZADCAv64YhxpPPqWcdTMY18zr5PF+seXch9MpXLAG4/3/MdlL8f+Ap4BKgEtgMuHd2jS+Gz1KbsYIyj4bsutC7rT8c/tly5rACmANdgHG1sxTjtegn4DTjRQfmxwCjgWuBtjII9oU2ZY6bz7RBCxGMcMXREiiRJs9u8HgQYJEnKanPskKmvHSKEcAP+D7geowh0xgLgM8n07ekKSZIahBC3Ah8AfzK1cydgj3EkBDCC9vfpYt0jOI/7ZGrHz3Tt0fM530H52cCfgZuAXRjf938xiupHnHuPoB9/lkxtrAEeApQYTTIJ51uXla65IgVYCOEKPAuMlCQp33TsK+BuSZJyhBAqjNPQ1vJuwPNAtCRJVUKI3zBOz/4gSVJNm6prgICz2+tAPLrCBePUri1VgGsX17wJfCJJUr4QosMCQohQjF+WrgT6bH4FBgghbAGBccbwdBsBV2G6Txf5HsF53CchhB2wCdgoSdLxnp7vhDcwjl4VgE6SpMNCiN1AiOm8ikvrs4QkSU8JIRYDk4CpQOP51mWla65UL4hrgRxJkk62OeaB0d4Fxmlk2w/WdUCWJEk5ptf2GD98H5xVryugvcC+1QJuZx1zo82XuC1CiNHADcB73dT7IMYR0mlLOyJJUh3GKfIA4HEgV5KkH9sUaXufLuY9gp7fJwXGKbMeeKan5zup0xeIBb4DRgIHTad8gFLT35fMZ6ktkiQZJElKAYKBJy+kLiudc6UKsA/GLwYAwjhsvA2INx3KwDjlaiUQKGrz+nGg8KwRC8BQjNOydgghtrVx4zr7Z9tZxbMAWyFEVJtjo+h8SjwV4wJSnhCiGKMJ5XYhRPpZ5R4ENnZSR1dkYxSZZcArZ51re58u5j2CHtwn0//3E8APo223qSfnu8AbaMYoQqOAQ0IIZ4wP+ARTmUvps9QRtsBAmeqycjZ9bYTuix+M9rc6YDRGO9cqYC9gZzr/AvBRm/K3YBwdB2C0050GygD7NmUcMNqSA2Xo32aMdlZn4Gq6WG0GnAD/Nj9/wWiD9GlT5iqMC2quHVy/AdjQRV/WA3nAFx2cM9+ni32PenKfgLUYFw9dOqmnu/Md3iPTva8FbsW4aPYIRi+BNR3do0vgs+SLcTHXBbABZpg+N3N7Wpf1x8L/T193oM/eOPwB40hEbfqCebc55w0UYPLdxDhN/AzjlDAHiDF94VLaXHMnECdT3zyBb0wf/jzO8rcEtgFLO7n2dc7yggD+Cfyrk/JJwGNd9GUpRhtgRAfnzPfpYt+j7u5T6z0CwjAucDXQ3h/6PlO5Ls93d49MgpUPtJg+S6vOEtNL5rOEcWaYbOpbNXC47fvuri7rT89/hOnGWjkLIcRKoFSSpL9ZWP434BFJko70asdkRAhhj3GaO1LqZNothHgWuEaSpDs7OW/xfbqM71EAcApwljr4Ql0JnyUr54dVgK10iRDin0CeJElv9XVf+itCiGnA3yRJOsdtzIqVrrhSF+GsWM4IjFNRK50zGOMilZVLFCHEeiFEqRCiw1mHMPK+ECJbCJEhhIiVpV3rCNiKFStXOkKIazHa/j+TJGl4B+dnAYuBWRgXT1dLkjTh7HI9xToCtmLFyhWPJEm7MHqedMZcTLtIJUlKBVQm2/8FcUnvhPP29pbCw8P7uhvnjSRJdLZz7WKi1+uxtbVFobA+j3sLnU6HnZ0dLS0t5nvd1/e7psboeuzk5ISNjY2sdVdUVGAwGLC1tUWpVNLU1ERFRQXV1dXo9frWYhpJknzOp34hRE+n7kcxerq08pEkSR/14PogjN4urRSYjql72I92XNICHB4ezr59+/qk7V9//RV3d3eam5sBGD16dI/rOHHiBIMHD5a5Zz1r/9ixY5SVlQFw7733AuDs7CxL/b/++isADg4OjB07VpY6L0W++eYbKioqUCqV6PV6lEolgwYNOq/PjFwkJiZy+vRpHB0diYyM5Oqrr5at7s8//5zDhw+j1+txd3dnzJgxHDhwgHXr1lFbWwuAjY0NBoMh90La6cngRZKkBkmSLuRD2FFjF2y/7XUBFkLYAPsw7vaZLYTwxOisHo4xGtRdkiRVmsq+itGZ3QA8K0nSdrn7c+rUKby8vFCpVOd9/eHDh3FycsLW1paQkJDuLzJdBzBw4EByc3Npbm7G0dHxvPogBydOnODf//43jY2N1NbWEhwcjEajAeQR4NzcXLKzs7G3t0epVAJga2tLfn4+9fX1rV9AgoKCGDBgAACBgYHm65OTk/Hy8sLPzw8fn/MaJPUaer0ejUbTrr+t6HQ6mpqaUKlUfPfdd1RVVZGfn49CocDJyQk3NzfCwsLaie+vv/7KoEGD8PHxoaioqMN65USr1dLY2EhoaCjXXnutbA9cMD5sfvvtNxQKBY6OjtjZ2bFjxw4SExMpLi6mubkZBwcHvLy8KCoq6r7CLuihAF9QWxhHvG2/7MG039F4XlyMEfBzGCM7te4jXwIkSZK0SgixxPT6FSFENEan9mEYt2vuEEIMkiTJIFdH9u3bh1qtxsHBgZiYmB5/sbdt28agQYOIiYkhLCzM4uu+/PJLoqOjza/PvvbgwYN4eHj0qM4L5d///jc6nQ6lUklwcDCenp6ytm9nZ8eCBQsA+Oc//8mPP/5IQ4NxBtjc3ExDQwMRERHs2bOH6dOnm6/x8fHhxIkT1NXVMWVK/wy0VVdXd45Itn62ysrKaGhooKmpicrKShobG1EqlVx//fW4u7u3e6DodDp+/fVXhg83rvlcrBmRSqVizpw56PV68vPzGThwoGx1GwwGhg8fTmBgIJs2baK6upqsrCyKiopwcHDAzc2NgQMHEhMTw0cf9cQCcC49MeG0tLRcUFvAt8AzQojNGBfhqiRJuiDzA/SyAAshgoGbMUbResF0eC7G+AVgjE2wE2OMgbnAZkmSGoHTQohsYDywR46+6HQ6IiMjsbW1JSsriwMHDtDY2Iibm5vFX/SIiIgef1jLysoYNWpUl18sX1/fXh/1nM0DDzxgfi9PPvkkOp1O1vrbvp/vvvuO6upqmpqacHR0xNHRkZaWFnJychg9erR5FlFSUkJTUxO2trbMnDlT1v7ISUezp6NHj6JUKlEqleh0Ourq6nBxccHDw4OJEycyYcK5C+bFxcU4Ozvj7e0NcNFH+h9//DFXXXWVrHXefvvtAKxevZq0tDRqa2vR6XTY29sTERGBg4MDd955J/Pnz78gARZCyGpDF0L8G6MueQshCoDlGGOGI0nSWoyxPWZhjI1SBzwsR7u9PQL+G8ZQfW2jQfm1PjkkSVKbIkqB0aCd2qZcq5G7HUKIxzEGMCE0NNTijrROs0aPHk1VVRU7d+5k8ODBNDVZGneF8xqduLu7d/vFarWLXUxaxXft2rUcP34cV9feiSr4zDPPoFarMRgMODs7mxdkRowYwaOPPkp0dLR5Ucbe3r5X+iAXOp2u0+n6tGnTaGhooLm5mcDAQL755htqampwdHRk8ODB5OXlkZeXh5OTE7GxsWRkZNDY2Ehzc3OfvO/U1FT27NlDfX09RUVFzJo1S7a6dTodX375pfm75ebmhqurKyNGjADAz8+PI0cufJOfnAIsSdI93ZyXgKdla9BErwmwKVB1qSRJ+4UQUy25pINj5xhuTCuXHwGMHTu2x4adnJwc9u3bh4ODA8HBwd2OaPPy8nok9G1Rq9UEBASg1+uxt7cnJyeHiIiIc8oNGjSog6vlJzk5GTAugEyePJnVq1ezefNm9Ho9FRUVpKcbA6jFxp6fj/nZ92rBggUUFBRQWVlJUFAQo0aNora2lhdeeIGRI0eay/V34W2lK1tpaGgoKSkpGAwGjh8/TlVVlXmB9ttvvzVPgVtHnG3ff2+jVqvZu3cvBoOBefPmAfDaa69hb29PeXk5e/bsOW8BTk9PJykpidraWkpLS9m1axcAhYWFGAwGFAoFbm5ueHh40NLSgoODA4cPH8bPz++C3pPcI+C+ojdHwFcDt5gcmB0BNyHE50CJECLANPoN4H9xU3vFyH02e/fuRalUMnjwYCZP7jixQFZWFmfOnAEgMjKyR/VrNBqqq6vx8/MjIMDoJmhvb09mZiYuLi4X1PfzJTEx0fwFHD16NH5+fqjVao4ePYq3tzeOjo4oFIrzFl6ALVu2YG9vj52dHQB/+MMfOHLkCIWFhbi7u+Pl5UVBQQHffPPNOdfm5OSgVCrN9+tSIjU1lerqaioqKszeJDY2NoSEhODu7o6Hhwfe3t7tHkwX8lDvKVqtlm3btqHT6YiOjiYnJ4enn34ajUaDk5MTZ86c4aabbjqvupOSkti5cyc1NTXU1tZy/PhxtFotzc3NGAwGlEol9vb2jBgxghtvvBFbW1ucnZ254YYbCA0N5YUXXui+kS7oDy6cF0qvCbAkSa8CrwKYRsAvSZJ0vxDiHYypcVaZfm81XfIt8IUQ4q8YF+GiMIaI7JQWWqjpJB506wJH6wJPK1ffdLXZhtfRtampqUycOJED2QdwcHCg3ra+0zY6wsHbgRPHT+Dk50QLLea+hESHdNpmb5GamsrBgwcpKCigpaWFkSNHct3c69BqtXyy8RPOlJ/B0c2RsEFhPPzww+fdN51Oh8JdwYSrJ+Ds7Myzzz7LT3t/MtqVleAR6sF7H7+Ht7c3P6b+SEVFBT4+PkRFRXHkyBEaGxsZM2bMRb03lqDX6ykuLj5HLBMTEzlx4gQNDQ0oFAp8fHxQKpVUtVSh0+nw9PTkxnk3otFoqKiowCPUg0M5h6itraWuro7Y2NiL9l5tVDaERIcQGBhIaGgoU6dOpbi4GCEEk0ZP4tEnH2XcuHHn1R+tQYuNyoaKigpOnDmBulJNRVMFDfUNuLm7ofJT4efnR8y1MQyfNJyJEyear5Xj/VtHwOfHKuBLIcQjGEPa3QkgSdJRIcSXQCbGINdPn48HREJCAg4ODjg4ODBkyBDz8c6+TGfT+iFxcHDAx8enxyOVrKwsqqur201XW//OysqitLS005G33FRUVKBSqfDy8kKpVDJ79mzUajU//vgje/fuZd68ecyYMcO8CNQZraaUznB2dmb69OmsWLGCAwcOcOjQIRwcHLj77rtRqVTccsstVFRUUFpair29PVOmTDHfk/Hjx/dbE4S9vf05//93332XyspKXFxccHBwICIiAi8vL2xtbdFqtXh4eHDLLbeYN7fY2tqSmZmJWm1cMJ84ceJFf7+tn+l58+aRk5NDUFAQ119/PStWrOjyulbTWWdMnz6d4uJi0tLSaGxspKKiAiEELq4u+Pn5MXv2bMLCwhgxYgTjxo2T9T1dLiaISzoWROzYWCl5X7L5dUdCkZeXx8mTJzlx4gTh4eEW27qysrJ6bJvVarXY2dmZxaXVs0BOP8sLYcuWLezYsYP8/HxmzpzJ4sWLZat79erVfPzxx2i1Wvz8/Bg+fDjPPvssw4cPN9u/bW1tL9rUuzdotZE7Ojqi0+nMopKWlsaxY8eora1l6tSp+Pr6kp2djY2NDc7Ozu1cEPuCrVu3sn79ejIzMwkODubWW2/lueeek6XuJ554giNHjpCbm4skSfj4+ODt7c3gwYMZPXo0NjY2jB07tkObt5tw23++myNsbW0lN7ezMyR1TmVl5Xm31Ztc0jvhWtFqtahUqg5HaUVFRZw4cYKysjJmz7Y8n+H5LIyd7Z5UWFh43nXJTUZGBvv37+fEiRMIIWQVX4AffviB6upqmpubcXZ25qabbmpnU/bz86O8vFzWNi8mWVlZODo6niOmer2e6upqqqqqcHV1xdHRkby8PBobG3Fycupz8QWjH/qZM2cIDg5m1KhRPPnkk7LUGxcXR2FhIYWFhVRXVePt482QIUPw8vLC09OToqIiWlpazN4PcnM5jIAvCwHualfbtm3bUCqVTJ48+aKPvvqD8AKsXLmSnJwcDhw4QHh4uKy+n4mJibz11lsUFhaiUqnw9fXlrrvuYv78+e3KOTs795uZwPnQ0f8yLy+PrVu3otfriYyMZMCAAURERKDVasnLy7uong6dMW3aNOrq6vDw8GDevHmyPXj1ej1/+MMfqKmpobqqmrDwMO6++25iY2MpLS3lt99+o66uDnd3d9nND2A0QVgX4fo5GzZsYNy4cUyfPr3f2hl7kw0bNrB7925Onz5t3ln217/+Vbb6FyxYwK5du8zeD9dccw0Ajz32mGxt9HcGDBjAmDFj2s2+VCrVeW91l4t33nmH77//noKCAlQqFQ8++CALFy6Upe7x48djMBjQarUoFAq8fby55557WLp0KXl5eWRkZBAZGcnixYt79XtnHQH3c2699dY+/yL0Jfv27ePgwYPU19czd+5cs1+qHDzxxBMcOHCA2tpaVCoV4eHhPPvss7LVfykQGhra72zaer2eBx54gMbGRrKzs4mMjOSuu+6STXyXLVuGTqdDo9GgUCgYOXIkd999Nw899BDwv3vSE3Pf+WIV4H7OlSq+KSkpbNq0iR9//BEhBOPHjycwMJCnnnrqguvOyMhg5cqVHD58GI1GY95gMXTo0H5jcrlS2bRpE8eOHaO4uJjCwkL8/PyYMmVKl//37jwd2vLMM8+QkJCApkyDykNFUFAQ27Ztk6v7PUIIIXsIzb7gshbgK5XXX3+dEydOEBkZSVhYGBs2bOiyfFdbbM/m22+/5fjx42g0Gnx9fbn66qv58MMPZei1lQshMTGRuro6fv31V4qKiggJCeHdd9/tdnONJeKr1Wp59dVXSUlJQVOmIXxAOFFRUXz11Vdydf+8sI6ArfQr0tLSePnllzl16hSRkZEsWLDAPDXsCkvFd8mSJcTHx1NbW4urqytPPvmkLKNqKxeGWq3m7bffRqfTUVtbyw033MDvfvc72WYkTz/9NLt376ZcU070sGiWLVt2UUwMXWFdhLPSr3jiiSc4dOgQJSUlPPjggwAWia+lrFixgq+++gpJkpg+fTqTJ082t2Olb0hJSQHgzTffJDc3F29vb9kfis8++6w5vkNMbAzr1q3rN6Ym6wjYSr/gpZdeYs+ePWazQHc7nHrKrFmzyMrKQgghSxxXK/Kwfv166urqOH78OBEREWzZsqXbXY2WsnTpUgC++uorbGxsuO6661i6dGm/EV+rDdhKn/H000+Tm5trjmJWVlaGq6srs2bN4uOPP5atnQ0bNrBmzRpOnz6Nq6srY8aMYcuWLd1el5SUxPXXXy9bP6wY0el0fPTRR2RlZaHRaDh58iRVVVVcffXVfPHFF91ev3btWgICApg7d26X5V544QV27NiBRqPBwcGBoUOHdruOAMbPi5yzru6wjoCtXHQyMjJQq9Xk5eUBxsSKM2bMkH1U+sQTT7Bv3z5z5K5Jkybx5z//2aJrreLbO3zzzTfm/4lSqSQwMJDo6GiLxBeMW6i7E9/ly5fzzTffoCnT4OXtxd13382qVau6rTstLe2iii9Yo6FZuQikpaVRXV1NWVkZP/30E6WlpRQVFaHVamlpacHb21t28X3++ef59ddfKSsrw83NDZVKxaRJky7pnWyXImlpaZSVlaHT6di2bRs5OTlIkkRzczP19fWsXr26RyFEuxPImTNncubMGTRlGpROSjw9PS0SX6BXdrt1hdUEYaXX2bx5M8eOHUOhUFBeXk5RURE6nY7i4mIAPDw8ZM1mq9Pp+O6779izZ49ZfAcMGEB4eDj33XefbO1Y6Z64uDh2795tDmheXl6OJEnmqGPXXnvtBcVvPps1a9ZQWFiIVqvF2cWZwYMH4+DgIFv9vYHVBGGl19i8eTPp6enmrLp1dXUUFhbS3NyMm5sb11xzDQ4ODvztb3+Trc3p06ej1WopKysjNDSUAQMG8Oc//7nDLB5Weo/nn38etVqNn58fubnGzO0GgwFbW1vs7Oy48cYbefPNN2Vpa9OmTezfv5+vvvqKyopKPL08iY6OJiEhQZb6ewvrCNhKr6FWq6mpqSEoKIjo6Gi+++47vL29iY6OpqWlhZtvvplFixbJ2t7vfvc7SkpKqKmpwd/fn+nTp7Ny5UrZ2rBiGa0hL0eMGMGhQ4ewtTV+RaOiorj33ntlHfW+8cYb7Nu3jyNHjlBZUYmvny8TJkzg8ccfl62N3sQ6ArbSK9jZ2XHo0CG+++47AAICAggLCyM2NpZFixbJ5moEkJmZyauvvsrhw4exs7PDxcWF4cOHWyy+ycnJXWaVTk1NJS0tjXHjxrXLiGClYxISEjh9+jQ7d+4kODiYkJAQ/P39ee6553q0tV6j0fDf//63ywf1Tz/9RF5eHuWackJCQxg3bhzPPvusRSKfmJh4TraZtixfvhxPT0/GjBlDYGAgzc3NVFRUcPz4cYYMGdIuWcL5cLkEZLcKcD/is88+Y/PmzRQUFGBra0tMTAxOTk4EBgYyYcIE7rzzTlnb++Mf/0hqaiqnTp3C1taWYcOGERkZydtvv21xHV2J70svvUR+fj41NTW0tLRYBbgTMjMzAaPniVKpZOTIkURFRREQEMDgwYNRKpU9Et+0tDSOHj3aqenos88+4y9/+QtVVVW0tLQQMTCCqVOnWmzO6s488fjjj3Pq1CkADh8+zIgRI9BoNIAx04yjoyNOTk4Wv5/OsApwFwghHIFdgIOpnf9KkrRcCPE68BhQZiq6VJKkBNM1rwKPAAbgWUmStvdG33oS++Bi8MEHH5CUlERRURFOTk7ExMTw5ptv9mqkrTlz5pCTk0NDQwPOzs5MnjyZf/zjH7LV//LLL5Oammq2J4Mxa4ZcmRguF5KTk1m+fDkA11xzjSxmn3HjxnXqlTBv3jxyc3MpKSnBzs6O8PBwHn/88R7tauwqq8ycOXOoqamhoKAALy8v6uvrsbOz480330Sj0cg2e7PagLunEbhOkqRaIYQdkCKEaA2d9J4kSX9pW1gIEQ3MB4ZhTMq5Qwgx6HzywnVHfxLfGTNm4OjoyODBgwkKCsLX19f8hewN1qxZA0BJSQkajQZPT0/GjRsnq/hu3ryZPXv2oNfrcXd3JyYmxiq8Z/HHP/6RzMxMPDw8zEHcb7nlll5rLysrizfffJO8vDyKiozJxq+99lref/992UTxpZdeIjc3l4qKCqZNm0ZQUFA7NzY5TWeAVYC7QjImm6s1vbQz/XSVgG4usFmSpEbgtBAiGxgP7OmtPvYlSUlJbNu2jfHjx+Pt7U1LSwuNjY0sWbKkV9tNT08nLS2N4uJi/Pz8CA4OZv369V1ek5mZaVFqnfT0dDZv3kxiYiIqlYo//elPuLm5XXQf0f6MWq1m7dq1SJJETEwMBoOB/Px8Bg0a1GsmmpycHN5++23S09PJz8vH2cWZiIiILjdwpKenU1dXZ3EC2WXLlrF9+3YcHByIjo7m0Ucf7dI8daFYR8AWIISwAfYDkcCHkiT9JoSYCTwjhHgQ2Ae8KElSJRAEpLa5vMB07Ow6HwceBwgJDenN7vca6enpREVF4ebmhpeXF9nZ2QBdLmrIwYoVK0hNTUWj0TB58mSuuuoqXnzxxW6vszSv2R/+8AcKCwuJiIhg06ZN/Wqm0V/w8vJi5syZVFRUoNPpKCsr4/rrr++1TNmJiYls3LiRvLw88vPy8Q/wZ9y4cWzatKnL6yz1ttDpdKxevZrExES8vb3NKY96U3xbse6E6waT+WC0EEIFfC2EGA78A3gT42j4TeBdYCHQ0d08Z8QsSdJHwEdgzIrcOz3vXVo/3K220YvhZ7tkyRJ27NhBU1MToaGhssdyvffeeykrKzNnYLCKb8fY29ubR7p5eXmUlJT06gzhp59+Ijk5meqqagYNHsRnn30mW6LQd999l/3795OTk4NSqSQqKop58+YBxoW6tLQ0gF4xqVlHwD1AkiStEGIncFNb268QYh0Qb3pZALQd0gYDRRfadk8i/l+OpKSkkJOTg1qtxtnZmfXr18uaLPKDDz5g27ZtFBUVMWnSJGJjY89JyGmlY3o7pdHChQvZunUrzs7OhA8IZ9++fbLUm5WVxbZt2/j3v/9tXmR95pln2nnpnDp1ijFjxvRq3GC5BVgIcROwGrABPpYkadVZ56cCW4HTpkNxkiT934W02ZteED5Ak0l8lcANwJ+FEAGSJKlNxW4Djpj+/hb4QgjxV4yLcFHA3gvtx5UsvgAHDhxAo9FQW1vLLbfcIqv4btiwgc8//xw3NzdcXFys25X7EU888QQZGRk06Ztw9XflpZdekq3ulStXotFo0Ov1eHl5oVAouPHGG9uVmTFjRq+GrpR7BGwyl34I3IhxMJgmhPhWkqTMs4r+IkmSbE+V3hwBBwAbTW9MAXwpSVK8EOJfQojRGM0LZ4AnACRJOiqE+BLIBJqBp3vDA+JKIiUlhb1791JaWsqGDRvMvplyEBcXx3/+8x/Ky8sRQhAdHd1rdkwrPeOBBx6gtLSUrBNZDBo8iIULF8oWPD8tLY0zZ86Y3dgmTZrEe++9d85A52LEDZZ5BDweyJYkKQdACLEZo2PA2QIsK73pBZEBxHRw/IEurnkLeKu3+nQlERcXx/vvv4/BYODWW29l//79XU4HdTod1dXV7dKrd8Ybb7zBzz//TGFhIWFhYUybNk02u6KV8yctLY1vv/2WX375hYryCoKCg7j55pu7zJCRkpKCk5MTKpWq27WI9evXs2bNGtRqNV5eXgwdOrTP8gGex044byFEWxvMR6b1pFaCgPw2rwuACR3UM0kIcQijefQlSZKO9qQTZ2PdCXcZ8vzzz5Oamkp0dDS+vr4WeTo4OztbtHC2bt06c1jMAQMGMGPGDIvqt9K7xMXFkZ+fT1xcHA0NDQQFB/HTTz91+0C1dNby7rvvEh8fT1lZGTExMYwZM4Y33nhDjq6fF0IIc5wMC9FIkjS2qyo7OHb2In86EGba2zAL+AajqfS8sQrwZcbtt9/OyZMn8fPz4/rrr5fVLrt582bWrl2LEILAwEA+/vjjXl1EsmIZaWlpLF++nIaGBhobG7nlllt48MEHLZrNWMIbb7zBjz/+SGFhIVOnTmXjxo2y1HuhyGyC6NYJQJKk6jZ/Jwgh1gghvCVJ0pxvo1YBvkyYPHky9fX11NbWmqeccopvWloab775Ji4uLsyfP5/58+fL9gW3cn4kJSUB8Pvf/57cM7n4B/izYMEC2UJVgnFkHRcXR0tLC+PHj+834tsLbmhpQJQQYgBQiHFX7r1ntekPlEiSJAkhxmNc2yq/kEatAnwZ8MQTT6BWq3FycmLs2LGyb/t95plnOHjwIJ6enjz99NNWN7N+wr/+9S/UajXZJ7MJDQtlypQpsonv5s2bAVi1ahXNzc3ceOONssaevlDOwwTRJZIkNQshngG2Y3RDW29yDFhkOr8WuAN4UgjRDNQD8007fs8bqwBfwqxZs4YdO3Zw9OhRYmNjCQsL4y9/+Uv3F/aAzZs3k5SUhKurKzNmzLCKbz9Aq9Vy6623kpOTg7ZSS0xsDH/4wx9k3Unp6enJypUrUSgURERE9CvxbUXunXCmoGAJZx1b2+bvvwN/l7NNqwBfosyePRuNRoNGo8HPz4+YmBhzKnE52LJlC//+9785duwYISEhzJw5k9/97ney1W/l/MjIyOCee+6hsrISV1dX3N3dSU5Olq1+nU7HHXfcQXZ2Nm5ubtx33339cpFV7hFwX3Hpv4MrkAceeIDMzEwkScLOzg6lUsm1114rW/2JiYmsWbPGHMPVw8ND1g0cVs6PZ5991uzX3SpAPdnGnpWV1aV/bmJiIuvXr+fkyZPmOCVz5syRo+uyYxVgKxed9evXk5qayr59+zAYDNjZ2REQEMCjjz4q6yaIXbt2UV1djV6vJzg4mBkzZlhTzfcxL730Etu2baOiogJbW1v8/PzIyMjoUR3dbY6Ij4/n5MmTODs7Y29vz6BBgyzeUNEXW/6tsSCsXFS2bt1KcXExdXV1ODk54evry7XXXiurXXbt2rUcO3YMnU6Hra0tI0aMYOrUqbLVb6XnqNVqMjMzKSsro8XQgru7O6NGjZK1jQ0bNpCZmWkW0SFDhvQootnFFl/rCNjKRSMpKYkPPviArKws6uvrUalUxMTE4O3tLavL0bp163BycjKnRIqMjOStt96yRjbrQ+bOnUttbS3Hjh3DxcWFkJAQHnvsMRYuXChrO2VlZTQ0NKDX6xkzZgwfffRR9xf1MdaURFZ6nXfeeYf//ve/aLVaamtrcXV1JTQ0lA0bNsjaTkpKClqtlurqapqbm/Hx8SEyMtIqvn3IhAkTzFvEW3MEzp07V1bxVavVnDp1ioMHD2IwGEOvjBgxQrb6ewvrCNhKr/PEE0+QkpKCTqejqamJ4OBgAgICeiV+8JEjR0hJSaG+vh57e3tGjBjBNddcI3s7VrrnvffeIy0tjdOnT9NQ34CLqwtDhw5lxowZso98N27cSFFREWfOnKG2thZPT08GDhwoaxu9gVWArfQqCxYs4ODBg+h0OpqbmwkPDweMMWT/+te/ytqWRqMhPj6egoICGhsbGThwIDfddFOXyRetyE9mZibPPfcc2dnZtLS0mMVXoVDw2muv9TjLRHdeDzk5OaSlpVFYWIher8fPzw9fX1/c3Nwu9K30OkII7Ozs+robF4xVgPsZ8fHxfPjhh+Tn51NbW4vBYMDDwwMXFxfuuOMOHnvssV5ps6ysDAcHBwYOHMjkyZOt4tsHPPzww9TX19PU1ER1VTUuri5MmzYNOL8UP915MCiVynZxfceNG8ewYcMuibCi1hGwFdlJTExky5YttLS0UFtrzGdqb2/PvHnzejXylKOjI83NzURHR+Pk5MTTTz/da21ZOZeffvqJ/b/tR9GsoKy4DGxAKASTJk3qNnfbhXDq1CkUCgU+Pj5Mmzat1xPCyo11Ec6KbHz22Wd8//33ZGRkoNPpsLOzw9vbm2HDhvWq+Gq1Wr7//ns8PT3517/+1WvtWOmY+vp67rnlHkJHjGLE9Dk079zJ6Yx0IgZFyJ6372zWr1+Ph4cHt9xyCw899FCvtgWQnJzMsGHDZElPbzVBWJEFvV7Pa6+9xqlTp8jMzKSqqgqlUsmoUaPYvHlzr/pXvvvuu6Snp5OdnS3rNmYrllFbW0uwezD3r1jB7S+/DEBTYyNvzr6ZpS882attL1++nLS0NPbv339RfHiTkpKYOHGibF41VgHuBiGEI7ALcDC1819JkpYLITyB/wDhGFMS3WVKS48Q4lXgEcAAPCtJ0vbu2okMj6Q0t9T82jfMl+wz2d32LzExkfDwcIt2+rzzzjsUFhYyZMgQxo8fb3HK7s7IysrC0dGR0NBQli5dyv79+1Gr1eZV6KFDh/Lyyy/36hdj06ZNlJSUUFVVhZubG3q9vtfauhJJT0835+IbPXq02XNFrVbj6+vLfzb9h0ULFvH8xo1c94AxSYwQAjsHB6ImTOTwwcPMnD1T9n6tXbuWgoICfvnlF/z9/S+K+Kanp1NRUUF5eblVgM+iN0fAjcB1pujxdkCKEGIbMA9IkiRplRBiCbAEeEUIEY0xBucwjEk5dwghBnWXF640t7Rd3PpSUdp5YYziFxcXh0ql6nC1t+3K8Zo1a9i2bRtarRYAW1tbFi1aZNGbj4uL4/RpY/LUtsFMtmzZQllZGSNHjqSkpIT9+/dz/PhxbG1t8fX1Zdy4cfzzn/+0qI3zYe3atVRUVBAcHMyQIUM4ceIEDz30kOw7q65UEhMTOXr0KIWFhVRUVODr62tO1f7ee+/R0tLChg834OgXwFs//8wI0+KaEILWyIZHf9nFnUvkCymq0+lYvXo1Go0Gd3d3AOzs7NiyZYtsbXTFDz/8gI+Pj6zB+60C3A2mOJm1ppd2ph8JY6K7qabjG4GdwCum45slSWoETgshsjEmytvTWRuKlo6Pu9Z03q/cPUcIsVHhq/TlxmET4ayyA5W+OJTr+eCDD8hIT6c4/SRDhgwB4IboCe3q1mq1qFSqc9qIj4+n+nQhQ3yCzumPk9aAZ4Md6v1ZZGdnU3WsAI8GW5RKJSO8B/DUrQ902f8LpfG0BndbW+w1jeSfOoVzVQsLb7rTOBLqxXavBNRqNXmpRynJzkar0eDt6cn1Q8fjWmO0fzqU61n1978z/s47eeof/0BhimXQVny/evtt6vMLuH3ydGxk+n9sWf8fhnuGUax3QFuppbqklGCFOyE2qovyP/dtVhLlEiDr57oXArL3Cb1qAzZlRN4PRAIfSpL0mxDCrzUtvSRJaiGEr6l4EJDa5vIC07Gz63wceBwgNCTk7NNdkpmZibu7OzNmzOh0KvTFF19w9OhRDh06RGlpKYMHDyYmxphb9M4772xXtnVbcCubN2+mpqaGhoYGYmJizO48Wq2W+vp6kpOTOXHiBIcOHaKyspLy8nL0ej2RkZEMGzaMv/9d1lCj7dDpdLz99ttUVVXh6+tLWFgYu3bt4s4777zo+/gvR9LS0khJSSE7Oxt3d3ceeeQRBg8eTHV1NXFxcZw4cYI333uPa+6/n6fXrkUoFOZ4tq3i+9O//sW/X3uNqsJCWcXl7rvv5sSJE3h4eHDw4EGAXl/ga0WtVjNp0iQmTpwoa73WEbAFmMwHo4UQKuBrIcTwLopbkhQPUybTjwDGxsZKefn551x0Nq2Rmnx9fTvM3tt6Pi0tjZ9//pni4mKcnJwIDw8nODiY5cuXtyuvVqspKChg3LhxpKWlAZCamkpFRQV+fn4sXry4Xfm9e/dSVVVFSUkJmZmZHDlyxOho39DAsGHDuOmmm3o91m5mZiZNTU2EhYURFRXF1KlTZQ3gfSWRmJiIk5MToaGhHD9+nIqKCtRqNWVlZbz44otme69er2f//v0UFxezbNUq7n/rLW5/5RWEEO3EV5Ikvvrzn/n3a69RkZeHUqmUtb/Ozs7ExsaSlZWFj48P48aNk7X+jkhOTmbQoEG4ubm129qs1+upq6vrcObYE4QQl8XA4aJ4QUiSpBVC7ARuAkqEEAGm0W8A0Gq07TYp3vnS+o/y9vZGq9Xy7bff4u/vbxYge3t7Vq1aRW5uLvn5+Tg6OtLY2Mjw4cN56KGH0Ol05hHz5s2bSU9PZ/jw4Xz99dfU1dWZ677nnns6XNQLDw/n73//O7m5ueTk5FBVVYWzszPBwcFMmTIFFxcXUlJSetUB3mAwcP311zNq1ChZ3ICuVDQaDbt27cLJyYm6ujqzv7a7uzsrV65Ep9MRHx+PwWDA3t6ehMRE/rFhwzmLbWAS35YWPly0iINbt1JVWCi7+LbF0tCSctDY2EhTUxOnTp2ioKCA8ePHk56ebp6BNTc3X1D91hFwNwghfIAmk/gqgRuAPwPfAguAVabfW02XfAt8IYT4K8ZFuChg7/m0nZOTg5+fH87OzmRkZNDY2EhlZSXTp09n7969jB49ul2A8XfffZeDBw9y5swZhBD4+/tz//334+3tTVRUlFl8W0fIdnZ27Nmzh+rqaiIjIwFYtmxZh33Jysrio48+4sCBA9TU1FBeXk5QUBB+fn6MGTMGMH4xekt8t2zZgq+vr2z+l1c6H374oVl4m5ubCQsLY9SoUTg5OQHGmZBSqcTR0ZHb7r8fz6ioThfbWgwG1jz5JCe2byf/2LHLYkSXk5PDwYMH8fX1RavVUldXh4+PD+np6Rw6dAhHR0cA82Lg+aJQKHBwcJCjy31Kb46AA4CNJjuwAvhSkqR4IcQe4EshxCNAHnAngCkB3pdAJtAMPN2dB0RnFBUVkZSURG1tLY2NjQDm0W7baXdSUhJJSUns3buXiooKPD09GT16NCtXrqSpqekcO/E//vEPysvLaWpqwtXVFSEEjzzyCEVFRSQkJDBr1iy0Wi15eXn4+Pjw448/UlVVRVpaGnl5eej1etzc3Bg4cCDTp09HpVIRHh7ea1PChIQE8vLyOHPmDMOGDeuVNq40mpqa0Gq1eHp6UldXh8Fg4NixY4wZMwadTkdUVBR7UlN59LnnuOqeezpdbDuSnMzGV1/FvrqarAMHLgvxTU1N5bfffqOiooJrr73W7GpZX19PQkIC9fX1eHl5YWNjY45tciFYF+G6QJKkDCCmg+PlQIfpFSRJegt460LbdnBwYMaMGRw/fhwnJycCAwPbRRBLS0vjjTfeQAhBYWEh1dXVeHl54e3tzZIlS7C3t6epqclc/qWXXjL7MXp5eaFSqbC3t+fee+/F398fME5NV6xYYbYxZ2dn89133wFQUFCAg4MDrq6uDBw4kAULFmAwGIiKijKXT09Pv2D/4rZotVoyMjKor6/H29u7X41+WxclKysriY6OJisrC/jfFLkz75K+ZtWqVZSXl2Nra2uOGOfp6cl1111HcnIyv/32Gy4uLixbuZKr7rmny8W29x9+mC83bqRRpyMuLu6ySHbq5uaGn58ffn5+ODg4cOTIEQCqq6spKysjOjqa8PBwWeKMWG3A/ZjWEWWr32FaWppZgNetW8cXX3yBWq3GwcGBxsZGrr/+esaOHUtQ0P+cLlpHvwsXLuTUqVPY2dnh6+vL2LFjGTlyZDuTQWhoKFu3buXBBx8EjB/EV155hZycHDQaDSqVioEDB9LU1MRdd91l/pC29u/JJ5/E3d0dlUolS6jJdevWUVBQgMFgICIightuuKHTsmq1mtzcXNlXqVvJy8sjIyMDV1dXpkyZQlxcHHq9nrKyMpydndm7dy8tLUZ/wrS0NAwGA+7u7sydO7dX+nM+rFu3joMHD+Lp6UlsbCyjR4/G2dnZ/FBes2YNKpWKlpYWVv/jH3gNHsxT//jHOeLbutj2xWuvUVVQQFFREa+99hoLFy7sk5Q+chMdHY2vry+HDh0iOzub8vJyFAoFOp0Og8FAbGysbIMMqwniEkGn05n/3rJlC2vXrqW2thYhBJGRkSgUCgYMGNBhlLHk5GR0Oh0qlYqQkBCmTJlyjitaa+LKVs+H+Ph4PvjgA44dO0ZTUxM+Pj7ceuutjBs3DrVaTXBwMFFRUebrly1bRl1dHf7+/rLF+W1oaGDQoEFcc801HTq/63Q6kpOTcXZ2Rq1WY29vL7sAaDQaqqurOXz4MCqVChsbG1JSUrCxsaGiooKGhgZaWloYPny4eSErJiYGLy8v2fpwoWRmZvL9999TVlZGSEgIixcvNj+Yc3JyOHbsGDY2NowdOxZvb2+m33YbIaNHs3jdOhQ2Nh0utu3dsoXqwkK2bNlCeno6bm5uF90bpTfEvrXOvXv34uDggJubG3V1dbi7u5OSkoKdnR06nU62doUQVgHua5rbmAk6Qq1Ws337dnOG13Xr1lFeXo6LiwuBgYG4urry9NNPm93JysrKzNOjJUuWoNFocHV1ZciQIXh7e3c4Kms7tV+2bJnZ3iuEwNvbm5iYGK677jrKy8spKirCzs7OHFowPj6e3Nxchg0bxi233CLbfTnbDa4tW7duJS8vDxsbG3x8fFAqldxwww2yfyFbzR5tHyqZmZmkpaVRUlJCQEAAHh4euLm59buEn6mpqfzyyy/mNYSamhpsbW3N4puSkoJarSYgIAC9Xk9DYyMzbr+dyfff387mC+3F90BcHEd/+43GxkZ27dpFU1NTn4R+lPt/rdVqaW5upq6ujpaWFtzc3KiqqsLR0REbGxuqq6vx8fHB1tYWjUYjWzAeazjKPqa6urrL8wEBATz00EOsXr2ad955h+LiYvOod+bMmfzud78zj5BbEx3GxcXx7bffmo/7+fkRGxtLVFQU5eXlBAQEdNjWfffdx5kzZwBwdXU12wbHjh1Lbm4ulZWVhISEmKNOZWRk8P333zN+/PguBVNu1Go1er2eESNGXNSRV2JiIgUFBfj6+jJlyhRZ7d1yc+bMGRQKBUqlEqVSiUqlYsGCBebzAwcOxM3NjcDAQLZv387CxYuZ+tBDHdp8W1paWLNoEelxcRRmZWFvb8+yZctwdHRk2LBhvRLf+WKjUqnIycmhtLTUvCjZGu/E1dUVHx8fRo4ciYeHh2xrEQqFwuxRcSlzSQuwJaugK1euJDk5mZKSEsLCwggPD2fatGncd999wP9svY6OjhQXF5OQkIBOp8PPz4/o6GhuuummLk0DSUlJnDx5Er1eb86pdfPNN+Pn58f06dPJzMwE4NFHH23nVfH1118TFhZ2UcUXMGdTvtjTXgcHByZPnnxRfVHPl3nz5pGdnY2jo2OH//uAgAACAgL4z3//ywNPPcUDnW2wMInvgW++MYtvSkoKjo6OTJo06bJJ+bRlyxZsbGywsbFBqVRSVVVFdXU1w4cPp6amhjvuuEP2kb4QwirAfY1DN/+AnJwcDh06RElJCfb29nh6ejJ27Fiz+LZl586dgDFIicFgwNHRkYkTJ3YpvllZWWzdupXKykry8vIYMGAAALNmzWLcuHFs3bqV0tJSXF1d24lvYmIidnZ2srji9JSIiAgCAwN7pe7WoEVnezBoNBpCQkJoaGjolXblxt7evsMdk22pra3lnkce4bkNG7rcYJG2ZQvFp06Zp/179uzB3d2dQYMGyRqcpq9IT0+nuLgYMM4MmpubKS8vx2AwEBAQgIuLS6+YWVpnKHIihLgJWA3YAB9LkrTqrPPCdH4WUAc8JElS+oW0eUkLcGdPwLy8PN5//3327dtHUVERCoWC4OBgnnvuuU5TuxQUFHDkyBFycnIYOXIkS5Ys6XK6pNVqefnll1Gr1TQ2NuLiYkyc2LptOTMzk2PHjjF8+HBmz55tvi4uLo4ff/yRIUOGcNVVV13Au+8Zer2e9PT0XrM5JiUlMWDAgHYPrK1bt5pXwmtqanB3dyc6OpqMjAxOnz5NVVUVnp6e7e5Pf8dgMBAXH89dDzxwzu62szdYpMfFkfj119jb27Nu3TpOnjxJaWkpN91000XZDtzbbNq0iaKiIsrKypgxYwYhISF8//33nD592uz10lvILcCm/QofAjdi3JWbJoT4VpKkzDbFZmLcIBYFTAD+Yfp93lzSAtwZCxcupKamhsLCQoKDg/H29iY+Pr7T8q1PcZ1OR1hYGB999FG3bdx7770UFxdjZ2eHg4MDc+fO5fe//z1gHBkXFBSYU7y0eh3s2rULtVpNREQEer3e7EPcm+Tl5bFv3z4MBsM5HhxykJGRQWVl5Tnim5SUxNGjR7Gzs2PgwIHtMi64uLgwc+bMS87tSq/XM2TMGOwDOg8l2brBoqGwkMKsLI4cOcLq1avZt28fer3evIaQmZlJRUXFJZF/7Wzy8vIAqKqqwsfHh/HjxzNlyhQyMjIoKSmhsrISX1/fbmq5MHphK/J4IFuSpBxT/ZsxRmhsK8Bzgc9MkR5ThRCq1rAK59voZSnAZ86cobm5mSFDhvDqq692u8puMBhoaWnBx8eHjz/+uMuymzZtYv/+/VRXV6NQKMxJE1vFt7V9Jycn85T8+++/Z//+/ZSVlaFUKpkxYwYAxcXFvToNTU9P58iRI5SUlDB16tReaaPtlu5WtFotR44cQaFQ4OzsbI6H24pc7nYXE71ez6CYGAbPmNHp7raf/vUv3n/oIV594QVmz5jB5s2bUSqV5Obm4uDggJ+fn9kv+1IVX/iff/1TTz3V7nhrRheFQtHri6ytny0ZCQLaRvYq4NzRbUdlggCrALfF2dkZNzc3Hn744Xbim5mZ2aFt7+jRo4SFhXUrUu+88w67du3CxsYGb29vJk6c2C6RYU5ODklJSecscjU1NdHc3IyLiwsLFizotU0PZxMQEMDevXuxt7e/qFNeJycnnJycmDBhwkV7r71JfX09wUOGMPb227vd3Zb2889m8bGxsSEtLc0cctTZ2Znw8PB+tzNRLkpKSgAYOnRor2fVPo8RsLcQYl+b1x+ZIiuaq+zgmrOjMVoUsbEnXJYCPHnyZG6++eZzPgSdLazY2dm1G8GeTetmi19//ZWamhr8/f0JDQ09J4tsSkoKAQEB59g0W1paCAwMZOjQoRdVkOzs7HBycrroKebt7e0ZNGjQZSG+tbW1eISEcO+bb3YbSrKqoAAXFxfztZWVlRgMBvz9/S+5jMPnQ6tv/cKFC/u6Kx2hkSRpbBfnLYnGKHvERtH6BO+ykBAKYBTGKGX1wFFJkkoupGE5GBsbK+0/cKD9M0iA1I1/cFvy8vK6NQO02i8PHz5MdHQ0N9988zl79zds2MDIkSPbTb1ag5PU19dfEV9AS1m5cqXZaX/YsGEEBwcDRtt5q2nn4YcfRqVSmTfIxMTEmDNFXyzq6+txDQhg8aefdhtK8syRI+0WhXQ6HevXr6ehoaHLh/uVjHBz29+NKHbK2LFjpX379nVfsLUtIbpsSwhhC2RhjFNTCKQB90qSdLRNmZuBZzB6QUwA3pckafz59L+VLkfAQoiBGNMF3QCcBMoAR2CQEKIO+CewUZKkTpID9S6FBQUXdL1are5yISwuLo6vvvqK48eP09jYSGRkJB988ME5blYZGRk0NTW1E9+EhAQyMjIA44KdFSNZWVns3r0bDw8P3N3dzbOL3bt3o9VqsbOzw8XFhXXr1qFQKMjPz6elpYXi4mLzlt+u0Ov1stjW9Xo9XuHh3LdiRZeeDmeHkkxLS+PMmTPs378fOzs7czorK8i2C643kCSpWQjxDLAdoxvaelOExkWm82uBBIzim43RDe3hC223yxGwEOLfGF0tfpHOKmhKJXQvUClJ0sYL7cj54OsbIJWVFZ8zAq6uviCzDADPPPMMJSUlnDp1iqCgIBwdHS1OYvjCCy/Q3NzM3Llz+9022/5AfHz8OWaalJQUGhsbiYqKIjQ0lCyTB0F4eHiPFnTk+JLHx3/HokUPMvGuO3j6n/80mx3O9nSwqdaxa9c+s/jqdDr+8pe/UFpairOzM2+//fYF9eNyx82t61FpV8g9Au4ruhwBS5J0TxfnSoG/yd2hnhAcHERZWbHs9cbHx3P0qHHm4ePjw+23397Ojaoz1Go19fX1AEydOtUqvp3Qkd/v2R4BgwYNOq9dc3KMsP7855UERA/hyTVrzhHf1sW2jRu/ZPbsW9vtxkxNNaY09PDw4NVXX73gflyq9Ndwov0RixbhhBB3Aj9IklQjhFgGxAIrLnQXiBz4hvpRKkravb4QdDodpaWlTJ06laamJlxcXCwSX8Ds4/nMM89ckq5WVzq33347AMePZzLnd0+haOPtAHA4OZn3H36YgoKqdottYDRD1dbWYmNjYw5XeaViFV/LsdQL4o+SJG0RQkwGZgB/QYZdIHKQfUTeEbCzszMLFy4kJSUFBwcHc863nmAV30uTVjcqBwclh3fupLmpCTuFAoBmvZ7PXn2VjRs3nyO+YPSHdnR0pKqqql/FMrbSv7FUgFtTA90M/EOSpK1CiNe7ukAIEQJ8BvgDLRj97labrnsM44IewFJJkhJM17wKPGJq71lJkrb34L3IyqXqJN+fSElJoa6ujl27dpnDcz766KNkZmaye/dumpubsbe3NwtWSUkJUVFRfTKCSktLQ6PR0NzcTHMzFB09ztJp07jOFGT/5399jq+dktmz53Vax/maTaxcuVgqwIVCiH9iSqwphHDAmOetK5qBFyVJShdCuAL7hRA/ms69J0nSX9oWFkJEA/OBYRjd3XYIIQadb144K31Lamoqu3btIjc3l8LCQmJiYqioqGDt2rXk5OSYQ4l6eHjw+eefm4Nr93TDiCVuhJbw9ttvU1RUiIuLK5GRkSQnJ/PDD/F8881/kSRY/uzL3HTT7MsiD5mV/oOlAnwXxpTyfzFlOQ4AunRuNO2PVpv+rhFCHMO4ba8z5gKbJUlqBE4LIbIx7s/eY2EfrfQjJk6cyJAhQ0hJSUGhUJg3g+j1egoKCjh48CDh4eFERERQVFTUbfSxzmjNRny+xMfHs2zZMnJyTjFhwkSmT59u9tu9+ea53Hyz1ZxgpfewSIAlSaoTQpQCkzH6AzebfluEECIcY4LO34CrgWeEEA8C+zCOkisxinNqm8ta91mfXdfjwOMAISGXfji/yxmVSnWOx4O9vT0RERHt7OQXYnK4EK+HlStXsnv3bnJyTjFgQAQ//vhj9xdZsSIj3ZkRABBCLMe4IaPVt8YO+NzCa12Ar4DnJUmqxrh4NxAYjXGE/G5r0Q4uP8ehV5KkjyRJGitJ0lhvbx9LumDFyjk8+eSTfPTRR+zalcyIESOvaLcxK32HpSaI2zCOYNMBJEkqMtl1u0QIYYdRfDdJkhRnurakzfl1QGucSNn3WVuxcjYffPABAD/++CMKhYKoqEH8+uuvfdwrK1cqlgqwXpIkSQghAQghunVyNEWP/wQ4JknSX9scbxs/8zbgiOnvb4EvhBB/xbgIFwXstbB/VqxYxIYNG9BoNCiVSkaNGsV//vOfvu6SlSsYSwX4S5MXhEoI8RiwEFjXzTVXAw8Ah4UQB03HlgL3CCFGYzQvnAGeADDtu/4SYwDkZuBpqweEFTkZNWoUp0/nEBYWzj333MPSpUv7uktWrnAsXYT7ixDiRqAaGAy8JklSlysWkiSl0LFdN6GLa94C3rKkT1asWMoDDzzA6dOnOX06h3HjxjNu3Dir+FrpF1gcD1iSpB+FEL+1XiOE8JQkqaLXembFygWSkpLC66+/TlZWFlptJePGjScpKamvu2XFihlLY0E8AfwfxljALRhHthJg3XNrpd/y8ccfc+rUKbTaShwdlZdcDjorlz+WjoBfAoZJkqTpzc5YsSIHa9asAYw58crLNfj7BxAWFsaiRYv6uGdWrLTHUgE+hTEAsRUrfYpWq+WHH34gODgYg8FwTurzZcuWcfDgQcrLyzlz5jReXt7cd999LF++vI96bMVK51gqwK8Cu0024MbWg5IkPdsrvbJipQ3vvPMOZWVlaLVa6uvrcXJywt3dHTCGD501axafffYZn3/+OZmZmej1ehoa6hk2bDiPPfZYf81RZsWKxQL8T+An4DBGG7AVK+eNXq+32B772Wef8fPPPxMdHY29vT2hoaG4urpSUWFc//X39yc+Pp7169eTk5NDfX09Dg4OuLq6MWfOHKv4WunXWCrAzZIkvdCrPbFyxdCTxbCkpCSUSiWenp6MHTuWAlMewCeffJIdO3awdetWUlJSOHnyJA0NDeb67ezsuP/++3ul/1asyIWlAvyzKQjOd7Q3QVjd0Kz0GsnJyXh5eTFy5EjuvfdeKisrmT59OgCbNm2iuLiY+Ph4CgoKqK+vw8nJmaCgIEaMGAEgS5hKK1Z6E0sFuDWtb9uIJVY3NCsW0xOzQyvl5eU8//zzZiFtTfmUlZVljulgzMNXh7u7Cnd3d3x9fYmMjGyXIt6Klf6KpTvhBvR2R6xc3lgivhqNhp9//pni4mIKCwsJCgpi3rx56PV6tm7diq+vLxkZGaSmplJaWkpLSwu1tTW4uroREBBAcHAw06ZNQ6fTWX1+rVwSWLwTTggxHIgGHFuPSZL0WW90Si7kypZg5eKQkJDAqVOnyM/Pp6WlhSFDhpj/hzY2Nnz//fccO3bMnFFDr2/EycmZ8PBwoqKiuPbaaxk4cCDbtm2jtra2r9+OFSvdYulOuOXAVIwCnADMBFIw5nzrt1jF99IgPd2YXLukpISDBw9SX1/PwoULueqqqwgNDUWtVpOamsoPP/xAfX09FRUVCCFQKp0YN24cgwcPZs6cOQwbNoy1a9ei0WhobGwkKSmJ66+/vo/fnRUrnWPpCPgOYBRwQJKkh4UQfsDHvdctK1cSGo1xg2VJSQktLS24uroyf/58wGg73rZtGydOnKC2tha9Xk9Tkx5XVzc8PDyIjY3FxcUFGxsbsrOzqa2tpbGxkaamJsrKyrpq1oqVPsdSAa6XJKlFCNEshHADSrEuwFmRAbVaTXNzMydPnsTe3p4RI0YwYIBxyWHTpk2kpKSQnZ1NRUUFdXV11NXpcHNzx8/Pj4iICO677z5++uknYmJiWL9+PbW1tSgUChwdHYmNje3jd2fFStdYKsD7hBAqjDGA9wO1WIOlW+kCS70eAgIC2LNnD+Hh4dx4443tknPGxcWRn58PQFFRES0tLbi4uOLn58dDDz3E4sWLycjIQKPRsG/fPoKCgqiqqsLDwwOVSiVrinjreoKV3qBbATZltviTJElaYK0Q4gfATZKkjN7unJVLl67Ed+vWrcycOdNcZt68eR2W02q11NXVmU0J/v7++Pr68uqrr7az7dra2nL06FHq6+tpaGggODhYVrFcvnw5SqWSqVOnEhYWZnaHs3JlIITwBP4DhGNMInGXKZHw2eXOADWAAePmtbHd1d2tAJtSEX0DjDG9PmNxzy8y1lFK/ychIYGkpCQOHjyIj48PgwcP5uTJk9jY2ODg4ICLiwt6vZ4dO3aQl5dHSUkx9vYOBAUFMW3aNEaOHMnEiRPR6XSUl5dz+PBhtFotdnZ2uLm5MXPmzPNOcd8Rzz77LAUFBTg7OzNv3jyr+F4ASUlJ2NranhNA6RJgCZAkSdIqIcQS0+tXOik7rSdRIy01QaQKIcZJkpRmacVCiBCMXhL+GONHfCRJ0uquniZCiFeBRzA+QZ6VJGm7pe2B1evhUuDEiRM0NTVRVFREeXk5eXl5NDc309zcjKOjI3Z2dhw9epQ9e/ag09Xi4eFJdHQ0c+fOZejQoTQ2NlJSUsLx48cpLS3l+PHj5hxvL774ouz9zc3NZcCAAQwcOFBWk8aVhk6n4+DBg0RGRp7Xppw+Zi5GLzCAjcBOOhfgHmGpAE8DnhBC5AI6TAHZJUka2cU1zcCLkiSlmzIo7xdC/Ag8RAdPEyFENDAfGIYxKecOIcQga164ywe9Xk9ubi75+flUVVUB4ODggCRJ6PV6bG1t0el0aLVadLpaPD29GD16NPfddx933nlnu7q2bt2KWq2msLAQgDlz5sje3y1bttDU1ERYWBju7u7mkbazc7c5aa2cxTfffIO/vz9z587tqy54CyH2tXn9kSRJH1l4rV9rImFJktRCCN9OyklAoil58T8tqb9LARZCDJAk6TRGv98eYepwa6drhBDHgCA6f5rMBTZLktQInBZCZAPjgT09bdsSsrKycHV1tU4pLyIFBQVUVlZy5swZGhsbkSQJOzs7FAoFNTU11NfX09zcjMHQjK+vH2PGjGHAgAH4+v7v856SksJvv/1Gbm4uTU1NBAQEoFAoZPd4SE1N5bfffsPNzY2SkhJOnTpFWVkZ1157LePGjev2+uTkZLRaLV5eXkyePFnWvl2K2NjYmF0L+whNVzZZIcQOjLP1s/lDD9q4WpKkIpNA/yiEOC5J0q6uLuhuBPxfjLbf9ZIknbdHuxAiHIgBfqPzp0kQkNrmsgLTsbPrehx4HCAk5PxNDkFBQTQ1NaHValGpVOddj5Xu0el0ODs7c/z4cQoLC6mrq8PGxgYbGxtsbW2prKxEr9fT3NyMEAJbWzvc3d3x9PRsF9MhMTERvV5PbW0tsbGxREZGMmTIEOrq5M0VkJCQwPDhwwkICECj0WBra4skSYSGhlokvmq1GgA/Pz8aGxu7KX35k5ycjMHQvyeykiTd0Nk5IUSJECLApFcBGN1wO6qjyPS7VAjxNcYB5AUJsMK0C26QEOKccJSSJP21m+sRQrgAXwHPS5JUbXSq6LhoB8ekDtr8CPgIIDZ27DnnLcXZ2ZktW7ZQVlbGU089ZdE1KSkp1tFMD0hOTkan05GRkcG4cePYtWsXpaWlKBQKHBwcsLGxobKykoaGBpqbm83HJcn4bzUYDDQ2NnL69GkcHBzIzc2lpKQEPz8/BgwYIPv/4t133yU/P5/GxkZmzZqFUqlEqVQSGxtLaWkpY8aMsaiegIAAvLy8+Pzzz6mqqroUF51kofXhlZubS1FR0aU82PkWWACsMv3eenYBIYQzoDDN9p2B6RjzaHZJdwI8H7jVVM61Z30GIYQdRvHdJElSnOlwZ0+TAiCkzeXBQFFP2+wJZWVlPXoyBwYG9mJvLi8yMzPZs2cP9fX1aDQabGxs0Ov1Zm+F+vp6amtr0el0NDXpsbW1w9bWFhsbG5RKJa6uxo9bVVUVTU1NBAYGkpWVRWBgIL6+vr0iamlpaeh0Om6//XYAfHx8GDJkCNdddx0//fQT5eXlRERYtv+ooKCAmpoaFAqF7P28FIiLi6O8vJygoCDc3d1pbm6+VMUXjML7pRDiESAPuBNACBEIfCxJ0izAD/jaNMC0Bb6QJOmH7iruUoAlSToB/FkIkSFJ0rae9NjkP/wJcOyskXJnT5NvgS+EEH/FuAgXRS9u9khKSsLFxYWxY7t11TNj6ZfvSkej0dDQ0EB9fT3l5eVoNBrKy8sxGAwolUqKiopMwttEc3MT9vYOKBQKnJ2dcXBwAIy+vc3Nzfj6+nL11VcTGhrKgw8+2Kv91uv1qFQqhgwZAhh9mcPDw9m/fz979+6lvr6eUaNGWbSCX15eTnh4eK/2t7+SmppqHu3GxMSgVqvNuxsvRSRJKgfOMcGaTA6zTH/nYAzX0CO6fDwLIe4XQig6E18hxEAhRGfzwKuBB4DrhBAHTT+zMArvjUKIk8CNptdIknQU+BLIBH4Anu5ND4iKigq8vb1l9Rm90snKyiIrK4uKigqKi4uprKykrKyMkpISCgsLKS4upqysjKamJpqamjAYDCgURltwqwuawWCgubmZmpoaGhoaaGlpMWe66E3UajW1tbXY2dlRVGSceOn1ekpLSzl48CAajYbTp0+zc+fObuvKycnhzJkzVFZWYmNj08s973taY3mAcfOMk5MTNTU1hIeHExAQQE1NDXl5eWRmZvZhL/sn3ZkgvIADQoj9GLcgl2EMRxkJTAE0GN3IzkGSpBQ6tutCB08T0zVvAW91323LycnJobm52ezDmZ6ezqlTp6ivr+fGG2+Us6krnubmZsA4/c7IyODo0aNotVqqqqpQKBQIIZAkyTzytbGxxd7eHicnJwDzwpwkSSgUCnx9fXFycuLMmTMMHjy4V92/cnNzsbW1xdbWlt27d3Pdddfh7OzM/v37zaYRV1dXc0aOVjIzM80bSVxdXbGxseHAgQM0NDTg6uqKXq/vtT73F7y9vc1/q1QqcnJyCAkJMS9Y7tu3D7VaTUFBASdPnmy3C/JKpzsTxGohxN+B6zCOaEcC9cAx4AFJkvJ6v4sXhp+fn/nvnJwcjhw5glKpJCgoiJycHGxtbRk5sit35v+RkJDArFmzetyHrKysK8KJv3U2sW3bNo4ePWo2RbS0tJi9ARoaGmhsbMDRUYkQAhcXF3x9fZEkCUdHR9zd3WlpaSE4OJiRI0fi6+vLoEGDOhRfOe+rk5MTbm5uODk54e3tjUqlYtasWZw8eZKrrrqKxsbGcxb9NBoN1dXVqFQqPDw8KCsrY+/evZSXl6NQKDAYDFfECLgtWVlZHDt2DKVSSUZGBunp6Rw8eLDd5osjR45YAyWZsGQrsgH40fRzydH6xc3KyuLQoUMolUomTJhAaGgoaWlp7Z7ebUlMTCQ8PLzdF/x8xBe4IsQXjF4iYNztptFokCQJg8GAra0t9fX1NDY2mke+KpWKlpYWHB0dUSgUxMTE4Obmho+PDwBRUVHMnj27y/bktLGGhobi4OCAv79/u//zc8891+k13t7eeHt7o9fr2blzJwUFBcTGxlJfX4+3tzdhYWHY2dnJ1sf+jkaj4dChQxQXF6NQKNi5cycajYaysjLs7OyorKwkMjLyUl6Mkx2LM2JcyqjVavLz8ykrK2Py5MmEhoaSkpJCTExMp9Pa2traK0Y4X3jhBcaMGcN99913QfXs2mV0eVSpVDQ2NlJZWYnBYKCmpoa6ujokqQUHB0fc3NxwcDAuvLWOjB0dHbnttttwcHCgubnZohGSnNNYlUpFQEAAw4cPx9HRka1btzJixAiLFl6bmppwcHCgpqaGgICAS2ZdISEhgauuukoWQUxLSyM4OBhPT0+0Wi0NDQ3k5uZSXV1NfX09Hh4eeHh4kJ2dzcKFCy+885cJl70Ax8XFUVtba/Z2GDlyJGq1muzsbPP0tiM6i9B1ufHQQw9RWFhIfn4+11xzzQXF02h101MqlVRUVJg3TbRuO1YobHBycsLZ2ZnGxkaEEAghcHJywt7eHhsbG0aOHIlWq5XjrfUYPz8/s53S19eX4ODgbq9ptQEDDBw40Cy+KSkpBAYGtlt/6E8sXbqU/Px8du7cyc0333zBbn2t9l4vLy8AXnnFGCpBp9Ph7++Pn58fwcHB2NnZmTfmWLnMBTgtLQ2DwYC7uzvR0dF4eHigVqvZtm0brq6u/fKLcTGZN28eubm56PV6BgwYcMHBjFpt6V988QVHjx4lL8+4RGBjY4O9vT1KpRI3NzcaGhoQQmBnZ4eTkxNCCBoaGsjPzyc2NrZPpqgpKSmcOnWKhoYG3NzcGDduXJcj7MzMTCoqKti/fz9OTk4EBQXR0tICGM1XtbW12NramhcY+xMrV65k7969Zu8FnU4nW90ff/wxa9eupaGhAQ8PD5ycnPD09CQsLIyioiL8/PxoamqSrb1LHUtzwvkBK4FASZJmmgLnTJIk6ZNe7V13KFrAteacwxkZGZSVlVFYWMiNM280xXuowU0hSEhIYPLMUSbxPffaVrKysmhubr5kppM95Z133qFaKsRGVUe4nx+hwzw6vJc9oc6mBIDDZ3ZTWn8KvX05tna2eHt74+npiSRJ1NdXY+8s4e/vj4ODA87OzrS0tODs30K9bSkZp39lyJAhF32V3NajHrcggY+DipCQEIaMC+rwfmg0GhISEigvL6e8vBw/Pz98Q724YeYEduzYQZZ6Px6htkR6DWxjvriw+yonL7/8Mr/++ivFmmLmzJlDSEgIs+6+Bjn6uHXrVj78bBXVzdU4uTnhE2E0x/zud7/j1KlTZ5n8+s896UssHQFvAD7lf4EpsjCGlOxbAe6A5ORkwLiL6eyEjJmZmedE1eqMqqoqi6agHdHfw+3l5eWxZ88elEold9xxB1FRUbIkrzxy5AharZbc3FzKSsvw8vbC1dWVmJgYwOimVltbi4+PDzNmzMDe3p4zZ85QX1/P9ddf32WchbOnrXJPYydOnIhGo+l24e/bb7+ltLQUGxsbhg4d2s5uHhkZSXh4eL/93+fk5LBnzx5KS0uJjY3lvvvusyi2hSVkZmaydOlSysrKcHZ2ZsqUKdxyyy3m+2kNetUxlgqwtyRJX5ri9SJJUrMQol9G1+jKluXu7m5xPRfyweyvX0CAFStWcOTIERwdHVm0aNE5fq0Xwr/+9S/Ky8spKizC28ebSZMmERsbi4+PD1VVVTQ3N9PY2MjixYvN4pmamoqTk1O3roBni21v2BC7E1+AhQsXolarKSsrO6fP/Vl8ExMTeffdd3F3d8fFxYWHH35YNvFds2YN8fHx5t1/0dHR3HzzzRbdzysdSwVYJ4TwwhQcRwgxEajqtV7JjFqtZs+ePVfMwlpHZGZm8sEHH3DmzBmUSiULFiyQVXzvvvtuTp06RZ2ujqDgIBYtWmR24UpLS6OkpISKiop224n1er1F4tvfCAgIOGdEp1ar++0ob+XKlfzyyy80NzdzzTXXsGzZMtnqXrJkCTt27KCwsBAnJyfef/99wDgIiY+PByx7sF2pWCrAL2CM1TBQCPEr4IMxVX2fo9FoOvXlBeOX/+jRo1f0qqtGoyEuLo6CggJaWloIDw+XNTD28uXLOX36NHW6Ojy9PBk4cCB33XWX+XxwcDB1dXXnBDOyt7e/5MS3I3JycigqKuqXApyUlERKSgr29vbExMTIOgh58sknOXXqFE1NTTg5OTFw4EAmTpwIGIPZ5+Xl4ebmJlt7lyMWCbApq8UUYDDG7cUnJEnqF0uZXYlvXl4eSUlJ+Pv7W2z7vdxYvnw5J06coKSkhMrKStauXStr/VOmTCE/Px9NmYaoQVEMHjyYL774ol2ZjkaMlwtbtmzBxsamX86unnnmGQ4cOEBtbS2BgYGsXLlSlnrj4+P59NNPOXjwIDqdDjc3N2666SY+/PBDwOhRcujQIcLCwqyj326w1AvCBmPUn3DTNdNN+/q7jQfcl/z2228sWdJhqIorgmXLlvHdd99hMBjMO7xaRygXSkpKCm+++SanT59GW6nF28fbHMbxSkGtVjNq1Kh+586YlJTEtm3bSE5OpqWlhQkTJrB+/XrZ6l+7di1qtZqamhp8fHwYPXq0WXwBYmJiMBgMV2wc5J5gqQniO6ABOIwxweYlwZU66tVqtaSkpJCamooQAkdHR2JjY3nhhXNi6p8XGzZs4JNPPqGwsBBJknBzd2PevHksXbpUlvovFby8vPrdyH7t2rV8+umnKBQKVCoVNjY2somvVqvlkUce4ejRo+h0OlxcXPjzn/98zii31QvCSvdYKsDB3STgtNKPWLZsGQcOHKCsrIxRo0Zx1113yfYwWrp0KV999ZU5zsM111zDuHHj+N3vfidL/ZcS/cnjQa1W8+mnn7J7926qq6tRKpXMmzeP3//+97LUv3z5cnJzczl16hTOzs74+PiwaNEiq4nhArFUgLcJIaZLkpTYq72xcsG88MILfPPNNzg4ODB06FC2bNkia/1ff/016iI1Hp4eTJ06lSlTplj39vcxGo2GV199lfz8fEpLS4mJiWHMmDG8+OKLstT/zjvvEB8fT2VlJZ6entxxxx1cc8011lGuDFgqwKkY020ogCb+l5beusTZT1i1ahXHjx9nz549hIQYMzu1ugHJwaZNm/jrX/9KYUEhEQMjiIiIYOPGjbLVb+X8+e9//8uxY8ewtbVlzJgxbNiwQba6n3/+eRISEtBqtfj5+TFnzhxZ3diudCwV4HeBScBhqTVjopV+Q1JSEt9884056tiJEydkrX/p0qVs2bKFkuISgoKDuO+++2Sb2lo5f1asWAHAjz/+SENDA88995xss5HWHaU//vgjWq2WqKgo5s2bJ9uo2ooRSwX4JHCkJ+IrhFgPzAZKJUkabjr2OvAYxswaAEslSUownXsVeAQwAM9KkrTd0rauNLRaLZ9++ikAp06dIi0tjYKCAvz9/XnvvfdkbevRRx/lxx9/pLKiEg9PD5577jkWLVokaxtWekZGRga7d+8mKSmJ+vp6WlpaeOihh2Q1BX3xxRekpaVRVlbGgAEDeOqppy44XKmVc7FUgNXATiHENqCx9WA3bmgbgL8Dn511/D1Jkv7S9oApuM98YBjGhJw7hBCDejMn3KVKWloaq1atoqCgAEmS0Gq1uLm58dZbb/HQQw/J1k5SUhKvv/462dnZ2NraMnHSRO69995u28jJySE4OLhfLVBdLqxYsYKcnBzy8/OpqKhAp9Ph5eXFRx99ZNGGlg8++IC5c+d2G/Vu7NixlJSUUF9fz5QpU/jqq6+6rTs+Ph4/Pz/ZtjdfKVgqwKdNP/amn26RJGmXECLcwvrnApslSWoETgshsoHxwB4Lr78i2LBhA3v27KG6upqmpia0Wi0eHh787W9/OyddzoWwevVq1qxZQ11dHXZ2dsTExBAVFWWRwFszR/cOGRkZZGZmkpOTg5ubG42NjcyaNYvXXnvN4vCdixcv7vJ8YmIia9euRaPRmFM0WSK+YMz60t/8oS8FLN0J94aMbT4jhHgQ2Ae8KElSJRCEcaGvlQLTsSueTZs2UVdXR3V1NWlpaZSXl1NWVkZNTQ16vR5JkmQVX4Ddu3ej0+morqrGXeVOQ0ODVVj7gKSkJGpra1Gr1aSmppKXl4ezszMVFRV4eHiwatUqWWcaH3zwAfn5+dTX1+Pg4EBYWJjF11rF9/zoUoCFEH+XJOkZIcR3mALxtEWSpFt62N4/gDdNdb2JcXFvIR1nT+7Q3iyEeBx4HCAkNKSHzV86pKenExcXR3V1NQUFBZSWliJJEi0tLTg4OODt7Y2bmxtxcXGytXnnnXeiVqs5c+YMQgj8/P2YNm0aixYtsiZRvIjEx8fz22+/oVarqaysBDBnXK6trSUgIEDWzMIbNmzgyy+/5MSJE9TU1BAVFcXy5ctlqdtK13Q3An4QeAb4SzflLEKSpJLWv4UQ64BWP6kCoK2aBgNFndTxEfARQOzY2MvWIyM5ORk7OzscHR0pKSkxBzbx9vZmxIgR3HTTTbI6wT/zzDPs3r2blpYWhBCMGDGCqKgo/v73v8vWhhXLOHLkCI2NjQwbNszsUubh4UF0dDSTJk1qF1HuQkhNTeWDDz5Ap9Nx8uRJFAoFvr6+TJs2TdZIeVY6pzsBPgUgSVKyHI0JIQIkSVKbXt4GHDH9/S3whRDirxgX4aKAvXK0eSmSl5fHlClTePrppwFoaWnh+uuvx9vbm1tvvVV2k8OSJUv4/vvvqamuwcPTA39/f0aNGsXbb78tS/1ypo+/EoiNjeVPf/oTarWa0aNHA/Diiy/KvsC1fft2jh49SnNzM0IIYyaQIUPM7m0XSnp6OoMHDzZHIkxKSgKQJfj/5UJ3AuwjhOg0gEBXXhBCiH8DUwFvIUQBsByYKoQYjdG8cAZ4wlTPUSHEl0Am0Aw8fSV6QGi1Wr799ls2b97M6dOn8fT0RKfT4e3tjbOzM0899ZTsttilS5eSnJxMuaYcH18fnJycePPNNy0aAbWGYezqgfD888+j1Wrx9PQkJCQEX19fqzvTWWRmZhIdHc38+fM5efIkNjY2ODs7ExkZiaOjI4MGDeqR+Gq1WuLi4tDpdJ0uvK1cuZKvv/6agoIClEolAQEBLFq0yKIt62vWrEGhUHTpjjhnzhzs7e1pbm7G398ff39/AgIC0Gg05OfnExkZafH7uZzpToBtABc6ttF2iSRJ93RwuNMURpIkvQW81dN2zofExERsbGz6zZM4OTmZrKwsvvzyS4qKinBwcGDUqFEMHjwYNzc3xo4dK/u2zw8++IAdO3Zw4MABtJVacxB1BwcHi6efERERXT4Q5s2bR1FREW5ubgwYMIDq6uorImaETqfj8OHDgDFeRERERKeeCgkJCXz++ecAVFZWMnbsWCZMmIBCoSA2Nva84iWrVKpOfYLT0tL4+OOP2b17NyUlJYSHh+Pr60tCQoLF9d90002d/t/VajUPPvggVVVVuLu709LSgru7Ow4ODuzbtw8XFxf8/Pz6ZbLSzhBC3Am8DgwFxkuStK+TcjcBqzHq5seSJK3qru7uBFgtSdL/9ay7/Zs1a9bg5OTErbfe2qvtWJqz7NFHHyU7OxuDwYCrqyvjxo1jxIgRvbbjqDVTcWJiIhkZGWgrtXh6eTJ06FBzBgs5WLhwIQUFBahUKiZNmnTFLOqkp6eTlZUFGD0D6urqOhXfZcuWkZuba84SLHe2irNZtWoVKSkpZk+aAQMGsHr16h6HKO3qofviiy/i5OSEv78/4eHhPPbYY9TW1vLhhx/i7e1NaGgoo0ePvtQWdY8A84B/dlbAFLL3Q+BGjGtaaUKIbyVJyuyq4u4EuMcj3/5MYmIi48aNuyjO4paI7wsvvEBNTQ3Tpk3Dz88PjUZDdHR0rwb3/vzzzzl06BAZGRk0NDQQFh7GjTfeyJtvvilbG++++y45OTnMmDEDgEceeUS2uvszycnJHDx4kMDAQMaMGYOnp2eH4vvuu+9y9OhRvL29uf7668nMzKS0tLRXxTcxMZHa2lry8vIoKSnB19eXZ599Vrb40GAUeCEE/v7+/PWvfzV/B+Li4ggKCuL+++/vdhNIf0SSpGMAQnQph+OBbEmSckxlN2Pc33BBAtw/5ugycb4ruzqdDpAvEeTWrVspLCwkMjKSyZMnYzAYUCqVXHfddb26WLVq1Sq2bt1KSUkJer2eoUOHMmXKFN54Qx43782bN3P06FF2795NbGws99xzD9HR0bLU3d/Jy8tDrVZz2223dSkySUlJjB49Gjc3NwwGA1VVVQQFBcm6i/Fs4uLi2LVrFzt37sRgMBAZGcmtt94qqy0+KyuLo0ePEhISwnPPPdfuuzJ27Nj+kDHEWwjR1nTwkcmjSi6CgPw2rwuACd1d1KUAS5JUcYGduuTYsGEDjo6O2NjYYG9vj0qlktX+un79ek6fPo1KpaKlpQW9Xo/BYOj14PGrV6/m008/RV2kxl3lzpQpU1i5cqVsi3p5eXkkJydz5swZRowYwV//2q+TpchOaGhot6O7LVu2oNfrKS0tBYypmjw9PXvV5Wvz5s2kpKSQkpKCwWBg8uTJPPfcc7I+GN955x2OHDmCk5MTq1b9z+yZkpKCWq3uL4kRNJIkje3spBBiB+Dfwak/SJK01YL6Ld7L0BZLtyJfsuj1erKzsy36wGVmZjJgwAAaGxsZMWIETU1Nsk6ZNBoNtra2eHp64uLigkqluigfzieeeILk5GTURWpCw0K56qqr+Ogj+R7+GzZs4Pvvv6esrAwvLy8ef/xx2eq+FNDr9d1uisjIyMDd3Z3a2lomTZpEWFjYRcmmsWPHDn7++WecnZ2ZMGEC//xnp2bMHrNq1SoyMjI4ceIEXl5efPzxx6jVatRqNVu3bqWqqgoPDw/Z2utNJEm64QKrsHgvQ1suewHesmULPj4+Fglwb0+Xvb29efDBB0lOTra4TxfKkiVLOH78OI2Njbir3Jk/f76s9W/YsIHvvvuOrKwsAgMD8fX1vWLMDlqtlv379+Ps7MyQIUO6jMkwcuRIBg4cyFdffSWr3bU76urq8PHxwcbGRlY7f1ZWljk+hb29PdHR0Xh5eXH48GHS0tIoLCy8oj4LQBoQJYQYABRiDC52b3cXXbYCrNVq+eKLL3jqqaf6uitm8vLyqK2tZdiwYV1mc5aLe++9l927dwNgMBi48cYbZV3oWb58OXv27EGr1RISEsJVV10lW939mfT0dOrq6sjKyrJ4RT81NZXGxsaLJr55eXmsWLGCqKgoJkyYIKuHy6pVq8whUIODg/H39+dvf/sbYBxx19bW4ujoKFsW5r5GCHEb8AHgA3wvhDgoSdIMIUQgRnezWZIkNQshngG2Y3RDWy9J0tHu6r4sBbh1pbc/hcbLzMzE0dGRsLAw2RbzOqPVp/PXX381x/EdP348QUHyxTdat24dX3/9NfX19YSFhfH8889fEdtXc3JyUKlUaDQaampqGD58eLflMzMzMRgM2NjYXJQ0PrfffjtnzpzhjTfeICIiQtZR6JYtW/j0009paWlh/PjxvPjii+YHkFarxdnZGYVCIdt26f6AJElfA193cLwIY7b41tcJgOUO1VymAmzJgkgrlvrrng9arZbS0lIOHTpEQEAAw4cP73XxBaPL0c8//0xlRSUjRo7A09PT4rCCljBz5kwaGhrQarUsWLCAG2644YrID5aXl0dERASpqalUVFQQGBjYqe1Xq9Xy448/UlRUhJ+fHwEBARfFHrpgwQIKCgpwdXWVPWGmVqvlxRdfxN7eniFDhnDfffe1G/2rVCrmz5+Pm5vbRfmcXw5clgJsCcnJyRw9ehQnJ6decwH69NNPqaioYODAgbLHb+iMrVu38vXXX1NZUUn0sGjuuOMOWXefPf7445w+fZrq6mquvfZaQkNDrwjxBcwP9YkTJ3ZpSkhNTaWoqIgjR44QFhbGnDlzLoogrVy5kiNHjjBt2jTZ696wYQNr165FkiQmTZrE3XffzaxZs84pdzEWFi8nrjgBVqvVNDU1oVarmTp1aq8tEmzZsgVbW1tiYmIuig+kXq/nT3/6E+vWrUMIwdDooSxevFhWX8/58+eTnp5OaGgoc+bMkdWueDmg0WjMu9rs7e155JFH8PLyuijiO3XqVEpKSoiMjGTs2LGyLrauXLmSDz74ADs7Oz755JMrwtR0sbjiBLj1Cd3bO3Iupu9jZmYmd911F9XV1SgUCv74xz9iZ2cnq/iuWrWK9PR0PDw82LFjh2z1Xk60Lqxe7FHglClTKCgowMfHh++++07Wut977z3WrFlDREQE8+fPt4qvzFxxAny5odFoeOSRR8jLzcNd5U5ERASPPfaYrG3MmzePM2fO4Obmxv333y9r3VbOn5UrV3Ls2DEKCgoYPXq0rDOt9PR0AD755BNCQkJ49dVXZbcpW7EK8CXLqlWrzIs8hQWFDBk6hICAANmzIv/xj38kPT0dlUrF448/bs2I3A947733iIuLo7KykqamJiZOnMimTZtkbeOzzz4jISEBg8HAbbfdZhXfXsIqwJcgW7Zs4ZNPPkGr1dLY0Ii3jzeRkZF88cUXsrWxYsUKkpOTyczMxN/fn0mTJlnFtx+wbt06/vOf/1BSUoKLiwuhoaEWi68lgfEzMjJYvHgxeXl5CCGYMWOGrBs4rLRH0dcdsNIzUlNTeeutt6isrKSxoRFXN1fZ4wkkJCSwbds2jh07hr29PV5eXgwePFi2+q2cHytXrmT9+vWo1Wrs7e0JDAxkwIABFl/falbojPj4eF544QVOnTplzAno52dx2EiNRmNxP6z8D+sI+BLizjvvJDc3l9zcXBwcHHB2diYwMJD58+fL5kqXmJhIQkICxcXF2NnZERQUxNSpU7njjjtkqd9Kz9Hr9bz88svEx8fT1NSEg4MDAQEBbN++3eI61Gp1l6PfuLg4Pv30U3JycrCxsUGlUhEWFmbxekJRUdFF2d15uWEV4EuEd999l2PHjqHVajE0G7B3tcfd3Z0xY8bI6g6WmJjI3r3GdHw2NjYMGDCAsWPHWv07+4itW7dy7Ngx0tLSzC5unp6ePU7pk5uby5AhQzo9n5qayunTp3FxcaGxsZGgoKAeZeO4FOP89gd6TYCFEOuB2UCpJEnDTcc8gf8A4Rhzwt0lSVKl6dyrwCOAAXhWkiTLH++XOY8++ig7d+5Ep9Ohq9Xh5e1l3lUnV9bi1NRUvvjiC2pra6mvr0epVOLn58dtt91mdT3qI1atWsX27dupqKhAo9Hg6urKqFGjuOqqq3r80G0NrdoRa9as4dixY+h0OgwGA6NGjeLPf/5zj3zkuwpEZKVzenMEvAH4O/BZm2NLgCRJklYJIZaYXr8ihIjGGD1oGMasyDuEEIOuxMScbdm0aRPr16/n1KlTNDU1YTAY8PXzxc/PT3Z/z+3bt5tT1eh0OnMSzX4QSPuKJC0tjU8//ZTGxkb0ej3Ozs6MGTOGxYsXn1dAn45suWq1mq+//prMzEwKCwsxGAw4OTlx3XXXXUlRzPqUXhNgSZJ2CSHCzzo8F2OmZICNwE7gFdPxzZIkNQKnhRDZGFN87Omt/vV3kpOTee+998jLy0Ov12NnZ0dERASurq7nlaixOwoLCzl8+DCSJOHk5ERsbCw33XST7O1Y6ZrMzEz+9Kc/UVxcTENDAwqFAicnJ2bOnMlLL73Uo6l+d14PTU1N7Nu3j8LCQqqqqlCpVAQHB5tTSVnpfS62DdhPkiQ1gCRJaiGEr+l4EJDaplyB6dg5CCEeBx4HCAkN6ajIJc/SpUvZvn07eXl55ihagwYNQqfTMXbsWNmzTWzevJmDBw+i1WqxtbXF19eX2NhY6+j3IrNs2TJ+/vlnKioqkCQJhUKBp6cnAP/3f//Xo2l+Wloap06d6lKAt2/fzrFjxygpKcHOzg4/Pz+8vLy6Hf3m5eXR0NDQq+mzrhT6yyKcxek8THmcPgKIHRvbbcqPS41nnnmGPXv2UFhYSGNDIza2NgQGBmJnZ8eSJUtkD6gOcOzYMQwGA97e3gghGDVqlNXr4SJiMBhI3JbIfzf9F8lGoq6xDhsbG7y8vMyBjnpqY7WxsenWhUylUlFXZ2zLx8eHiIgIixb3nJycrItuMnGxBbhECBFgGv0GAKWm4+eVzuNy44knnuDQoUNUVlZSp6vD1c0VhULBvffey9KlS3ut3dbYtkFBQXh6ejJhwgSrS9FFQq/XM3zAcDwHDOSq+Q9yOHknmoP78Qj0YMqUKec92ykuLu4wWllbfv31V5qbm3Fzc2Py5MmMGTPGokhqtrb9Zdx26XOx7+S3wAJglen31jbHvxBC/BXjIlwUsPci963PUKvVvP766+zbtw+1Wk1tTS3uKncCAwOZNGlSr4rv2rVryczMxNPTk40bN1pHNhcRg8HAlHFTGDlrNk9++CFCoaBZr+ePN96IfYvugkxN3YlvUlISmZmZBAUFceedd/YofojV40E+etMN7d8YF9y8hRAFwHKMwvulEOIRIA+4E0CSpKNCiC+BTKAZeNpSD4jI8EhKc0vNr33DfMk+ky3nWwGMgdvr6+tlHxkmJCSwceNGysvLKS4uprm5GZWHitjYWN5///1eFcT33nuPnTt3olare70tK+fy97/9nWYnV55cswYbW1uEMFrihk25liFOvTs2+vrrryksLOTo0W6z5ljpRXrTC+KeTk5d30n5t4C3etpOaW5pO2txqSjtvLAJjUbD9u3bGThwYLcuPZ999hm//vorZWVlXH311dx2220Wp3LXaDTU1dXh7+/fLnNCQkIC4eHhREdHs3HjRo4fP05hYSGSJJl3nr3//vsWtXG+ZGVlUVFRQW1tLS4uLuj1+l5t70ojIyODw4cPAzBu3DjzglVWVhYDBw7kj6/+kb+/83dW/vyzWXwlSQJJ4tgvKdz5ymLZ+6TX60lMTEStVnPy5El8fX27v0gG0tPTsbW17RXvnUudK86YEx8fz4kTJ/Dw8CAwMLDLsgsWLODQoUPmgNpKpbKd+CYnJ3eaDWL58uW0tLQAtAtmsnr1ahwcHAgPDycpKYnffvuNivIKlE7GjQ+PPfYYixfL/+VrJSEhgaamJurr66mvr0er1XLbbbdd1Ey9lzObNm0ybxeXJImhQ4cyaNAgMjIy2LJlC56ennz49od4hEfw1s8/M/ysz89v331HWXYW02fKt/lFp9ORmppKfn4+VVVVqNVq6urqeOWVV2RroysOHjxIcHCwVYA74JIWYEVLx8ddazq/RqhrGeIWhI+nD8M8QqGTskuXLqX88BkaT2uYMt24mvzKA0+1K6+sbD6nLZ1Ox+rVq0FTwwvPP39Of6JVIbi7u6Mo1rH90/9iV9aAH04EugUyInIESx9a3Gmf5KDxtAZ3d3fKctQU7TuBaw38+TlTpuRebPdypKN8gjUniqjIzsWuoYGhQ4ey5NHfQQ1UHSsg0smPN1b8hdh583jqH/9AYWMDYB79Si0txL3zDv949z1UdTay9dOgbWLW6GvYqd3JLwdPUX7yDN56B+6ZMvui/M/HBAxi+NDhXX4vr1QuaQHuKXq9nujo6C6fxFqtlueff559+/ah1Wq5+uqrefzxxzsse80117Sre+vWrRgMBjw9PTtcOMvIyKC+vp4ffviBjIwMioqKUCgUXHPNNUyePLlXR74Ab7zxBk1NTQQHB6PVaqmtrb2sstdebNqKb+v/v7a2lquvvrrdfc3MzESj0fCHP/2JiXffzdNr1yIUCrPNV5KMNrSv3n6bhsJC5nazgNZTWhfNwsPD2bt3LwaDQXZf8q64WPkQL0WuKAEGup0GPf300xw/fhx/f3+CgoIYO3Zsp9Nze3t7tFot+/fv58iRIzg4OBAcHHyOr65er+f48eMAHDhwgB9++IGWlhYkSeLGG29k2rRpvZYYtJWkpCTKysrw9/cnODgYX19fXn75ZWv22gtEp9Px888/c+zYMZydnXnjjTfanddqtWi1Wh5+9lnuX7GC2195BSFEO/GVJImv/vxnvnzjDSry8rCxkW/02xZPT09iY2N55plnrJ4M/YTLXoAzMzPNO3taF8LWrVtHUFBQO1edtWvXsm/fPk6dOgVgDsU4atQo8vLy2nkIxMfHo1Qqyc3NNcdY9ff3Z86cOWYbcUZGBqGhoahUKuzt7dm3bx+JiYkcOHCASq0Wdzc3YmNjCQoKorKykq1btzJ37txeuQd6vZ78/HwGDhxIREQEfn5+jBs3rlfauhLYunUr9fX1nDlzBq1WS0tLC7GxseYHb1JSEgaDATc3N3bt3s2S11/nuY0bue6BBwDai29LCx8uWsTBrVupyMtDqVT2Wr+9vb27dU+Tm7PNNFqt1ir+bbisBTgtLY2ioiIOHDhAWFgYkydPJikpiauvvrrddsvNmzfzyy+/kJubi42NDUOGDGHIkCFMnz6d0NDQdq5nK1aswN3dnYqKCrKyshg2bBhgTNfetlzrSDshIYEffviBvXv3Gl3NSkoYFh2No6MjKpUKPz8/nnzyyXZeEnISHx/P5MmTe32EfaWwfv160tPTaWpqQqFQmB9o48ePByAlJQWDwUBgYCDX33ILHpGRrPj5Z0aYFtvM3g5Ai8HAmief5MT27eSbgt9f6uh0Ovbt28fYsWNxdnbG2dnZLLparRYnJyc0Go11o4+Jy1KAN2zYQEtLC5WVlSiVShobG80Zgq+/vr0X3JIlSzh06BAnT57ExcWFq6++mrfffrvDqfnmzZspLCwkPz8fd3d35syZw1VXXcXp06c5ffq0+UOVmZmJi4sLCQkJ6HQ6MjIyKCwsRFtVhaeHB35+fqhUKubNm9dro14wemkoFAr279/PqFGjrB96GcjOzkav15tHsQB1dXVERESg1+vJyclBoVBw50MPMXH+/A4X2wCOJCez8dVXsa+uJuvAgctCfDUaDTt27KCuro6QkBDzbFClUpGTk4OtrS0VFRUWu3FeCVyWAtw62lOr1ecEEm87BXr88cf57bff0Gg0+Pn5ERoayocffthhnampqWzbto2mpiYCAwMZNWoUAwcOJDQ0lNDQUDIzM1m/fj0jRoxg3LhxqNVqs7CfPHkSIQT+fn6MHj2a6OhoBg4c2K0b3IVy7NgxHB0dGT9+vFV8ZWDt2rV4eXlhY2ODUqkkLCyM4OBgpkyZQmpqqnFHobc3jyxezMT58ztdbPvpX//i/Ycf5suNG7l19uxes/lebLy9vTuNVXIpi64Q4k7gdWAoMF6SpH2dlDuD0a/EADRLkjS2u7ovSwFupa34arVaqqurCQ0NZcuWLXz++eeo1WoqKioYOnQojz76KHPmzOmwnrvvvpvq6mrq6+u57bbb8Pb2No+oW6moqGDhwoXm15988gl79+4lNzcXpVJJZGQkLi4uvP/+++zfv5/AwMB2wVK68inuKVu2bOHIkSPodDquu+66HmdPsNKetLQ0EhISzEkwH3nkEQCqq6tpbm5m8+bNaLVa9Ho9L772GuPvuqtD8W1dbPvij3/k8O7dREdHk5iYyIgRIy7bjCN6vZ4jR45YnFuun3IEmAf804Ky0yRJsjhB3mUtwK20LsQ5OTnx6KOP0tTUREFBAQDR0dFd5tZ64YUXqKmpwc3Njeuuu67TTARtXW1eeOEFvv32W0pKS1G5uzNu3DgWLVpEbW0tmZmZTJs2rZ2J48knn8Tb21s2Af7tt9/M7m0Xe9HlciMjI4P9+/djY2PD1Vdf3c4jprS0lKqqKgYOHIiHhwcxkydz1b338lRH4mtabNv31VdUFxWZF9sSExMv64wj6enpl/wmH0mSjgHtzE5ycVkLsEajIT4+Hg8PD6Kjo3nooYc4ceIEAL6+vowaNYprr70WMC6eFBQUtFvJ3rNnD8XFxYwbN47g4GDuvffeLtvLyclhxYoVpKenU1Jaiq+PD0OHDuWFF15ArVabbcNt7dBbtmwhMjKSF198Ubb3PXPmzHNs3VZ6zqZNmzh27Bi1tbU4OjqahUSv1/PLL7/g4eFBfn4+Nra2zLv/fq66996OR74m8d375ZeU5OSY7b2pqancfffdffb+eoO2Xg8ajeaSF98eIgGJQggJ+KcpdG6XXLYCnJGRwZAhQ8z24KSkJDQaDba2tgwbNozXX3+9nWvZ2c7in3zyCQBRUVEsXry4WxuqRqPhnXfeIS8vj1M5OYSGhLB48WIeeOABnJ2diY+PZ9y4ceeMSNPS0vjd734nwzv+H1bxvXDUajUajca8ZrBixQrzufLyctzc3DAYDMydO5cnnn0W57AwnvrHP8zie7anQ3pcXDvx3bRpE42Nje3MVpcD5eXlFBYW4ujo2N+CO3kLIdrabj9qK5BCiB2AfwfX/UGSpK0dHO+IqyVJKjIlmvhRCHFckqRdXV1wWQpwenr6OTan3NxcWlpaGDFiBNB1FtfExETs7OwAGDRokEULWF9++SW7du2irq4OpenDt2jRIvN5pVJ5ThzVzz77DFtb28vW/ncpExAQgJOTE/b29uekZsrKyiI3N5eRI0cSc/XV6N3cWLxuHQobm3bT1LyjR/ngsceoPn2awqwss/i2xmLw8vK6qO/pYtDc3IytrS0NDQ193ZWz0XS1KCZJ0g0X2oAkSUWm36VCiK8xplW78gT4bPF97733+Omnn3B3d+euu+7q0uYWFxfH7t27aWpq4u677z7HTawzR/K4uDjy8vPxUKl4/PHH242YkpOTzxmVJiYm0tLSwi233HIe7/D8UavVuLm5WXfAWcDs2bPbxcltDWQUFRXFpEmTGBQTw+AZM9q5moHR7NBiMPDBY48xyN2dO197rZ2b2fbt27Gzs2PSpEkX9f1cDC5lb4cLQQjhDCgkSaox/T0d+L/urrssBbgtDzzwANXV1SgUCubPn9+p+G7atIn8/Hz279+PVqvl+uuvP0d89Xr9OeI7Y8YMCgsLyc3LY/LVV2Nvb99OfOPj488Zbaenp1NUVISvr+9Ft5FZR9uW03qv8vLyzP/HuXPnotfrzeLbkc23paWFNU8+SfXp00x/7jkeMO2AA+MuzEGDBsm24NofyMnJuayFVwhxG/AB4AN8L4Q4KEnSDCFEIPCxJEmzAD/ga9PnwBb4QpKkH7qr+7IV4K1bt/Lhhx+i0Wjw8PDglltu4c477+y0fE1NDS0tLeh0OsaMGcOSJUvOKXO2s/ycOXM4duwYlVotYaGhXHfddfz+979vV2b27NntXufk5LBu3TpGjRrFDTdc8KzHykXAycmJe++9F5VKRX19PWHDhhFz662dLritWbSIYz/8wMFffzXH3E1LSyM1NZXm5mZ8fHz68u3ISmpq6mW/0CZJ0tfA1x0cLwJmmf7OAUb1tO7LUoBnzpxJc3MzJSUlDBgwgNGjR6NQKLq8xtfXF5VKxZEjR1i1alWXZePi4nj//ffJzs5GkiTcTKnizxbfs0lJSSEpKQlvb29zjAi1Wk1TU1N/W7Cw0obWNYDa2lo8Q0O55//+r+OgOiZvh6zERLIPHjQ/sHNyctiyZQu2trbo9XoKCgoIDAwkJyeHgICAS3pW0lPxtW5Dbk+fCHBHO0aEEJ7Af4Bw4AxwlyRJledTf2uksVmzZnUrpq20Oox/8cUX5mMdBRJ5/vnnKSkpISsrC4PBYA5q0/a6jnj55ZfRarXY29ubPSFSU1PR6XRWr4V+jMFgIP6HH/j3f//Llq+/5tlPP+0yqM6J7dvP2Vq8ceNGsrOzUSqVxMbGUlBQQE5ODq6urpf6BoUekZqaSklJCSEhIbi4uPR1d/oFfTkCPnvHyBIgSZKkVUKIJabX5xWyPygoCH9/f1auXGlR+cTEROrq6rjqqqvIycmhtraWkSNHnrNQ9frrr7Nr1y4kScLe3p6oqKh2ng6dsXz5ctRqNTY2NkRERNDU1MTRo0cZPXq0daNEP8VgMLA1IYEnn38ej4EDue7BB3nzmWe6DapztvhmZmZy5MgRnJ2dCQgIoLy8HFtbW6666qorKiKdVqslPT0df39/6urqrqgHT1f0JxPEXIxJPAE2Ajs5TwGOiYmxKLi5Xq9n586dJCQkMHDgQKDjVdycnBwaGhr49ttvKdNo8PL0JCAggKuuuuocG29bdDod7733HmfOnGHSpEkcOnSI4uJi7OzsejXTcSt6vZ7i4mKreaOHGAwGps+di7qpiUfWrmXCLbe083KwNKjOAw88gF6vp76+nubmZmxsbIiJiTlnN6XBYGBbYiIHMjKIGTmSmdOnXzbxIVpJTk7mhhtuMOfGs2KkrwS4ox0jfpIkqQEkSVKbnJnPQQjxOPA4QGhISIeVWyK+CxYswMHBAbVazahRo7p0iF+wYAH19fWUaTQMHjSIkSNH8uSTT3Y7gnnllVfMUdays7MJCQlh2bJl55RLTEzk0KFDzJgxQ9a8Wfb29lbxPQ+2JSaiAVb89BN2Dg4A58R0gK6D6syfPx+1Wo2dnR1ubm5m//O28UYMBgOPPfkk3/zwAwHDhjHs2mv5eu1a/rZmDdu/+eayEWG9Xk9tba1VfDugrwT4nB0jll5oEuuPAIZERUlKlSP14n9O336h3Wd6XbJkCRUVFbi6unLbbbcxaNCgDv1i09PTeeqpp8jNzaVWp2NgRATXXnutRelcXnjhBWpra/Hw8CAyMvIcv+BWNyS9Xs9bb71ldvq3Ji7sew5kZDDoqquwtbM7J5KZoamJU+npJH7yCcn/+hdVBQXn2DNbH7IOJvEeMmSIOVNG69rB1oQEHnriCQKHD2fRJ58wYc4cEIJmvZ4Vs2ezLTGR2TNnXqy33KvExcXh7u7e193ol/SJAHeyY6RECBFgGv0GAN3mly8uLqZJaupR2+np6QwcOJCBAwcye/bsTleg161bx9atWzlz5gwKhQJPDw8mT55skfhu3ryZ5uZms89wW/FNSkpizJgx5tc333wzhYWFDBgwAA8Pjx69l/MhLy8PJycn60p0FwQHBPCfzZtpbmrCzuQ9YzAY2Pvtt8S98w4lx47x9OOP811hYTuTQ15eHmvWrKG0tBQHBwf8/PwYPHiwWZATEhIYNGiQeffcMxs3tjNvtIp91MSJHDx8+JIWYK1Wy/Hjx9HpdMybN++yiHfcG1x0Ae5ix8i3wAJglel3t/uv7ezsaK6sNRo0TJSIrnXb1taWoUOHEhMT0+VusI0bNwKg+//2zjyu6Svd/++DyiIqi2JFVHArrVhFRMXdWv2VolbH7WpntO21rXfsZm2ny4xztdN1HP11tbVX61W7DLe0Kh2udauK0oIiEXFBEEFZ1aJEBJUInPtHEgRJACEhJJ7365WXyTdfznOenPjJ+Z7veZ7n2jX8unbFw8ODxx9/vL4uAXDjxg2KioqYO3durQ331Xc8rF+/nosXL+Lr60uHDh2YOHFig9pvCkp866e1EFw7d44/P/gg4w3FNfds2oQ2M5PPP/iARyMiTC4PZGVlce3aNbRaLT4+PvTr16/Geu+ECRNMRs/dvpsi9cAB5lu5QKu10Gq1gP7HKCcnB2dnZyW+dWCLGbDJiBEhRCLwnRBiAZANmI+aMODq5gZFDd+ptm3bNtzd3QkKCjIrvjt37mTjxo3k5eVReOkS93TuzLPPPkuPHj3qXfPdtGkTubm5pKenm93hoNPp+OSTTzhw4ACZmZm0adOGIUOGNEuV2ri4OC5dusSgQYPU2nAd9OjRg2kPP0yiRsPmv/6V9u3asfL995kcHm52XTYmJoY9e/Zw8eJFnJ2d+eKLmqljCwoKGDhypNkADoCK8nI+f/ZZWmm1PGKnKSozMzMpKioiLS0Nb2/vqrVvhWmaXYDNRYxIKS8Bd7Qh1tvLi9z8/HrP27ZtGykpKbi7u9d7g2716tUcP34cIQT+PXoQHh7eoJt6ixcvpqioiLZt27JkyRKza7nR0dFERUVRUlKCi4sLs2bNYvbs2fW231QKCwvZt28f3t7edO/eXQlwHYwdO5bk5GQ8PDx46YEHmD59utlzdTodkZGR/PDDD1RWVtKzZ89aIezXr1+n14ABzF62zGwAR0J0NFv+8Q+cr14lMTbWbm/AhYSEUFhYyLlz58xWx1DcoiVtQ7tj2tRzaZOZmcmqVau4efMmZWVlDBs2DNCv0d5777019iIuXbqU5ORkTpw4QVlZGUOGDKlKwl4XL7zwAgBpaWmEhYUxcODAWuKr0+lwdnYmJSWFtWvXUlRUhI+PD2PGjCEiIqJZ4ug1Gg19+vShW7duBAYGWt2evWMu8X51UlJS+Prrr8nIyADAycmJjz/+uMY5p06dYuykScxetoyZhvB2k7mCo6JY/9lnZpc37IXS0lJOnTqFv7+/rbtiF9i1ANfF+vXrSUxMpKCggIqKCjp27MiiRYvYvHkzAwYMqFEV+ZNPPmHr1q0UFhYihGDYsGH06tWL5557rpY4FhQUEB8fT7t27diwYQNeXl4UFxfj7+9fdae7OsYdD0888QQZGRlcuXKFLl26MHDgQEaMGNEsux6ys7MZN26cWouzELGxsWzZsoW2bdty6NAhnJ2d8fHxYe7cuYD+8+7RowdRmzfzHy+9ROiMGcx4Tb+l3VwAx4UzZxxifI4dO8aRI0esWmzWkXBIAY6OjqaiooI+ffowatQoKioqmG+4mXL75eTixYvZvHkzRVotHb29GTx4MJMmTTK7L3jLli1kZGRQWVnJkCFDmDFjBllZWbRv377WudnZ2YA+DPnIkSOUlpbSsWNHpk2bRq9eveoM4rAURjFQNB2dTsfy5cvJycnBxcWF8+fPEx4eTufOnau+L4WFhXTp0oWDBw/yh6efZs6bb9ZYdnDkqsigzw3h7+9v1/ktmhOHFOCG/vr+8Y9/ZMeOHRRptfQMCKBPnz689957ZjeMv/LKK5SXl9OzZ09Gjx5dtYRhFLjbxW7fvn0AxMfHU1lZiZ+fH0OGDGHAgAHNkv+hsLCwKrG8ouloNBp69uyJj48Pvr6+hIeH10hPqtVq6dSpEzqdjilz5zLnzTdrLTsAHIuN5a/jx/Pdpk0OVRXZiCnxTUxMvKtCrxuKQwpwfRjz/V64cIErxcXc27dvVeJtU+K7dOlSzp07h1ar5Z133jG7bHD7THPXrl2kp6dX5Znw8vJqcHKgpqDVatm1axc6na5W9WZbotVquXz5MuXl5dx7771V6RmLi4sBiIiIMJvw3tZERkbi7e1d595xT09PfcTXoEEMmjbN5LJDuU7HpjfeIHLTJmZMnUpiYiL5+fkOe8mekJCARqMhICDA1l1pkdx1Arxs2TLi4+MpKCjAycmJ4IED6devn9mkOmvWrCEpKQkXFxcmTJjQ4DXbyZMnc/LkSby8vJg2bRrz5s0zO7N+//33KS8vNxmm3BjWrFmDm5tbg24kNQfbtm2juLiYOXPmkJSURFhYGOvXr8fFxYWrV6/St29fQD9LKi4upmfPni0uwXdISEi9obQJCQlEzJ7NkJkz9VvNzCw7dG3ThumG5acNGzawcOFCq/e/udFoNBw6dIjLly/TtWtXKisrbd2lFonDC/DatWtp27Ytv//975k3b17VsoAQgtDQUNzd3eusdnz8+HECAwO57777GpT5LDs7m02bNgHg7u7OoEGDWLRokdlZ0wcffMDFixcJDg6+Y9/Mcf/999c7o8rMzKS8vByNRkNFRYVVZsoajYbU1FTatWtHt27diIuLY/DgwRw5cqQqP3NERERVbbSWWiapsLDQrPjqdDoOHTpEr169+H+/+x3D586tVZYeai87nDlzht27d+Pr6+tQ4eexsbGUlZWRn59PWVkZ/v7+pKamcunSJQYMGKDuR9yGXQuwtp4gjNLS0qqlhWXLlnH48GG0V67g3rYt/fv3Z+DAgSZ3LoB+1lZaWkrPnj0bXDI+NjaWEydOkJiYSG5uLlOmTGHJkiVVkWfp6em1qsV6eHhYvCpyfeKr1WrJyMigVatWDZrZNZaQkJBaaQdTUlLIzMwkKCjIbtYET5w4YbaE0NmzZ7l+/Tr9hg5l+Ny5pksUVVTUWHYA/XehS5cuLFq0qNn8aA5MfU7Lli3Dzc3NIYuQNhW7FmBPLy84e9bs+8XFxZw5c4ZPP/2UAwcOkJuXx+CQEKSUPP/882aFKiEhgV9//RUvL68GzXoBoqKiiImJ4fz585SWljJ8+PAa+Yizs7NrCV16enqzlyUvLS3l+vXrdRYmtQY6nY6zZ8/StWtXu5rxlZaW1lm/LTMzk3976inCZs+uIb63bzVzLi6uWnaA2qWqHA1jMYO4uDjuv/9+xo4d22KvcGyJXQtwffj6+vL9999z9uxZiouL8fL0pFWrVowZM8as+Kanp5OYmIirqytdunRp8JcmLS2NgoIChBCEhYXVmDVrNBqTCahtkZ6vuLjYJjOR3bt34+7uXlUjzV6oa/xLSkqY8thjPGaiRBHcKkvfWqtFExfncLsdzKHT6cjLy6O8vLyqYKfalmYahxbgqKgo1q5dS3FxsT602N+f1157zezsLyYmhh9//JHc3FyCg4MbtC767rvvkpeXx9GjRwkLC8PV1bVG9jOglvhqtVry8/NrBIM0F9b8j5Ceng7U/mFJTEykdWv9V61t27Y13mupux7qo6SkBM9u3XjeTIkiY1n66SNH8tbSpVXiu3bt2hql7h2BhIQE8vPz6d+/P66uruzbt68qsVRYWJjKA1wHDivAq1at4vPPP+fCxYu0b9eOwMBAVq1aZbYUyqZNmygqKiInJ4cHHnigQeWMFi9ezJEjR7h69Srdu3fnhRdeqPcmQ3R0NKmpqQwfPrxRfrVUIiMj6d27d4113bi4OMrKysjJyaGyspL27dtXBRxoNJqqkGx7EuCKigo2x8Twb/Pn8+KGDTXE19Syg1F8o6KiyM/P5/Tp07bsvkWJjo7mxIkTaLVaZs2aRUBAALt376aoqIiioiL69eunZr714JACPHHiRLKysrhw8SJB/foREBDAjBkz6qxD9dtvv5GTk0Pv3r1ZsWJFvTZef/11kpKScHNzw8XFhSVLltQQX+OllxGNRsPevXvJy8sjODi4znVFS/PNN9/UiAa0JDExMVRUVDBixIga/mdnZ5OZmUllZSVjxoypta3MHmuC6XQ6Bo4YgfT25u09e0zWh6se4XZ4/36KiorYsWMHR48eRafT1bo6skdSUlIA/Q1If39/5syZUzW+qamp5OXlcfr0aZtc4dkbDinAx44dq8rjm5CQwMmTJ+v9MkycOJG0tDRmzao7C2ZiYiJJSUnEx8fj6urKa4bN9tUFNT09vdZll6urK15eXvj4+FhFCOvqr4+Pj9Vuupm7mXThwgV69epVb95le0Gn05nN5VtXiaJOnToxadIkPDw86NSpU4uZ7RsTRDUG401UUzdTnZyc8PX1ddjAEkvjkALs7OyMn58fSUlJAA36JR4wYEC9d+d1Oh0bN26kQ4cO9O7dm3HjxtUKKc7MzDS55uVlqKjR3OthxiCH5sbDw8Nh1v6uX7+Of1BQnbl893z1FZ88+aTJEkWenp4tbteDNXJPaDQaWrdurdJQ3gFOtu6ANQgMDGzQMsKd8tFHH5Gamsq5c+fo3LlzrZlsYWEhbm5uJv/W19fXJoLk6elpk1mXMUWjvVNSUoKHnx+PLFnCs198UUt8pZR8//77rF6wAK0J8b2bKCwsZNCgQWrd9w5wyBnwrl27LNpeXFwcoN9V0bdvX0aOHGlyA70q9XMLU9VA7I3r16/TvmtXFm/caHKngzGX78516wBo37VrEy26IovrLYXYYmnuveWOQIsTYCFEOPAR0ApYJ6U0m73mUmGh1fvz888/s2LFCgoKCggMDOStt95qcXkKFJZHp9MR0L8/8999t86dDjvX7QLKAEtc0ov6T1E0O0KIfwBTAB1wBnhSSqk1cV6DtctIi1qCEEK0AlYDjwD9gLlCCLMLuAXnz1u1PwkJCaxevZpr167Rp08foqKi7kh8C5vhB0JheSoqKhg8ejTBU6eazGh2PDaW10aPZue6g0A6lhFfRQtmF9BfSjkA/YC/cfsJd6pdRlraDHgokGGoG4cQIhKYCpw0dfLNigrTrZhIjn4n6HQ6Vq5cCYC3vz8d/PyYOXPmHbfbqYn9UNiG/163DtmxI3/87LNaGc32fPUVHz7+JPAdMA39ZMeCWOk7YwwNbk4cpRiAlHJntZcJwEwTp92RdhlpaQLsB+RUe50LDKt+ghDiGeAZw8sS4BKCwtvOsXjHjGXqbUgn4G6ZUtvaV1/PLl26vjJ8eI1FgWtXr5Kvj/bTwExpATu1/LTGd7eFYGpMG104LikpaYcQ4k5uurgKIQ5Xe/1fUsr/aoTpfwf+x8TxerXLFC1NgE19+2p80Q0fWtUHJ4Q4LKUMtXbHbM3d4ifcPb7eLX6C5X2VUoZbqi0AIcRuoIuJt/4ipYw2nPMXoBz4xlQTJo7V+yPd0gQ4F+he7XU3oP668wqFQtEEpJQT6npfCPE4MBl4SBrXo2rSKO1qUTfhgESgrxCipxDCGZgD/GjjPikUirsYw+6G14BHpZTXzJzWKO1qUQIspSwHngN2AKnAd1LKE/X8WWPWceyRu8VPuHt8vVv8BPv29VOgPbBLCJEshFgDIIToKoTYBo3WLoTp2bRCoVAorE2LmgErFArF3YQSYIVCobARdivAQohwIUSaECJDCPG6rfvTVIQQ64UQF4UQx6sd8xZC7BJCnDb861XtvTcMvqcJIR62Ta/vHCFEdyHEXiFEqhDihBDiRcNxR/TVVQhxSAhx1ODrm4bjDucr6KPBhBBHhBAxhtcO6adFMWZ0sqcH+vCjM0Av9HGgR4F+tu5XE30aA4QAx6sdWwG8bnj+OvB3w/N+Bp9dgJ6Gz6KVrX1ooJ++QIjheXv0oZ39HNRXAbQzPG8DHATCHNFXQ/+XAN8CMYbXDumnJR/2OgOuCvuTUuoAY9if3SKl3A9cvu3wVMAYgrcRfeyr8XiklLJMSpkFZKD/TFo8UsoCKaXG8Pwq+jvGfjimr1JKWWJ42cbwkDigr0KIbsAkYF21ww7np6WxVwE2FfbnZ6O+WJN7pJQFoBcuwFhS2CH8F0IEAIPQzwwd0lfDZXkycBHYJaV0VF8/BF4FKqsdc0Q/LYq9CnCjwv4cCLv3XwjRDvgBWCylLK7rVBPH7MZXKWWFlDIYfWTUUCFE/zpOt0tfhRCTgYtSyqSG/omJYy3eT2tgrwJ8t4QsXxBC+AIY/jVm67Zr/4UQbdCL7zdSys2Gww7pqxGpzx+7DwjH8XwdCTwqhDiLfjlwvBDiaxzPT4tjrwJ8t4Qs/wg8bnj+OBBd7fgcIYSLEKIn0Bc4ZIP+3TFCn+7rSyBVSvn/q73liL76CCE8Dc/dgAnAKRzMVynlG1LKblLKAPT/F/dIKf+Ag/lpFWx9F7CxDyAC/R30M+gzFtm8T030559AAXAT/QxhAdAR+Bk4bfjXu9r5fzH4ngY8Yuv+34Gfo9BfbqYAyYZHhIP6OgA4YvD1OPCfhuMO52u1/o/j1i4Ih/XTUg8ViqxQKBQ2wl6XIBQKhcLuUQKsUCgUNkIJsEKhUNgIJcAKhUJhI5QAKxQKhY1QAqxQKBQ2QgmwQqFQ2AglwAqLI4QIEEJcNyShMZsDuNr5nkKI74UQpwznDDccv+Ocz+Zy8NbVXh323Qw1wHRCiE4W+GgUihqoQAyFxTFkOYuRUvY3vPYFfKWUGiFEeyAJmCalPGl4fyNwQEq5zhBa3ha4ij7ScSL6yMBEYK7xb+qwLQB3KWWJIedEHPCi4e9NtmfKvtTnbjC2eRYIlVIWWuDjUSiqUDNgRQ2EEEFCiN1CiHQhxF+FEJ8IIYY0pU1pPgcwQogO6JPRf2l4X2cQv0blfJZ6TOXgNdleHfYVCqujBFhRhRDCFYhCP2McCDwF+EkpE6udc8BwWX77Y0IDbQRwKwcw6Kua/Ab8t6GczTohhDsNyBkrhNgmhOhqwoapHLzm2jNnX6GwOkqAFdWZAByRUp6QUl5HX+5pVfUTpJSjpZTBJh6762vcTA7g1uhLMX0upRwElKIvX1NvzlgpZYSUslYaQ2k6B6+59szZVyisjhJgRXUGARoAw8yyREr5S/UTGjsDNpMDGPQz0VzDLBXge/SC2OScsbJmDl5z7Zmzr1BYnda27oCiRVGGXpgA3kM/A66BlHL0nTZaRw5gpJTnhRA5QohAKWUa8BBwkmo5n4E89HlmH2uALR/gppRSWy0H79/NtVeHfYXC6qgZsKI63wJjhBBp6KvWxgshPrRAuyOBeegrJRhnzBHV3n8e+EYIkQIEA+9KKcuB54Ad6G/afSelPFG9UTNrwL7AXkNbiejXgGPqaa+WfQv4rFDUi9qGprA4t29Ds3fUNjSFtVAzYIU1qAA8jIEY9ooxEAP9VrbKek5XKO4YNQNWKBQKG6FmwAqFQmEjlAArFAqFjVACrFAoFDZCCbBCoVDYCCXACoVCYSOUACsUCoWNUAKsUCgUNkIJsEKhUNgIJcAKhUJhI5QAKxQKhY1QAqxQKBQ2wq7zAbdq1UpWVlYiuK1UgoWxdvvNZ8QWtm3pmOWwmhcOO+63cHZ25oEHHqh1PCkpqVBK6WP9HrRc7DoZjxBCSikRQiAPH7aendBQq7ZvtHFYWteGOUKF9WyHilAOH7bf75iR0FDrfMccedyN2iKEwJTOCCGSpJShVumAnaCWIBQKhcJGKAFWKBQKG6EEWKFQKGyEEmCFQqGwEQ4rwFJKAqZMQYSGIkJDSc3Kqnpv98GDjFqwALeRIxGhoYx75hmL2/g4MpKh8+fjPX487UaPJnTePH6MjW20P1MCphAqQgkVoQxrPYyHuzzMK797hdzMXAC2/3M784fMJ8w5jFARyvInllvN1mdLP+Ox4McY6zGWsR5jeXrM0yTHJTfaXmNJTv6Fl16awkMPdWTECFemTu3NihXPc/Omzuq26xv7PtOm4TZyJB3GjiV03jy+27WrUXbqG4tP3/iUyf6TGeE6gge9HuSp0U+RuCfRKraMJMclM7TVUEJFKC9Pe7lRthR6HFaA92s0nCsoqHr91bZtVc/Ts7O5duMG/Xv3tpqNzXv2cLm4mEfHjOGBPn1ISk1lxquvcuTUqSbZHD15NDMXzcTdw519W/fx9lNvA5CRkkGr1q3o3qd7k9pviK2fvv6JyspKHpr5EF16dOHIgSO8GPEiv+X/ZjHb9bFjRyQLF47lwIEY7rmnOxER8/Dz68UPP6zhxo1rVrdf19hn5eXRv3dvnpwyhUGBgSSlpjL3L38hIyen0fbMjUVeVh5BQ4N49N8fxa+XH8lxySyevJjrpdctbgug9Gopy+YvQziJRrevuIVd7wOui69/+gmAQYGBHElL49vt23ln0SKEECyaNYtFs2bx4bffcvjkSavYeP/55xkaFISTkxMVFRXcO306mXl57D18mEH33ddom1MXTGXctHHs/9d+ljy6hLOnzgLw3HvPAbBq8SqyUrPqaKHptt7+9m0GjhgIwLWSazzc5WFKr5ZyLOEY46ePt4jturhx4xorVjxLRUUFjzzyB958cyNOTvq5RG7uGVxd21q9D3WN/Qcv35oVSinxevBBrpSUkH3+PH26N+4H0txYvBf5XtU5Vy5f4aGOD1F2vYxLFy7RrVc3i9oCWPnCSm5cu8H0Z6YT9VlUo9pX3MIhBbhMp+P7n38GYNVLLzHj1Vc5V1DAfo2GsYMHN4uNsNs2nuvKywHods89TbIb/WU0iXsSObjrIADjZ1hP8MzZMoov6AWm/Kbet87dOlutL9VJTv6FK1cuA7BgwdIq8QXo1q1pVzUNoSHfr+2//sr/xsWRkpHBlZISRgUHMyo4uNE26xr37d9u5+ivRzkWfwyAibMnNlp867K1d8te/rXhX3wY8yEnEk80un3FLRxSgGMOHEB79Sqdvb0ZGxLC5FGj+GrbNr7+6SeLCfCd2FjywQfkXrjAiAEDmP7gg02yeyDmQNVzZxdn7h98f5Paa4qt8vJylj+xnJu6m0yYNYH+Q/tbrS/VKSq6WPXc19e/WWxWpyFjn3DsGJ9+9x0Ari4uRIwcSZvWjf/vVtdYJOxMIGZjDAAdvDowbOKwRtsxZ6vwfCHvPPMOM/5jBqMmjVICbCEccg3YeHk4ZfRonJyc+J1B9KJ276ZMZ5kbNA2xUVFRwVNvvcXHkZGE9utHzIcf0roJ/wkBVm5ZSWJlIhsSNlBZWclbC94iLyuvac40wtaNazd4eerL7N28l1GTRvG3r/5mlT6Ywsvr1ky7oOBcs9k10pCxX75wITcTEjj6z3/S2cuLP69ezbfbtzfaZl1jsXzDcuLL4lm7fy0VFRW8/fTbJP+SbFFbm7/YjLZQy7m0cyyevJgd3+4A4HjCcf62oPnG3tFwOAEuKi5m2y+/APBldDQiNJTpf/oTAFdKSvjX/v3NYuNGWRkzXn2VL6OjmThsGHvXrMGrQ4cm2wZ9aOd9g+/Dzd2NyspKcs/k1v9HFrR15fIVFk1YxC/bfmHS/Ems3LoSZxdnq/XhdgYOHEGHDl4AfPnl21RWVla9V1BwjvLym1az3ZCxv1paCkDr1q0Z0Lcv9wUEAPqbv03h9rHIychBV6YX/DbObQgeFYxPV31qhex0y9rq7Kf/0Tu89zBx/xtH9ml9+5cuXCLx58btulA44BLEd7t2obt5kw7u7jwYeivM/GRWFqezs/lq2za6dOrEuq1bOZmZCcCpc+d4Yvly7gsI4PUnnrCIjS379hEdG4uriwt9e/Rg6eefAzA0KIjHwsMb7V/0l9Ec3nuYU5pTXNVexbWtK30H9GXf1n3s27qPE4f0l4bJccksf2I5waOCmfbUNIvaemnKS6TEp9DBqwPtPdvz0SsfATA8fDgjwkc02reG4ubmzp/+9AnLls3np5++5syZYwQFDeW33/I5eHAXO3deoH17T6vYbsjYP758OQ8NGYJf586czsnh50OHcHJyYuKwxi8NmBqLdh7tCPcNJ3R8KN6dvTmlOcXZU2dxcXNh0OhBFrU15tExHH7qVt6IL5Z/wdo31zJ26lhWbV3VaFt3Ow4nwN8YLvMWTp/OihdfrDoem5TEuIUL+enXXxkTEsLGmJiq9y5cusTGmBjGhoQ0SIAbYiPQX782eaOsjM+ibt0tfnzy5CYJsHF9rr1ne4JHBbPwzYV4d/YmLTmtah0QIPdMbtXMuLECbM7Wb3n67WbFRcVEfhxZdX47z3bNIsAAjzzye+65pzsbN/6dY8fiycpKpXPnbkyb9rRVd0E0ZOwfHj6cpFOn2B4fj0e7dowNCeGVefOadBPO1Fh079OdoKFBJB9IprioGA9vD0ZGjOTJPz/ZpO2I5sZdYXlUNrSG2FHZ0JrWtsqGZhZHHneVDa1+HG4NWKFQKOwFJcAKhUJhI5QAKxQKhY1QAqxQKBQ2QgmwQqFQ2Ah73wXxG9D8oVAKhcIS+KuinHYswAqFQmHPqCUIhUKhsBFKgBUKhcJGKAFWKBQKG6EEWKFQKGyEEmCFQqGwEUqAFQqFwkYoAVYoFAoboQRYoVAobIQSYIVCobAR/wcINjtLPXp1igAAAABJRU5ErkJggg==\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Segment (alpha): [260, 306]\n", "Length of segment: 47\n", "Length of feature sequence: 409\n", "Induced segment path family:\n", " [[ 0 46]\n", " [ 47 91]\n", " [260 306]]\n", "Fitness: 0.3035996282\n", "Score: 110.0260581556\n", "Normalized score: 0.4668596900\n", "Coverage: 139, 139\n", "Normalized coverage: 0.2249388753\n", "Length of all paths of family: 135\n" ] } ], "source": [ "seg = [260, 306]\n", "fig, ax, im = plot_ssm_ann_optimal_path_family(S, ann_frames, seg, \n", " color_ann=color_ann, cmap=cmap_penalty, ylabel='Time (frames)')\n", "plt.show()\n", "path_family = libfmp.c4.check_segment(seg, S)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As a final example, we consider the segment $\\alpha=[42:181]$ (`seg = [41, 180]`, the interval between seconds $20.5$ and $90$), which corresponds to $A_2B_1B_2$ and is repeated as $A_3B_3B_4$. Again, note that, because of the normalization where self-explanations are disregarded, the fitness $\\varphi(\\alpha)=0.37$ of the rather long segment $\\alpha=[42:181]$ (`seg = [41, 180]`) is well below the fitness $\\varphi(\\alpha^\\ast)=0.45$ of the thumbnail $\\alpha^\\ast=[353:396]$ (`seg = [352, 395]`)." ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "execution": { "iopub.execute_input": "2024-02-15T08:55:22.550388Z", "iopub.status.busy": "2024-02-15T08:55:22.550156Z", "iopub.status.idle": "2024-02-15T08:55:22.996304Z", "shell.execute_reply": "2024-02-15T08:55:22.995679Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Segment (alpha): [41, 180]\n", "Length of segment: 140\n", "Length of feature sequence: 409\n", "Induced segment path family:\n", " [[ 41 180]\n", " [259 395]]\n", "Fitness: 0.3692019768\n", "Score: 253.9126575247\n", "Normalized score: 0.4112370308\n", "Coverage: 277, 277\n", "Normalized coverage: 0.3349633252\n", "Length of all paths of family: 277\n" ] } ], "source": [ "seg = [41, 180]\n", "fig, ax, im = plot_ssm_ann_optimal_path_family(S, ann_frames, seg, \n", " color_ann=color_ann, cmap=cmap_penalty, ylabel='Time (frames)')\n", "plt.show()\n", "path_family = libfmp.c4.check_segment(seg, S)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Further Notes\n", "\n", "* Using the dynamic programming approach, the running time and memory requirements for computing an optimal path family for a feature sequence of length $N$ and a segment of size $M$ is $O(NM)$.\n", "* Since there are $O(N^2)$ segments for a feature sequence of length $N$, the overall running time for computing the fitness of all segments and to find the segment of maximal fitness is $O(N^4)$. The article Towards Efficient Audio Thumbnailing introduces several (multi-level) strategies to substantially accelerate the audio thumbnailing procedure." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "Acknowledgment: This notebook was created by Meinard Müller and Angel Villar-Corrales.\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n", "
\"C0\"\"C1\"\"C2\"\"C3\"\"C4\"\"C5\"\"C6\"\"C7\"\"C8\"
" ] } ], "metadata": { "anaconda-cloud": {}, "kernelspec": { "display_name": "Python 3", "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": "3.8.16" } }, "nbformat": 4, "nbformat_minor": 1 }