diff --git a/examples/references.bib b/examples/references.bib
index bf45cc5d1..994c98ff0 100644
--- a/examples/references.bib
+++ b/examples/references.bib
@@ -373,6 +373,14 @@ @book{hernan2020whatif
   year          = {2020},
   publisher     = {Chapman \& Hall/CRC}
 }
+@article{hoffman2013stochastic,
+  title         = {Stochastic Variational Inference},
+  author        = {Hoffman, Matthew D. and Blei, David M. and Wang, Chong and Paisley, John},
+  year          = {2013},
+  journal       = {Journal of Machine Learning Research},
+  volume        = {14},
+  pages         = {1303--1347}
+}
 @article{hoffman2014nuts,
   title         = {The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo},
   author        = {Hoffman, Matthew and Gelman, Andrew},
diff --git a/examples/variational_inference/streaming_dataset.ipynb b/examples/variational_inference/streaming_dataset.ipynb
new file mode 100644
index 000000000..eb26e4d64
--- /dev/null
+++ b/examples/variational_inference/streaming_dataset.ipynb
@@ -0,0 +1,535 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "3a1d0597",
+   "metadata": {},
+   "source": [
+    "(streaming_dataset)=\n",
+    "\n",
+    "# Out-of-core minibatch variational inference with DataLoader and Trainer\n",
+    "\n",
+    ":::{post} June 2026\n",
+    ":tags: variational inference, minibatch, out-of-core\n",
+    ":category: intermediate, how-to\n",
+    ":author: Yicheng (Ethan) Yang\n",
+    ":::"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "928fd0f3",
+   "metadata": {},
+   "source": [
+    "`pm.Minibatch` random-indexes an array that must be fully resident in RAM, so\n",
+    "minibatch variational inference {cite:p}`hoffman2013stochastic,kucukelbir2015automatic`\n",
+    "only works on data that already fits in memory.\n",
+    "This notebook uses the streaming API in `pymc.variational.streaming`, which follows\n",
+    "the same structure as PyTorch's `torch.utils.data`:\n",
+    "\n",
+    "* a {class}`~pymc.variational.streaming.DataLoader` batches (and optionally\n",
+    "  shuffles) an out-of-core source into fixed-size minibatches without loading\n",
+    "  the whole dataset into memory. Here the source is a directory of Parquet shards\n",
+    "  read by {func}`~pymc.variational.streaming.parquet_source`.\n",
+    "* the model observes a `pm.Data` placeholder of one batch, not the whole array.\n",
+    "* a {class}`~pymc.variational.streaming.Trainer` drives ADVI, writing each\n",
+    "  minibatch into that placeholder with `set_data` every step. There are no\n",
+    "  callbacks to write.\n",
+    "\n",
+    "The unbiased-gradient rescaling works exactly as in `pm.Minibatch`. The `DataLoader`\n",
+    "is sized, so `total_size=len(loader)` passes the dataset size `N` to the observed\n",
+    "distribution, and PyMC scales the minibatch log-likelihood by `N / batch_size`.\n",
+    "Batches are exact-size, so each pass drops the rows that do not fill a final batch;\n",
+    "with `shuffle=True` that remainder is re-drawn every epoch instead of being a fixed\n",
+    "tail. The one extra requirement is shuffling itself. A streaming source is only as\n",
+    "well mixed as the order it yields rows in, so pass `DataLoader(shuffle=True)` to\n",
+    "shuffle through a bounded buffer.\n",
+    "\n",
+    "`N` is small here so the notebook runs in seconds. The streaming code is the same at\n",
+    "any size, and the last section shows what changes at scale."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0d67f52e",
+   "metadata": {},
+   "source": [
+    ":::{include} ../extra_installs.md\n",
+    ":::"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "c2f6955f",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:31.179963Z",
+     "iopub.status.busy": "2026-06-11T15:48:31.179761Z",
+     "iopub.status.idle": "2026-06-11T15:48:32.724808Z",
+     "shell.execute_reply": "2026-06-11T15:48:32.724317Z"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "import glob\n",
+    "import tempfile\n",
+    "\n",
+    "import arviz as az\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import pyarrow as pa\n",
+    "import pyarrow.parquet as pq\n",
+    "import pymc as pm\n",
+    "\n",
+    "from pymc.variational.streaming import DataLoader, Trainer, parquet_source\n",
+    "\n",
+    "RANDOM_SEED = 8927\n",
+    "rng = np.random.default_rng(RANDOM_SEED)\n",
+    "az.style.use(\"arviz-variat\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "90fd8ed5",
+   "metadata": {},
+   "source": [
+    "## Write the dataset to disk\n",
+    "\n",
+    "First, we build a logistic-regression dataset, write it to Parquet shards, and\n",
+    "delete the in-memory table. From here the streaming path reads only the disk copy.\n",
+    "We keep `X` and `y` around only to build the in-RAM `pm.Minibatch` baseline later;\n",
+    "the streaming fit never reads them."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "03ccf873",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:32.726747Z",
+     "iopub.status.busy": "2026-06-11T15:48:32.726448Z",
+     "iopub.status.idle": "2026-06-11T15:48:32.752265Z",
+     "shell.execute_reply": "2026-06-11T15:48:32.751894Z"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "6 shards written\n"
+     ]
+    }
+   ],
+   "source": [
+    "N = 30_000\n",
+    "b_true = np.array([0.4, -1.2, 0.8, -0.5])  # intercept + 3 slopes\n",
+    "\n",
+    "X = rng.normal(size=(N, 3))\n",
+    "p = 1 / (1 + np.exp(-(b_true[0] + X @ b_true[1:])))\n",
+    "y = (rng.random(N) < p).astype(\"float64\")\n",
+    "table = np.column_stack([X, y]).astype(\"float64\")\n",
+    "\n",
+    "shard_dir = tempfile.mkdtemp(prefix=\"streaming_demo_\")\n",
+    "for i, s in enumerate(range(0, N, 5_000)):\n",
+    "    block = table[s : s + 5_000]\n",
+    "    pq.write_table(\n",
+    "        pa.table({f\"c{j}\": block[:, j] for j in range(4)}),\n",
+    "        f\"{shard_dir}/part_{i:03d}.parquet\",\n",
+    "    )\n",
+    "del table\n",
+    "print(len(glob.glob(f\"{shard_dir}/*.parquet\")), \"shards written\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8bd75c32",
+   "metadata": {},
+   "source": [
+    "## Stream minibatches off disk and fit with ADVI\n",
+    "\n",
+    "Next, the source is `parquet_source`, an out-of-core\n",
+    "{class}`~pymc.variational.streaming.IterableDataset` that reads one row group at a\n",
+    "time and gets `n_rows` from the Parquet metadata, so `total_size=\"auto\"` resolves\n",
+    "`N` without a data scan. The model observes a `pm.Data` placeholder of one batch and\n",
+    "passes `total_size=len(loader)` for the `N / batch_size` rescaling; the `Trainer`\n",
+    "streams each minibatch into the placeholder."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "477343ca",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:32.753347Z",
+     "iopub.status.busy": "2026-06-11T15:48:32.753280Z",
+     "iopub.status.idle": "2026-06-11T15:48:50.379526Z",
+     "shell.execute_reply": "2026-06-11T15:48:50.379109Z"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Finished [100%]: Average Loss = 531.47\n"
+     ]
+    }
+   ],
+   "source": [
+    "batch_size = 1024\n",
+    "loader = DataLoader(\n",
+    "    parquet_source(shard_dir),\n",
+    "    batch_size=batch_size,\n",
+    "    sample_shape=(4,),  # 3 features + 1 observed column\n",
+    "    shuffle=True,  # the shards were written in order\n",
+    "    buffer_size=15_000,\n",
+    "    seed=0,\n",
+    "    total_size=\"auto\",  # N from the Parquet metadata\n",
+    ")\n",
+    "\n",
+    "with pm.Model() as model:\n",
+    "    b = pm.Normal(\"b\", 0.0, 3.0, shape=4)\n",
+    "    batch = pm.Data(\"batch\", np.zeros((batch_size, 4)))  # placeholder for one minibatch\n",
+    "    logit = b[0] + b[1] * batch[:, 0] + b[2] * batch[:, 1] + b[3] * batch[:, 2]\n",
+    "    pm.Bernoulli(\"y\", logit_p=logit, observed=batch[:, 3], total_size=len(loader))\n",
+    "\n",
+    "    approx = Trainer(\n",
+    "        method=\"advi\",\n",
+    "        dataloader=loader,\n",
+    "        data_name=\"batch\",\n",
+    "        obj_optimizer=pm.adam(learning_rate=0.008),\n",
+    "    ).fit(30_000, random_seed=RANDOM_SEED)\n",
+    "    idata_stream = approx.sample(1000, random_seed=RANDOM_SEED)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2a1d41c8",
+   "metadata": {},
+   "source": [
+    "The negative-ELBO trace shows the fit converging on minibatches read off disk:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "a3ff7cb9",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:50.380772Z",
+     "iopub.status.busy": "2026-06-11T15:48:50.380702Z",
+     "iopub.status.idle": "2026-06-11T15:48:50.538325Z",
+     "shell.execute_reply": "2026-06-11T15:48:50.537986Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABx8AAAJvCAYAAACqKB9SAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjksIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvJkbTWQAAAAlwSFlzAAAewgAAHsIBbtB1PgAA+PRJREFUeJzs3Qd8FHX6x/EnJECAQIAAoXcUpNnAflawYa+nJ5bTU0/PUyyn51nO/vesh3qeXTl77x0OuxQFpffeCTWQQMr/9f3p7O0uu7O7yWazST7v12tf2d3MzszOzs7M/p7f8/wyysvLyw0AAAAAAAAAAAAAKqleZWcAAAAAAAAAAAAAAAQfAQAAAAAAAAAAACQNmY8AAAAAAAAAAAAAkoLgIwAAAAAAAAAAAICkIPgIAAAAAAAAAAAAICkIPgIAAAAAAAAAAABICoKPAAAAAAAAAAAAAJKC4CMAAAAAAAAAAACApCD4CAAAAAAAAAAAACApCD4CAAAAAAAAAAAASAqCjwAAAAAAAAAAAACSguAjAAAAAAAAAAAAgKQg+AgAAAAAAAAAAAAgKQg+AgAAAAAAAAAAAEgKgo8AAAAAAAAAAAAAkoLgIwAAAAAAAAAAAICkIPgIAAAAAAAAAAAAICkIPgIAAAAAAAAAAABICoKPAAAAAAAAAAAAAJIiKzmzAQAAQE2zevVqmzp1qq1Zs8Y2b95sW7ZssQYNGlh2drbl5ORYfn6+tWnTxjp06OCeA9LBG2+8Ydddd13U/99555124oknpnSdAAAAAADA/xB8BAAAqEMWLFhgL7/8sn344Ye2fPnyuF5Tr149a9++vfXo0cP69etnAwcOtF133dVyc3OrfH0B1BwXXXSRjRkzJuZ0xx13nN19990VXs6SJUvs0EMPjTldVlaWu9WvX991oGjWrJm7tW7d2jp27GhdunSxAQMG2M4772yZmZkJr8fhhx/ujqnRnHvuuXbttddaZfz1r3+1119/Per/tf6vvvpqyHNa5ptvvhn1NYMHD7ZRo0ZVar0AAAAAwA/BRwAAgDpg69atdu+999p//vMfKy8vT+i1ZWVlrrFft7Fjx7rn9tlnH3vmmWdivnb69On22WefRf3/YYcdZn369ElofQCkn4KCAvvyyy/jmvbTTz91mdaNGzeu0nUqKSlxt6KiItu0aZPL9o5Emd5Dhw61k046yfbcc8+4568g6oMPPhj1/x988IFdc801rgNHRWzbts0++eQT32mOP/74Cs0bAAAAAKoSwUcAAIBabt26dTZ8+HCbNWtWypet4ONDDz0U9f8q6UrwEaj53n//fRfoi4cCjwpAKniXDlR2WuV8dVNW4PXXX2+9e/eO+bpjjz3W/vnPf0bt0LFy5UobN26c7b333hVar//+978uaBqNMjqPPPLICs0bAAAAAKpSxbpgAgAAoEZQ5swFF1xQLYFHAHXH22+/ndD077zzjqUjBQtPPvlke+yxx2JOq9Kte+yxh+807777boXX5b333vP9/wEHHGAtW7as8PwBAAAAoKoQfAQAAKjFNK7Xzz//XN2rAaAWmzt3bsLHmW+//dZWrVpl6Wj79u2uTPV1113nyk77iZW9qbKp6gRSkWxMZT5WZtkAAAAAUF0ouwoAAFBLqcH73//+t+803bt3txNPPNF23XVXa9u2rTVq1MiND7l+/Xo3xuPUqVPtp59+sh9++ME1yAPVTfurbkgfFcliLC0tdVmBv//97y1dqQxrVlaW3XrrrVGnUdnT2267zYqLiyP+f+PGjW6s3CFDhiS0bAUto81TmjVrZoccckhC8wQAAACAVCH4CAAAUEuNHz/eNmzYEPX/xxxzjN11112ucT1cp06drH///oHxxDTumBrQFSz44osvqnS9AdQcGu+woqVFVaq1qoKPzz33nO21117uvjpO6Fio8W+nTJnijo0ffvihG3sylldeecWNS3vGGWdE/H/Tpk3t4IMPto8++ijqPLR9Eg0+xiq5qmNzgwYNEponAAAAAKQKZVcBAABqqe+++y7q/xRwvP766yMGHqM1sA8bNsxlUiojh8wzAKJA3tKlSyu0MWbOnGkzZsyo8g1Zv359a9WqlfXq1ctOOOEEu+OOO1xJ04svvtgyMzNjvv7//u//fN9jrPKnWpbKqMZr9erVvsdvOfbYY+OeHwAAAACkGpmPAAAAtdTy5cuj/q958+bWokWLCs1XWZG61RQlJSVuPLrZs2e7crIqR9u4cWMXQNV2iIdeM23aNFu2bJmbh0opZmdnu9fn5+fbwIED3TyTQeUoVfJWn5/GxFPGVlFRkVsHlcXNyclxN5XM7dGjhwuspGo7Tpw40RYvXmxr1qyxJk2aWM+ePW2PPfaIKwNLQSZ9DmvXrnXbrmXLlq7cb+fOnS2d6fOYNGmSe9/6PPRe8/Ly3Lon63ugwJbKG2v+ytLTPqXtMmDAAMvIyLB0puxFP8cff7y99dZbvq/v3bu3pVpubq5dfvnlLjvyT3/6k8vujkbfv/vuu8+NAxnJAQcc4I6nyqyMROVTE+m08cEHH7j9LpqOHTu6711NtWDBApszZ447tulYqlLfOn7qeKr9XkFidXipbEbuvHnz3HikWo6O23pOy9Bn361bN7ecVH6/dPzTe1+5cqXVq1fPHQP79evnjqM1XUFBgcsq1l9ta31nVBpY3wtt65133jll21qB/h9//NEWLlzo9i+N26rOB6effnrM1+o4oA4VK1ascPPRsb5du3a2++67u/NWXdyeqTgHepSNrvetDhje8UHfFV1vtG/f3rp06eLef2Xfu445Wo6+i1qOPncdg7R99Xnr3EtmOQAAlUfwEQAAoJaK1hAuCh6pUUuNn8n0/fff2/Dhw+Oe/rrrrnO3aO68884dGuw1zlm0LKQOHTrY6NGj3f3CwkJ77LHH7KWXXnKNd+H22Wcf3+Cjgn0ay07BkcmTJ/uOv6YAoAKQp5xyiitnG082lUeNnCplq3E1NcamGqe17Hh4y1WAR2UYFZSM11lnnWXjxo2L+v/PP//cBTnU6Pmvf/3LXnvtNbffhFODqD7zCy+8MGJjnQJPer3eVyQKAPz5z3+OuyylxuFLdJ/xKKh76KGHRn2tsuJUiljU+Pn444/b+++/H/F9y0477WSXXXZZwiU1PSpl/PDDD7v9KxI1tmqdVJpUja+x9n8vmzBV9J34+OOPo/5fAaS//vWvriSp9qNo5UWvuuqqhL4zyaTjwN13321//OMfXXAqGpVpvfLKK91nEul7ePTRR9t//vMf39Kr8QYfY5VcVdZjugelg2m7fvnll/bqq6/ahAkT3LnHj96bAnL777+/264qAR4vZYy+8MIL7lwU6bgfTEHIQYMG2W9/+1u3rEQo+BPN4MGDbdSoUe6+AlhPPfWUOw76nbe0/2n/UKAlEgV+dJyJto8q+1b7cbx0DNWx1C8Ars4tftRBRuWNldmrQK8fBXUUpNexLJHOBokcszXtgw8+6L6r4WNU61jkF3xUkFodDHRMjjS+tc6thx9+uMuW9gJu1157rb355ptxlX6OR7ptz1ScA0Xz1nWarjl0/vLreOFdc6jzhd7HUUcdFTg3xqJrQh2DtG+rM5nfOOYKNO+555525plnMrYuAACVQNlVAACAWipWSVU1Pqr3d22krBo1xj766KMxG6Ajef31112Dk0rTKkDnF3gUNWKpUf0vf/mLHXHEEfbtt9/GvSw1AN9www2uEXPWrFlxBx6Dl/u3v/3Nhg4d6jvuXEVofdQYqe0YrfFRmQkPPfSQnXfeeSHbWo3uf/jDH9w2iRZ4FGWkXnrppXbrrbe6DJV0oMZJBXOfffbZqO/b2z5a99tuu803cBVO+9OIESPc9okWeBRl2io4qYC2AtPpRo3FfhmDGgtRAZ4DDzww6jTKpEnk+1IV9F2PFRhUg7hfUDBW6VUFw9SYH8uiRYtcFqyfWMtKJ+pYoe/SBRdc4LI/YwUeRd8lHReefvrpuANqymLSZ3j22We7gHg8x31lPH322WcugKOAbqztnijt13rv+g77dRjQ/3SuUSeMaMd/BbwUKI/m008/jWsMU1FHAL9OAwq6+AUedWzX9cNhhx3mAquxAmVeZyh15lFHGQWr/DpHVYTej46TWoZfUCmSJ554wn2ntC9Ee63es64LtIxY2d6JSsftWdXnQO9933jjjXbQQQfZyJEjXUAwVuDRu+YYM2aMu+7R9z4WXVeoI5qWo85JOufG2kf0Hfnqq69csFnbePr06Qm9NwAA8AuCjwAAALVUrKxG9a5XY5cad+MJsNUUCnQpq0+ZIolS462ysJStFU+gIFrwQI3ZTz75pKWSypmq8VqNbMmg7XfuuefG1RAqKlWnAKQaD5VhoCCAskjipayxRx55xKqbtt8VV1zhG1QLpywnBRjioUZNNWgqmyReCk4ouzTZwZHKitUIr0C8KGOoMvNJhfPPPz9q1plHDd7RqEyfygFGo+9FPJ+5MiT9qNRh165dLd2pcf/22293Afb58+dX6bL0/VNWW2UC9Mq4OuOMM+yZZ55Jyjppn9Y+peB6vBScvemmm6L+/9RTT/U9d+n18VCgUsfoiixH21jBGHWaUSnuRClApUDhSSed5IJNyaD3rRLK8QZfg91///32j3/8I+6ApTpsqUPNiy++aMmQjtuzqs+BogCg3vfLL7+ccLA4EersoO+hSmYraFkRCjzq+KJgLwAASAzBRwAAgFqqb9++MadRj3YFyRSs05hGynJTT3SVplIZsppGWSMavy2e7JpIwQE1YMZq/I93XgrqJqshOxEqHecXJImXGh/9Mh6iNaTqPSuLI56MhHDKsKzqQIUfBUujjeuXrHVXhufXX3+d8PyVJaLSjPqbDhTsVmZINCqFp7KAooyThg0bRp1WGUcVCRwkk8ZQ7dOnj+802qf9MpNjZSTGKqcqsQKUytBLdwqIqCSlSkgmmg2VqOeff95lXSUjgKF5KDPKK5laUcoG0zGwIsEkBaFUOjYSdRbSWHuVDeL7TadS5F6ngXC6JjjnnHMq1LEnUocKdW5RZ53KdjZSMLAiWfPaDjpuJ0r7tI7jGleyMtJxe6biHKjjaLLetx+dUxR4rMj5NlKnoWuuucYF7gEAQPwIPgIAANRSavhPZFwwNZSq57x6oquclcbS0ThYathTkCFdSmL6UbaiGn4rQo3OiWTqxUMByFSXlFTDqAKQlW30r2gZNzVc+pX0i9X4n6yMkoqoSNA6kXVXUFhjZ1Zm/1apyHSgIJlfcEUlV70xQBWI9BtTL5Gsraqk8f/8KPDo12AeayzGn3/+2bcEsY6/fp0+NLakjsvpTtlk8QRaK0vHVmVXJpvOBd98802FX6+Sr/GUj4wmWvBTn786CEWjoOXKlStjHkP83pv24UgdBfSeVDq3otljkWieygKvTPl3BQAr0nFB20kBxIrS5+v3XY4lXbdnVZ8DNca1sqFT0dlEY/Qms1y5rqkUgKzoNSYAAHWR/0BAAAAAqLG6dOlihx56qMsqqig1VL711lvu1qFDBze2j0plRStPmJ+fb2eeeWbgsUp2+gXfNIaVMo6i8Rt3Kh4Keuy9995uvCyt8/Lly91YYhoTMrzknsp++undu7cry6eyhy1atHANfOrBr4biaBkQaqBU47jKdcUq6dixY0c3b2Wsan3btWtnTZs2tezsbDd+p8riKttM21TjqGl8x2jBHzWOqYF5v/32s8rS9lMZ2V69erkybK+88opvZlBwo7say5XhcPTRR7sywFp3lbjzaxD88MMPXdnb6qZSoaeddprbP9VQqowHZXb4NezGWvcHH3ww5nJ32203l8Gyyy67uH1G+6u2t1+WYXXQMcGPxgsL354aI9IvC0nHlurUs2fPmNNoHM5oxyUdIzVenkoQR6OgnI6jkcTKuv7Nb37jjj3pTN/xeEpOK4NP5T11fGnfvr0LeCmwruCrtp+C237jNqozzB133BEzyKcguEom6jPT90nHfmX2+2UwaZ4KQGqfjHXcjkXvTyWTd955Z7fOX375pf3zn//0fW8qia5MKx37w2mbaftG6lyi+etco6CW3/7nt810zIvk3//+t++4ldpOw4YNc8FxfY/0eer6QcctjWMY7f3q89DxTQGpZNA1yF577WVt2rRx21DnfAX9w4/b2oaxyorqHKz10vxycnJsyZIl7juqc2Blgss1ZXtWxTlQnZN0HROLxglWBq6uh5SNq44fKmGscq3qxKOhAvxomtGjR8fsbHLKKadYv379rFmzZi4QrGs5VW/QeTcSbQd1KtM4oQAAIDaCjwAAALXY9ddfbxMmTPBt6IyXGspUSk5l4RREiVT+TWORqWyrR9P6BR+VZXHiiSdaVdD4g2rkVwAynEqDBTfi6/34ZQoqoKps0PCGaAV4FWS5+eabXcZoJGrEUkP6McccE7WRTeX0dtppp5jvSY30/fv3d+UdlQGjoGA02u6VDT7qvSmLycvmUmOstoO2X6xgmLbVyJEjXeN/cMOwGn31fqOVD1UDo7JSNG110X6uoGkwBS9UltOvUddv3RVQ0dhRfvSZKugRnD2ngJbKlqrBM9XjiEajAJFfALlx48aBkqsedYRQJmS0sqVe1lZ1fu65ubkxp4kVsNB30y/4qOBFpOCjAkcffPCB77yrOzgbjwceeCBmuVEFVVQqtVGjRiHP67PXcVDHHZVtVZDw+++/jxrkiJWBdPXVV7uyi8H0fdIxVwEGfdei0by1DHWcqCiNexseiNG5ZI899rCTTz45aqlYrwqBSqGH0zlHwbBopVljBR/9Sq5qvSIF4HVcU3nbaPQ5KiilQGswBQAVPNJ3Qh13ogXbdFzT/xXgqygdc3S9o+uJSAHjSZMmBe5rvMvXX3/dd34KFiuIF3xM0P6pbaRjm8q7VzQAWRO2Z1WcA2fMmBGzg4UCjTqGqGNaOHXK0vbXtZ2CrPfcc0/UTHPNw4/G9g7/nrRu3dq9R12X6vNVJ4BI1IHghx9+iPj9BAAAoSi7CgAAUIspWKVGrGRmy6hhXQ1bsRrhq5PKbalcbKTAo3Tr1s01col6u0drZBI19kUKPHr0vBo9/cbi8hvHTfOPJ/AYTmUs1VgWjRrHKkMNlwqqRmrcO+SQQ2K+XsHW4MCjR/tipOeDVee4jwqMhje6ehS0UAZoRdbdL+vPC4pE295eY6myIdNBrLHl9PmGl27U/uQXDFfwLRnjrVZGPI31fmM+irJ1/Ma3VKlGZWJFOq6qJGE0CoIoCJ3O1KEgVraR9g0FDcIDj+EUqFagTp0fKjI2pgJE4YHHYPqOK5PUT6xl+NFxXQHUaFn0sTqG+B0Do2UnekFTBS6j/c+vA4SyKiNRSWRl3kejsZLDA2XB2rZt64JZ0ahzVGXG5VNlgMcff9wFdKOdp1VVwKOAtt/YuZmZmW4fjdYZQR0pgis8JCrdt2dVnQOVdevXyUuVEnS9GinwGE5Bck2rYGSk/VyBzmj0+fkF6HXsUVa11qcqjg0AANQlBB8BAABqOZVxVAZJrIbWRKgB/ZZbbrF0pBJafg1LkTIEY5Whi1V6T8GGSI1gwY2d0bJcgqmcmcrk3nXXXXbhhRe6QIYaqfUZqsFa2RjBN5Vhi8YvkBGPoUOHBgK04fxK5cZqyJZYwdbqDGxru/vR51CRdVcGsh+VhoxUZtGjfVDlG6ubGo9jBQlVri+R5+MNala1ePY7v8CiqFRyrOB8pO0Xa4xEHQu8MTTTlY6lfsc5NebfdNNNCY1FrEBQpMzAaBmRHmVHxaLyxn5U2rGi2W0KfPqdNyp6HPGCQypjnWhJZL9SyQq0hZdK9vhlueszVenKWGIFlCoTLDvrrLNcueN4xToWa11jnaOU1ZrIflyTtmdVnQNVLt7PSSed5K51EhHp+BCrKoM6z8WizmR++0Blti8AAHUJZVcBAADqAJWrUmaASrW98MILLjslnmCYHzWWX3LJJa7UajpRRkIijYLB5dgiURnZ4FKyFaFxgjQWmgKGkahM2cMPP+wah/0yIhKhjM7KUGm/aGJl0qoB1a8RMVpQM7gsXnVo1aqVDRgwwHeaWO892rrHKhHpl+ni2Xfffa26KeijcQ/9yh8qOyZaxon2jWjHHi8zS6X9qoPGHIxFwcVYVBZRJTujUXlVZcV5wSltD2VDxZpnuos29m3w/qvSzZWlY6lf5pqCtIMHD445H03jVwpYQRQtK1amVzh9rrGyVGMdR2K9P5Xg1dh/0bKyrrnmGpcRGJxZ7BfgVqnJaIF1v3Ok9t1klJ/U+MkVlWgWYqxjcTyZdxqjuXPnzrZw4UJLVDpvz6o6B6pjlcbWjhV8TIZY13R+5erjpexOfUcrU9oWAIC6gMxHAACAOkQBjn/+85/2zTffuL/KtlLDakV68Ksx89NPP7V0E08QJ1hBQUGVrUs8y1G20FFHHeXGjExW4DFW43U8/LIbY2V/KdgdKSPBE9woHm3fqg7xlL/1y06Mtu5qeI312caTTapxtBTcq06xshOVYR1tGzVr1ixmw351Zj/GCkpIPMEzlfz0y0xTxnLwmH3KCPIbl1dBDr/M6nQR61gaXPqyMtauXev7f3WIiXWMEU0Tq/NMRc4POv7F+p7GOo74laeMVXp1zZo1O2Rm6TyjTi6Jzk+Zn8kYM7qyn6lfEFDbOxGx3o/KsscjnmN2TdueVXUOjLU+CqirVHEyVPc1HQAA+B+CjwAAAHWQggAqgfj3v//dZUOovJwyI1WuNJGso1iZLqmmccQ0zmVNaajSuETa5pUNFFak8ToWvx79fmMheftXTRQrI1PiCWokmoWq4H+08UnDVWemRVFRUcwMPZUH9ROr9KqORxUtdVlZEydO9P2/gu4KBMazjxx99NFxl16NVcZWWWk1QaxjqbKqkmHdunVJO/7EymStyPmhqo4jwRQ09cvuDC+x6hfUV5Z6tOxOBcoqey6pyvNwjx49kl5eOd5jbLzH7Jq0Patq3421PlquX4elZC4rWQg+AgAQG2VXAQAA4BprlbGk21VXXeUygO69917773//67t1/DIpqkM8JRGrS6TMN41/Vtnyt1WlMuPLpfvYdBXN6JRY439WZHuoMVrj2MUK6kp17i+ff/55zEC5svjGjx9f4UCssgKVma3swVSaM2eOK/nqR5k58XxGXpnUUaNGRf2/ssZvvvlm97mPGTOmVgQfUbXHkUjZiuo45PddVSBNZb/9qhSoAkJ1i1b6tirO+bGOx/pOxiNdz92V2Z6p2ndrg2RWqgAAoLYi+AgAAICIpbceeeQR+93vfmc//PBD1C1UFRl7lRFvYCCYX3nEqjR37tyYYxMpI+Xss8+2gQMHuqyh8EbTQw45xJYuXVrFa4rKys3NddmNfhkvyqJQWdVYDcqVHcuzMuIpifrGG28kZTmpDj4+9thjMTOSDj744Ljn179/f1eWUWMGRsu+UucOleRVRqnfMaBLly5WE8Q6lqocaDLEGnMuke9IrCy46jo/xGPo0KEuYyxSGU8FRjTu6CmnnOKylRWAjHZs8stW1vxjHbtq2jlf7zkZGW0V2Z/TfXtWlVjfI+3DynhPRvajlqXrKwAAUP0IPgIAACAiNQKdfPLJvsHHipQdq2mNYipHq4zQZPvqq698/7///vu7gIhfY1yshnOkB2WKtGnTxjdTWCV4YwUflZlXXSVJI40jV1U+++wzKywsTNnxRcGZWIFVfQ+HDRuW0HyV/Xj//fdH/b/KrfoFHmta1mOsY2mszhbxysvL8/3/woULXfZarPKQmmbBggU1NviozignnHCCPf300xH/r31awcd33nnHd//yG8NP+70CZtFK3apTTKqOC8nStm3bmMfZY445JmbWo47ZiaqN2zMZ31l1rJk6daoNGDCg0suK9Z394IMPKlSuFwAAJI56CQAAALWUgoaLFi2q0ka6WOMDqYd/uovV2KUSkFVhxYoVvv9XKTy/wKN69ldnFhwSoww2P35lEYPHQ6wuWna85QgrS9mAscaWTBZ9v6+99tqY0x155JEJjyerwI7fMXDs2LH27bff+mZ1HXXUUVZT7LrrrjG3dazjXjyUUeo3Lp+y/vxK/3o0jV95SpX01LLS2amnnhr1fxMmTHDXAX77mN/r4zlHqlPCzJkzrTYdi1WyNlZmor676iBREbVte8Y7Hrcqavh5/fXXa/Q1HQAA2BHBRwAAgFpKDY9qML/uuuts8eLFFZrHlClTfP8fqxygX0aFRCoXl2r77befb5Dv1VdfrXCDubKa9HqNJ5doydpY4wkpIxM1x1577eX7/7feesu3VJz+99JLL1l18cueqq4Sr5WxYcMGl5V4/vnnxwwi6Dg2YsSIhJehYOWgQYOi/l+BL7+A7kEHHRSzg0c62XfffX2zDZUtduuttyZUcjLS9tEyYn2fomUDBnv22Wd9/z948OCklIGsSgqORtvHtJ01hnNZWVnUIFysgJDEKoH80EMPWUXp2iTV57K9997b9//Khn3ttdd8O0doTOyKqm3bM16xKkgo+Dh58uSE5hnp+BBr++rYUNHAsapN6PVr166t0OsBAKhrCD4CAADUYmqY0RhsGhvqvPPOi6vMn+fnn392ZT9jNTb7adasWczyV36ZJ6mgdfRrrFKQ8IILLog7gKsGX5UX/L//+z878MAD7W9/+1vEhqpY2+bNN9+M2kj/5JNPuv+j5lAZv8aNG/sGZn7/+9/bjz/+uMP/9Jy+v9X1XVHwXCXxUun777/3LVOb6HFQ30G9D31v1CFDgb1HH300rjK2f/nLX6xDhw4VWrZKr1ZUZV5bHZSNqHFoY5XUVaZprPOQPjN9VldeeWXE/x999NExM9P8ApCjRo2yMWPG+M4j1jLShV/2ot+YwPFkPcqQIUN2GG84mLKU77jjjriPTzqn6tx/8cUXu2uTF1980VJJJTf32GMP32n+/ve/R8zEU0ekP/zhD1HHcq2L2zORc6BfJrjOgRdddJE79sei67HLLrss4rAAO++8s/Xq1cv3O6HlxBtA1LFI2cM33nijC6DeddddLgANAABiY8xHAACAOkCZDxpHSDeVvxo4cKBrfOvXr58bY0hj5DRs2NCV8VSG1ejRo12ZRTUGRaPx2GJlEHTr1i1mgPPQQw+1Pffc03Jzc93YeB41Hv32t7+1VLj88stdY3W0YN+sWbNc+UON+aaAotZNJfkUuFAGlRqxVCpNARo1nK1evTrmMmONOaQxIRWMOvvss112i7aNxph64YUXYo4XifSj/eWkk05yQY9oli9f7srt9u3b12UkqaFW+16sDOTqzkLcfffdK9TgreBatHHTdMxStqUC/xUxfPhwSwYFaM4444wKv/6II46wW265JWYmczhlPOpYU9PoWKqylX5BXWX56lykbasMRmWIeucfZZ1NnDjRnX9WrVrlsg8jUVa/gsf6fkSjIIEqAGg5PXv2dN8nnd+Ujf7xxx/7vg99/7SMmkD72O23355QJQF1fom3pK/Kr+tc7Jcpqv/pc9cYlLq26Nixo7vW2LJliztHLlmyxI2lqGNZrHK3qXDuuee6/SwaXfv89a9/tYcffthliKrjiN6D1t3vuqiubs949O7d2wX0/cqHFxQUuGsedQ7Rfr3LLru4Y6GOn7rO0jXjF1984a6BdI4488wzox6HLrnkkqjLGTdunAvUHn/88a76ha5V1XlC21HHIR17tH2nTZvmrunSoUoHAAA1EcFHAACAOkY9tr/77jt3qwxlYimg4keNyvn5+b4ZTGrkUa/9cGp8SlXwsU+fPi7o4xdAUaOUskh1SwZlW6qkn18jvRcwRu3w5z//2WV+KcjoR0HsVGcaRqMGXmVMV0WGmIIf0YKPUpngYzKceOKJLgOqMtSgrQ4WkY5xftTw7pcdla7UqUKBnSeeeMJ3OnXQUGBHt4pQZwwFh9RBw+8Yqu+bbonQcVnZscGdYdKZ9hMF8mOVkQ0fjzRWWfRgyqpTwNavBLkCYiNHjrSaQNmHytJVRys/ypLzyx6tqNq2PeOlTGZd06xbty7qNOoEpqzkWJnJfg477DCXpahApV/G6H/+8x93AwAAVaNmXE0DAAAgrSgrRA3M8Tbg1wQqjxprrKBkat26daW2jXrrK7iLmkPB+rvvvrvCQSUFGCpa/rOilPXhFyxVoEaBsoqIlXmlrDZlnqRa/fr1XSP5nXfemZQAlLJrUvGadKFxBuPNqquMffbZx66//vqkz1eBx1glxdPNaaedltD08ZZc9bRo0cKNJRirXHhNctttt1X4HKrqBwpgVlRt3J7x0PZWOX+/EuTJojF9lTkJAACqD8FHAAAAJKRz585uzEGVXY2HMlNUMizdZWVl2T//+U9XWjVVFODo0qVLwq/r2rWr3XPPPb7jJyE9qYzkI4884spMJkIlIDUOmMaf8guaJZuyD2O9H5VurohOnTpZ//79K1XyNdn0fl577TU3rluyqKNAXl5e3NPrmKBSjzWVjksa87Yy5WrjpbKL6jiSjH1f81Dg8ayzzrKaJp5xDD277rqrGxevIp2ONI6mrgFqA30nVQY70Q4dCjw+9dRTMTNHY3UyqW3bM14DBgxw15BV3ZFGWefKwNbxFwAAVA+CjwAAALWUMjdUDjFZPevVoKwyqG+++aa1adMmoWwvNbBpfMl0p9749957rwvsaVymytI4Qpdddpkr6xot+0GNcGrMjJfGA3zuuefcOJ2omZRh+8orr7ggQDz75LXXXuuyOBQg13hf0SR7nygqKoo5Nl5FS656YmXIvf/++75lNZPVSK2x1Z5//nkXjNDYZMmkzy2R7aSSmDWdAi833XST/fvf/3adJaqSgoUqma1jY0UpGKfxdM855xyrqeLNZkw06zGYzuMas1Nlyisb8NX3QqUxr7nmGqsu6hilzgbxZBrrGujkk0+2l19+2V0DxRoHMJ7jcW3bnvHSOMHqWKLtWRWdZoIDzLrGUqcCjR1ZWTrGaF6JXAMDAFCXMeYjAABALaVGrfvuu89lSv344482YcIEmzx5sv3000+2du3auOahxjZl4SjrSiVCK9pDX69TA99XX31ln3zyiRvPbtmyZW7Mne3bt1u6OeaYY9x7/vTTT90Yj9p+mzZtivk6ZYMqY2mvvfZyAaZoQcfw7C8FopRJocBHtAZNNXadffbZ7laVjXVIDQW4XnrpJfed0HiA+m5q/FN9X1WSV9+ZoUOHupKmubm57jWzZ892AcFoKpqBGI3GyissLIz6f+2HWsfK0PdMpWg1zle0sQG1jQ488MCE562SsGqQ161Ro0auI4a2pbavgg46tinzUp+Fpq1KKpmrTgN1JfgYPHavPjuNvfbqq6+685DfeG/eeUcdN3QMjTcTXZ+jjtXffvutC0SqXHCs4JD2hUGDBrnATypLblcVLzvar4OCOgNVtiSuznMaC/VPf/qTC8TpODFz5sy4Ogko203ZxXvvvbcLlKVDJxqtgzJ1VUpe49tqH1KpaZ3zdUxVRyTtH/pe6nwtOl7FKgkdb7Zzbdue8dK+ePvtt7uxkHUu/Pzzz905Ltb7VmcRZflqLN14OrXpeKJOBeo8p0x+3X7++Wc3/nksClhqWdq+2gd0XAIAAPHLKI/2Kw8AAAC1lhonly5dakuWLHENtFu2bHENMRrfTA07agxT47wa5fW4rtMl85w5c2zevHm2ceNGt/0UNFVWmrZPfn6+K3vXrl27Si1HjW5qFNNYd2qgV6OZGhP1OajHPWVW6zYF6ZTF4ZfRdOutt6Z0nYBE6Tg6d+5cd+5RgEcBdQWH1dCvYIrKUVY2S0nHbC1Dy9JyvECkAo66KYig5XBMTQ5dQ6hT0cqVK935UZ+rrid0LaHtrc4U3bt3rzXXE2PHjvUtzaxs31hZ437q2vb0qLPLlClTXEccXWvppo4het+6zlKHEV1rVXYsXnXyUYB30aJFgWu6srIytxxtU13LaTnqqAIAACqO4CMAAAAApIiC/AUFBQmPd6VMrvPPP9+2bdsWdRqNJalsEABAbOpU1LNnz4Q2lQKCyqJTB65olGmn8pwAAAB1GWVXAQAAACBFVPJ4yJAhbkxWlSxVqTy/jFk1dKuMpDIe/QKPytDQPAEA8VH2orLdVN5X5YH9xntVVp5KZD/wwAO2Zs2aqNMpm1al2wEAAOo6Mh8BAAAAIEVU6jg8O1GBQ5V40/hiaghXSTiV2VPZyPnz57tycLForLeTTjqpCtccAGqXQw45JCSDUSU3lQmpDiG6r/KeGpt64cKFbjzC4uLimPM8/vjj3RiSAAAAdR2ZjwAAAABQjVavXu1uFXXQQQfZiSeemNR1AoC6RoHGSZMmuVtFaKzsq6++OunrBQAAUBNVbpRmAAAAAEC1UanVf/7zn67UHwCgeihb8tlnn3UZ7AAAACD4CAAAAAA1ToMGDexPf/qT/fvf/7aGDRtW9+oAQJ119NFH2+uvv+4yHwEAAPALyq4CAAAAQIrk5uba4Ycfbl9++aVt2bIl4dc3a9bMNXQPHz7cunfvXiXrCAB1wVFHHWVvvfVWhcpe169f35W8PvPMM22fffapkvUDAACoyTLKy8vLq3slAAAAAKAuKS4uth9++MGNLTZ9+nRbsmSJrVixwgUk9b/s7Gxr2rSpu3Xu3Nn69u1r/fr1c43cZDoCQHKoSWzq1Kn2448/2s8//2yLFi2y5cuX26ZNm6yoqMgFGXNyctxNpVX79Onjjscqed2yZUs+BgAAgCgIPgIAAAAAAAAAAABIinrJmQ0AAAAAAAAAAACAuo7gIwAAAAAAAAAAAICkIPgIAAAAAAAAAAAAICkIPgIAAAAAAAAAAABIiqzkzAaoWUpLS239+vXufvPmzS0zM7O6VwkAAAAAAAAAAKDGI/MRdZICjxdddJG7eUFIAAAAAAAAAAAAVA7BRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEQAAAAAAAAAAAEBSEHwEAAAAAAAAAAAAkBQEHwEAAAAAAAAAAAAkBcFHAAAAAAAAAAAAAElB8BEAAAAAAAAAAABAUhB8BAAAAAAAAAAAAJAUBB8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEYCvDcXb7bsV623W+kK2FAAAAAAAAAAA8JXl/28AddnWklJ7asZSKyotdY8P79TKdm+dW92rBQAAAAAAAAAA0hSZjwCiGrdyQyDwKB8vXsPWAgAAAAAAAAAAURF8BBDVos1b2ToAAAAAAAAAACBuBB8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEQAAAAAAAAAAAEBSEHwEAAAAAAAAAAAAkBQEHwEAAAAAAAAAAAAkBcFHAAAAAAAAAAAAAElB8BEAAAAAAAAAAABAUhB8BAAAAAAAAAAAAJAUBB8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEQAAAAAAAAAAAEBSEHwEAAAAAAAAAAAAkBQEHwEAAAAAAAAAAAAkBcFHAAAAAAAAAAAAAElB8BEAAAAAAAAAAABAUhB8BAAAAAAAAAAAAJAUBB8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEQAAAAAAAAAAAEBSEHwEAAAAAAAAAAAAkBQEHwEAAAAAAAAAAAAkBcFHAAAAAAAAAAAAAElB8BEAAAAAAAAAAABAUhB8BAAAAAAAAAAAAJAUBB8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEQAAAAAAAAAAAEBSEHwEAAAAAAAAAAAAkBQEHwEAAAAAAAAAAAAkBcFHAAAAAAAAAAAAAEmRZdVsw4YNgVt5ebk1b97cmjVrZrm5uZaRkVHdqwcAAAAAAAAAAAAgXYOP48aNs6+//tp++OEHmzJlihUVFUWcrmHDhta/f3/bfffdbd9997W99tor1asKAAAAAAAAAAAAIN2Cj6WlpfbGG2/Yc889Z3PmzAk8r0zHaBSUnDBhgrs99thj1rNnTxs+fLidcMIJlpVV7QmbAAAAAAAAAAAAAFI95uO8efPslFNOsRtvvNFmz57tAo5e0FFlVf1u4k2v12oep556qs2dO7eqVxsAAAAAAAAAAABAOgUf3333XTvppJNs+vTpLoAYKaioW4MGDSwvL8/ddD/4f+K9To+nTZvm5vn2229X5aoDAAAAAAAAAAAASFBWVQYer7nmmpCgo+5rLMfDDjvM9t57b+vbt68rp6qAY7Bt27a57MapU6fat99+a5999pkVFxcHApcqyXrddde5+8cdd1xVvQUAAAAAAAAAAAAA1R18VKbj9ddfHwg86m+jRo3s4osvtjPOOMNycnJ8X69gZJ8+fdzt5JNPts2bN9sLL7xg//rXv1zgUfMsKyuzG264wXr16mW77LJLVbwNAAAAAAAAAAAAANVddlWBR2UveoHHAQMG2AcffGB/+MMfYgYeI9Fr9Nr333/f+vfvHwhqahlaFgAAAAAAAAAAAIBaGHz86KOP3LiMXonUQYMG2TPPPGPt2rWr9Lzbt2/v5qV5euNBzpgxwy0TAAAAAAAAAAAAQC0LPr766qvur4KDzZo1swceeMAaN26ctPlrXvfff7/l5uYGApwvv/xy0uYPAAAAAAAAAAAAIA2Cj1u2bLHvv//eBQV1u/TSSy0vL8+SrVWrVnbJJZe4AKdu48ePd8sGAAAAAAAAAAAAUEuCj1OmTLGSkhIXEMzMzLSjjz7aqsqwYcMsKyvL3S8tLbWff/65ypYFAAAAAAAAAAAAILZfondJsnDhQvdXWY877bSTtWzZ0qqK5q1laHxJWbRoke21115Vtry6QIHjH3/80ZYuXWqrVq2ynJwca9u2re26665J/SwXLFjgxupcsWKFlZWVuWV0797devfunbRlAAAAAAAAAAAAoIYHH9evXx+436ZNG6tqWoYXfAxediooaDZ37lz76aefXNalbjNnzrTt27cHprnzzjvtxBNPTGi+S5YssUMPPbTC6/Xggw/aEUcckdBrtm7dao888oi98cYbtmbNmh3+X79+fTvggAPs8ssvt5133rnC6/bJJ5/Y448/7rZZJL169bKzzz7bTjnllAovAwAAAAAAAAAAALUk+KjMOU+TJk2sqgUvQ6VXU+Gjjz6y559/3pWYrQ3jTM6ePdsuu+wymzdvXtRpFFAdPXq0ff3113bdddfZb3/724SWsW3bNrv55pvt9ddfj7kuf/vb39yy/vGPf7jMSwAAAAAAAAAAANTRMR8bN24cuK+ynVUteBmNGjWyVJg4caKNGzeuVgQetf1+//vf7xB47Nu3r8ue3HvvvUMCvMXFxS6I+O677ya0nJtuummHwGN+fr4ddNBBLsuzU6dOIf9T8FFZlqkKKAMAAAAAAAAAACANMx/btWvn/paXl9v8+fOtKoUvQ+MGVqemTZu64OvKlSuTOt/hw4e7UqTxysvLi3v7KeMxeH01hqYyDoPHXty4caMr5fqf//wn8Nz111/vplGZ1FheeuklV841uISrgpEqR5uZmRlYl08//dRlVW7evNk99+WXX9pDDz1kf/7zn+N85wAAAAAAAAAAAKhVwUcFrzxr1661CRMm2J577mlVQfMOHp8weNlVLTs72/r06WP9+vWz/v37u1u3bt1csEy3ZGrWrJl17NjRkk3jL/7444+Bx1qGAoy5ubk7LP+GG26wjIwMGzVqVCADUgHJWO9VY0mOHDky5LkHHnjADjvssJDnNO+hQ4e6ALJKunrle5966ik744wzrHXr1pV+vwAAAAAAAAAAAKhhZVe7du3qAkgKJskTTzxhVeXxxx8P3G/Tpo0L/qXCxRdf7EqvKqNP4xMed9xx1r1798B7rinCA4c33njjDoHHYFdeeaV16NAh8FiZitOnT/ddxosvvhgSID7yyCN3CDwGGzBgQEiWZ1FRUcjnDAAAAAAAAAAAgDoUfJSjjjrKldHUbezYsfbWW28lexH25ptv2hdffOECfrodffTRliotW7a0rKykJoym3MyZM23WrFmBxz169LADDzzQ9zUaU/P0008PeS7W2I/vvfdeyONzzz035rop+Fiv3v92y/fff9/tSwAAAAAAAAAAAKiDwUcFjxo0aOCCggoaKTvwww8/TNr8P/jgg0AZUM1fYwgmMiYizMaMGROyGY455pi4Nkv4dKNHj446rcaSnDp1auBxly5dbODAgTGXkZ+fb4MHDw48Vubk5MmT+dgAAAAAAAAAAADqYvBRwaMLLrjABQYVINT4fVdddZXdcccdVlhYWOH56rWax9VXX+3m6c3/97//vVsm4vf111+HPI53XM527dqFlF6dP3++LV++POK033zzTcjjPfbYI+71C1+f8PUFAAAAAAAAAABAHQk+euMiKoDkBQhLS0tt1KhRdvjhh9uDDz5oCxcujHtemvaBBx5wr9U8NC+v3Opuu+1ml1xySVW8hVptzpw5gfsqcdqvX7+4XxuevTh79uyI04U/r/EcK7qM4PUFAAAAAAAAAABA+qqSwQs1JuLDDz/sxvibNm1aoESqSmg++uij7qYsul122cWNN5ibm2s5OTluuk2bNtmGDRts3rx5rmynl1nnjfvnzat37972yCOP1PjxF2P5/vvvbfr06TZjxgxbu3atCxY2b97cZSAOGjTIDjnkEOvfv3/c89O2LSgoCDzOy8tz4znGq2PHjiGPlf34m9/8Zofp9HywTp06VWoZAAAAAAAAAAAASH9VFrlTQFGZirfccou9/fbbLmgYHERctmyZCyx+/vnnUefhTSvBr9fYgzfddJMLWNZ248eP3+G5LVu2uO2n/ykAu++++9p1111nO+20U8z5LVq0KOSxgsCJaNu2re/8PIsXL67wcsKXkUimLAAAAAAAAAAAAGpZ2VVPkyZN7P/+7//s/vvvdwGl4OzF4GBipFuk6dq0aWP33nuv/eMf/6gTgcd4aXzFU0891T744IOY027evDnkccuWLRNaVosWLUIeK1M1kvDnw1/np3HjxpadnR0SbC0rK0toPQEAAAAAAAAAAJB6KalZeuSRR9rQoUPt448/thdffNF+/PFHKykp8X2NF4DMzMy0XXfd1c4880w37qMe1wUKwO2333629957W69evVyQUCVm161b50rZfvLJJ64kq2fr1q121VVXWbNmzWz//fePOt/CwsKQxw0bNkxovYKDgl5gMJLw5yuynKKiopD5JRJwXrFihe//tR0BAAAAAAAAAACQXCkbMFFBw6OOOsrdFCibPHmyu61cudKNQ7hx40YXcFS5Vt1at27tgo4DBw50gbi6Qu/1xhtvtBNOOCHi++7WrZvtvvvu9rvf/c5lPCrgqLEgpbS01EaMGOECkxoXMhJt+2ANGjRIaP3Cg4jh84v2fKLBx/D1SjT4eOCBB/r+v379+tanT5+E1gkAAAAAAAAAAABpEnwM1qhRI5fRpxtCKcNRWZ7x0FiPGlfztNNOC5Q5VSD3ySeftCuvvDKueXhlbeMVPn3wuJzpsBwAAAAAAAAAAADUseAjkqdHjx529dVXu2xJz+uvv+4yICMF/BT4DVZcXJzQ8oJLoUq0rFQtZ/v27SGv0xig8Qpfr0ReK2PHjo1ZdjV4mwEAAAAAAAAAAKDyCD7WAieeeKLde++9LutRVIZ15syZ1rt37x2mDQ8WJhp8DJ/eL/ioUrrBr6tM8DHR0rtt27aNWXYVAAAAAAAAAAAAyVUvyfNDNVAgbdCgQSHPKfgYSfi4icoATERBQUHI46ZNm0acLvz5RJaj8SKDx4xU4LFePXZVAAAAAAAAAACAdJe2mY8//fSTvf/++zZlyhQXuMrOzrY2bdq4cSKPPPJIy8/Pr+5VTCsdOnSIK9jXuXPnkMfLly9PaDkrVqwIedypU6eI0+n5OXPmhLxOJWLjEb5O4esMAAAAAAAAAACAOhp8XLhwoX322WcuE09Zcw0aNLCOHTvaXnvtZQcddJBlZmaGTL9+/Xq7/vrrbfTo0YHnysvL3d/p06e7sfwefPBB++Mf/2gXXHBBVa9+jaHgbDzlVJs3b24tW7YMZDCuWbPGZRmGjwUZzZIlS0Ied+/ePeJ0en7MmDGBx4sXL45r/oksAwAAAAAAAAAAAHUk+Lht2za7/fbb7bXXXrOysrId/j9q1Cjr2rWr3XPPPda3b1/3nMYsPOuss1zGnAKOGRkZ7nnvr+h5Bcvuu+8+W7lypf3tb3+rqrdQo4RnOirIGE3Pnj1t3Lhx7r4+G2WXhpdtjWby5Mk7zCuSXr167fC6008/Pe6s12DxZkwCAAAAAAAAAACgetWrqsDjH/7wB3vllVestLQ0kLkYTM/Nnz/fhg8fbgsWLHDP3XTTTTZ79uxAwFHThN/0vPe/559/3j755JOqeAs1zs8//xzyWCVqo9l3331DHk+YMCHucqhLly4NPO7WrZu1b98+rmVMnDgxrmVEWp/9998/7tcCAAAAAAAAAACglgUf//Wvf9l3330XkrUYHkT0/ldYWGh33XWXzZgxwz766KNAYLFdu3autOrIkSPtqaeecpmO5557ruXm5gaCkPqr19Z1c+fOddvPo1K2u+++e9TpDznkkJDH7777blzLCZ8ufD7BNCanl9Hqld8Nz5qMRNms33//feBxq1atbODAgXGtHwAAAAAAAAAAAGpZ2VWN2fjkk0+GBB379+9vxxxzjHXp0sU9t2jRInvvvfcCwagvvvjC8vLyAvM47rjj7LbbbrP69euHzPuoo45yAclLLrkkUDZU2XhfffVVnc2O0/b9xz/+EZJduuuuu7ogbTQ777yz7bTTTjZr1qxA8FJjaR544IFRX1NUVGQvvfRSyHP6TP0MGzbMpk6dGnj89NNP2wMPPOD7mmeffTakTO/RRx8dUnYXAAAAAAAAAAAAdSjzUUFFlV31nH/++fbqq6+68qoKbummcR1ffvllV5pVFDh766233P3u3bu7sSLDA4+epk2b2j//+U9r0aJFICilwFltoGBsIuVJFaT7v//7PxszZkzI8xdddFHM11566aUhj2+99VY35mY09957b0jJ1cMOO8z69Onju4zf/va3LnPR8+GHH9pnn33mO9ajgo+e7Oxsu+CCC2K+FwAAAAAAAAAAANTS4GPweH29e/e2q666Kuq0I0aMsF122cXd19iQCiaeccYZlpXln5DZvHlzO/744wPZfsHZdamwZMmSiLeNGzeGTLdu3bqI061evTrifJWBqPd/5plnuvEy16xZE3E6ve/x48fb2Wef7bIJgw0ZMsR+85vfxHwPQ4cOtd122y3wePHixfa73/3OZs6cGTLdpk2bXGDyueeeCzzXsGFDu/zyy2Muo1GjRvanP/0p5LkrrrjCBaP1eQe/n08//dTOO+88KykpCTyvMrutW7eOuRwAAAAAAAAAAADU0rKr06dPD9w/8cQTY06vaVRi1TNo0KC4ljN48GAXeFPgKjgjLxUOPfTQuKa7++673S3Suo8aNco3gKvbDTfc4Ma+7Natm8v4VDaoytpOmzbNCgoKdnidxkZUCdZ4KND74IMP2sknn2yrVq1yz6kMq0reaqzGTp06uWUpG1HjcgbT59WrV6+4lnP66ae7jM433njDPVZW7N/+9jd76KGHXOakxqfUclWKN9gBBxywQ+ASAAAAAAAAAAAAdXDMR0+/fv1iTq9AV7AOHTrEtZzg6ZSdV1tpTEvdYlGJ0+uuu85lJcYrPz/fjc952WWX2fz5891zCuZOmTLF3cJp3tdee60de+yxCb2Hv//97+6vF4CUFStWuFskBx98sAuiKjAJAAAAAAAAAACAOlx2dfPmzYH7LVu2jDm9xm4M1qRJk7iW07hx48D9rVu3Wm2wxx572DnnnOMyAmOVnpWcnBw74YQT7O2337abb745ocCjZ6eddrI333zTja2Yl5cXcRplXB5yyCGuXKrKwiaqQYMGduedd9rIkSOtf//+Uafr2bOny6p89NFHXaYnAAAAAAAAAAAA6njmY/BYfgpaxRLPNNHKhlaX8HERk6VLly4ue1GKi4tt9uzZrqSsxojcsmWL27YKyuXm5rqypwoc1qtX+fixxmbU2Jwax/GHH35w41JqvEkFgtu2bevGhownkBzPOJO6KctyxowZtnLlSveelIGpwKPGCAUAAAAAAAAAAEDNlfTgI5JDWYwqWxtP6dpkUbalxqPUrSppDEvdAAAAAAAAAAAAULskvewqAAAAAAAAAAAAgLqJ4CMAAAAAAAAAAACApCD4CAAAAAAAAAAAACB9x3zMyMhwfydNmmRLly71nXb16tUhjydMmGDl5eUxlxH+OgAAAAAAAAAAAAC1MPgoCiBeeeWVCb/mrLPOSijIGU+gEgAAAAAAAAAAAEANDj4mEhj0MiUlkWBi8OsAAAAAAAAAAAAA1OLMx6qYNhmvAwAAAAAAAAAAAFADgo+DBg1K9iwBAAAAAAAAAAAA1MXg46hRo5I9SwAAAAAAAAAAAAA1QL3qXgEAAAAAAAAAAAAAtQPBRwAAAAAAAAAAAADpWXY1VUpLS23lypWBx+3bt6/W9QEAAAAAAAAAAADquhobfFyxYoUddthh7n5GRoZNmzatulcJAAAAAAAAAAAAqNNqbPBRysvLq3sVAAAAAAAAAAAAAPyKMR8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEQAAAAAAAAAAAEBSEHwEAAAAAAAAAAAAkBQEHwEAAAAAAAAAAAAkBcFHAAAAAAAAAAAAAElB8BEAAAAAAAAAAABAUhB8BAAAAAAAAAAAAJAUWZZkffr0sVTJyMiw8vLylC0PAAAAAAAAAAAAQAqDj6kMBir4CAAAAAAAAAAAAKCWBh+FoCAAAAAAAAAAAABQ91RJ8JFSqAAAAAAAAAAAAEDdk/Tg44wZM5I9SwAAAAAAAAAAAAA1QL3qXgEAAAAAAAAAAAAAtQPBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACRFliXZjBkzAvd79uxpWVlJX4SzYcMGe+SRRwKPr7vuuipZDgAAAAAAAAAAAID4JD0yePzxx1tGRoa7ffbZZ9a+ffsqCSJu3rzZnn32WbecRF4HAAAAAAAAAAAAoGpUSVpieXl53NNWNoioZXmvBQAAAAAAAAAAAFDLxnysSDAwkYAlgOpTuL2UzQ8AAAAAAAAAAFIXfCSQCNRek9durO5VAAAAAAAAAAAAdSn4CKD2GrusoLpXAQAAAAAAAAAApCmCjwAAAAAAAAAAAACSguAjAAAAAAAAAAAAgKQg+AgAAAAAAAAAAAAgKQg+AgAAAAAAAAAAAEgKgo8AAAAAAAAAAAAAkoLgIwAAAAAAAAAAAICkIPgIAAAAAAAAAAAAICkIPgIAAAAAAAAAAABICoKPAAAAAAAAAAAAAJIiy6pARkaG+ztixAhr0KCB77Tbtm0LeTx8+PC4lhH+OgDJV8ZGBQAAAAAAAAAA1R18lPLycps8eXLCrxk/fnxCQU69BkDy6bu1rLCITQsAAAAAAAAAAKo/+OhlPyZ7WgCpsXATgUcAAAAAAAAAAJAGwUeyEYGab+GmrdW9CgAAAAAAAAAAoK4HH5977rlkzxIAAAAAAAAAAABAXQw+Dh48ONmzBAAAAAAAAAAAAFAD1KvuFQAAAAAAAAAAAABQOxB8BAAAAAAAAAAAAJAUBB8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFIQfAQAAAAAAAAAAACQFAQfAQAAAAAAAAAAACQFwUcAAAAAAAAAAAAASUHwEQAAAAAAAAAAAEBSEHwEAAAAAAAAAAAAkBQEHwEAAAAAAAAAAAAkBcFHAAAAAAAAAAAAAElB8BEAAAAAAAAAAABAUhB8BAAAAAAAAAAAAJAUBB8BAAAAAAAAAAAAJAXBRwAAAAAAAAAAAABJQfARAAAAAAAAAAAAQFJkWRqYNGmSjRkzxn744QdbunSpbdiwwbZs2WIZGRk2bdq0HaYvKChw00jDhg2tffv21bDWQO1WbuXVvQoAAAAAAAAAAKCGqdbg48SJE+2uu+6yKVOmBJ4rL48d8Pjpp5/s4osvdvezs7Ptyy+/tJycnCpdVwAAAAAAAAAAAABpWnb10UcfteHDh7vAoxdw9P4q49HPQQcdZF26dHHTFxUV2XvvvZeSdQYAAAAAAAAAAACQZsHHp59+2h544AErLS0NPKcMxkGDBrnAYjzZj8OGDQvcHz16dJWtK1BXZZh/JwAAAAAAAAAAAIBqDz7OnDnT/vGPf7jsRt0aNWpkN9xwg33//fc2atQodz8eQ4YMcX8VqBw/fryVlJRU8ZoDAAAAAAAAAAAASKsxH++//34rKytz95s1a2b/+c9/bKeddkp4PnqNApdbt251pVfnz59vvXr1qoI1BuqmcoudgQwAAAAAAAAAAFBtmY+bN2+2r776KpD1+Ne//rVCgUfR64ODjfPmzUvimgIAAAAAAAAAAABI6+DjhAkTXHlUlUrNzc214447rlLzy8vLC9xfs2ZNEtYQAAAAAAAAAAAAQI0IPq5YsSKQtThgwAD3tzJycnIC9wsLCyu9fgAAAAAAAAAAAABqSPBxw4YNgfvKfKys4uLiwP2srJQPXwkAAAAAAAAAAACguoKPTZs2DRn/sbJWr14duN+8efNKzw8AAAAAAAAAAABADQk+Bo/ROGfOnErNa/v27TZ9+vTA43bt2lVqfgAAAAAAAAAAAABqUPCxf//+7m95ebktWbLEZs+eXeF5ffbZZ1ZUVBQoubrbbrslbT0BAAAAAAAAAAAApHnwsX379tazZ8/A4wcffLDCYz0+/PDD7n5GRobtvvvulp2dnbT1BAAAAAAAAAAAAJDmwUc588wzA/c///xze+ihhxIut3rttdeGlG0999xzk7qOAAAAAAAAAAAAAGpA8PHUU0+1bt26BcqvKoPxoosuChm/MRJN+8UXX9hpp51mH330kct41E3lVg866KAUrT1Qd5RX9woAAAAAAAAAAIAaJyvVC8zMzHQBx9/+9re2ceNGF1QcO3asu3Xo0ME6d+4cMv2IESNs3bp1NnXqVNu0aVPgeb2uVatWdv/996f6LQAAAAAAAAAAAABIh8xH6d69uz3++OPWunXrkGDikiVL7Ntvvw157sMPP7TvvvsuEKj0nm/Xrp2bR35+fnW8BQAAAAAAAAAAAADpEHyUAQMG2DvvvGNHHXWUK58azCupGvx88P0hQ4bY66+/bn369EnpOgMAAAAAAAAAAABIo7KrwZo3b2733XefXXHFFfbSSy/Z999/78Z+LC0t3WHarl272r777uvGjOzdu3e1rC9Ql4R2CQAAAAAAAAAAAEjz4KOnU6dOdvXVV7v7RUVFtnr1atuwYYOVlJRYbm6u5eXlWbNmzap7NYE65ZcixwAAAAAAAAAAADUs+BgsOzvbBSN1AwAAAAAAAAAAAFBzVNuYjwAAAAAAAAAAAABql5QHH+fMmZPqRQIAAAAAAAAAAACojcHHYcOG2SmnnGLPP/+8G9cRAAAAAAAAAAAAQO1QLWVXp0yZYrfddpsdcMABdtlll9mYMWOstLS0OlYFAAAAAAAAAAAAQJJkWTUpLy+3bdu22aeffupuLVu2tGOPPdaOO+446927d3WtFgAAAAAAAAAAAICakvl4zDHHWHZ29g6ByLVr19ozzzxjJ5xwgrs999xzVlBQkOrVAwAAAAAAAAAAAFBTMh//8Y9/WGFhoX300Uf29ttv2/jx493zGRkZgUDk9OnTbcaMGXb33Xfbb37zGxeMPPjggy0rq9oSNQEAAAAAAAAAAADEUC3RvCZNmthJJ53kbsuWLbM333zT3nnnHVu4cGFIILKkpMSNB6lbbm6uDRs2zAUi+/btWx2rDQAAAAAAAAAAACCdyq6Ga9++vV1yySX28ccf24svvminnnqqNW3a1GVAenR//fr19vzzz9vJJ5/sSrc+/fTTtmbNmmpddwAAAAAAAAAAAABpFHwMtttuu9ktt9xiX331ld1333124IEHWmZm5g5lWWfPnu1Ksh500EF24YUXuhKuAAAAAAAAAAAAAKpXWg6i2KBBAzvqqKPcbe3ata4k61tvvWUzZ84MBCIVhFRZ1rFjx9qXX35pRxxxRHWvNgAAAAAAAAAAAFCnpVXmYyR5eXl27rnn2ttvv+0CkGeffbZ7LjwbEgAAAAAAAAAAAED1SvvgY7DevXvbiBEj7KqrrrIWLVpU9+oAAAAAAAAAAAAASPeyq5FMmDDBZT5qfMfCwsLqXh0AAAAAAAAAAAAANSn4uHjx4kC51aVLl4aUWPXGfZRWrVpV63oCAAAAAAAAAAAASMPgo7IaP/zwQxdwnDhx4g4BR0/9+vXt4IMPthNPPNH233//altfAAAAAAAAAAAAAGkUfFRw8euvv7Y333zTRo8ebUVFRYHnFXD0shx1GzBggB1//PE2bNgwa9asWXWvOgAAAAAAAAAAAIB0CD7OnTvXBRzfeecdW716dcQsRz1u06aNHXfccS7o2KNHj+pcZQAAAAAAAAAAAADpEnxcv369vf/++y7oOHXq1KhlVRs2bGiHHnqonXDCCbbvvvtavXr1Ur2qAAAAAAAAAAAAANI5+KjxGUtLS0MCjsFlVXfbbTc3juORRx5pOTk5qV49AAAAAAAAAAAAADUl+FhSUrJDwLF9+/Z27LHHuizHLl26pHqVaoTly5fblClTbOXKlbZ161bLz8+3rl27Wv/+/UMyRiujuLjYJk6caMuWLbOCggJr3ry5W84ee+yR1EDwjBkzbN68ebZixQqX0dq2bVvr3bu3ez8AAAAAAAAAAACouaplzEcFHBs1amRDhw514zjus88+VtOUlZW5MSt/+ukn+/nnn91t5syZtn379sA0d955p8virIxx48bZww8/7P5qmeE6duxop59+up133nmWmZlZoWWsW7fOHnjgAVcOd9OmTTv8X5/VYYcdZiNGjHCB4op65ZVX7LnnnrPZs2dH/P/AgQPtggsusCFDhlR4GQAAAAAAAAAAAKhDwcdBgwa5DMfDDz/cmjRpYjXNRx99ZM8//7zLQtyyZUuVBmgVEHzsscciBh09S5YssXvuucc+//xze/DBB12mYiLGjx/vgoqrVq2KOo0yLd99910bO3asC6gqEJmIjRs32jXXXGNjxozxnW7y5Ml26aWX2qmnnmo33nij1a9fP6HlAAAAAAAAAAAAoI4FH0eNGmU1mcqSKguxqinb8dFHHw15rkWLFta3b19r3LixK1s6Z86cwP9+/PFHu/DCC+3FF190mYrxmDVrlntNYWFh4DllTw4YMMCVQl2zZo0LCG7bti0QRLz88svtqaeessGDB8e1DI3vqdd8/fXXIc937tzZdtppJ1eGd9q0aSHBT2VIyq233hrXMgAAAAAAAAAAAFCHy67WRk2bNnVBQY3JWFlffPGFPfTQQ4HHGtNRAbxzzz3XGjZsGHheQdCrrroqsMzp06fbTTfdZHfffXfMZRQVFdkf//jHkMCjAop33HGHderUKfDc6tWrXRDw448/do9VVvZPf/qTffDBB5aXlxdzOcreDA48ajvdddddduihhwbGqlQA8vXXX3fL8crWKgCpMqwnn3xyzGWgapSzYQEAAAAAAAAAQILqJfoCmGVnZ9tuu+1mZ511lgv0ffjhh6586SmnnJKUcqsqo6q/nuuuu84uuuiikMCjFyxUCVgF9DzvvPOOzZgxI+ZyXnjhBVu8eHHg8a677mpPPPFESOBRWrdu7QKIwaVW169f78rBxqKg6LPPPht4rDKqTz/9tJuXF3iUrKwsO+200+z+++8Peb3KyBYXF8dcDgAAAAAAAAAAANIDwccEXXzxxa706ksvvWR/+9vf7LjjjrPu3buHBNMq45NPPrGZM2eGBAWHDx8edXoFC6+44orAYwUtR44c6bsMBfQef/zxwON69eq5rMPw4Gbw/2+++WbLyckJPKfyrsqK9KMAZXDw8Oyzz7b+/ftHnX7IkCF2xBFHBB6rFKu2MwAAAAAAAAAAAOpg2dVly5bt8Fz79u1jTpMM4cupKi1btqzS+b/33nshjxWwixXYVGlSZSdqTEYZO3asbdq0KSQjMpjKoBYUFAQe77fffm78RT/KgDz66KPt5Zdfdo8VVFSg9Mwzz4w4fVlZmSvNGjyWpF8Q1aPSsh999FHg8bvvvuu2AQAAAAAAAAAAAOpY8PGQQw4JCZTp/rRp03ynSYZIy6mJtm3bZl999VXgscaQDC53Go0yFpU1qHETReMmatxIBQsjGTNmTMjjY489Nq7103Re8FFGjx4dNfg4adKkkADnXnvtZfn5+TGXoUzPzp0726JFi9zjKVOmuAzINm3axLWOSJ7kfksBAAAAAAAAAEBdUCVlV1X607vFM00ybrWBAnZbtmwJPFaJ0gYNGsT12j333HOH7MZowv8X/tpoBgwYELI+48aNc4HOSL755puQx3vssUdcywifVp9t+LyQGrXjWwUAAAAAAAAAAGp08DGeQGBtCRYm2+zZs3cI9sVr4MCBIY/nzJkTcToFN4NL37Zq1SrukrUKPPbp0yckU9PLUKyO9wIAAAAAAAAAAIBaXHb10ksvTco0ddX8+fNDHnfq1Cnu13bs2NF3Xp4FCxaEBH8TWYa3nMmTJ4csp0ePHtXyXgAAAAAAAAAAAJBeCD6mkcWLF4c8btu2bdyv1biPLVq0sHXr1rnHGzdudPf1XLDwTMVEliHh4zZGy3wMfy/t2rWLexnh6xRtGQAAAAAAAAAAAKgDYz6iYjZt2hTyuGXLlgm9Pnz6zZs3V/kywucnpaWlIWNXNm7c2LKzs5O6DAAAAAAAAAAAANTyzEdUTnDAzstmTER4gK+wsLDKlxE+v0jPaazIRISvU6RlxLJixQrf/3sZogAAAAAAAAAAAEgego9pZOvWrZUKDIYH+cLnF+m5ygYGIy2jsgHOZAQfDzzwQN//169f3/r06ZPwfAEAAAAAAAAAAJBGwcfhw4e7vxkZGXbfffdZXl5eheazZs0aGzFiRGBezz77rNU2el+Vmb68vLxWLAMAAAAAAAAAAAA1Q8qDj+PGjQsEl4qLiys8H71W86pNwapGjRqFPC4qKkro9eHbU2MtxlpGop9B+DpFWkb4c4m+j3iWEcvYsWNjll298cYbE54vAAAAAAAAAAAA0qzsqrLlakvAMJkqGxgMn75JkyY7TBMeyKvsMuIJcG7bti2hZYRPX5HgY9u2bWOWXQUAAAAAAAAAAEBy1Uvy/FAJTZs23SE7LxEFBQUhj3NycnaYJvy5yi4jfJ0lKysrJACpMRsTyX6MZxkAAAAAAAAAAABIPzU2+FhSUlLrstg6d+4c8njFihUJZQsGB+2aNWtmLVq0iLmM5cuXJ7SO4esUPj9Pp06dfF/nJ3ydwucFAAAAAAAAAACA9FRjg4/Lli3zLS9aE3Xr1i3k8eLFi+N+bfi04fMKfj645O2SJUsSWsfw6aMtp3v37r7rl8gywucFAAAAAAAAAACA9FRjg4/vv/+++6tAWm3JjOvVq1fI48mTJ8f92p9++inkcY8ePSJOp/ET27dvH3i8evXqkEBurOzK6dOnh2ScRst8TOZ76dmzZ9yvBQAAAAAAAAAAQPXJqoqZvvXWW3FN9/HHH0csDRrN9u3bXbDs+++/t3HjxgWeHzBggNUGu+66qwsOaoxE+fnnn13Ar0GDBjFfO2HChJDH+++/f9Rp9913X3v11VdDXnvsscfGFRTU+nj22muvqCVv99lnHxs5cmTg8cSJE2POP3h9PAoua14AAAAAAAAAAACoo8HHa6+9NqS0ZyTl5eV29913V3gZer3nuOOOs9pAQUYFDT/55BP3WEHIzz77zI466ijf1xUXFwdeIwoI/uY3v4k6/SGHHBISfHznnXfiCj5quvD5RLPbbrtZy5YtA+NQKmC8cuVKy8/P913GpEmTbNGiRYHH/fr1i/kaAAAAAAAAAAAA1IGyqwoQht9i/d/vFsk555xj/fv3t9pi2LBhIY+fffbZqO/d89prr9nGjRsDjw888EBr2rSpb+ajAoOer7/+2mbNmuW7DGWceqVupWHDhjZ06NCo09erVy8kaFpaWmrPPfecxfL000+HPD7mmGNivgapV1RaymYHAAAAAAAAAACpCz7GCphVdJ66ZWZm2u6772733Xef/eUvf7HaRAG9nXbaKSQT0C9ot2TJErv//vsDj5Vxeumll/ouIzs72y644ILA47KyMrvhhhtCSqoG0////ve/2+bNmwPPnX766da6dWvf5fzhD38IKRmrQKpKyUajLM+PPvoo8FjzP+2003yXgWqS/K83AAAAAAAAAACoBaqk7Gq0YJkCh2effXYgSHbvvfdaq1at4pqnplcgKycnxzp16hTXOIhVRQG/SIKzD2XdunURp1XWYLTAnd7n1Vdf7QJ3XgD3zjvvtKKiIpflqdd6xo8fb1dddZVt2rQpJFOwT58+Md/DGWecYS+88IItXrw4EOT8/e9/b3fccYfbvsEZj7fddpt9+umngeeaN29uF154YcxlqFyq1vmxxx4LjNl53nnnufdz2GGHBaYrKSmxN954w2655ZaQ1//5z392gVKkIf+qygAAAAAAAAAAoI7KKK+KFEUfvXv3DowH+fnnn1v79u2tptl5550r9frBgwfbqFGjfKd56KGHbOTIkSHPqVSqxkBs1KiRzZs3z2bPnh3yfwUdFVBs3LhxXOuhUqvKLNTYkp6srCwbMGCAtW3b1tasWeOCksEZkRpP8sknn7S99torrmUosKgsy2+++Sbk+S5durgMT/1/2rRpbjzIYKeccooLelaVtWvX2kUXXeTuP/roo5aXl1dly6qpRi9da9+vXB/xf1cM7GrZmZkpXycAAAAAAAAAAFAHMx/9BAcbVT4VkV1yySUuU1BZgyp7KgUFBfbFF19EnH633XazBx98MO7Aoyj4p/mPGDHCVq1a5Z5TMPCHH36IOH2zZs1cZmS8gUcvmKn1Ujbnf//738DzCxcudLdIFHi86aab4l4GqkZquyUAAAAAAAAAAIDaIOXBx9GjR6d6kTWSskOvuOIK22+//VwW5Lhx4yKOo9mhQwc3/qJKplYkmDto0CB755137IEHHrD3338/pISrR6VPVSb1yiuvrFCmqoKW//73v+2VV15x4z7OmTMn4nTKuFS52SFDhiS8DAAAAAAAAAAAANTBsquomGXLltmUKVNceVKN/9imTRtXunTgwIGBMraVVVxcbBMmTHDLUpZlbm6utWvXzvbYYw831mayTJ8+3ebOneveiwKmGhtS5Xi7detmqULZ1dgouwoAAAAAAAAAANI+8xEVo4zDqh4fs2HDhi7TsqppbErdAAAAAAAAAAAAULvUq+4VAFADkS8NAAAAAAAAAADSOfNRZT7nzZtnGzZssM2bN0cc39DP8ccfX2XrBtRFFGQGAAAAAAAAAAA1Kvi4YsUKe/755+2DDz5w4wxWBsFHAAAAAAAAAAAAoI4GH19++WW78847rbi4OOEsR09GRoZ7rf4CAAAAAAAAAAAAqIPBx6efftruvvvuiIHD4MfhQcnw/1U0aAkAAAAAAAAAAACgFgQfp02bZvfcc09I5uLQoUPtkEMOsczMTLv66qsD/3vuueessLDQ1qxZY5MmTbLPPvvMjQmp/7Vs2dKuueYaa9++farfAgAAAAAAAAAAAIB0CD4++uijVlpa+svCs7Lsvvvuc8FHWbp0aci0gwcPDtw/5ZRT7IYbbrAnnnjCzWPdunUue/LJJ5+0Pn36pPhdALWfXzXj2Ru2WP+8pqlcHQAAAAAAAAAAUAPUS+XCioqKbPTo0S5zUbfzzjsvEHiMR3Z2tl166aU2cuRIlyVZUFBgf/jDH1wgEkBy+VU1/mjRajY3AAAAAAAAAACo3uCjSqeWlJS4UqsKHp599tkVms/BBx9s559/vruvkqwPP/xwktcUgJ8SxlsFAAAAAAAAAADVHXxcsmSJ+6usxx49elheXp7v9ApURqPgo8q2KpD53nvvBUq5AgAAAAAAAAAAAKgDwccNGzYE7nfp0mWH/yuYGGzbtm1R55WTk2MDBw4MzHfixIlJXVcAAAAAAAAAAAAAaRx8DM5O1PiN4Zo0aRLyeO3atb7zy8/PD9xftmxZUtYRAAAAAAAAAAAAQA0IPgYHF7ds2RLx/xoLMt6AYnAwU2M/Akidx6Yttk8Wr7GSsnI2OwAAAAAAAAAASH3wsUOHDr5ZjRoLMrgc6+TJk33nN3v27KglWwFUrbVF22zi6g02fd1mNjUAAAAAAAAAAEh98LFHjx7ub3l5eUjgMNguu+wSuP/uu+9GnZfGeJw3b17EEqwAUue9havY3AAAAAAAAAAAIPXBx06dOlmbNm3c/cLCQps1a9YO0xx++OGB+3PmzLF77rlnh2kWLVpk11xzjcuUFP3dc889q3TdAQAAAAAAAAAAAPhLea3Sfffd19566y13f8yYMbbTTjuF/P/AAw905Vk13qMyJJ988kn7/PPPbb/99nNjQi5YsMD++9//2rZt29z/FXjUa1q3bp3qtwIAAAAAAAAAAACgujIf5cgjj3R/FTh87bXXdvh/gwYN7IYbbnD3FVjUdPPnz7fnn3/eHnvsMfvkk0+suLg4MH1OTo5dd911KXwHAAAAAAAAAAAAANIi81EZjH/84x+trKzMPV65cuUO4zUedNBBduutt9rf//532759e6C8qscLSjZv3tweeugh69y5c0rfAwAAAAAAAAAAAIA0CD5mZWXZZZddFnO6k08+2QYNGuSyHceOHWtr1qwJGTtSY0Oed9551rJlyypeYwAAAAAAAAAAAABpGXxMRJcuXez2229397du3WqbNm2yZs2aWXZ2dnWvGgAAAAAAAAAAAICaFHwM1qhRI3cDAAAAAAAAAAAAkJ7qVfcKAEhP238dlxUAAAAAAAAAACBeBB8BRDRt3Wa2DAAAAAAAAAAASAjBRwARFZeS+QgAAAAAAAAAABJD8BEAAAAAAAAAAABAUmRZih166KFJm1dGRoY1btzYmjZtanl5ebbLLrvYgAEDbO+997Z69YirAgAAAAAAAAAAALU6+Lh06VIXNCwvL0/aPDU/+fTTT93fVq1a2emnn24XXHCBNWjQIGnLAQAAAAAAAAAAABBdtacHKnAY6Rbv/yU4kKn7q1evtoceeshOOukkW7BgQUrfDwAAAAAAAAAAAFBXpTzzcdCgQYH7s2bNso0bN4YEEHNzcy0/P99ycnJs27Zt7v/KliwtLXX/9wKPKrHapEkTKy4uDkyzffv2wDSa3+zZs+33v/+9vfzyyy4bEgAAAAAAAAAAAEAtCj6OGjXKSkpK7N5777UJEya4IGH79u3td7/7nQ0dOtQ6duy4w2sUYJw4caK9/vrr9sEHH7jnCgsL7ZZbbrF+/fq5xwo8Tp482V555RV777333HOa97Jly+z222+3+++/P8XvFAAAAAAAAAAAAKhbqqXs6l/+8hd75pln3P0TTzzRBRTPO++8iIFHadiwoe27774uYPn8889bixYtbNGiRXbWWWfZzJkz3TT169e3Pffc0+6++2579tlnXeaklwH58ccf29y5c1P6HgEAAAAAAAAAAIC6JuXBR2Umvv/+++7+kCFD7I477rDs7Oy4X7/77rvbE088YZmZmbZ161a77LLLXHnW8NKu9913nws8egHITz75JOnvBQAAAAAAAAAAAEA1Bh+ffPJJ91dBweuvv75C89B4jyeccIK7rwzISIHFAw44wPbZZ5/AWJLjxo2r1HoDAAAAAAAAAAAASKPg408//WQLFy50gce+fftafn5+heelrEnPO++84zuNApBaLgAAAAAAAAAAAIBaEnxcsGBB4H608R3j1aFDh4jzDaYAp2fDhg2VWh4AAAAAAAAAAACANAo+rlq1KnB/+/btlZpXSUlJIKsxeL7BcnNzA/eLi4srtTwAAAAAAAAAAAAAaRR8zM7ODtyfO3dupeYV/PqGDRtGnCY4wBltGgAAAAAAAAAAAAA1MPjYtm3bQLaiSqVqDMiKeuutt9xfjR/Zrl27iNOsWbMmME2LFi0qvCwAAAAAAAAAAAAAaRZ83Hvvva1+/fouGKgA5E033WSbN29OeD5vv/22ffHFF24+st9++0WcbsqUKYH7nTp1qsSaAwAAAAAAAAAAAEir4GNOTo4NGTLEBR4VOJwxY4YNHz7cZs+eHdfrS0tL7fHHH7frr78+EMDMzMy0E088MeL0//3vfwP3d9lll6S9DwAAAAAAAAAAAAA7yrIU+8tf/uKyFgsLC93jadOm2QknnGAHHHCAHX744S5IqDKqTZo0sZKSEtuwYYPNmTPHxo0bZ++8844tW7YsELzU7ayzzrIePXrssJx58+bZxIkTA9mRgwcPTvVbBQAAAAAAAAAAAOqUlAcf8/Pz7eGHH7aLLrrIioqKXHBQQUZlKQZnKkaioKN4WY+HHnqoXXnllRGnHTlyZOA1TZs2tX333bcK3g0AAAAAAAAAAACAaim76tlrr73smWeesc6dOweyGEX3o93Em05/zznnHHvwwQctKyty/PTmm2+2b7/91t3GjBnjxpoEAAAAAAAAAAAAUIsyHz0DBw60d99915577jl75ZVXbNGiRb7TKwCpQKOyHc8//3zr37+/7/S5ublJXmMAAAAAAAAAAAAAaRl8lAYNGrhAom5Tp061SZMm2cyZM239+vW2efNml63YrFkza9++vQs27rHHHtaiRYvqXGUAAAAAAAAAAAAA6Rh8DNa3b193A5AeMizDyu2XkscAAAAAAAAAAABpO+YjgPRH4BEAAAAAAAAAACSK4COASistJ0MSAAAAAAAAAAAQfASQBLPWF7IdAQAAAAAAAABAeoz5uGLFCvvyyy/thx9+sCVLltiGDRtsy5Yt7n+fffbZDtOXlZVZSUmJu1+vXj3LykqLtwHUWVtLSqt7FQAAAAAAAAAAQBqo1qjdwoUL7f7773cBxtLS/wUvyn8t4ZiRkRHxdR988IFdffXV7n7Tpk1d4LJhw4YpWmsAAAAAAAAAAAAAaTXm4zvvvGMnnHCCffzxx4EsRgUddYsWdPQceeSRlp+f76bdtGmTmwcAAAAAAAAAAACAOhh8fP/99+0vf/lLoLSqKJDYvn1769OnTyDzMZrMzEwbNmxY4HGk0qwAUsf/GwsAAAAAAAAAAOqKlAcfly5datddd527rwxHjdl43nnn2ZgxY2z06NE2cuTIuOYzZMgQ91eByu+//z5mwBIAAAAAAAAAAABALRvzUWM8btu2zd1v0KCB/fvf/7Z99tkn8P9YJVc9/fr1c6/XvDZu3GgLFiywbt26Vdl6AwAAAAAAAAAAAEijzMfi4mL79NNPXYBRtxEjRoQEHhOh0qs9e/YMPJ47d24S1xRAIuLrMgAAAAAAAAAAAGq7lAYfJ0yY4AKQKpHauHFjO/PMMys1vzZt2gTur1q1KglrCAAAAAAAAAAAAKBGBB+XLVvm/irrceDAgVa/fv1KzS8nJydwf/PmzZVePwAVw4irAAAAAAAAAAAg5cHHdevWBe7n5eVVen4lJSWB+/XqpfStAAAAAAAAAAAAAAiT0oidSq16tmzZUun5rV27NnC/efPmlZ4fgIr5YfVGKysn/xEAAAAAAAAAgLoupcHHli1bBu4vWLCgUvMqKyuzadOmBR63bt26UvMDUHFrirbZp0v+1xkAAAAAAAAAAADUTSkNPvbp08f9LS8vt3nz5tnSpUsrPK+vv/7aCgsLAyVXd99996StJ4DE/bB6A5sNAAAAAAAAAIA6LqXBx27dulnHjh0Djx999NEKZz0+/PDD7n5GRob17dvXmjZtmrT1BAAAAAAAAAAAAJDmwUc55ZRTAtmPr732mr3xxhsJz+Ouu+6ySZMmBR6fddZZSV1HAAAAAAAAAAAAADUg+HjOOee48RmVsagA5PXXX2+33nqrFRQUxHzt3Llz7aKLLrJRo0a51+vWo0cPGzZsWErWHQAAAAAAAAAAAEB0WZZiDRs2tAcffNDOPfdc27ZtmwtAvvDCC/byyy/bHnvsYe3btw+Z/t5777V169bZ5MmTbc6cOe45vUaaNGni5qUgJAAAAAAAAAAAAIA6FnyU3Xff3e6//3676qqrbOvWre65kpISGzduXMh0CjI+8cQTgfviBRpzcnJc4FGZjwDSw+wNhdYrt0l1rwYAAAAAAAAAAKgrZVc9hxxyiBvvccCAAYHAoscrqRqc0ejd17S77LKLvfLKK7bffvulfL0BRPfa3BU2Y91mNhEAAAAAAAAAAHVUtWQ+erp27erKrX733Xf20ksvuczHaGM/NmrUyAYPHmynnXaaC1wCSE9vzV9l17bIqe7VAAAAAAAAAAAAdS346Nl7773dTRYsWGArVqywDRs2uFKsubm5lpeXZ7169bKsrLRYXQA+yi00kxkAAAAAAAAAANQdaRfNUzakbgAAAAAAAAAAAABqlmob8xEAAAAAAAAAAABA7ULwEQAAAAAAAAAAAEBSEHwEkHRl5Yz7CAAAAAAAAABAXUTwEUDSbdxWwlYFAAAAAAAAAKAOyqrOhW/atMnee+89Gz9+vM2YMcMKCgps8+bNVlpamtB8MjIybNq0aVW2ngASk/Hr3wmrNtjopWutYWY9O7ZrG+vWrDGbEgAAAAAAAACAWqxago9lZWX22GOP2aOPPmrFxcXuuXLKNAK1SlFJqX2+ZK2VWblt0f2la+18go8AAAAAAAAAANRqWdUReBwxYoR9/PHHgYCjMhd1IwAJ1BIZZtPXF7rAo2f11m3u77LCInt7/ioXkDygXQsbnN+8GlcUAAAAAAAAAADU6ODjc889Zx999JG77wUcdevcubN1797dcnJyrH79+qleLQBJ9NLs5bZT8yYR/zd6aYGt37Y9cH+XljmWU79aK0ADAAAAAAAAAIAkSWmLv8Zy/Ne//hUSdDz66KPt0ksvtW7duqVyVQBUoYLi7fbdyvUR/7d489bA/XIrt8lrNtl+7VrweQAAAAAAAAAAUAukNPj4448/2oYNGwJlVs866yz761//mspVAJBm/leYFQAAAAAAAAAA1HT1UrmwefPmub/KeGzSpIldeeWVqVw8gDSUUd0rAAAAAAAAAAAAambwcf36X8owKutxt912s4YNG6Zy8QAAAAAAAAAAAABqS/CxefPmEe8DAAAAAAAAAAAAqPlSGnxs165d4L7GfgSADOquAgAAAAAAAABQa6Q0+Dho0CDLyclxYz5OnTo1lYsGAAAAAAAAAAAAUJuCj9nZ2Xbssce6+wUFBfb555+ncvEA0hCJjwAAAAAAAAAA1B4pDT7KiBEjrFOnTi778fbbb7e1a9emehUAAAAAAAAAAAAA1Ibgo8quPvzww9a2bVtbtmyZnXHGGfbTTz+lejUApIkMch8BAAAAAAAAAKg1slK9wPHjx7u/11xzjd1yyy22cOFCO+2002zPPfe0Aw44wLp3725Nmza1evXqJTyeJAAAAAAAAAAAAIA6FHw866yzLCPjf6O86b5KsE6YMMHdKkLzmDZtWhLXEgAAAAAAAAAAAEDaBx89Cjh6QcjgYKSeBwAAAAAAAAAAAFDzVEvw0QswEmgE6pYf12zc8cn/9T0AAAAAAAAAAAA1XMqDj3feeWeqFwmgAnZvnWs/rN6Q1G33yaI1fBYAAAAAAAAAANRiKQ8+nnDCCaleJIAK6NCkof2wOrmbrsx2LKtcXFKW3IUAAAAAAAAAAIBqU6/6Fg0gnaWqGuo3K9fZ50vWpmhpAAAAAAAAAACgKhF8BBBRZkbqBmMct2q9Ldq0lU8CAAAAAAAAAIAajuAjgIi6NWuc0i3z/OxlfBIAAAAAAAAAANRwBB8BVHvmIwAAAAAAAAAAqB0IPgJABEWlpbZ66zYrKStLaPss2VxkY5cW2OwNhWxXAAAAAAAAAECdk1XdKwAAnpVbii07s541bZBlpeXlVlZu1jAzdX0kysrL7fuV6+3HNZtsw7bt7rnWjRrYmb3aW6OszLjW//lZy6zMys1Wmp3Yva3t3LxJCtYcAAAAAAAAAID0QPARQNp4asaSkMcZlmEHtGth+7VrYdtKy2zDthLLbZBlDaooIDm1YLP9d1lByHPKfvxh9Ua3DrF8tmTtL4HHX709f6Vds1v3KllXAEDtp3NfcWmZ5dTPtAzKoQMAAABIojVbtwXawQ5q39JaNWrA9gWQNAQfAUSUmQZDPpZbuX2xvMDmbdxiSwqL3HMtGta3M3q1t2YN4jt8bS8rszVbt1tuwyxrHCN78b2FqyI+r3WIJ/i4aPPWkMfK3kTVmrOh0FZt3eYyTPOyuUgGUHssKyyy1+ausMKSUuuZ28RO6p5v9QhAAgAAAEiC8vJye33eCiso/qXyl/5e0KcjnR5rqVVbi62opMw65WTzGSNlCD4CiCidMiy8wKOsK95u361cb0M7tQopd6owX9vGDUNet7Wk1P4za5mtKdrmAo+n92xn+WHTiIKbY5aGZjwi/f28dlMgYPzNivX2h106xR2UBoB0N3ppgQs8eh0t5m7cYr1yKeUNAACA9DVrfaHrmN29WWN3qws2by+xrIwMy45juJ50sq64JBB4lLVF2+zpGUvt8E6tXDtZ4/qZtlurZnSArCZFJaX2yeI1rsN9v5ZNba/83Aq31U5cvcE+XbzWJXmoY+spPdomfX2BSGilBVDj6KTpBR+/WFZgX69Y5+4r+624tNyN17hrq2auZKoCj7KlpNQ15P62V7uQeZWUldtb81e6snbpYub6Qnt3wSqXtTkwr5kd2blVWgWDI/WWU5mOn9ZuslbZDeyYrm2iBgEXbdrqLpy6NWtkTevHX0K3pKzMPly0xmZvKLQOTbLdMoIzVbWtvlq+zo7q0tqqmy7YV2wpto5Nsi23Yf3qXh0kSJ0ZMjMy6nS5maWbi2z5lmLr2rRRnd4O1W1xWDb99ys3EHys4XS+nLB6o2vM6ZLTqFINCHXtmKROZh2aNKzx22t98XZ7Z8EqW1uka9WmrrxZTX9P8dJ+//XyddakfqblNWxg369ab/Xr1bNjurZ2jWAA8OPqje7YoKFWjurcmt9SNZB+ryuTTsav2mBn9mpvnZs2CplG7TUfLFzjzomD2uTanm1yrSb7bMka9151ThvWpbX1bpFj6UrtKpPXbnLBxP4tm0as1rVya7E9N2tp4LGuWYI7/yN19Ptv6rrN7v6YZWtdxmKHnOwKzUtBTI86tqrKTvsmFZsXkAiCjwBqLAUOvcCjF7TzjFm6dofpF2zassNzyqKsSOBxeWGxC3jFU65AvZXGLiuwzdtLbe/85r4XC5rvG79erMvktRtdMEsXe42yMiMG9XQBOXvDFmuSlbnDhX0qLC0sdttR1MMxPDPVM61gs729YGXIcwoYayxPje85pGNe1G2j9zelYFOg8WrSmo07TLNgU2hDfbIbjBdvLrKS8nLr1rRR1M9cF3DPz1rmpmtQr56d07sD5WBrkM+XrLVxq37Zlw9o19L2j6Pcck2kY4p+fOjY95v2LUN6JM/fuMVenrPC9YhU791ze3ckAFmB44XOR1tLy6xvi5wqG6e4InS+UBODGkfSiRo+dE5vUC8j7QIx+jxnrC+0wu2l1rdljjsXV9SsDVtcA5V3LtM5fZeWOUn7bL9evt6mrdts+7ZtbgNbNavUMWLGukJr3aiB7dS8SYW3m64P6tfLiFh1oiLH5d1b57qe+DWZKjUs/bWih66XdB1UFxp+dL5RGen/NXL+cs1eWlrqxkwn+IhY9Ltr4upfrv93b9Usrc6tSI4Nxdvto8WrA9WO1Ln1uG75bN4aRp2og320eI2rUBTsi2XrAm0zny5Z436LtMz+pdOuOpDP3bDFVbXq2iz1bRuJUgBVgUfvOPXBotVpHXx8de6KQLvJ/I1b7YA4fu8Gd/5PlrLycje2vY7l3rASakvROqmNLZ52LW1vlRDNqZ+Zdr8fvGthtQupY3NFh874ZuX/2jtFQWH9fmjZsL7reJ/boOKd3RdtTjz4qN9DUws2WdMGWda7eZNKb3dVq1MSga4P92nbvFLvJ5x+W+o3jdY11hBcqFoEHwHUWEWlv5Sjq0yG05fLC+K+cNAJURcNajjyXte3ZVM7tmubqK9Tpt+ktZvcCdprbLxsQFdr+OsPZq2DGl108XVoxzx3P1LZ2admLHEXLcr4a96gvv2wZoM1q5/lTtDPzVzmeqfJoR3ybHB+8wptD7/3ruCf1lGNkOEXTl5DaqyL0/DAY3jA+N2Fq+zCXToFLmB0Maofns0b1nfZqcEUzE1lg9l9k+cHHvdpkWPHR/kh/Mnite4CU7aVlbn1PLE75SySSQ3sXyxfZyu3qPRIjrsl48eGsqO9Bm7Rd1ydBbLqxT/vjdtK3HzaNGqQ1qVplEXsZdW9OW+l/XlAF8v6NRj1wcLVLvAo2pfV8HNyHCVZ9LlMWLXRisvKbM/WuS6zpa4as6zAvv+1Q4Z68A/fuYPbPjru++2rOseoY8X20nKXEZVs09dttvcXrrbSsnI7uEPLpJ8rKmrN1m322rwV7nivQMyxXfMT+t7FKoGlsupFpWW2b4zOP8GNnzo3aRzhHrmN3fHmm187Oqlx6cK+nSr8/X477Fym82Kk4KN+LGt/8K4V4g3S/fBrxxw1fK3fVmIHtm9ZoW32zIyl7hwmR3Zu7apJBNP1wOQ1m9zxbmCrppZTf8eflMruUyBUDm6fZ3u3TXx/03VA8HH5h9UbrHFmPdcxJPy7tGlbib05f6UtKyx21yrKpku3ILvXqSyYqnL8bqf2cb1WVRX0PVZQWB0bEjn3qQPXR4vWmF6hxipdy1T02vvdBavd9a2qWBzdpU1c+6lK5UcbC13ffV1rVkfDoZarc7caPyvTsaC66FyuRjt9BvoNUJsrbrw+d6XN/zVYMW/DFjvj1++NPkNVa9D3Xd+NytCxrazcTKcgHe9VinBAy6YVzjJBYr799drJo3NIRYKPv5w7NrjrnUH5uWnX6Kxs/k3bS6xnbuPA9XdtEt6xW43/Hn1flcXldSr2fLW8wI7tlm8FRdvtmRlLAr+n9Tta14bBgQ8FK3Xe2KN1rg3IS/71cqKC2zOkohW1dPzR9aqO52r3qapAaXCH7ZnrN7vfrfHQtamGgGicVc865VQuKKxrNgVB1YalalH6rbm+uMRGzVxm6iapjumn9Wxr3XxK9qoDl5dw0LVpYzulR35afZ/0eSoQr+NYi4b17eTubZPWoVfXLbqp02Gkyl/6rN5f+Muy2zRqaCd3z498fRD5sizEjHWb7buVG6xp/Uw7rGOeG9Zq4/aSpHXWVlvEwl/bJbRvBrcHVoYSQJ6ZudQdK7IzM+3UHm05l1cjgo8AaiT9qFBWTmVECvRF89q8la40QevsBrY66AJaQUX1FtMFRSSjl651P4g9upBWIEvz0cWHGpGCT7zKlIlGjTYKwuli1GvA0cW7Tqiez5eujdqgrIt9lTzVBa0CeokEKrzGsp2b59iJ3f/3I3Dhpq0h768y9D6UHaqeSbqYemH2MvdcvL2fVL5FWaP7tW3hm2WhfUc9rLSceBqQgwOPos/s4PYtd7iA0wXm8i3/G5800o+R2kBBFH1O6mEYfoE/ec1G971S4ECB8s452TZrvX6kZCbcc1U/jvQjpH2ThiG929WY4AV21GNWpX7bNYn+eauB9Mc1G90+v19bBRPrhfw4dAHuBlkuABKpkTWnXlbcJYbemrfSfcdVTvH0Xu0qFYDU91WNMMo+0vtTo2JFevnrPSojuUWD+oEfPMHlPBVg0H6qjhTi/ZgIfl/xUFDLCzRMLdjsfpjq2KDPXh00Fm0qcv/Xd1OlpCvayKsfEmuKtrve0Zq3AkVvL1jlekkrCKLtVN09X739U/Sj+h+T5gWOn8d3axPYL7TuOmZr+6sMtY5LXmPLpLAAhffeK7M/fbhoteshLFrugFZN3Y+xRIJS+k6qkSKZGSffrFwfOI9pW6iTTni2nYIuXgO7AqeRgl3Rzl86d3vHguBAeyQKpj01Y2mgc5NKvnmBR1m/bbvbjmrwUo98zVMdZ/Q9U4BNPbU/CmQVt3BlcnVfgTg1lv+Sd+pPvb5fn7fSbe9ESq97gUeP1lsdAWZtKLS8hvXjrozw9Yr1gcCj6P2GBx91beP18te+enHfziHHOzU0escDr0xUpOCjzttq6I90DaUGSI0ZFe6rFetcj+/wzE69fy+jUI1pxXN/qU6h0n3ad7Tth3RsFdi3dC5buKnIzauyAYvK0HczHjpePDdzaeD6T50UdmsdX3arXvPZ4rWB7/+ni9e4xtxEz1E6Rulz975T+r52a7bZjQUViXeN2LNZY9egWlV0Xffl8nXuukHfzXivN7Tt31242l3Lq1qFOpapw0E60bVlxq9jcqmz3/ayctepoGNOtmtcfGXOisD3Vcea3/aKL5Dt0bXPum3brXNOo4Q6OySb9tHN20rc2GKROg3oeOgFHkUNlfo+aHus+rUDphzcIc91HKsIHc/VgWFriY4W//te/rRmk/2xX2f3uyH4fKzfYDoX7ZTbOCnXHdoG238NPFR2ftq3fy5Qh6NfMtA0rldNyC7ephNChPNyosFD/V5WkET097w+HS1dqJOuV/pQQYFze3dIqw6LOp7qWtTb5joGbU9iZQp1yv5o0S/ZrcG0DPlieUHgWtjrtHXNbt1DOod67Sfav/VbM952Db0Xiba99X+1Qeg3SqTjob5XqmCxZXup9WnZJKFr6Fh0TFEwTp3O1U50Ss+2Sc0A8+g3fLh4O+Orbca7zop0rFUFL1Ul02+EQzq0dJ+pNrl+s+n6Qb8VdU2g6w+1Z3md5/WeNXyPrim862Qdg9Uh9pL+XSKui34rBFc6U5uAqpRoTMTgaj6qwNGjWWPftoKK0PXrhl/bKbzfFbqWUDuI3psXEPSuhfU7R9evXgd2nbfUIcyrQhQcYE+E2ugiBR91vvSWrWWNX73BDuvYqsLnRVnuKrgU7rDvJBp81PdIx3pVRtF5zws8ettJ+0hwNqb7rbVglbuOPKRDXlzXvmoXeGL64sBj/a77dMlaV5UM1YPgI4AaSeUGfpfgD+xwapCPl9fQEhx49KgR59SeoWNJeqIF5jSf8HmFN/pHE9xzPDjwGMt7C1e7Xob1LMMO79xqh8bEHbIa1m5yDQHBvfTVoLdpW57LalJDT3CjbDKU/9ogqPl6700Xd/HSRavKo/6pf5fADwtllyojRE0zKpmnRkg1Ymg8O/0Y92uIVuNnJAp+hAcfIzWSBl/86v+6kIqUqacAj8s4KzfbuUUT98NADaZqvG5YL/FGCK8xM/x1ahAevUTZmWYHtm+RUJkN/WjQhdzHi9e4sVQV9DutZ7tAKWA1auqiTn+3lZm9PEeXqP+jca32adsiMOaZPqdeuY0jbg9tL5Vn049PZR+dvXOHwI/A8KxXNcadtXPkC0l9zi/OXh74IaP9ap/85rZ/u5auMUMlchVMUBbxAREyhNRjWplr+gGknrXR9hW9J62vRxfRj0xZ5HoHqtFDjQt7tm4W9XPUPj59XaG1yq4faBxSo7j3XvXjTAFSbb9wupj+ZNEa9+NKjeoHdWgZ6C2rBoSnZywJfJf0g2yXCNku6pTQ1yonONDgfpD9enjTdn4paF/Q9m6dXT/i9o5FDQKjZi11f0UBi+BjoBrFezfPcY2y6VhyVMfP+RubWpemjdy+GFw2XD3gg0U6tuucoeOEGp4TzSzVeSO8J/Yv38EmgR7IaqxTkE+lopTtFpx9qB9+z89eFnis8ln6bsai+WrZfg1DXmWA4MCW9mXt2+NXbnDBKQXivcZg7VM69gR///Qfb223lKjxNsN9X71zt6iBftq6Qvdd9nrh6z2o8V3XAxpP+OeCTSFVFZRBGE4NJLop0K0grkel4hTE0nFS3pm/yi4b0MU1kgWvR7g35q10+4YocKLX69glOv8qCzbWsVrH50genbooEJhQhlqkDAE1eGnfUKakAsv6rGPxAo+i7+M9k+a7YGx2Zj1X+UDZh7FMWLXBNdLo01MDlhqyPEs2F9mLs38pYR6JPpfw4GP49YgaosLL7SuQqkwT7VPqDa1trWuiY7u1qXAmYLIpgKbrK339ftOuZSDgoQzc4Os/7W/6jqrhK9Z3cdWWbbY1aL8uLCm1L5etsz3aNIs7kK9OPDrXhAfQ1YgcKfioax5vzK8vlxXELEEX/B32q0Ci81v4uVQZ+mrQF1XquLR/57jel67RveOPvie6vrm4WSd3Tow21EG8dP7VZxjpukHXScoe9b73Cvx5jd49mzdxxxZ9tmOXFrhya3ov3jFBtF11natrx+COAtGGH9C8x63cYGuLt7nxvbyOCHr9G3NXBj5TfQf0/Q0P9OiYq/fi19CuRmdtv5LyXxoHtR/rsaix1e/7pfemTAqvw6GyXcL36UiZRF4jeDA1Rut4omO8V7YxuOOFArZ6L5GCD7rm0nsNp+2j8/URnVsHtqeXTSE67u/ftkWgOoyuNXSM8Trp6Lug3yK6vtf1YCRqHH5y+pLAYwVbEsl80jrpuKnPTh3NdH73Ao+ioIYyrNs1znbnI+3fuhbxXqtGcJ3X9F7UyTRap9p4uN8Di9e6zgfKjta1cLzZSAURfmc/NX2JazDW90DnKw3XoXOMPttIAQUt3ws8etfQ2v6VKf8dvi9mRvluJzrmmj53HbO8wIP2T2UAetf00X47qDPKyq3b3PE/3mO49m39rgkOoofTZ/ZLZ+hS951V0EbXZN73QseAto0buOsJ7/hY9uuxWZ3C/X6v6vul332RAo/B5mwIPW+HZ8zrXBTsX1MXuWOmxoz0vn86x6kUp66Vve2jjoH/XVrgfk/qnB885IR3zNZvVwXC1CFQbQTh+4zOzd61u6oyXLDLL5UwvKBmPLR/Tlqj8RbNdTDzrrXVWUDL9q73dZ2VSLDIDUe0fJ3b9/W90HFY17aqOqDf7Pu2bVHpqiLBx1zvWBv8OWn7edcawdf2anNRpqTXodU7XwfTfhYuvG1Mx1J10NNvZQUyw+m6xgs+alneb3O1/egYomNGMizYuNVVbNFnqUCxKtxo0+o46p2T9T1Xp6Zg+j3gBR8/dr/df9me6jx/5cBuoZ2tV6537SUtGsZ/HaLznjKDlamv34vBtD/F0zFH+5Eyi9X5Wd+VRNrhgq/XVClObYlq21Bw1PseapvpGKP9VPulOquHU5uIrvObZGXZ0V1au98K3r6g7aqKMbE6SwX/xvaEJwkgtQg+AqiR1GD0YlhwIx5qZExWuQOPTp7JDsIlm36Ae+VN9CNaDXDtGjeM+kNM/9eP0EgemrIw5vJemr3c9Wbql9fULUc/ZuLxcBzzjkU/kPQDXhkOCpLpAsoTnPGlRprwHnLh1Esskki/rd6eHzq+hUcZL94FuDIz3lu4ygV+dcmkH2kKLgbX8g8ez1IX52oAOKm7GmLqu8CAxvb0K6vl/bjS56zG3L3a5AZ+DGpdNA9ZM2+bXdqvc+B/akRRY0X7xg13+I7oIluNXcGNXNq2XywrsGG/XjRqfAYvqyISNQ6qwSD4R6Ma3vQDLzhLQQHf4GCVLqS1Lw5qkxtxvl4g6n8NAgo6ZAQGaA9vKFUQQ8GJzHoZLhDm5rG9JKSRxvPWglXuR47owlc/nnTxrMyr4Mw9Zd6FUzDJ6ylo9svnr/egxiEFebxgk36YBG/X3r9mxymbKnz7RQo+6seLF/jTD+HOTbMDASXtS8FBLB0DwssceWVr9IMkWjaiArzazvoRXJnGWI96fgYHH/VDR/ul5u03f312wZ93pACdgpMKTIU3KujHtxo81CB4WKc8F+QJpsYKNZT+vHZzIBiiIF+8vdHV8KAGJB33Yr13/RjzGp0TpWCUGmr1Q27iqo1u/1ZASQ1C+v4lktWuxjttezUu6VjjNXpoH1FDmBq+vB7vwYFH0TE2uCd6JGoEUQ92GdymuSstHi8F0x6esijiMUW9nb0OFlo3TaeAofYdBZZ1zFADthqOIp0fgtdLDRheYFBBqHgyEz3BgUePF3gUfa+VeRkeXA0XvC9EapBRqaMTu/8v+KjPX40LaiTW/qltpcbZSIKPLTrGhQcf9SM+eJk6v0TKPPGWq4YVlSEMp0YnrxFF1w86xvrRfhV8btYxSPuIF1TXukYLPFaGPn8FSdYWbw98VvrMFSgODo5oOp0P1WgST5DdjzrT6Liv81ykMmX63qlBTdkbahzSMcKzeuv2QC/tSIEWXU+owXe/di1cMLdJ/SzXoUfBLGWfa3uqgTzSYUzXHcpaVUeCaMd+V9GhsNj936/BWMcRNfQp+KxsUx1bvMCj+395ecRzT7DyoPO4GoFVXUEdvXScz2/cwN5fsNp17FHw5OQe+TZ7/RZbVbTNlZ8N3od1/B7580I7a6cOUTuieNTYG0yNbMo61nvRtYSy9v2Cpvoe6nPRd9EL1mjfUdlizUvndzWshWdyq0rCnKDgSHBnSJUW7pKT7YJU3rVhcODRW4auYSNlk+qaReNAZQd9psHDRegcd3G/Tu4YqUyC4GOel02kxlEdV/T/4CDOcV3zI5aJ1vH4mZn/OwaFdz7TNY+CJJGyeLS+z85cGnis7aZS2fGUe/erWBIcrNQ1tD4Db/xYXXcqK1yZi9pfhnZs5QKU3jkwEl276rh2VJdW7no3+PpDxxJd7+kYog5e2m46N6kyjs5L+p561zAqJ6dreHXM8cb91XE8vBrPC7OWRexYp9epo1/h9hKXTa7rPW3/F2YvD/mNE4muF9XJJPi8eu1u3V1lCq+jqYJh365Y7zprajt5Y/B510naBvpu+l0bTSnYHJjfj2u2u449/X899ygIqGvubk0buWOTvkO6bvPrTKvrZX2nw+mcfeqvZRkVlNT2VdaqjkHh9Hw8dC7Tbw4d97RcBXPVeTI7q547zukY7TK76tVzv3/0XVXAQNtw7PIC69CkoRsexes8qGs1XWPp95/2m44RgqX6LiurUJ2Eg7/rOi7ot5+uZzU/HeNzG2a5TpT6Pec5r3fHiL/n9V3S738dS/R7R8EIvb+98pu7wFQkOs57wSMdD4IrNImutbS+3/wajNcyXpu70u03Cuwc0am1C05Gotf4tZf4VVIaNfOXMe6iBeN0PaabSm+GdzrStaDKenq/jfX+1HH8wr6dQ6ZT1RbvGKDPXpnVPXKbuMoROsfqt2VwQEPXQjq2asw7/UaLROcy7Zc6l3ZrqgoApW5ab39X+4g3PEv4MDba17TvBQe5tQ/qWKQgsrK/vACM2rceD8ryUsBd03rn3ZlW6L6zrmS9VY3Zbqz7yB3htM9E6xwTi1eSXaU/vd/V4eWZPdrPdT2pz0TXdYF5WLnrjKvjvjpDaH7frlgX+Nx0LPltr3aBjio6zurzjnRtpI41L875328iBYr13VVHweBrbn3Gfp04ws83Ogd4GYz6HeP9xog3+Kfvqt6z3++Y4A4uftf5XpA43iGuvHHitc/Wz8xwQUWd37z11zllyK9DMilRwLuuUFvSexHaULzhFlbbth3aBvX+1NalbNFI1I6kZYcfuzy6tqmJJfZrg4zyeGu9ALXI2rVr7aKLLnL3H330UcvLi79BrC6584e5VtuofvwFu3R0mWUqHVEbXbd7jx0uJCL9aFPj1Jk7tXc/RHTRqh9FurDv06KJ3fXjL2UCa7tGmZmugUMNBwrgqKekftR5P5LVYy5S2UkFNvXjWxeV+mGoHwZV+X1Rg6U+o+CLejUKnN6znfv89KNWjX4KPIQ3EKph8pzeHd0PEF2Qh1NATd8HXQB6FBzVj2WVH1MD3L+mLAoE6sKp8f64rm3cj4rgxsZ4qTFK5axEASj9OAinhr3z+3TaIfvK80sv9+3us9J2UUOTen2+PnfFDuVBkkHbzCttFl7Ww4+yl3TRm0jGcrA/9u1speUKRGqcWGWvtoy4bO2fKkGqbKR4aRsqIBerA4AaG1wmcXm5yzxSL+bwhsZ4XD6gq7v4V2PWqFnLAkFxrYN+wGVl1HON5xpDUj0f1c51dwLHJW+MXP2wV+NW+H5w1a7dQjIgFZz0GggjuWJA15DGXFED3nsLVtuMCgYSq4KO4SqBqR93rqErJ9saZSkDcEtIx4ZgCq4Ej8fjCQ+Mh1OjpQKpCmApyKGApRrqFDBQwCU8g0TjgqhRXOcYHbf0I1UNxNEaESpLZW7Dg7wq/xjc6F8T6H0M7ZTn9tEf12wKaYhQAPib5etCAlZ+dKzVeLn5jRq4ziCRjqfRjtNqiIm0n8TrnJ07Bholg8u5e5Q9pwCBMgLiCc4HB0LUKBsczPSjoJTKuIX3qPeunYLPM7pmVBUC7csKnuiYEi3LRPuyGvvUYKxGPpVi1HkxuDONshzC37dH351IWVc69+kcqCzWipw79s1v4RqDoo2VrYZoZf8rkKXgdPA5QA2bFR27KlEKeCn4reBfrAavWMcmjxqcleWjxkbvc1MgWOdtXd+osS+efU3nk8v6d3HnAAVPVFrZ65gUvF8pc/j+yQtCGv90TXLpr2XjvAyzaPtAMHVWCs4wjvS56tgeqROEd93YqUm27duu+Q7XAgrYqgLEv6ctivhafReCG2bD/xdOFSKUiRsPlXg/olMrt42UjRkpO0EO79TKdg/qJKRjz2PT4rvWCqdzvSpoRLtW03lT11ep+O2ja2adh+LZXrre0jWMGsL13VSDubJhIlXhqSgFrtSBLtKxXfu9xlvUuV2ZQ14AW8E1lQrVeUfXGio53qpRfdeZUoHVSL+HlNWjTgLePqVtflHfzu43S7TOrvHQsUtlGdUBxyvhGI0CY1quGr4VJFSgWNtUxzgFo+Zv2uoCM+GVKKpapIBZsPxGDWO+N+97qYzIR6ZG/l4HU4azAtcKGi7aXOSCwTpWBQewYtEygzvYejTfeIdsCHf1rt0DQxVEog6aybrmvnrXbu6aSkOG6NgeK3iv78pTEX6jRqOOkPH8PlKg5vw+HSMefxRs1blAvzd0Tf9L567SQBvG5QO7uvu6pgmuQOP3mWm8UVUQSwatm67rdf2m3wNav2RLtGNgLDoX6BogvLKJjmsX9e3kOnd4HZrUsVOddYKz/hXACy717Se8Qo+oo51K0kb6vL3vcbyfp0ftQcHXmonQUDG6HtZv5XiuqyrKe2/3Tppf6eXo3KSOc/pdoutBDcmjjmDa1vFcJyvxYFiX1tU+TEtdQ/ARdRLBx7obfKwLdFGlbEOd1tVzUz1KE7lYUYNFrAvw2kyNjeoZqEbDeLeDLkrj7R1Wk0S6aI5GP1or+oNdF6TRguSe4LFO4/k81HD19gIv8zD51JilhosGmRm+DYSI3iisXv360R+ptGVVU4BWJZoSOd9d2q+La8hWlpLK8unHazoFHlNNQYvg8S3jpc4NygqJlRGI2kedWnR+Dc50qgx9J9VQlEgmcfjY3Z4TuuW7hgiVv4pGZcvUyKYShmqy0H11llAjUXAmSrKpxF2k7HzERw3W+7Zt7rLDvGu1ily3XTGwq708Z8UOgUdRdpGuUyKV9VLnnkidvyqrMtddsa7Jop0T1WFPQXY3vvev41pVpDOnOkDGGv7iL7t1d4F8NfSqg5kyKKvymiDeDgx1jbKN4ylXpyCBglrxBqKTZcTAbi7bNVFeOWAFpYOrFiA+KuGsUvV16XeLX3Z0VVGnzEhVMbyAujJ94w0a79dWnVYyd8iyhAUCupGyN4d1aeOqSmzaXpqUKl2JLr82cB1oLCOuzlepcGav9iEl2VH1CD6iTiL4GB+CjwBSIZHAImoP9aZXr/nqoB9A1+7e3WWiqFxWcAmdaNTQqoBlVTQiA6iZogUzgZpKlQke+GlBda8GEBe/jHIAACKJNR41kosxHwEAQLUi8Fg3VVfg0Rv/Q+XzEimhpHHLCDwCCEbgEbUNgUfUJAQeAQCJUhl8jQ9K+dXU+N9gNwAAAEAdkUjgEQAAAAAA1Gwqvb81ReOag+AjAAAAAAAAAAAAajlG/Ukdyq4iYPny5TZlyhRbuXKlbd261fLz861r167Wv3//pKUiFxcX28SJE23ZsmVWUFBgzZs3d8vZY489LCeHessAAAAAAAAAAAA1GcHHNHTttdfam2++WaHX9urVy957772EXjNu3Dh7+OGH3d+ysh3Tjjt27Ginn366nXfeeZaZmVmh9Vq3bp098MAD9v7779umTZt2+H+jRo3ssMMOsxEjRlj79u0rtAwAAAAAAAAAAABUL8Z8rMPKy8vt/vvvt7PPPtu+++67iIFHWbJkid1zzz125plnuqzIRI0fP96OPfZYe+mllyIGHkWZlu+++64dd9xx9tlnnyW8DAAAAAAAAAAAAFQ/go91mLIdH3300ZCgY4sWLWz//fe3oUOHWs+ePUOm//HHH+3CCy90gcJ4zZo1y71m1apVgeeUPbnbbrvZkUceaYMGDbIGDRoE/rdx40a7/PLLXRYmAAAAAAAAAABAMpRbORsyRSi7WgN8/vnncU9bv379uKb74osv7KGHHgo81piOCvqde+651rBhw8DzCgJeddVVgYzH6dOn20033WR33313zGUUFRXZH//4RyssLAw8N3jwYLvjjjusU6dOgedWr15tt956q3388cfu8fbt2+1Pf/qTffDBB5aXlxfnOwcAAAAAAAAAAEB1I/OxBtCYi/He8vPz4yq3qjKq+uu57rrr7KKLLgoJPHrBwueff96aNm0aeO6dd96xGTNmxFzOCy+8YIsXLw483nXXXe2JJ54ICTxK69at3XiQGvPRs379envsscdiLgMAAAAAAAAAAADpg+BjHfTJJ5/YzJkzQ4KCw4cPjzq9goVXXHFF4LGCliNHjvRdRnFxsT3++OOBx/Xq1XPZjeHBzeD/33zzzZaTkxN47sUXX3RZkag+w7q0YfMDAAAAAAAAAGq89cUl1b0KdQbBxzrovffeC3l89tlnu7Krfk4++WRr1qxZ4PHYsWNt06ZNUaf/+uuvraCgIPB4v/32s5122sl3GcqAPProo0MCmAqUovr0a5ljh3bIs+7NGluDevWsXeNsPg4AAAAAAAAAQI3TvCEjEaYKwcc6Ztu2bfbVV18FHjdu3Dik3Gk0ylgcMmRI4LHGZdS4kdGMGTMm5PGxxx4b1/qFTzd69Oi4XoeqoaD04PzmdlrPdnblrt3snN4d7Opdu7O5AQAAAAAAAAA1SpOszOpehTqD4GMdM2nSJNuyZUvgcf/+/a1BgwZxvXbPPffcIbsxmvD/hb82mgEDBoSsz7hx41ygE+mjnn+SLAAAAAAAAAAAqMMIPtYxs2fP3iHYF6+BAweGPJ4zZ07E6RTcXLZsWeBxq1atrH379nEtQ4HHPn36hGRqLlq0KO51RNWrl5FhHZtQfhUAAAAAAAAAUHPEGn4OyUOB2xrgtttusx9++MEF9DZv3mw5OTnWsmVL69evn+211152xBFHWJMmTeKa1/z580Med+rUKe716Nixo++8PAsWLLDy8vIKLcNbzuTJk0OW06NHj4Tmgap1XLd8e3jKQjYzAAAAAAAAAAAIQeZjDTBq1CibOnWqrVu3zpUg1d+5c+fa22+/bX/961/tkEMOsccee8zKyspizmvx4sUhj9u2bRv3emjcxxYtWgQeb9y40a1LuPBMxUSWIfn5+b7zQ/Vr1iDLmtWn7wLw/+3dB3gcxf3/8XGT5SLbcpEtdxtXbGxTjAOYXoLBECB00yGUOJDQfoE/JQQIJbQQSgIJxbQkNBOaaaHENIMhuIB77733ov/zGXmP3dXe3a50dzrdvV/Po8c+6cre3c7szHxnvgMAAAAAAAAA8CL4mANWr15t7rvvPnPhhReaNWvWJLzvunXrPLe1gjIK//21EjPdr+F/PmSHs3qES6ULAAAAAAAAAADyB0uXsli3bt3MIYccYvr06WM6depk061u2rTJpl8dM2aMGTlypCfY+Pnnn5vLL7/cPPnkk6Zu3bpx92P0r2aMorDQu9ffhg0b0v4a/ucLY/HixQn/HrRiE9E0rV/PHN+5xLw1Z5nZ4UqzG0bDunXsY/85fVHS+17Qq73578KVZvraiufBIW2bm48Xrqzw+zO6tTX/mP7jvqMAAAAAAAAAgPw1rDuLaTKJ4GMWOvDAA81ZZ51l93QM0qtXL5tqVYHG2267zbz22muxvyko+eijj5orrrgi8LEKXlYlMFhQUJDw+YJ+539MMv5jCnqNZA4++OCEf69Xr57p3bt35OeFV5/mRfZn/bbtZtHGLaa4oJ5Nyfrp4lVmxaatZkCrJqZ70/L9SJds3GLem7fcbC8rM4e2bWE6FXmDzEGO6djKtG5YXzsBB/69f8smFYKPjevVNZ2bNEj7V3VOj3bmk4UrzZz10c9P5PeK4QZ16pgWhfXM+BXrzNtzl1X3IQEAAAAAslDnooZm9rroE/JTbc+WTcz/lq+t7sMAgCo5ebc2pmNR+seM8SPSrmahY489Nm7g0U0rIe+++25z2mmneX7/9NNPh17ZVytOUCfs/ctCrHjLxGugeingpyBjywYFpqBObXNYuxbmlG6lscCjKIh4ds925vxe7W1wMMx5ocCmFNYJX1WVNChIuOLSr32jQlPasND8pHWzSAGkdo0LTbtGyQOoyWbbXL/XbiYTrtuzq+ld3Dgjr1WTHd2xVVqfv0PjBrac6PxX8PzKfp3T+nrIfr/s29Ec3q5F0vud17O9nbhR01zUu4OpV5vmZpDm9euZmqBOiOt1kW8f6FqmlmkUcM2NZ//WP+4pHlXXJg0r9Ti95vC+nZLe72edvXuR57OjO7QyR7ZvGbvdrKD6z+G6EfsZuWJQ62amIEfq1mTtU01GVPvJ74Qu1V82L+zd3nRO8QBWcf16pmVhtMm7iVS1v5Itrh3QxfabNGiYavuWhO8HoubTtb+kQbRJ+NVF4yEndimxffnqUr9ObZuNSv3kK/bobC7ZvWOVt+ApbVjfDCxpWuH3amM0zYK2RWU4Y1ep0KlxA3P5Hp3MKWmo7/KR2gupvlYn67f/OoPjPEFlKSxNjM90+zWdNMaqOipRP3S3SvYdUXmsfMwBN954o/n000/NggULYqlQ3377bTNs2LAK923QwFvhbt68OdJrbdmyxXO7YcOGSV/D/5hk/McU9BrJfPLJJwn/ruDszTffHPl5kfoLz5glq+P+vW7t8gGlA9oUm4kro+39eVznEvPG7KWeBtwxnVqZx76fZ3aa8oD2UR1amr1b/Xih/tJ3LD2bNbKBVR3H+OXrTKsGBea0bm1M3V2DPY0Lgi9o/Vs0MTPWbrQrQhPRc0ehQaZ9SpqaiSvWmbVJnttPwS41ug5t29w8+v3cSI/NB/u3KTYHlhabrTt3mncyuBqxMMngvAI3m3fsMM3rF5jHf5hn/59pasApcD9q7jIzf0P4a0bPZo3NlNXr7Xnbr0WRmbNuk1m2eWvKjkvPq+/Lr1uThoFpmitTP2nuS/emDc3z01Kbxln1juofx4CWTcxni1cn/H5LG9U3zQvrmXnrN0V6f/u1bma+SFDPVoaOfc3W7Qnrb4fqzbN7tDVjl62xq32zmY512abUnaPJdGvayBzevoW589sZce9T29SKXbPcQcuVW7aZTNGA0wfzV5hZCWbd/2L3DubTRavMzLUb7WSePVoU2clj/5i2yGYIUDDypx1bmtYNCsy/Zy01q7d6j//gds1t5oR35kWvf3/etY1Zu3W7eeyH5Ne21g3qmwt6t4/0/L2KG5lVW5qbaWs2mLaNCs1h7ZrH2gFbduw094+blbAe+d+ytaZRvTq2nC/duMV8v6rifulB1AaZsrri9gZBKyLWbN1mVlXxnBjaqcR8t3xt3Hr+0HYtzJ6tmtj/6/vVey+qV8d8u3ytzWxRVX2KG4f+bESTGlS3qNz+efwcsykD10cFhZJ9zrqPgvHbdu40fZsX2eB9lPNa38Obc35svyaq27X9wEszFttyVxVDOrayWyik4nuMp86udugrM71bY2hwVe1hBcg+mL/cMwBdr3Yt+3lqdrqCcc9MLe/rOno1+3GSo39ioerWEVO89w87CKdzqVezxuboji3tdS7RNVT312TL2eu8mVAU2FBgcsO2HebPE2YHHmMze67UsZ9Np6IGdmLj9p1lpqB2Ldtu12t/uGCFqQpdL4b1aGuWb9pqP2fVRYmuOdU1CevRicnrb6fe1eet65Ky31T13Jer+3ex19Svlsb/ngvr1DHD+3Y067ZtN9NWbzQfLVxR4e9h2ugKcn67bI3NBJTsfomOJyp99y3q10uasadJvbq2nL42a0nkvmbQKjmdx3q/ySigpXNz3PK15p25yyu0e8JMYFi8cUuo66DqFLU3VD5l1tqNobaCidemOLVbG9tP0nGPW7HWtnfUTti9eWPz0YKV9lqgsQzdR2MTUbImHdG+ZZWCCmH7efH6OeqvdG3SwNZn+sxE35N+mpt6dmKIP4uPrmFLN21Nev7WrlXLvr9JqzZ4xkzU5pL3XdeDKNTnDOpvxPt90Lm4vWynmbhivfnvoorb+ySa0NalSQPzfcQxq3jtOgUdNQbVrWld88s+HQPHb1R3rty8zcxcuylyfXF6t9LI572uW1pooMCNxqQ2bd9h/jS+4vUtnlaFBSkdCwiitsIZ3UvttffDBStt20blUe131RMqj+/PW2HLalW/owZ1a5tJcdqt6rc7Ex0emTin0q+j+njbzjIzd90mm2lueZzPT2Xp66VrAicNqm0V75rjLIRYummLeXvOcrNoY7T4QGUkWiwSleobfdfKSKfrhsb0nBWNQzuXmH8FnOMHlTa39Q8yi+BjDlBaU6Vp1SpI9/6PYYKPUQOD/vs3alSx0+cPFlb1NSoTfGzTpk3StKuofoNKmplxy9cl7axpwH334sbmB9/FPdElo0fTRvZn6poNdkDj+C4ltuN1ds+2tpGri17f5t6Z1lpV5HQmNePu+M6tYwFQreYMasR+vGBlhQCIHqPVmusT9H86Nm5g31fYwaxTdys1uzUtLwsKPlZlr86wtCpPDUQN8I2YvCBpR/DsHu1My8J6dnakvqtVW7ab0Qka7fp+2jcurPLAimhm04btlRt4VONeQQBn8ED/n76mfMBXM+3dMztfnbnEBtTctHI2HQ01zfB2BlrkxC6t7UCLBuPm+jquOsfVqdXgeFWoUe5vRHdoXGgHD7RyWZINWikQoIBd+ermH1clzF+/2Tw/dWGF80jniwaxE1HDWUVR33Gf5o1toEErbp6YNL/CgLNWXSc6xhaFBWaFr+GuQawnJ8/3/K5f8yK7StXpmI+ctcRs3L7D7NWqaaiBlETBpP3beGf8qSN3Qa92ZtyKdeazxasqrLrSAIbzWen9ydYdO+379wdx/IMGh7RrYQaXNredrZ1lxj7/vPWbbScjDJV/Pdb/vPo+wwQfnZXvx3YqMT2aNTIvz1ic8Pzbo3mRad2wwH7Hr8xcknQSR2VosCRoILtn00b2OqDOnTrWT0/xnhPpoqDOR756UPXQCV1K7CDWE5PmxYKNqqMu7N3BbN+506zftsNMXr0hVs9qcCje96oghQa3Ew3G+4Oa6lgPaFleDk7vXn7e6Zzzv4bKhM5NDfYfbn68VqoO0ON0vqmD7qwyUJ2qzqAzAKHgg+i1CurUskHMJgX1QqUYUwdS11y9fjI2AOoK+jtUzyuoH0T3Vwf1gNJi++On1x3cptimm/fTd6i2g7/9cPyu1Vr+z1KByv1bN7OvpzpBFHx81RescWgFQteiBrHr+tOT59tzN4wDS5vbVQeqPzdv32kHzFQ3KqjopkEbTRzRNdamwXe9b+czV32gQKkGVZ0AjJ5b73viyvWhBnj02ge1bV4h+BhUNuS3e3a1bUAni8bPupSY/yxYYb/jRHWbyr7aXPqMn/WVf+d51abTNemrJWvsII+CQXqMrj2qD4Me5x/QU1DJoeBzUPtKAf0JAYOU+g6SBR81e7/BrslLp+7WxpZbTVDyC7q2+cuP+7zW+3srROAzKu3F7myJcO2ArraNpeNXneRuUy/asNnWF2rvqj5wt4HaNqpv9mrZxAa75dye7ez37267iwIKTl2j685XS1bbQXOVF7WRkgVYh+/R0X7nzuera6iCXa/OCr4e6RTs37LI1lvu9shRHcrLvQbpr+rfxf5d5bN3cSPTv4X3vbnVqfNj72bfkqY2uOYPbLrbW+5grgaGt+zcGZtEo6DmKbuV2n/d5dfdzk1mn1ZNzTfL1pqyOH2AZH0YTRTTALnOdw2e+gNaKrPqc6nOCFt/id7Pad1K7Xf1+eJVdnuPOes3e9orCtoEDca66RxRfZtohb8CR6qfpEWdAtOiTYFZsWVrLJCh1dd6j5oom+z8stfJ9i1sG2rhhs22XOha4KdjVz3krz/VhlSZVYBaExhVF7wwbaH9DnSeq0yrX6X2qrICqO7WJ3JQabEprFvbPDCuYqBAWYlUz+4oM7FB+uG7rssydfWGCpMGgupWvf5rs5bafr0m/R7ZoaX9XBvXrRML4tiJtK2ams+XrPLUmSonoqwwKv+69oxdujZ0QEUTSHUd+XLx6lh9oLED1fX+Ccya2OnWpUlDW0e9Omtx0v6IIyhzkSY5a/KVirBzXdZkaLcze7S15VMD5To2jUUk+2wrE0ALQ+VH54/KUBCVycGl5UHTePR9qQ+qsqvzq0OjQnteqqyH/e7UzlL7QQEWvZ7Ga9T3DGqrndy1tRm9aJWd0BjkpK5tbHtk7rrNFfpHKusqq841JN5nUn4u1rHlTIFQfd5jlqwJrAN1PdM5ozauvk+977DUjw3qw2kils4bd5YwnU+aEDPZNwahulM/OofVjgsKtMSjx+g8Vt2ZqJ3gn3zqvpboOpkoeO2n9v/YZWttv0XnjT5vXVfV9lA5VbmoioPbNrcTs/TZ6TxSENJPr6Wyqh9N3JMVm7fF/exUpiesWBcbZ1Id9qs9OsXawBrzeN214EEUAHM4QfuoVIeqvnCXffX9XpqxpEL/yLl+KSuAu5+t67MmDaqtpWtWoj642k3n9Sof61mzZZt5bupCe73WNUTXI/9WV4kk6lepH6tz+Q2zNPCz9i8GSUTtOqff0rdFYzvO4f6842XG2bukfCIlMovgY47Yf//9PbenTp0aeL+iIu+gQtj0rI6VK3376zWumCbH/7uqvob/mJE71Li7oHc7M3X1RjvQnKiRqtWE/uBjIupI/jwgTYU6VPoJ8pM2zUzLBvXMuq077OCAE3iMR40XNaL8DTZdODUb298QUwe7TUMFPYvsQIW7w/bCtEVxO/YaJHQCj05Dz99518CIE3jS/ccuXRMbBNHruanT5zTaNcP1sr4dzabtO81DE+bEjkGDzuo8OGmu1Dj8etmauIMVapQokOhPOzJ9zcbAwJwah3rMjp1l9nvVbFU1brSCwT3IokaUggTfLFtjGteta0oaFgQ2Dq/o17nSM7n9aXeU2uZ/y9fZjq8GurzHXewJPmqwWEHUf0xbmHQWczwKLv179pIKv/cPTGmA4rwmwQFANS7VUVEj+m8/zK8QLHJ3JJds3GoDgGqcaXXGog1bbKP0xK6tbUNaz6PBZgXQg8pKotnYChIqwBRE54eC/xpYVPC9cb06ZvGmLXbWsAbXtLLSXVacc1GBDQ1Cejphu9LxHNu5lachrdlvyWiljGbYO5MGNFM1KHDh/p1m0Gnmos4J/X791u12YkOiBvOgXZ0VDQC5y8WeLYsCVz2rY3nQrnKhAKEGA/Zr0yxu6iHVcef3amf+PGGOPS7pUtTQNKxb2w7g67zWwKWoLqtbu3zwwFlxqQCszoWgc8/tqv6dzd3/m+l5n9KuUX3PoKMGPdUBcH+P7pXlzvHpfkErq/Q9uwND+ow0iKwZ6ZqhmmxV0y96d7BnjGbr65g0ILJq87a4q/W0wiSI6lqnvtVKNa0Eq6p4AwzOqh19TzrbFmzYYgcwFIB1H9+QXTPL9T1rZqvzfaqOHtygwH6m67dvtwMus9ZuspMkVH7UGdUAvSbbOOXnV307mYcDZuBqAEqv899Fq+yx6FwM6jArRfDrs5fYTrg6kBrw22vXarggGoxxBxikqKCuOa9XeztAqeuZ02nWMTp7SYuCbv6Vf7qvc76rLnICJ2pPqKPuXlGg66Im+ei9qZ7Tv0Hfu847rXjfvGOnfd+6Limoq2CHzoFkDmzb3B7zP6YvtG0ZhwZ7Ezmucyt7bmsyia6J7sEKR4+mDW25CTqPtarEzRkADKLBw8qkf9LnHSatbfkgT1v73WggTtcPffeqO/XedI3XtT3etUPnnAJ27sk8an+oPvQHH3X98M9Y1iDaRbuOU/tRuTMYKDisAc1k6f51Pul5dV7rRwPnfsm2glCbxR14lJ7FjTwrH1VvquxqRvaxnVqZv34/L1Y/aOW+qA4ICkwGpXbT+9Kgu1575Mwl9jvQayigECYo76bBrd2LG5l7vou/mtc/0UH1hPu8F9VT/11YPjlPgQX3Xuyqv3oFpFnVsTqB+SB6nz/t2Mr+BLXdVcZVd7vbdHo/+nFfk1QvqA0Qr82ttn0938emekPBJQ3S+idwlN+/tm1LK0ij8191seo593tTwCkqvWe1vzWQrvrFT+0tTU5UthUN1jvpa/V96Ptp07B+YHtDk1E0SK7vp2BXX8YZUHavAtNnp3Kg512yaYsN7Pj7H1r1/sPK9ebtOcts21L1hQJg785bZj8LTSBQPe8EYfztV6dUarKCJhzpuhAlOKrzyQkMqnyq7lb7Scesv+m7UXDSoc9L7dHVW7bZsuqUEU2eVHvLGYx2t5vVFgu6bqhfopWQqqf0PDq/VPamrd4QGOBwt4l0DE6/yR1UdyYNqFwd0KaZbatpYFz1vc518Zdr3X+V6/2oD6vPIFl2FdHx6jxR/R2PzjO1gXV9UdBTwRh3n1yrmXSu6Hy9tE8HuwpF116nntaEO70fZczYo0VjW7+qzaIyq7rA30ZwztkDSpvZc0iTQPR9xZtkqoFs1YGiz8j5nBw699Qv0kQb9T2CrgWqo67Yo5NdKaWgsOoJTfz7a8BqM3fmEj9nYnEi6uu6UwcrgKV2VVCmA39a8UPatrDfrT7LBa62tLttFJYms5Z/R2WB/bl+LctX7iejCWrOZE2H6iOn7aKAuQI98SbV6Py6ol8nGzhwzm3VR+56QPWDk6pRdZJzfVbAapTqnjJNVmpu615Rpg33eIGzelTfna4ROn8Palts/69VYQt1LjZrVCEtp84r1WH60TiLu8+petTdvyr/TIOvuRrLcK9Y82fAEX2v+g6dle9+akvpHHGuXTomN9W9msy3fNM2O4kz0YpEfScOfdZhJn+oDgpasRZv3zzVtS/O8I7Z6H2pPaYfR/065c+pNrAmfmoChyZX+vtNGpNQoG/Gmo2maUHdwEwZQe3oRJw+tsbDlG1BdZuudW46J9TXiVen6pxT+9spj6q/NNbm5q7j7aKIziX2+VRu/NdEnZMKoAaNp2hs6PRubcxdrn65ONd+fX46T75fud6+jnOOqV4KUzfFPpf69WwffPnmbaZFYT17XgcFH1V33fOd91hEbQp9jxq/E30edWrXsitltfgk3hirzmm1xRWc3rqzzLRpUGBXjwctgND4pgL1yTLK+fuGomsQMo/gY45o1658QDpZwK9jx46e24sXh59ptXXrVk9gsEmTJqa4uDjpayxaFG05v/+Y/M+H3KKLvhqE6jDqoq0fBR6U6sitfeP6noa1AlVBe0FWdu8nh7MCLiw12JQKY9Tc5bZDrkFkdbDUuXFmSqkRoQEmDY4GLfFXo21Yj/LVIU56SzenIe1QQ8I9895JGaqAkRrtapxptqcG+fR8/k6zBkDUCFYnTkEdHZOOTQ11zczWZ6uZr/5OmX6CBvX2Ly2O2zk5ebfWdiDc3UGyqzt2NQ41wKMVk+pU6jNTJ1OzyxUUUadPHRc1UDSQ5XDSATncf6sMfwNIDbt4KW7UCVGwUAEJNcY0G9rug9G7vU0x4z4uv6AVN6LGZdh0TfHos3PK0zUDutjnWrB+S4VGvzrox3VuaBv1apQGNf6SDTarUasAiTq+/sBMsjQa/uC/MzinBqQGV3QOq+E8uG15I1/Hl6jjq4FalQelm1FHVytTEtFsZw2CaVBWs8lVVtWZUhlQENTpcOgY3QOHYjuYuzprh7Rr7gk+6rxQ0F8dUXXeVIcp3YpW0mkWpx6rcucMjiSi4wt7TqsTpA6MUraqrKve1OCTM8iXaLBdx6KfRMFHlVN9Nr/p19l2yFRPOA19PfeZ3dvadHQawNCKLR2Pvi+tklAnR/WSmz4HZ9b3U75JG0rr5Kf3pEF21VfuSR7uTrIGzjQ4q3+dgVCHriuaFHCvbyA93v5X7kkeojp8wfrNoVMEaQBQEy50LD2aNTQz1pavGNOxutMJOmmJFdwSfcb7Jtj/QteIS/vEbwu5B/ZVn6jTp3NZZckfCNF57Z9AoMFndfh0X51PidgBuhTsYaJzoXHtxN0Qfc/+wbhzerYzk3cF1zXA6qYJR1rlou9ddbh7kCBRnWZTA+9a2e1wVhuHpfNdq9N1vdNgtILpyepDvQelQ0wefKg40BCFvl8FljNB5d+/SkL1hsqW6uh46QadlZ56r5rEoZqrfAVJLVtv/31S+ao+XSuDMlG4qU7vvavdFGbw3REmA5O+D62OVOpg/woEDeC79zp36PPQ6jxdC/Q+3QOOth7tUWq+XLLGDjg67TWdv/7gowah9dnpddTG83NSUa7aus0GaMMEHjVT3E9tIK2GUCo3DXJOW7MxFgDW568JDc3q1421ZzUQ6G9r6DtId6pAf9s9bPtd130FRF+asSjSpDGdi6qT3GnjNCjpfM76179yOBX0OaveDcra4J8w436PiVZbqM67qHd7e23Se3JnRVEgX+e4BvOdNqITKNNEl6ddmVC08lKfi963rmeayKh/dcyaYBKvXeEOBiozjXOdu6xvBxsg0sByZeo8lSf//qEa9NTkB/3ob/os/StBncf+vGt5hhG9a10n9Z3qswpqK6ucKF24n641+tmvTbFdOeIEMxWcU7AjiCbRrN22w36/GqR2rh36XvzXpiCqV9zvR8dWt3b8ybzu1ZRHhAyKu88pnSO6pny1dI1diTOwpHylkdOGdVYNO/Q3f9nQKrRkVN8oaCUfL1hRIf2x6iKtlusYZzJZhUkCBYnrRH1u7vpZGUaCJOtrRKXzS+0HvRcFzJwJejpP/W0RXU9/tmuShoK8WlGqvrOCFfo+omxT4wTKguIAatOECTzGo+9c6WgVYNb5oOfTGIJ7PMCZIOk+FjeVR028UDl0Ulj6qf+oScCqkdzXPAW0NXlM2aJKXHtL6ri0EsxJIR/2XBS1qdUuUbtU/UT3czg0ydw9iUHXf62S02epfq3aDWqTKYjm5+5nBtFneGb3UptRQvXSPgGvrwCME4RRPRQvLbX/89R5Hy8ziiYeKCtJooCNf1VmmIl7QVSPOXWeJlSoT6Vrrruf44whRUnTn4zOC6WRVdkOCtomqlP1N/VNNm/fYcdG9D35z2d9vppQp8wEWu0cL0jtjG0k6oM4k5Jemr7YXot1Pjn9bV179fnEC8I6WeEcTt85iNrP7Rv/+J2f06NdhYxBiRZqHNm+hc3yoGPS+RqW6jitaHSobaBxg51lZWbTjp22fR8lk5vGX56avCAWtNc4HqoHwcccUb9+/VB7OXbp0sVze968iil64vHf1/9c7t+rUnRmB8+fHy1tmf/+8V4HucUZyNYKAH96L3eHQLNo1IDT//UY99446hQpmJZpugA6KekcutA6M+XCUINEP2qY+oOP/oaBVhzpedXB0ODlIK2aqVXLdgYdavi4L9z+YwtqMGuwyL+Swk8DXprZrcasBvD1PIk6JzbVbY+2tsOhlJIaPFGQwt9wcWZMFu4KDCWiATYN7Go2pNK7OLPatYLNncYnTOoGfbaJ0skEUYPQ3yh0AjlBwUfNpNP3EW/wW+//nJ5tA1OmxaPvSZ+pM3tRgyRu6iD4B6O8xxr6pQKP12nUNqhTOzYYoEGpeKvJwjynf1WAO/VYPKoDNHvUv8IuWYokfR/+sqlO8heLV9sB6P1aJ545qc9QG5mrvtLAnbu+cgdX9Xm4U6akg2aN+1eRJFvh46agabz0yHvvqic0cBC0ObzqXH9QXQNMiQZgVV9o8Fydfa2UUGdAq52cFT9BNBCnFXdOHaIgqyafaOBFxxav86PPoV6tWvaxSskoGnx06hsdg50xbcpsveJf6Vu8a+anUpEFzez0a+07F5yJI0HBS62wjvI9ReFeURFEZU11iGagKlCm78MfpMwGquc1IcYJEGhQVseqnyA2BWqS1YbppPNHq6PUQY6XUrEydJ44gSeHZmn7KbjnT0WmQSNNDvAPBlcHDZArBeQXS1ZVSCPpDIDrOP11swYttCebBh+a1v9xYC2RKEHHqNSGUqYApRVXgE71p76jgxOce1qJM7B1Uzus6J+EpoFKfz2qOs9J8a02sYI8mmSkSW6qL+PVeUErUIJogswBCVbWizNLXumY9aNBNQX8/Z+/BgXdE6g0kSdZ5pDqpmNW+up35y33pC/T5LZEnFTSSsGnt6gB/5pKfZd4A3g6R4O+wvKB4bY24Kz6zt3ODDO5ymlXOHtXabKMexDeWXWcShrI1EBtlFSuqaD2ka4HYahMaQuITLCTqjZstquw1Kdz9x2j1ueZvN4GtZfiZTJK3WtW/J3Oj3SsmtH7U4BHQd1RuzIxKKCQaBKJzm33xCJ9H5rI7F+9K+onaoKnM8lSdb/24Ba1V9xZB3TNCUp7GpXqEXe/UG3xl2cuthlhNLm4e5xMOe7Hu4NO8TiTl/x9DX02qZx4FWZsR9+jyrJW/pffv3lsnETXRX82qqgUiIm30tBP7/0/81fYa7MyHbknNGlswk19WSfbi+ojBX6jlC9NrFCaTU1sV/tUYz1aNVcV6oMlmsTkT0Hsz2SVaWp3xmt76rwIPakxRPNJY0q6Fi/ZtNV0KWoQug+nydOzd2U9sCuSI5QPTSxyr/B0vht/xgBNfrNvI2CST2Uk6veFoWNQ0F4T6fQe/BOUkDkEH3OEf6Vjs2bBnaHu3bt7bo8bNy70a4wfP95ze7fdKua6d/ZobNu2rVmwoHxmxLJly8zChQvt78Ksrpw0aZJnb0ZWPuYPNcoSrU5Qo8imdXF1QjTI7aT+Uwc20UyimvIZKNWJNlnXOOORHVpUGMTRe1fn0Ul/mEnqBGl/IQ2Bur+HRHQfDaLpJxX0eWigzj9Yp/ztSj2ljpUaoGoUBQUftYpCjWt1CoJSKVWFVjO507LamYkhZv9r0MY9a1ApqxJRx0fHr8HAga2axvZKyTStgrQz+XbsjJtutaZQZysozV48UVOYZCulzNRKMidApk6D6lnNzExXcEz0Ghr4UEdVr5notWrtWhnoXx3oX50ajx6nAYytO8rs5A33MShoro65Ot7xXrtuiI/hvJ7xJ01otegH85bHgkdKAZPOzzaM8nM3u89fXW80m1gBNQ2c+PdKzUblg/ap/2610sG9L5SzGsRNdbB7VZZWUFd2r5l0cbIorNu63bw5Z5ld6aNyGK/8uQcXKzu5JaxalSxHziqUZKKsIrEpRju0tFkodD45j40y29uhlaLuvbU1eUiBx7ABch1Louu7jk17bio1ltM+qwn03Sk7h/Yi1P576n8ErVwNaitUJoVqrki0fUUYdlJiiNV8SN95r1WvUfpx2UBBa00QdFavKHiVbhpTcKeYblDHu09tOmjySGXLhzNhTmmXtXpYwaZxyzVpr55tP6m9b1MYlpXZ/pv7u1fWC03S0yRobTGSjnaMruGX9PlxdXNNOfeiUt8/aOufTFP7r3yiVHkmGH3+WomqiZH+lY/6LhSsVIBY333UCUQax7ho9w52QpZSWCsIpzEKv1R+48q+NW31Rrsthybo1ZS2R3Vei/U9aV9HrSjWORAlqFe+UKTUpnRV/8yZYKsAplK1O9t9hF1Nr4njTlpW8ac8TqUoQXukT3b1SFFpEyZM8NwuKQlOHzBgwAAbHNy4cWPscQr4FRQkn604duxYz+3Bgwcn3IPypZde8jz2+OOPDxXg1PE4Bg0aZAOQgCOoMZxslUdNo9WEzuyoqPv1ZIIaH9nYXdCAhk17saN8L7KghrM6ZuoURAkyRaG0m5pFr2CcGmAKJId1QpcSM3d9E7tKJVljUudFoj1HMnkuxFthWZ2cVdKOeKuAUZ4SSB2RVVu2m6KC1K86SCQVMyLDijcjNmwQzj+xQHtMauWAVm8EdeT9gRMFKJQeS4NX2VB2awqdI2FTYuUyTa5SOmUNGHRp0iBwtYraR7oGKk2VVsllc0dbEwfO8GWMyPgx1Cvfr80RlMq0uqViQp2Cu5rxrVUJmrWuQfxUrsx1p5qsaeplMCVxVWX5YlLUINnaj0vWx9OgtvZE14RLrSjLBG2bohXSCsppMkhVUpFmgpMqWPyZYZRZ4KRdwaigtnAmVt+mY3UzkrPpZm2Gq+irSKP0Jxu60nRqbE6raJ29WhXoqsrzx9uXUGmKlS0i7Or7fOdk66oMtR37+7KkaUKhtvvQikgFEMP27ZU5QvsjK6NHmC0VUPMRfMwRo0aN8tweOHBg4P0UZFTQ8L333rO3FYT84IMPzDHHHJPw+bds2RJ7jCggeNBBB8W9/2GHHeYJPr7++uuhgo+6n/95gHyUjUHHmiDZPmLpnhWnztuFvdqbeRs2mzYN6odKf+buGKR7Rq0GrfOB0trMWbfZpvbSe3ZvSI7gzkSU/Rjy0UGlxWbl5q12Jq/2qXHSG/pT3sajlUM1fXUwqo+uD6rXkqXs0sBO1P0q85X2uX591lKbfkqZNTI5GSKTNOA8rHup3TNdA+e5utok1x3VvqV5ftrC2O0w6eZrMq1q+dS1N2SylLjIfcokEyabTCopYKK9JQFEowlxJ3VtY/fRVdtDewCmYyJb74LcavM2y7KMJWFUZiGIvjsFj5U6Vquzw2ypgJqNbzgHaLXg22+/7fndIYccEvf+Q4cO9QQSR4wYYYYMGZKwM/ryyy+btWt/3Jz84IMPNkVFRQlXPjZv3tysXFm+j9Rnn31mpk6danr06BH3MUrP+tZbb3n2sTzqqKPi3h8AklEqiFFzl9tUEErtmM6UDmH20ck09552ynGfq4OrQcH7eLN7gcpQoFEphQDkBqXYHN63o9m2syx0Cueaytk3FDWXBva0nYDSKGqVx6DWuR18VBaYOes3m4UbNlfYGxIAUDOuW1oVV1MErcxsmeYVlc5em6LsOH2quC9oTaL0vOmeeI/skds9rRroxRdftKsQGzcON4Nj+vTpZvjw4Wbnzp2e1Kr77bdf3McooKcgoIKB8t1335lnnnnGnHvuuYH3nz9/vnnggQc8Hdhf/epXCY+rsLDQ/OIXvzB33323va3ju+mmm8yzzz4bmOJVf//9739v1q//MaXZ6aefblq1qhmpcABkJ6WGUApTbW7foXFh3s34V7q1jo0LbRpYDVYBAIAfBz5yJ2k/cn0VifZx1U8+0CqImjRoDQCo2bRC88DS5mb0ovIFNEpVm+7JaUe2b2n3QNVY1T6tmkbebxOoKcjrl2X++te/msMPP9zcfvvt5ttvvzXbt1fcqFfWrFljHnvsMXPKKaeYpUuXxn6vwN4NN9yQ8DU0+H7ttdd6BuHvvPNO+3xKr+r29ddfm2HDhpl169bFfnfccceZ3r17J30vZ555punQ4cdVAgpyXnjhhWbevHkVVjxeeeWV5v3334/9rlmzZuaSSy5J+hoAkIzSSWpWVdB+ofmgWf16dsVjvgVeAQAAAAAAkhlcWmwu6NXepltO93Y9omDj/m2K7cQijdkAuapWWVlZWXUfBLx7HC5YsMCTerR79+52BaBWQ27evNksXLjQTJ482ezYUb55r6NOnTrm3nvvTbp/o+Phhx82Dz30kOd3SpXat29f06BBAzNz5kwzbdo0z98VdHzhhRdMw4bhUp9odeVpp51m95Z01K1b1/Tr18+0adPGLF++3AYlt27d6tlP8oknnjCDBg1K26mxYsUKc+mll8YCvi1asMEtAAAAAAAAAABAVZF2NctpJeLEiROT3q+0tNQGHvfZZ5/Qz610rdu2bTOPP/54LG2r9mj873//G3j/Pffc0zz44IOhA4+i9K56/quuuiq2QlOrObWqM0iTJk3MHXfckdbAIwAAAAAAAAAAANKDtKtZ5rLLLrNpV4uLi5PeVyn0evXqZW655RYzatSoSIFH5/FKdzpixAgb7IuXkq9du3bm6quvNs8//7xp3bq1iWrgwIHm9ddft3s4FhUVxd0jcujQoebf//63OfLIIyO/BgAAAAAAAAAAAKofaVez2KJFi8ysWbPsv6tXr7arIJWGVasDFQTs37+/adq0acpeT+lctcpyyZIlNr1rSUmJ6dSpk32dVO0VpvcwduxY+1paZanj16rNvffe26aVzRTSrgIAAAAAAAAAAKQeaVezmIJy+smUtm3b2p90UvD0gAMOSOtrAAAAAAAAAAAAoHqQdhUAAAAAAAAAAABAShB8BAAAAAAAAAAAAJASBB8BAAAAAAAAAAAApATBRwAAAAAAAAAAAAApQfARAAAAAAAAAAAAQEoQfAQAAAAAAAAAAACQEgQfAQAAAAAAAAAAAKQEwUcAAAAAAAAAAAAAKUHwEQAAAAAAAAAAAEBK1E3N0wA1y44dO2L/X7VqVbUeCwAAAAAAAAAAQHVq1qyZqVOnTkqei+Aj8tLatWtj/7/++uur9VgAAAAAAAAAAACq01//+lfTokWLlDwXaVcBAAAAAAAAAAAApEStsrKystQ8FVBzbN261cydO9f+v0mTJilbSpxrli5dak455RT7/5deesmUlJRU9yEBqCTKM5A7KM9A7qA8A7mD8gzkDsozkDsoz9GQdhWoooKCAtOtWzc+xyS2bdtmf6S4uDhlS64BZB7lGcgdlGcgd1CegdxBeQZyB+UZyB2U5+pD2lUAAAAAAAAAAAAAKUHwEQAAAAAAAAAAAEBKEHwEAAAAAAAAAAAAkBIEHwEAAAAAAAAAAACkBMFHAAAAAAAAAAAAAClB8BEAAAAAAAAAAABAShB8BAAAAAAAAAAAAJAStcrKyspS81QAAAAAAAAAAAAA8hkrHwEAAAAAAAAAAACkBMFHAAAAAAAAAAAAAClB8BEAAAAAAAAAAABAShB8BAAAAAAAAAAAAJASBB8BAAAAAAAAAAAApATBRwAAAAAAAAAAAAApQfARAAAAAAAAAAAAQEoQfAQAAAAAAAAAAACQEgQfAQAAAAAAAAAAAKRE3dQ8DYBcsn37dvO///3PLFiwwCxdutQ0btzYtGnTxgwYMMA0b968ug8PyFuLFi0yEydONEuWLDGbNm0yrVu3Np07dzZ77LGHqVWrVkpeY8uWLeabb74xCxcuNCtXrjTNmjWzr7P33nvbuiBVJk+ebGbOnGkWL15sateubeuYXr162fcD1ESUz2hUv3z33Xe2Dli/fr0pKSkx7dq1M3vttZepU6dOmr4lIP/a27Nnz7bXXJW1nTt32tfo2rWrveamisqw2g5qn6xevdoef9u2bW3boX79+il7HeSOjRs3munTp9u24KpVq2z7s6ioyLRs2dK2a3X+pEqulIGysjIzfvx4M2fOHPs6DRo0sO+lT58+prS01NSk9gxySybLcybkUn+Y9jaiWrFihZkxY4Y9/3U903WgoKDANGnSxHTq1MnsvvvuKSsH9J/zozwTfAQQo4vKo48+al599VWzfPnyCp9MvXr1zIEHHmh+85vfmJ49e/LJIW9dd911ZuTIkZV6bPfu3c2bb74Z6TFfffWVeeSRR+y/GjTxa9++vTn99NPNBRdcUOlGhzqKf/rTn8xbb71l1q1bV+HvGuA44ogjzFVXXVWlDuSLL75onnnmGTNt2rTAv/fv39/84he/MEceeWSlXwMQlRV1nDRQN2HCBPszZcoUs23bttgHdOedd5qTTjqpSh8Y5TP6QMuDDz5oRo8e7fkuHK1atTInnniiGT58uCksLORkRlrL8/z5883hhx9e6U9Z5/LRRx+dle3t9957z/ztb3+zn1m89si5555rTjnllEq/hgKn999/v/nggw/M5s2bK/xdA1XHHnusfS8avEV+U5l99913zWeffWbL8I4dO+LeV4Pvw4YNs+en2qD5XAY0UeHJJ580//znP+3r+SloMWjQIHvdHDhwoMnm9gxyR7rL85gxY8w555xT6eN7+eWXbeAzX/vDtLcRltrSI0aMsAF3XS+D2qb+a47aqSqfgwcPrtQHTf85v8pzrTJNnwKQ99TwueKKK+zMq2Q0e/P66683Z5xxRt5/bshPmQo+6hKtDtDjjz8eOAjgt+eee9pGiWZmRvH111/bTpRWXiSjQRQN8KrjFcXatWvN//3f/5mPPvoo1P1PPfVUc/PNN9tBWCCKd955xzz//PN21r5mYidSleAj5TN6+dT3ctddd5mtW7cmve9uu+1m/vznP5tu3bpV6vtBbkh3ec508DET7W2Vr1tuucW88soroe5/2GGHmXvuuSfyLHYNOt9www2BA7R+apcoQLPPPvtEeg3kjtNOO83O1o+qS5cu5t577zV9+/bNyzKg1Q2//vWvQ312GhC+9NJL7f2zsT2D3JGJ8pzp4GMu9YdpbyPqeVnZiSuaXHP77bebhg0bhro//eeb87I8E3wEYBtYJ598sk2t4qYULh06dLBL7TWbbcOGDZ6/q+F43HHH8Qki72Qq+Pjwww+bhx56yPO74uJiWzbVwNPgpVLcuPXu3dv84x//CD2rdOrUqXYWs7t8azZzv379bOoXzXwbN26cp7GjBpNmYO+7776hXkMzYTV7UzNj3Tp27Gh69OhhZ3T/8MMPFTp76nDddtttoV4DcPzhD3+ws4nDqErwkfIZrXyqzlTd6aaBXg0MNW3a1MybN898//33nr8rjZwGj5S2C/kp3eU5k8HHTLW3FbDUqko3BQnUPtD1Xdd9lTc3zWB/7LHHQq9m+uKLL8xFF11kr9/uYKlWa7Ro0cKm0fKvhFF512oPDYwg/2hlns5xN51vagfq/NT5ob9r1YUGQt0aNWpkV2WEDSTkShlQXaA2uo7X35dQ2lj9XRMz/J/rlVdeaYOQ2dSeQW7JRHnOZPAxl/rDtLeRiuCjrmNasawU4qrnNQFw7ty5NguJf5WzysATTzxhU7MmQ//51LwszwQfgTynmSeaUa09Zxxq/Gj2p3sfDF2QNMDy3HPPeTpYmlGqDhCQz8HH//znP6Efq45KmJnC//3vf83FF19sy6hojxWlbDr//PM9e8coZcU111zjGcz82c9+Zv74xz8mfQ2lhxo6dKhnAEaNxzvuuMMOhDqWLVtmG0ma4e1Q6qi3337bNkyTue++++xsaof2ANHsLQ34OnvHqMOl+kSv404loYFnDdYCVQ1W6LzTIJq7rFQ2+Ej5jFY+lSpGqbbcgzZnn322XZ2h78V9P9Vn7jRUWiWiGZ/IT+kuz/7gowY6lYYxLF0Dwwy+Z6q9rbSMv/vd7zxtDt3W5+IEVXQs77//vg3QaL8Yxy9/+ctQK6Y0MKpgqHvgeciQIebGG2/0DHRokEqrwtROcWivIE3ACjNAhdwMVtStW9cccsgh9pzU7/yrDdUefO2112w70b2iUPsaaSW0Ahf5Uga0CkvpHx0KgqhN7V49qba8AiBa6eDuM+h3+++/f1a0Z5B7MlGe/cFHrRj86U9/GvoY9Rphylku9Ydpb6My1PY86qijzKGHHmon4mhv03jjVSoHTz/9tHnqqac8QUil+1Rmj0ToP+dvea5d3QcAoHppLwz3QIj2ctCAh3sgxEktcdNNN9nKzr0RtwZIgHynchP2J0zgUQMAWungzoyuARLNYnYPBDidIzUs3A2Q119/3TZCknnhhRc8Ha0BAwaYv//9756OlpNDXumY3Kll1OF0d6Di0SCFZre6B4HUWNVzOR0tUedVKXweeOABz+NVx6iuAaLQXgdKC6ZrlgbGRo0aZdMpVWVfJwflM3r5VP3h7jgpuKNBWne9JWp7qA3Srl272O/Gjh0baYIHck86y7Of2rtRrulhV/1kor2tvST9q5dU9vQ5uVdz6dqrQSZdi3XtdShYoUGlZLQ6zB100XMpnaR/hrVWc6hNoZVgjjlz5tjgEPKP2n9q53344Yd2X0ENuAelOdU5qUE5nScqD+6An87ZfCkDWjXpDjzqs1B735+2VfWjgqbulRFOOyVb2jPIPZkoz35ajRvl+hx2kksu9Ydpb6MyVK9rRa4m8R1zzDEJx6tUDq699lobOHfT+Ry077GD/nN+l2eCj0Ce07J3N+WU1/LteK6++mpPpaZZo5MmTUrrMQL5RoOUU6ZM8XSCEqWdUedIKZbcjTv/4IufGjx/+9vfPPvEaJalf7DB/Xftn+PuWCrdUrJBGnXI3I0rNZoSpcA58sgjPSns1DlloBJRXHbZZeabb76x540a6JqZr/Rk7s59VVA+o5VPpYRz722jNoRWc8SjWeTuVSuSrD5D7kp3ec6l9rauyUoN516JlWg/KqWTc6/y1KCRu10QbwBVaSMdahOovKqNEERtCrUt3H9XuyDMvjXILTpvbr311tD7BGq/Iq10cnvjjTfypgwooOOmdr4CKvHofbiDnErD9sEHH1R7ewa5KRPlORNyqT9MexuVpTZ12JTjjuOPP96udnYoLeuXX34Z9/70n4/O6/JM8BHIY+psuPeQ0P4TBx98cMLHaIa38uFnW8MRyCX+PSHVQUk20KpZpe4ZpZ988oknvY2fZretXLkydvuAAw6wKeAS0Uw3bSruUCdKDcl4du7caVPRONSoDbN3h1I9uVHHIArtTeFeyZBqlM9o5dP/eakNoZUaiagt0qVLl9htBV38+00hP6S7POdSe9tf1vzX0iBqX7iDIlpp5V4F5adrvjtoolR1yfaU6dmzpyf9owZptV8e8kvbtm0jP0aDm+7VxbNnz/YEF3O1DKxZs8aMHj06dlsTFX7+858nfA31E/wpo6PWGenobyA3ZaI8Z0Iu9YdpbyPTBg8e7Lnt30vZjf7z+Xldngk+AnnMPZNCtHdFGP77Kd0GgNTQgMann34au609rRLN2nZohqZmSTq0T4Ty6oct/+oQhuG/X6Ly/91333k6dJodF2aGrGZeK1WVe+aXZogB1Y3yGb18Vrauoa2BXJGJ9rZWY2mlk3tfOfcqqHh0TVY6RYcGgseNG5f28pxNqaCQvdS27dy5s+d38a43uVQG1H7X3m/utK7xVmK5qb/gDu6oPxFvhWWm2jNAZcpzpuRSf5j2NjLNn8FDqx+D0H82eV+eCT4CeUwzvdz8e0jEU1pa6kkFNWvWLLNo0aKUHx+Qj9RBcTfclJIl7J4V/jLsL+OpKP9KUeU+nq+++soOPAT5/PPPPbe1eXlY7vtqBrr/uYDqQPmMVj4XLFhgZ7Y7lDKuTZs2Ka/PgHxvb1flehu2rOlar2u+Q22BRGnjEr0G13SE5U8F5w7K5WoZ8P8tbJ2h4I77eNavX2/Gjx9fre0ZoDLlOVNypT9MexvVwd8m1argIPSfy+VzeSb4COQx9xJspZvp27dv6Mf6Z5JOmzYtpccG5Ct/WVLnprLlMl6aBQ02LFy4MHZb6aLCps9RR6t3796emWxz586ttvcCZBLlM1r59P8+Sh2g+7rTv1EHoKbKRHs7E3XTnDlzPIOraguEDVZo4MSdmnL+/Pl2fz0gEQ3O6Vxxa9GiRc6XAf97CRvgzLY6A6hsec6EXOoP095GpulaOGrUqFDBe/rP5fK5PNfsDTQAVJr2knCnf1DDz52mJRn/hveajX3QQQfxjSAv3X777ebbb7+1HRjNMtYm9NqnSgOMSq2iDeMbNWoU6rlUltw6dOhQpXIZRDOp3PvZRHkN53XcKan0OtrDqjreC5BJlM9o5bMqn5faJBoI0t5YTio87StVVFQU+XsDwhozZozdI2Xy5MlmxYoVNljYrFkzuwJx4MCB5rDDDosUCMhUezsTdVNVXsN5HWd/L7VB1Bbp1atXpOdAfhk7dqxZvXq1p/zECw7kUhlwv44GEbPtvdBGR7rLc5B3333X7p2oQIauq/Xq1TPFxcU2xbJSJysdcFB/NJ5c6g/T3kYmacXyrbfe6lmdd+ihh3rSBLvRfy6Xz+WZlY9AnvLPzFJqpyj8y77jzfQC8sGzzz5r95lZtWqVnQWmf2fMmGH+/e9/m//3//6fHax8/PHH7Ybzyfg36g6bYsFJt6ROmGPt2rX2WPz85TXKa4h/n4p45d//XqLUM9QxyEaUz2jlM9V1jVadAOn09ddf273YlPJIq5KclRH6/aOPPmpOPvlkc/7555upU6dmVXs7ldfbeOWM8ozqaF+7HXzwwZ4Z/blYBhRU0URGhyYzhl1dWZU6Ix39DaCy5TnIxx9/bEaPHm0WL15sVxpu2LDBrqRUWsEHHnjADB061FxxxRWhtwPKpf4w12ekm9rDWkX3j3/8w5x44onmxRdf9KRbvfnmm+M+lv5zuXwuzwQfgTzl7tQ4HZso3B0O0WwKAME0y/O+++4zF154oV0FkYi/LEUtm/77+8t6Ol4jqPzv2LHDs5dMw4YNTWFhYUpfA8g0yme08pmJugbINO3Rcuqpp9oVGNnS3vb/3v+4RPzXZ127gyZLUZ6RSV988YVd6eRQkOKcc86Je/9cKQNVrTOq6/oc1N8AKlueK0NlVq+hwMiXX36Z9P651B/m+oxUO+CAA0zPnj1jP3vuuac59thjzS233OKZgKf0w5pYkGgVM/3ncvlcnkm7CuQpzRTzz2CMwt9ocjeqgHzRrVs3c8ghh5g+ffrYlC9Kt7pp0ya7SkKp20aOHOkJNmrA8vLLLzdPPvmkqVs3+BLsL0tVLZv+sp6O1wgq//7fRZm1HXRM1DHIBpTPaOUzE3UNkAoaENRAy09+8hPTvXt323HXdVqreX744Qfz3nvv2eu6Q9f6a665xjRp0sQMHjy42tvbqShr7v3n9Hxq06T6NRI9H+BQubvuuus8H8hJJ53k2WMtV8tAptrPmWjPAJUtz25Kfa5043vvvbftezsTC/S8Snv61ltv2QxEDv3+sssusyu0EqX2pjzHL89cn5GMtiBQJhBtL1SnTp2E96X/XC6f+88EH4E8pUGTVHZs/M8H5LIDDzzQnHXWWXZPxyDq6CjVqgKNt912m3nttddif9PgpVK3KS1MEH9ZitrY8JfloLKZifJf1UYTwUdkI8pntPKZic8LqGrQUamitFJC//fr0qWL2Wuvvew1XxOIFHDUXpDOioarrrrKBiY1OFqd7e1Ul7WgwEtVX4O+A8JwypVSK7pTjvmDF7laBjLVfub6jGwuz6J9y+6++25zzDHHxL127rPPPjaz0BtvvGGv5c75rn/VD1eGAu0NGSSX+sOUZ2TaxIkTzXPPPWfP0SOOOCLhfek/l8vn8kzaVQBWlHz7Qfd3b9YN5DqlnIgXeHTToIU6Taeddprn908//XTovVEyUTZrwmsA2agmlJ1sKp+0NZBttMJx2LBhgYFHv/3339+mlioqKor9TtkNnnjiiawrAzWhrNF3QBBN2lOg36HAwf33329XGUeRK2WA6zPytTzvtttu5oQTTggVFDzuuOPM448/7gk0at+0V155JfSx0t7+EddnvPzyy3YPdP188MEHNqPXgw8+aM444wzTqFEje458++23Zvjw4XaCgfZhzaWyxvU5tQg+AnmqQYMGnttbtmyJ9Hh3WhoJM2gD5Ksbb7zRtGvXzpOaKN5eUf6y6S9ryfjLclDZzET59/8u6vugjkE2onxGK5+Z+LyATNJg6LXXXuv5nQY34w1SZKq9neqypoGlVL8XyjOS+ctf/mJTJTpq165t/vjHP9p0i8nkShnIVPuZ6zOyuTxXxsCBA80FF1zg+d1LL70U9/651B+mPCPVSktLTfv27e1Phw4dzO67725TrGrPRwUkDz300Nh9lfpYmUHiof9cLp/LM8FHIE/5KyAGEID00YxNpWxzc88CTefARtDgSSbKv/99RJkNF3T/bGg0AZTPaOWTYAVykfaqatq0aey20rBOmTKlWtvbmShrlGek07/+9S/zpz/9yfO7m266yaZcDCNXykBV28/VVWcE9TeQv6panivrvPPOs0FOh/ZsXr16deB9c6k/zPUZmaR9Vx9++GGz3377xX737rvv2iBkEPrP5fK5PBN8BPKUfw+LsCkgHStXrvTcdqegAhCcrs1t6tSpgR+TvyxVtWz6y3qmyn/dunU9DSfltI8ya4s6BtmI8hmtfKb686KtgWygtG5aYeEWL/iYqfZ2Vcqa9oJx7wejQQr34G0qXkMoz4hn1KhRdjWF25VXXmnOPPPM0B9arpSBqr6G//6Zuj4H9TeQn1JRnquSSr1nz56x2zt37jTTp08PvG8u9Ye5PiPTdG5rQoHbU089FXhf+s/l8rk8E3wE8lTHjh09txctWhTp8e5Nw0VL8QHE5067mqgR4S+b/rKWbDaVu7Gh/TQ0My3d5d//fPHqhSjvxX9M1DHIBpTPaOWzKp9XlLoGqCnX9HS1t1N5vY1XzjLVdkB+GT16tE1lrCCBQ6kTL7300kjPkytlQMETd1BEK6ujrJbKxPU5bH8D+SdV5TmV1+d4Kx9zqT9MexvVtRVBjx49YrcnTpxo90L3o/9cLp/LM8FHIE81a9bMdm4cy5cv98z4TGb+/Pme2127dk3p8QG5pn79+p7b8WY9dunSxXN73rx5oV/Df1//c7l/795U21+ek/HfP97r+OuFKO+FOgbZiPIZrXxWpQ5Qm0SDro6WLVvaAU4gGxQWFoZKiZSp9nYmrrf+31el7aA2SLy2A/LHN998Yy6//HKzbdu22O9OOeUU89vf/jbyc+VSGXD/XvvJRnmdsO8lE+0Z5JdUludUXp8T9blzpT9MexvVpVOnTp7r1YIFCyrch/5zuXwuzwQfgTzWrVu32P81O00zVcIaN25c3OcCUJF/VYQGJIN07949YVlLZPz48RVmowVROqm2bdvGbi9btswsXLgw9GznSZMmedLPxZtNlcr3Qh2DbED5jFY+/b+PUgdMmDDBM3M+Xn0GZPM1PVPt7UzUTbrW65rv3ksr7IosDUYp8OpemeIfIEZ+0flzySWXeILxQ4YMMbfeemulni+XyoD/vfiPr6bUGcgfqS7Pmbg+51J/mPY2qjP9qlvQNZH+c7l8Ls8EH4E85t+DbuzYsaGXi7tntGgmi7vhBiC4IeBWUlIS+DENGDDAsym0Hhd2YMNfhgcPHpzy8q9Gk/t4Bg0a5BmIcXNvQu7MiA3LfTyalep/LqA6UD6jlU8NrrpnxGrG55IlS1JenwHZek3PVHvb/xqVvd4mKmsFBQWevS7VFvB/DmFf44ADDgh9fMg9M2fONBdeeKFZt25d7HcHHXSQueeeewL3Wgwjl8qA/5oats7QCmz35IZGjRqZ/v37V2t7BrkvHeW5sjTo/v3336f9+pxt/WHa26gu/n5dixYtKtyH/nO5fC7PBB+BPHbYYYd5br/xxhuhHue/n/95AFQ0atQoz2334IV/YMPdSNDG9B988EGoAYf33nsvdlsdIHX84vGX29dffz3U1+a/X6Lyv+eee3rSzY0ZMyZUw+m7774zc+fOjd3u27evad26dajjA9KJ8hm9fPrriH//+9+hPmvaGshWM2bMMJMnT47drlOnjtlrr72qtb2tMtinT5/Y7Tlz5oSaKa1rsq7N7vRM8YIVqWw7HH744aEeh9yjlUXaA869Z6DaxA899FDcwfswcqkMHHzwwZ7VJGrfx0vt7Kb+gvoNjgMPPNC2W6qzPYPclq7yXFmff/65Z49HrXpMlD0nl/rDtLeRaevXr/dMwNE2Q0HnKP1nk/flmeAjkMd69uzp2SBYgymffPJJwscoZ/4///lPz++OO+64tB0jkAs0O/Ltt9/2/O6QQw6Je/+hQ4d6bo8YMcLm0E/k5ZdfNmvXrvUMXBQVFSWc6enuCH322Wdm6tSpCV9D6WjeeustTwPzqKOOint/zXY95phjYrd37NhhnnnmGZPMU0895blNHYNsQvmMVj6PPfZYz221IeLtv+NQW0Qz6R29e/cm9TKygq7FWs3hviZrRnfTpk2rvb3tr5v819Igal+40zOpvLr3wPLTNd8dzHjzzTc9qSSDTJkyxQ4Iu4M7ZDPITwpQKFChVb3uAfW//vWvKUnDmytlQPWJAoeONWvWmFdeeSXha6hO0nupSp2Rjv4Gcle6y3NUWol4//33e36nwHii1Ze51B+mvY1Me+KJJzx7vOq6Fm/CC/3np/K6PBN8BPLcr371K8/t2267zXZw4rnvvvs8KaCOOOIIW6kB+eLFF1+0s7zCmj59uhk+fLhnYEMDlYkGHdSBcQ9UauZjok6KUjE88MADsdsaNPGXbT91Cn/xi1/Ebuv4brrpprgpl/T33//+9573fvrpp5tWrVolfJ2LL77Y0wjVwEaiFFWadf3OO+/Ebuv5TzvttISvAWQS5TNa+dxjjz3MoYceGrutNoR/cMhNbRD/HkHJ6jOgMrQqKkr6M10H7777bvPRRx95fn/ppZdmRXv7jDPOsEENd8aFRCuZNDHKHazwtwuCaEb7qaeeGrutNoHaBu42jn+VlNoW7r/72wXIDzpXLrroIjNr1izPPlB///vfTePGjVPyGrlUBtR3cFM7310n+Ol9uFd6ahVoshXGmWjPIDeluzyHCQK6qf963XXXeVKuKuinfSjzpT9MexuV9eSTT5oNGzZEeowm1j/22GOe3yU6R+k/v5PX5ZngI5DndBFQKgjHvHnzzFlnnWVnaLoph78GStwdEs3y+s1vfpPR4wWqm2ZzqjN/++23m2+//dZs3749bgNADbJTTjnFLF26NPZ7dTxuuOGGhK+hzvy1117rmXl955132ufzp136+uuvzbBhwzz7bGgmVZhJAWeeeabp0KGDZ9BBe3aoHvDP8LzyyivN+++/70ljk6xD5wzSnHfeebHbmh2nWbL+wSB9jgrs+uuUX//619UyexY1mwbIgn7cs/Vl1apVgffTOR8P5TN6+VS59g+6qA71T+RQKku1QfQdOPbZZx8beEH+Sld51gpEXQd1DdX1J97qJa0E0rX23HPPrbAS4cgjjwyVcjAT7e0GDRqYyy+/3PM7Xbtfeuklu9LC/X50Pde12N2GOf/885MOoIqu/e6VnkrBeNVVV1X4/PQeNTjtDohoDxsFiJBfNJB/2WWXeQIDxcXF9lzXgGe8Mh70k2iANJfKgAYe3SsfVN+pvvLv56R+wV/+8hdz1113edop11xzTcIVnJnsbyC3ZKI8q499/PHH26CdUpDHm6yjcq0JQepvu1cjyjnnnBNq1U8u9Ydpb6MyHn30UTu+9Yc//MGe//HGt0TlXtcNlQX3dVVZvRKlH6b/nN/luVZZsrwKAHKecs6ffPLJngCJLg6aMamGmPLma2aov3GotFNqFAL5RI0q98xjDQpqpqcGKzTTU+kQtP+FGgHuBpmzL9S9997rSb2SyMMPP2z3zHBTahiltNEAi9IqTJs2zfN3DQK88MILpmHDhqFeQ7NKNfPKvUeM9pnp16+fadOmjR1IUSPUPQNUe3gozcagQYNCvYYasJpV6k455QzAaMa1/v7DDz9U2P9CHUk1sIColOawKvbdd1/z7LPPJrwP5TNa+Rw5cqSdle6mOlN1jQZxta+NeyBLSktL7aBxmMFg5K50ledXX33VXH/99RXOuS5dutg0grrWqQ2s65N7PyuH9oXTQICux9nU3tZ70ntz0/Vc7QO1Q3Tdd+8jJUrxqICD/h7GF198YYMq7gEqtYf0mWjlmdLw6b2420Eq7xpU3W233UK/F+QGDYilap9PBcdOOumkvCgDqgu0qsq/AkxtZ9VTartPnDjRTrzwD1gqOBRWJtozyB2ZKM86H3VeunXs2NH+6PqsVY0rVqyw7UZ3QNyh49PjE6VczdX+MO1tRKVAlbsc6VqmwL36XypvCpZrAoAmzAW1h1VONDkvzKpn+s+n5GV5JvgIINbguuKKKzypM+LRxUgVoGaJAfkefAxLjQAFHtW4C0vzg/70pz+Zxx9/PG46JzetqnjwwQcTbkYfRLOZNWPbPSAaT5MmTcwdd9xhV3tEoRnbmiX38ccfh7q/Olq/+93vbMcOyMbgI+Uzevl87rnnbNrKeOms3DQ4++c//zlr9qpAfgQfw9LKJT1WbeJsa2+rfOn66Q++xKO0TgpwRt237d1337WZHIIGfv3ULlG6qChtIOSOTAcfc6kMKIipYKICH8k4aSajZibKVHsGuaG6go9haPLAL3/5Sxt8DzuRIBf7w7S3UZXgY1iaQKcJMjqnGzVqFOox9J9/l5flmeAjgJhNmzaZRx55xHbUNJPMT5WkZoWqQ1PVgSCgptIsIqV3UToY/0zjoAaZyooaZSeccELo1RF+X331le2A6d+ghAXt2rWzr6EUMVE7Wg69Fw08KGVNUONTqSGUuuHqq682bdu2NZWlGd9aKaK9MINoFpdS7ETtzAGZDj46KJ/RaFW46prRo0cHpvXRihENRGnwqLJ1JnJLusrznDlz7MqdMWPG2FU9idJMOTONdW1S6rRevXplfXtbqSAVTIi3r5QGJvReNLhZWc4eNEodp8wPfgrmKHWk3ovS8iE/ZTr4mGtlQHWT9uT65z//GTgBUv0Nrb7S/k4DBw40lZWJ9gxqvkyUZ63kUZ9R5+Ls2bOTBsVVtoYOHWpTrWp1ZGXlUn+Y9jbC0kr+//znP+bLL7+0WxKEKW9Dhgyxq4Ur2x6m/5xf5ZngI4AKVJkpsKKGpVJMaBaL0k1olqNSsAD4cTayVi/oX6VL0/4oWqmg2ZCaDazUS+49YapK6VyVXknpWDTAUVJSYlO16HWS7esSlt6D9pPRaymtho5fqzb33nvvUKk0wpo0aZJt3Oq9aABDn5car0ojBdRElM9oFHTRSo7Fixfb1HLqNLVv397stddeNtUVkEm69ikAqYF97euk1GtKl6jAga6DSq+utGhhU7hlU3tb7RQNWuh6q/ek662CLlUJoPpp/5lvvvnGtoeUmkvHr4FZzaaPujoUSLVcKQMKCGr/SE2c0OosBUL0XrQ/pNrqNak9A4SlNqKuzzovda10UqPq+qwgiMpx165dU/qB5lJ/mPY2ol7LlKVDbVOd+5owp3NT5U3XNaXcrkqA34/+c36UZ4KPAAAAAAAAAAAAAFIidVM3AQAAAAAAAAAAAOQ1go8AAAAAAAAAAAAAUoLgIwAAAAAAAAAAAICUIPgIAAAAAAAAAAAAICUIPgIAAAAAAAAAAABICYKPAAAAAAAAAAAAAFKC4CMAAAAAAAAAAACAlCD4CAAAAAAAAAAAACAlCD4CAAAAAAAAAAAASAmCjwAAAAAAAAAAAABSguAjAAAAAAAAAAAAgJQg+AgAAAAAAAAAAAAgJQg+AgAAAAAAAAAAAEgJgo8AAAAAAAAAAAAAUoLgIwAAAAAAAAAAAICUIPgIAAAAAAAAAAAAICUIPgIAAAAAAAAAAABICYKPAAAAAAAAAAAAAFKC4CMAAAAAAAAAAACAlCD4CAAAAACotDFjxpiePXvGfl599VU+zSzm/q6uu+666j4cAAAAADmI4CMAAAAAAAAAAACAlCD4CAAAAABIu/nz53tW3T300EN86pVw9tlnxz7Dww47jM8QAAAAQNYh+AgAAAAAAAAAAAAgJQg+AgAAAAAAAAAAAEiJuql5GgAAAABAPho0aJCZMmVKdR8GQuK7AgAAAJBurHwEAAAAAAAAAAAAkBIEHwEAAAAAAAAAAACkBGlXAQAAAAA11saNG823335rFi9ebFauXGkKCgpMixYtTN++fU2XLl1S+lpTp04106dPN8uWLTObNm0y7dq1M8cdd1zCY5s2bZqZNWuWWbVqldm8ebMpKioyzZs3t8fXsWNHU9OsX7/ejB071ixZssSsXr3aNGrUyH7e/fv3N23btk3pa61Zs8Z88803ZtGiRWbDhg2muLjY9OrVy352tWrVSulrAQAAAEidWmVlZWUpfD4AAAAAQB4ZM2aMOeecc2K377zzTnPSSSfFbp999tnmq6++ivScCup9+OGHCe8zfvx488gjj5jPP//cbN26NfA+nTt3Npdccok54YQTTO3ayRP/HHbYYWbBggX2//vuu6959tln7f9feeUV89RTT9lAopsCiQrEuenxb731lvnoo4/MhAkTzLZt2xK+T312p59+uiksLIx7v1dffdVcf/31JqpnnnnG7snp1rNnz9j/TzzxRHPXXXeFei69lwcffNB8+eWXcd9Tjx49zMUXX2yGDh0aKjj40EMPmYcffjh2+z//+Y9p3769Wbp0qbnnnnvMu+++a7Zs2VLhcW3atDFXXnml/V4BAAAAZB9WPgIAAAAAagwFvm699Vbz4osvJr3v7NmzbdBu5MiRNlDZpEmTSK+loOa1115r3nnnnVD337Fjhzn88MNN2Dm+ClQqWKvje/TRR20wMtvovSgQ+OSTTyZ9X1oZes0119jvRkHFpk2bRn49Bap/85vfmBUrVsS9j1a5/va3vzXff/+9ueGGGyK/BgAAAID0IvgIAAAAAKgRtAru0ksvtasd3ZT6U6k4W7ZsaQOGM2fONDNmzPAEtM466yzzr3/9yzRo0CD06/3hD3+IBR61km/33Xe3AUL9f968eWb+/Pme+ys45w7Q6X5aydepUycb+NRtpV+dNGmS/dcxefJkc8EFF9gVjnov2eTGG280L7/8sud39erVMwMGDDAlJSVm7dq1ZuLEiZ7343zezz33XKQApL6zq666yqZ2dVY49u7d234mSnX73XffeVZCamWnvvef/exnKXmvAAAAAFKD4CMAAAAAIG3uv/9+GzDSarVhw4bFfq90o+eee27gY+rWrRs3GOgOPJaWltqVdkcffXSFxyig9/vf/97uBylTpkwxd9xxh7nttttCHbcCak662OOPP95cffXVNhjm5g8+Oseu1Y86pgMPPNCmZvXbuXOn+eyzz8wf//hHu1rQWaV57733mt/97ncV7v/Tn/7UpoEVBefGjRtn/9+6dWvzwgsvxH0PrVq1MlWhFZnuwKOCp0qje/nll3tWkW7fvt3eVylcncCh3pc+f33/Yf3f//2ffbzSt2pF409+8hPP37XH5C233GJGjRoV+50+M6V5rVOnTpXeKwAAAIDUIfgIAAAAAEibeAEwBa+0KjCs0aNH25WLjl69epkRI0aYZs2aBd7f+ftll11mPv30U/s7pQNV8EzBrWQ2btxo/9Uehgo8BvEfvwJg77//vmnbtm3C59b+kwpM7r333ub888+3K/pEKx9//etfV3hPWvnnrIisX7++J9AZ5TOMYsOGDTbY63bdddeZ8847r8J9dRynnHKK/VwVUN60aZP9vfa+/PnPf24OOOCAUK+p4KJWVD7xxBOmcePGFf6uz0XBzCVLlsSCytofUt/vwQcfXMl3CgAAACDVaqf8GQEAAAAASLHHHnss9v/CwkK7h2O8wKOjoKDAri50p1p99tlnQ7+mUn5q/8GwtDIwWeDRrWHDhnZ1oGPz5s3mww8/NNngtddeM+vWrYvdVrA0KPDo1r9/f3PFFVd4fqfUqGHp+7rvvvsCA4/uwO3w4cM9v3NWqAIAAADIDgQfAQAAAABZbfr06ebrr7+O3T755JNDr/hr0aKFOeaYY2K3P/7449Cvq1V86U7nqRWa7vfipFStbm+88Ybntj+oGI9WljZv3jx2+5NPPrErGsMYMmRIqO9V6VgVqHSn2AUAAACQPQg+AgAAAACy2pgxYyrsgRjFPvvsE/u/0nTOmzcv1OMOPfRQkyra93LFihVmwYIFdq9I9497BefMmTNNddu6davd89LRoUMH069fv1CPrVevnt3v0lFWVhY6oKrVlWEozWunTp1it1etWhXqcQAAAAAygz0fAQAAAABZzdnfz1FUVGSDdlHSobopAKiAWiJKn5osrWsis2fPNm+++aYNnE6dOjX06r+1a9ea6qbj3bZtmyedahS6/wsvvBC7rUBmmD0Zd9ttt9Cv4U7Nun79+kjHBwAAACC9CD4CAAAAALLa4sWLPbdPOOGEKj3fmjVrkt6nuLi4Us+t4OHdd99tXnnlFbvqL6psCKStXLnSc9u9yjCMrl27em6HXZmooHJYWmHp2L59e4SjAwAAAJBuBB8BAAAAAFktTLAwio0bNya9T6NGjSp1nOedd5754YcfKnlk5WlKq9u6devirjIMw3//sN+ff4UqAAAAgJqJ4CMAAAAAIKu5U4CmQroCfHfddZcn8Fi/fn0zZMgQs//++5sePXqYkpIS07BhQ/v72rVrx+539tlnm6+++iotxwQAAAAAmUbwEQAAAACQ1fx7L44bN84UFhaabLJo0SIzcuTI2G0FGkeMGFEhBWmQDRs2mGziT38aNRWs//5NmzZNyXEBAAAAqBl+nGoJAAAAAEAWat68eaX2EMykTz75xLOi8tprrw0VeJRly5aZbP6858yZE+nxs2bNSsn+mQAAAABqJoKPAAAAAIC0q8p+fgMGDKiw8jHb+AN0gwcPDr1icunSpSabKEVsvXr1YrcnTJgQ6fH+76dv374pOzYAAAAA2Y/gIwAAAAAg7QoKCiq9j+N+++3nuT1q1CiTbfypRhs3bhzqcW+88UalP8etW7dGemyU13AHDBVYnThxYqjHbt++3bzzzjueoHO/fv3ScpwAAAAAshPBRwAAAABAxvcRjJJqVMGrPn36xG6/9957Wbf60f/+Zs+enfQxK1euNE8//XSlX2f16tVmx44dJh2GDh3quf3II4+Eetxzzz1nVqxYEbt90EEHkXYVAAAAyDMEHwEAAAAAaVdYWGhKS0tjt8eOHRspcDZ8+PDY/3fu3GlvT5kyJdIxaAXfmDFjTLpSlbo99dRTCe+/adMmc+WVV3oCdWF06dLFs3r0m2++MelwwgkneAKdH374oXn++ecTPmb8+PHmwQcf9Pzu7LPPTsvxAQAAAMheBB8BAAAAABmxzz77xP4/d+5cc8UVV5hPPvnEzJw508yfPz/2s3jx4gqPPfzww81pp53mWTl56qmn2mBXoj0TFdx79dVXzSWXXGKOPvpoM3r06DS8s/IVfg0aNIjd1mveeeedFdKxOoHXM844w3z55Zc2LWmzZs1Cv87AgQM9t6+77jrz0ksvmUmTJpl58+Z5PsctW7ZU+v0obaye2+322283d999t1m3bl2FVKuvvPKKueCCC8zGjRtjvx8yZIg58MADK30MAAAAAGqmutV9AAAAAACA/DBs2DDz5ptvmrKyMnv7gw8+sD9+7dq1syvt/G688UYbaPzoo4/s7c2bN5tHH33U/nTt2tV06tTJBs20F+KaNWtsUDNRYDKVmjdvbs4//3x7LA6lVH3xxRfNgAEDTIsWLWwgUqs1Fy5cGLuPHqP9FL/66qvQ+19269bNTJ8+3d5esGCB/VyCPPPMM2bQoEGVfk8nn3yyDZSOHDkytuL0ySeftKlV9Z5KSkrM2rVrzYQJE8yqVas8j9Ux3nLLLZV+bQAAAAA1F8FHAAAAAEBG7Lnnnua3v/2tueeeeyq1V2FBQYEN7mm149///ne74s6hQKN+kmnSpIlJl1/96ldmxowZ5t133439TisBP//888D7ayXntddea84999zQr6GVkg888IBdyekOYqaLVm82bdrUjBgxIhY0VnA3UbB07733tt9TlBWdAAAAAHIHaVcBAAAAABmjlX6vv/66ufDCC20wUisGFVQMq3bt2navxHfeeccG75IFuBSs69mzp7nooovMG2+8YS6++GKTLnXq1LGB0RtuuMG0atUq7v30vh966CFz66232vdTmf0l9V5uvvlmc+ihh9qVog0bNrTvNdX0nNdff71dwTl48GBTt278Oczdu3e3aVm1NySBRwAAACB/1Spzpi4CAAAAAFDDqEs7efJkm4ZUqT+V2lR7L2qFo9KwKv1ndQTCtm3bZsaPH2/TrCo1qdLBKiC5++67mw4dOpiaSvs9fv3112bJkiU2ta2Cni1btjT9+vUz7du3r+7DAwAAAJAFCD4CAAAAAAAAAAAASAnSrgIAAAAAAAAAAABICYKPAAAAAAAAAAAAAFKC4CMAAAAAAAAAAACAlCD4CAAAAAAAAAAAACAlCD4CAAAAAAAAAAAASAmCjwAAAAAAAAAAAABSguAjAAAAAAAAAAAAgJQg+AgAAAAAAAAAAAAgJQg+AgAAAAAAAAAAAEgJgo8AAAAAAAAAAAAAUoLgIwAAAAAAAAAAAICUIPgIAAAAAAAAAAAAICUIPgIAAAAAAAAAAABICYKPAAAAAAAAAAAAAFKC4CMAAAAAAAAAAACAlCD4CAAAAAAAAAAAACAlCD4CAAAAAAAAAAAASAmCjwAAAAAAAAAAAABSguAjAAAAAAAAAAAAgJQg+AgAAAAAAAAAAAAgJQg+AgAAAAAAAAAAAEgJgo8AAAAAAAAAAAAAUoLgIwAAAAAAAAAAAICUIPgIAAAAAAAAAAAAICUIPgIAAAAAAAAAAABICYKPAAAAAAAAAAAAAEwq/H9ihcyeVmqUogAAAABJRU5ErkJggg==",
+      "text/plain": [
+       "<Figure size 1800x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, ax = plt.subplots(figsize=(9, 3))\n",
+    "ax.plot(approx.hist, alpha=0.6)\n",
+    "ax.set(xlabel=\"iteration\", ylabel=\"negative ELBO\", title=\"Streaming ADVI convergence\");"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "9035a137",
+   "metadata": {},
+   "source": [
+    "## Compare with in-RAM `pm.Minibatch`\n",
+    "\n",
+    "For comparison, here is the usual fit that keeps the whole dataset in memory. The\n",
+    "prior, optimizer, iteration count, and seeds match the streaming fit; only the\n",
+    "minibatch source differs. On this toy problem the two posteriors land on top of\n",
+    "each other, with ADVI's usual mild bias relative to the dashed ground truth."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "b3b5d9fe",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:50.539450Z",
+     "iopub.status.busy": "2026-06-11T15:48:50.539367Z",
+     "iopub.status.idle": "2026-06-11T15:48:53.984891Z",
+     "shell.execute_reply": "2026-06-11T15:48:53.984550Z"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Finished [100%]: Average Loss = 531.25\n"
+     ]
+    }
+   ],
+   "source": [
+    "with pm.Model():\n",
+    "    b = pm.Normal(\"b\", 0.0, 3.0, shape=4)\n",
+    "    xb, zb, sb, yb = pm.Minibatch(\n",
+    "        X[:, 0].copy(), X[:, 1].copy(), X[:, 2].copy(), y, batch_size=batch_size\n",
+    "    )\n",
+    "    pm.Bernoulli(\"y\", logit_p=b[0] + b[1] * xb + b[2] * zb + b[3] * sb, observed=yb, total_size=N)\n",
+    "    approx_inram = pm.fit(\n",
+    "        30_000,\n",
+    "        method=\"advi\",\n",
+    "        obj_optimizer=pm.adam(learning_rate=0.008),\n",
+    "        progressbar=False,\n",
+    "        random_seed=RANDOM_SEED,\n",
+    "    )\n",
+    "    idata_inram = approx_inram.sample(1000, random_seed=RANDOM_SEED)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "6731e9d6",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:53.986146Z",
+     "iopub.status.busy": "2026-06-11T15:48:53.986070Z",
+     "iopub.status.idle": "2026-06-11T15:48:54.193017Z",
+     "shell.execute_reply": "2026-06-11T15:48:54.192668Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 2600x600 with 4 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "bs_stream = idata_stream.posterior[\"b\"].values.reshape(-1, 4)\n",
+    "bs_inram = idata_inram.posterior[\"b\"].values.reshape(-1, 4)\n",
+    "names = [\"intercept\", \"slope x1\", \"slope x2\", \"slope x3\"]\n",
+    "\n",
+    "fig, axes = plt.subplots(1, 4, figsize=(13, 3), layout=\"tight\")\n",
+    "for k, ax in enumerate(axes):\n",
+    "    ax.hist(bs_stream[:, k], bins=40, density=True, alpha=0.5, label=\"streaming\")\n",
+    "    ax.hist(bs_inram[:, k], bins=40, density=True, alpha=0.5, label=\"in-RAM\")\n",
+    "    ax.axvline(b_true[k], color=\"k\", ls=\"--\", lw=1)\n",
+    "    ax.set(title=names[k], yticks=[])\n",
+    "axes[0].legend(fontsize=8)\n",
+    "fig.suptitle(\"Posterior of b: streaming vs in-RAM (dashed = ground truth)\", y=1.04);"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "04dc1901",
+   "metadata": {},
+   "source": [
+    "## Memory usage\n",
+    "\n",
+    "Both paths feed the same `batch_size` to ADVI, but `pm.Minibatch` keeps all `N`\n",
+    "rows resident, so its array grows linearly in `N`. The streaming path keeps only a\n",
+    "bounded number of rows resident — the model batch, a shuffle-buffer fill, and the\n",
+    "source chunks used to build it — independent of `N`. The dense `float64` design\n",
+    "matrix dominates the cost. The lines below are array lower bounds, not measurements;\n",
+    "they also ignore framework overhead, PyTensor's resident copy, and the transient\n",
+    "copy `np.concatenate` makes while filling the shuffle buffer:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "6f25aa9a",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:54.194247Z",
+     "iopub.status.busy": "2026-06-11T15:48:54.194176Z",
+     "iopub.status.idle": "2026-06-11T15:48:54.488206Z",
+     "shell.execute_reply": "2026-06-11T15:48:54.487819Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABiAAAAPICAYAAABQMDPFAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjksIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvJkbTWQAAAAlwSFlzAAAewgAAHsIBbtB1PgABAABJREFUeJzs3Qd4leX5+PE7e5BBErLDhiQIKLgXYt2KqFgHigJatfivtVp/v1prtcP+tK2jWkdttZWl4gIVce89QQEhg00mJBBC9vxf99OeNAln5+x8P9d1rpyTnPc9T97znve873M/z32HdXd3dwsAAAAAAAAAAIAHhXtyZQAAAAAAAAAAAAQgAAAAAAAAAACAVzADAgAAAAAAAAAAeBwBCAAAAAAAAAAA4HEEIAAAAAAAAAAAgMcRgAAAAAAAAAAAAB5HAAIAAAAAAAAAAHgcAQgAAAAAAAAAAOBxBCAAAAAAAAAAAIDHEYAAAAAAAAAAAAAeRwACAAAAAAAAAAB4HAEIAAAAAAAAAADgcQQgAAAAAAAAAACAxxGAAAAAAAAAAAAAHkcAAgAAAAAAAAAAeBwBCAAAAAAAAAAA4HEEIAAAAAAAAAAAgMcRgAAAAAAAAAAAAB5HAAIAAAAAAAAAAHhcpOdXCQAAgEC0fPlyueWWW2z+/a677pLzzz9fQk1bW5u8/fbb8tFHH8n3338vu3fvlv3790t7e/sBz83NzZV333235/Hll18uX375pc11FxcXe63doYTtCF8arMc6AP5TUFBg829HHnmkLFmyxKftAYBAQgACABAwTjrpJCkvL3f4vNTUVPnggw8kOjra5df4wx/+4PQFwOLFi+Woo45y+TUABI73339ffv3rX5ugAxCo322nnnqqPPTQQx4L6LzzzjuSl5fnUjsBAAAAbyAAAQAIOnv27JFXX31VzjvvPJeWa2xslBUrVnitXQACyyuvvCI33XSTDNYR4PY6vX/605/6tD2w76233pJ169bJ5MmT2VSADzz44IM2/6Yz4ZghAwCA5xCAAAAEpSeffNLlAMSLL74oDQ0NXmsTgMChMx5uvfVWGaw02GpvhDwBiMDzl7/8Rf71r3/5uxnAoGBvxpGmyyEAAQCA51CEGgAQlNauXWtGi7oatAAwOGiqtZaWFn83A3DaJ598YjdoBAAAAAQjZkAAAILW0qVL5U9/+pNTz/30009l8+bNXm8TgMDw4Ycf2v17TEyMHHvssZKRkSGRkf89JR46dKgPWgfYngXx9NNPs3kAAAAQMghAAACCltaBuPnmm01RameCFcBgpyklBkNaic7OTikuLrb599jYWHnppZdk1KhRPm3XYJ+RAsdWr15tCqefeOKJbC4ACCL2zjsAYLAjBRMAIGi1tbXJ888/7/B5WohVO3QADA779u2Trq4um38//PDDCT4gYN1///3S3d3t72YAAAAAHkEAAgAQ1JYtW2ZGO9vz1FNPOXwOgNDR3Nxs9+/Dhg3zWVsAV23cuFFee+01NhwAAABCAimYAABBIyws7IBRoTq74b333pNTTjnF6jKtra02Z0lYW58nZ2ds2LBBKioqpK6uTurr603aF80vn5mZKYcccojEx8eLL3R0dMg333wjO3fulJqaGhkyZIiMGzdODjvsMImOjna4fFFRkSn4XVtba/4HTXk1ZcoUGTFihEfbqe3T6eu6vXQEu7Y7OTlZkpKSzGtNmDBBIiIixBf0tfV/Li0tNe3R91PfL01fRI2A/9L9esuWLbJ9+3bZv3+/NDU1mW0XFxcnCQkJkpWVJTk5Oeb9i4qKEl9x9LnWz76/VFdXm+PWrl27zGdKC2XrcUq3j342dbsNHz5c8vPzzWO4Rz+3a9asMdtbjyd6HNF9UY97uo0D3V//+lc5/fTTfXbMg0hjY6P5rtR9Zs+ePWY/0e9r3WdSUlI8enzSmlR67NR9U/dVpa+RlpYmBx98sPnpK4H8WdH3QbeTnh/od4wGl3V2m37HJCYmSnZ2tuTm5ppjZnh48I2t1P9LzzUqKyvN/6af96OOOkoOPfRQp9exY8cOKSkp6Tl30gE3ui/p7aCDDjLvpafod1ZZWZn5/tLzSf3+0pu2W7+v9KavV1BQYM7ffEU/T3r+qNtRP1+W/Vc/v47o/2E5R9Z9S7fb6NGjzXlu7/pQgUjfew1Y62dX3wdPt13X+eWXX5rtqvuXnv9r3axAODYACD6BfUQFAKCXY445xhSTtlbfwVYA4pVXXum5uO9NL5aOOOII+fzzzz22jbWT+uWXXza55b/77jvTqWiLdjZqEOLCCy+UmTNnutTJdPnll5sLAlveeecdycvLMxcOf/vb30wARi+w+tNOhrlz58qPf/xjq4GIF1980Sy/bds2q68zfvx4+dnPfiannnqquGv9+vUmN7y+r3pBa48GAPSiXLfZaaed5lJnw0knnWQ6fa3Rzot33323pwPqH//4h5lZY22/0X3wiiuuMMElazQ4ocWPtcCxM3SU8w033GDz77/+9a/N++0py5cvl1tuucXm3++66y6HNSI0yKBFct944w1Zu3atU0E83d/HjBkjkyZNMumP9LOnHUaepB0ezlqxYoW5WbN48WLTAeQJuq00/Zvm9ddOpk2bNklDQ4NTy2qQZOzYsTJjxgw577zzHHYi/fKXv7T5P7mzvTydy9rRccve6z344IPy0EMP2fx77/dMj+lPPPGEfPLJJ9Le3m51X9SOff3ceXof9KStW7ea9/OCCy6QQKHHh4ULF1r9mx7zvv76a5tBbT2eHn300VaPFyeffLI88sgjbh+39HvXlc9/f9qBp8d9HcxgbfaUftccf/zxctNNN0lhYaHbr6P7pB47v/rqK6vfL71pAPLMM880nxvtaHdWKHxW9u7da87r3nzzTdOx7gwdHKEDKyZPnmy+X/R7pn/ns73zgP70WGVvnzryyCMPqGvjyrZ/++23zT7//fffH/A8PS9zFIDQ7159/c8++0x2795t97kjR44058h67pKeni7O0n3igw8+MN9f+nqWARnO0vNQ3YdnzZplvsuc5ejzft1118lPf/rTnu3497//3bTP2ud2+vTpplacdsr3p/uWprvT72hrs6S1g/2yyy6Tq6++2unOdlf3GXe+I7Wt+t2g56l6XmGNO23vTQMa9913n/kM6nmMteO9fp4WLFjQc0x0tP87c34JILQF3zABAMCgNWfOHKu/1wswHf1kzZNPPmn193rirCPnPOWFF14w67z11lvNBYS94IPlwk47a/TC6IwzzjD/gyfphZVe9D366KNWgw+W0et6sXDllVf2uajUDtJrrrnGtM1W8EHpxaheCN5xxx128+3bGrWlF9k//OEPTaDDUfBB6UXQxx9/bIIeGoDQC2NP0s7hc88912wzexfZehFviy6nxdGdtXLlSrsdKtqeQKLbXzsL//znP5sgm7MziHR/14tn/Zxox8K8efNkMNBA1Y033mg6Hb799lungw9Kt63ukw888IDZ3/Wzaq2TEP+mAVfLvqUdSra2lf5eA9P62fL0cdddtkbpPvzwwyawHSiOPfZYm3/T7zw9Jtii33e2jhf6N3vfIfY65TSdmrvBB31N7YDUTjE9bttK3abP08CyBoPsHbNt0c++7m/6XfvWW2851Ymr3+H62dfzikWLFslg+azo+YB+x+jxztngg+V/0gENGuD5+c9/Lv/zP/8jgUg/z3pu9ZOf/MRq8MERnQ2iHcs6EEMDb46CD0pnKP7zn/80QYi//OUvTqck1dkO2k5d1pmgWX86U+Kxxx4zQXQ9T7TWkT2Q7ajn29o+a8EHy+dWg4q6rfrvv3ptoJ97HbBja3vo97WeD+ryOgMgEOg2vfTSS83/biv4MNC26yAqDRzpZ9HWe6bHex1Ao9vw8ccfd/n/ADA4EYAAAAQNvRDX0erOBhp01JatCzxbwQxX6cm5Xuj+6le/cupC0FZn/I9+9CNzkecJOo1cO8n1QtUZemGpHSN6EaYzALRTwpXOfR2paG/0an+vv/66CY588cUXMpD/UWdu3HPPPS4HP6zRQIuOPtP1OnLWWWfZDV5pzRFnaKoE7dSyRUee6iyVQKGzVHSba0oM+JZ2BOroQg34Uc/mQJqaRQOaOnLWWXqs09Gbzh4nvenaa6+1+ntN4ffMM89IoNCR5fZSqdk7ptsLIuix0N4MGHvL6qwKd2iqOB3Zr7P8nP0O0c/h//7v/7r03aXf63q+oakM3aEDBe68807Tqe6otk2wf1a0Tdo5r68XinQ/01k02rHrDg046KANPWdzhwZptFNaz/d8+T2ugUc9T9TAiSeCELodf/GLX9hMr2ptn9eZAJZAhc52+v3vf+90QF/3ez1HdzSwyNv0GHLxxRebgKazXG27ft+48hnU85G7775b7r33XqfbBGDwIgABAAgaOp169uzZVv+mF3T9Rxfbmv2g0/Q1lc5A6Ym3dmC4MyLS2rp0VLmt9Bau0BHXtmY92KKBGn1tHRWpowhdpRe1mjLEER1tpu1zZSS4vYtaHV2n222gI+l0Or+zF+SaV9deWiS9yHVmG2ogxt4F8EUXXSSBQtupI+600w7+o6OVudA/0G233WZ39L29Drnbb79d/E1H1tuqqaPHVk90PHuCpsHT3OK22OsYddRpaqtTX0f82kubc9xxx4m7NTY0jZw73zt6LHQmEKipYfT7yRPHzVWrVpngx0DrVgXqZ0W/f//v//5PQpmer2hKG3foeaZ2unuiA19Tb2lAydezq/Q8015qJWfpLBcdfe/qOYzuvzoDyZ3vUJ1lrZ9nf9KBRa6e27vSdj1G//a3v3XrGKNBHVffEwCDDwEIAEBQ0enE1vLr62gdnTZsobMRbHUu6PRlT9B8pp5OA6SdFQNNdaD5k92hF2XudMhYLu70otAeTdmks0U8MWOhN81h7cpozv50X3El1YPSUWj28uraCn71pqktbNF6CZrDOpBmP+hobPifpnPS/Mz4r4GM5tVOF3dHp3uKzirQ2S3WaIeT5o4PFPaC9zoy11qnpo5AdrSNbc1ysDf7wVFaKHvcnbGodKacpnaxR79LNd2NJ2nnqb0c68H8WdGAvCcGJgQyrbXhDv1c6SzbgQafetMglD+Cr/o+20sd5M1zXK31oik83aU1F/yZBnEgn11HbddrKJ35MJDzc1upcAHAgiLUAICgkpKSYvLJWutw1k5fS2olnUZs7WRbO421oOtAaboInVJujxZm02CHjhjVdusoVkvR5TVr1lhdRkdV6ihAnWrvSpFlW6kpdOq1FovWDqBnn33WbgG83iM6tUNs/vz5Zlunpqaaadw6zdpezmId/aQXyfaCK45G72mnu07TP+igg0wxU81drNtC3297F9/aNq2loSN0B2rIkCFm22nRTX0PNH+uBk80H3/v/UiDYRr8sEbzievFnBaltkbXaW9EsK47kNgbMav7in7uTjzxRFN0Mi4uzuzrekGrHXV6Uar7vTs5pF3RO62avra9NBca4LHVkWorH/9A6H6gxwEtkDpq1ChTUFqPCVrnQ/dzPVbpttHR3poPX/d3TUtjjXbwapovnUnUm/4/vfd/7ay0V1vFU2noAol+dvWYp+n69Lil+59+RrVQqT167BpIcWFPmDlzphkhrccaa2l8LrnkkoBIyaYzDnT2gK1R8joDrH/wVD/7jjq2LHUg+n/v2QtAaGHZrKwsGQg9fumMNs15rp99De7p+YOj9C66z2hOfWs03YmmTbL3naW1K/S7Tj+3mtJPv3/1e1aLy9oLTutIZk3D46gofbB9Vux9x2hb9XxEg036f+sxU88l9LxGzxF0u+nymnbTXhBDz/16fwfZGyiQkZEhp556qt3CzgMRFhYmkyZNMjf9XGtATI//1raD1k+wN1tBz0f0XFMLLmuaUv0MaWpPPQ/R8z5bM3B0X9PjyiGHHGK3rbr9Ld9f+pnT10hLSzPfXzogSNevqcJ0kIK+B/r9Ze+7R2fbemImn+63OntVP0PaFj1W6OfO3nlG7/NcLQh+1VVXmXNk3W+ee+45u/VWtCaGvoa7s648RWfLaZ02Pc7q/63n5Vovxt5AGkdt1+OdowLteh6j2+uEE04w9/U91hk9us38nZ4KQHAgAAEACDracWYtAKEdnTp7QE/KbeXN1toDejE1UHqyb69zQdv461//+oDOFL1o1Y4OneZsq43aAaXpFrRDyl36Gjr6Ui9ylXZwaHs0TZIWErZH26z55n/wgx/0/E47ZixFDG1d4OvFiHbeWOvA1RF89uodWLaZpoewtNnSbg0G6AWPdrja6sTSkWEaENKLsoHQ3Mg6EtnaPqLbTjuNLbRWhgZ0rF3ca0ec7qO6Pmu0g8nW/qMdYp4IknmSvWn/mtPaVmFu7aiydOLo/6ujZ3UEpK2ikQPRezSnduTYC0Boh4svRn9qEEtHr+sxKSIiwu5ztSNVt5d+xnR76j5ga8SjHuf6ByC0UGzvouV6HLHXCRQIqYc8STvE9BigwSULPRbpttdUefbSQwx0RK4n6HFXR+damwmhwSg9/vZ/z/1BOyETExNN5681GmywFoBwxFIHYsKECU4HIAbaEajBUt2uhx12WJ995uCDDzYdbPaKq9pLs6ezAauqqmz+ferUqSaQkJyc3Of32rl+/PHHm45kTbdkjQYrtW6FdkqH0mfF3nfMH//4RznttNOs/m3ixIl9Ope1A1/bb+3Yef311/d5bC8AoYFibx0j9TP0hz/8wWogR4PnvTuCNSBkb1/TdmrgqH9ASt9PrdmitaS009jW6Hc9l/3Xv/5lM7Ch+6kGfjToY4++nnbk676rr6cpJG3VDBlI/a/ewQcNruh3rIV+/2mKTK2X4kwtr/vuu6/P+aYOoNGOensBQN3//RmA0GOHvl+9BxtYzpO1ILQG5Fxtu6VGhz36Gvp56V2HT99z3Zf1HE/Phz1ZZBxAaCIFEwAg6OiIMT0Jt0ZPom2N/NULDU+M+tWRXpqL3Ra9ILYWfLDQ32sOae0EsEUDEO7Si0YNcPS+sLLQ0Y6OaOCjd/DBQjvfrf2+N1t1IBz9P/n5+ebiz1qblc5u0LRH9gxkm1k60nXWgq0AlY786z2jQS/ItF32OqFsBRnsXeDqxZxeXAcSe50Pzo7E1fdWOxe1E9XWzJFQoxfoOsrSUfDB2nL2Usts2LCBEYf9aFHR3h2qvelIb3ucqV/jC/rZ1w4dazSQpZ1j/qb78pFHHmnz79aCDdaCCBrEcNQx6aj+w0BrOemxqHfwwZV9Rjv6bAXEddaeLTpiWWeQ9A8+9HbOOefY/a7WIO5AitEH4mfFE98xum8eeuih5vwqUGvl6PmhfpZtzSLR8w89H3JmX9LvVB1oYm/7aMe0rdpplvSKtgLdei6pMxsdBR+sfbbtzR7R2R46Q2MgNMjQO/hgoe21dR7Zu322zpF1dlGgfldo0EEDRtZm+up75egc2VbbdeaEo/fjd7/7XZ/gQ28asB1IaisAgwcBCABAULIVSNC8zI888ojVv+nIH+1EHigdfWzv4l8vAhylT9Jp67Y6PiwdMe7mmtWRgrZS/9jqdHC2+HHvC2NrbI2KdZT7eO7cuWbkmj22Rtlb6OhZd3PkalDr6quvdnk5WzMclF7QffTRRwf8Xkem28ujHUjFpy00tZIt999/v1dmNIQaPWZomhntfNSLdR2tOW3aNHMc0JRjBQUFfW72glR6bHCnGGWoGjt2rM10OEq3p71OKQ0qBwpbsxx0dKkWpA4E9kYAa3rB3t9dOmNOc6/3pu+Fph9yFKiwN/tBO5u1c9VdmvZG08/YokFge+nYNLhs7ftOv4M0QGiL7qea3scRewFI3V/dnYkQqJ8VW52bStNS9k6BGMw0RZCzqSL1O8NeTTBNi6TfHQPZl3Q/drbumKb60WOQDtbQ0fY6M1VnWej5U//vL0dFj+3NEHJm5pKtGcIawHGURlFnhdgKADo6x/Xnd4WeM9j73/rPHnO27XpeYo9eN+nMFnv0vNUTKVABhDZSMAEAgpJeQOi0/P6dcHrBZi2Htidznms6IXt02v5Ap+5rZ5NOX9cLOVfpiGtbeqcQspX+x9bsEmUrsNE7fYC13zkqTqdpJxzR9FU64k1zVdu6kNaOcB0B5yrdNxyNmrM1mlFHA9vqKNNc/XqR3tvKlSttrk//v4F0qnmLdpRr0XVrdD/VmhW6b2iAS/8HzVGs9y03V0dQhhJNx6WpujRtwkCKSFrrTLDXaTeYOJqZpfufjhC1FSC1dtzyZ+e+rWOKFhLVoKfOvvInezMP9LtLU8ZYvke++eabAwL2+r2mgXJNJWSvDoS9AISOutX3dCDb2dFxSb8v7RV81/2mf0emfgfZS8+ogUV7wUVnaQBCO6BD5bOi35O2alRpXQGtR5Wenm46Q/U7Rs8H9L5+v2gaIkcDGAKBfq5dqZ+h3632Orw12OfOOaK1fUm3rzWaYlLTSWrgwVGNAFcMpCNfUyjqTCJ7wUN7AQ57M7jszUzy93eFo8+uo/N7W223VzvC2ZlmGnzQ45HOqAEAWwL/mxoAABsXyTrixtZsB2sjuN3pmLbGk52I3ngde7McdOaFPXphby9djKOLfGspKRz9HzpizdkOtXHjxtkMQDjzWrYMpNNfU1bY6ij74IMPTBoRywwC7Ziy1/mkHfnuBEK8TUfNaropTf1hixZ+1I4ivfX/rGpnoV7E6gg+a2kTQpUWG9f9w1EAzh32iq0ONs50wmmHla1OVXsdxv6aBWFtdL4Won344YdN/nh/0u8YPWbr/m0rDZMlAGHt2KijprUjVmch9O6I7F8Hwl4AYqDpl5zdZ1z9vvNVmix3XydQPysa5NbvCXuz6TR1j9767xfa+an7mwaVdGS8MzNM/MHe4BBr/L0v6WdzwYIFJogYSN9fjmYyOzrP1YCVvUE49vjzu8JR8MrR/22r7faKditnZ47r9wIBCAD2kIIJABC0NK+ts6PetDPHUVqkUAlA2BsV6ujiSjuEPG3v3r0ee01recMHus10Or+zOaat0anp2kFvq4NKRy1baOe8rVGEui9rWoNApZ2emmrBVdppqiObtbC5pv746U9/arPjMpToSPDLL7/cK8EHZSv//GDkaGaWCoYR0haax95WwHzFihWybds28Td7AYDetRysBRG0I1a/j62lIbQs66j+g720Mv7cZwL9/CBQPysaeNf0dO4EqPVYq2ke//znP5vPzW233eawU9UfbJ0nBOK+pN8v11xzjVeCD5b1u8vReaCj81xHywcqR59dR/+3LbaCjRbOzjQbyIw0AIMDAQgAQNDSXKj2chn3Hs13wQUXSLBpbW11a7mBpLsZjKlyBnoxqh0n9upTvPDCC6YT3lH6JZ1erykmAnk7aWFtHZ09bNgwt9fz5ptvmuChJ9M5BCJNL2Nvtg48x9HIT+WpALSv3HDDDVZnQ2lKFO2o9Td7AQANtGo7NeVH/3oI+j8dfvjhNlOhWAIW9mY/WNJ9DOZ9xvKdEkr/t86qefHFF02aMXe/lzXd17PPPmuCv4FU28VbAzy8tS/pNtQUT4FooOepwXqe6+iz6+7sWUfbQ4/l3jwmARg8AvesCgAAJ1grZNmf5rZ1ZtSfszS/LJznKC+tK50EjkZqufPeuDtqrDdNLZSWlmZzdOGrr75qLuLspTDS9EuBTi9UNSWDppb6xz/+YTqKNB+zqyPfND/zHXfcIaFKUx3oSHV7dNaNjtTVfUPrymjqmd63WbNm+ay9CDyahkjTnlmj+8zWrVvF3zMgbHV46Yh0DTxYAhG9jR8/vuc7wVoAwlIHwl4AQlM4eeK47Q2cHwyMfpfcfPPN8vHHH5tZc1qfSWst6UxFV2he+wceeEACiaszS/y5L+nACXv0nFqLUb/00kvmc15UVNTn++u6667zWVsxMI7qXjg7E6d/TT4A6C945iIDAGCjI0JzGusFz0CCFJ68KHzssccOKDw8mDnaXjpKVjuks7KyHK7LUTobf12wa8e8dpTYGpmsMwf0gt1WOirtjNYc2MFCO1I09ZTeLDQ3tyVtihbP1NQNmk7FVt7h9957z6Ri8ndBXW/Q45FuD1t0pot28NjbXx0F2xD6fvazn5kZQ/2LOOtnyt7+5Qs6Cyo/P9/md69+9q0Fl3sHHTTIoiPde+/rljoQ9gIQA02/5E22AtEWV199tfzP//yPz9oTrHTmqhYq15tln9fzBP1+0e8ZPRfQfey7776zuQ49xmowI1hHvDval8466yz5y1/+4vHX1c+tFqa2RYNBmlrSXm0Avr+Ch6Nz740bNzq1nvXr13uoRQBCFTMgAABBTzt+bdHChAcddJBHX08LJdpDEbYDi0w7yn380UcfOdzuO3bsMDdbdDTu5MmTxV+0zoitoqU6wv3++++3uewPf/jDgE734QztVNfP29lnny3XX3+9LFq0SP7+97/bXUZHO4ci7SizR0e22ws+6Ahwex1rGBy0g++8886TQGUvEKCFqPVmLwARERFhtQ7E8uXLvVr/wZv0O8heKhTOD9yj21SD1Zq+Sz8TOvpe0wT95je/sblMc3NzUHeK6uffXioqDdK1t7d7/HWrq6vtFlvW2U+OChPrOQ+Cp+aQPTobqaWlxWGQQgeeAIA9wX2lCwCAiJxzzjk2pxBrHmBPO+6440zHiS3PPfecww5IW/QkX5fftGmThBLdZvYsWbLkgFG+/WmHtj06E8bRiEFv0g5le2lzbI0i030p0GuUuNvJoTMk7KU/8/cobm9xNPrTUa5kzYM+0G3jKGWJo+LwCAyayiRQ0w3ZK0StwcX+nb/aiayzFnvr/1g988wzdgOdOvMiUOl3kL1BD99//728++67bq1bO4U1/d0777wjocbd/PGa/tAee8dRe8fIQChirTMN7X3GNOWNzkRwl85S1O8aT39/OZqZgsALQNj7jmloaJBHH33U5t/13P2uu+7yUusAhBICEACAoKcXkVoYV6ej977pKDnL9H1PFxK0ly5HT9Y1zYKzBWi1U0FHi/3pT38yHba//vWvpba2VkKJ1uGwR1Nu2LuAeeutt0wao4G8hi/Mnz/f5ZkMui85k37KnzTH8+mnn27qPrhSWFnzcNur8WFvlGUwc1RsVNNP2cqrrKNaPVEfw1ERV2sdTwg8mp5Ni7YHInu1GDS1Xv/A5bhx4w6Y+WOtDkRra6vN1zz66KMl0Dn6LvrFL34hn3zyidPr2759uzz++ONm5tQ111xzQGHvULBq1SoTwF+6dKns2rXL6eUczaKz9x1j7xhZWloaELMnHO1Let6odRhcmd3w5JNPyvnnny+XXnqpfP755y7XBNDzgW3bttn8zv/5z3/udHvgf/o5OPPMM+0+RwMQev7Xv6aPpszT91uDTgDgCDUgAAAhQdPf6M1XbrjhBjMS0dbFrV6EaRBE09FoUEELb+pJvo4U0hN2DTBop7uOhtQT91AdCW4xZcoUUxfjww8/tDsLQgsZ6qwVHUGqHVuacunll182uZw1LY0t2qnl6Vof7hg1apScdNJJ8vbbbzu9TDAUn1ba4XDvvfeaW15enhx11FFmJLKmYsjMzDTFQzUYqKMjNX2Kdmzoe2rvfQv0wIu7HKUc0w427fzRgt6TJk0yxwbd11955RV5/vnnD7jId4ejFBl//OMfTTFj7RTuPxJY90nNz4/AoPuJHgO1uHMgiY+PN2nX7NVrcDTbQQsMa5o+DVh4YjZdoKSFXLhwoc2OdB1hfuWVV8rxxx9vggr6faczJ3SGiP5NR9/rLEhLIW/tDA91ei6l/6/eNACrxyXdX/TnmDFjTM0R3U80zaGmVtI6EHo+oZ3p9tj7jtFjpK33SNuj55QaIMvNzT2ggLTW8dD939s08K+fET1XtEaDfBrQ0gEaM2fONN8nGRkZpr0a/NebpsbRGZhr1qwx63EU+Nfv95iYGJuBQD0OzZ07V3784x+bFGpaVF4DG2+88YZ5P/T9QXDRwTMrV660uW/o7/XcT2ciaxo0DVLpe67XL7zfAJxFAAIAADdo55yOSrU3Kl87YjWXtd7w71GfOlrRXiearbzhjviqM8AZV1xxhdMBCE0ncuKJJ0qw0c4fvQ2EdpAEw2hmd2gHjgYd7XUcbt261RRI9VdeZ7V27Vpz60873QhABA7tfNXArKOaKv6gKWKcDUBYm+1gqQNhLzgdLPUfLLST/JZbbjEzMx3lVtcbDqQBmIGmotT0f9p5b4sGz+yN3NZzOFvvj6ZG88U5hwalfvWrX5kOYnupEDW4oDdP0OCDfq7ff/99m8/Rzuff//73Hnk9+J9+TnQQjw4asUfTfr3++us+axeA0EIKJgAA3KSpkuylYkJf2iF7zz33eLzYsnb4axHnQKGjww455BCnnqvt7j+ycrDQ2UH+rNnhbZoqxV0amNKZUwOhwR0duYvQcNVVVzlM7eUPrsxIsBaAsDUzwhodCa+zrYKBzoDUmZLw70wUe7ntNQ2RvYLhgXRO8X//938+betAvr80MKP7P4KLDuSxV7/G0TlLoNcyA+B/BCAAAHCTdhz/9a9/NR2pcM7JJ58s9913n0nXM1B6Ma6dcjqzItBoUMSZ9g/WCzbtSLz11lsllGk6DHdq0Gg6pAcffPCAXPmu0kBfqG/jwUSDD5q2J9BoyhdnAiOalsxWwFHTuYVK+qXerr32Wvnf//3fgC0iHsq00163vz0jR46UefPmSTDQYtuaNs9XMz11VpLOunDne0frUuh3PIKLztx64okn7M4askaDwpqaSWfq2RMdHT3AFgIIdgQgAAAYAL0Y1LyoOrLfE/nsNSfx9ddfH9LpT7TY3YoVK5zudLJm+PDhJh2Jdu54ekaFJ2jHs6bhcZRKRP+PYKAXjp7azqeeeqo888wzATma25M0wHT33Xe7FITQoIMWm9XUIJ4K+GkbNHc6gp92lgbarCFNoWRrZoOzsxy0w8uZjlVNCxNsNEi+bNkymTx58oDXpccHHdXvqDBxsHZ+eoJ+T1188cWmI9WZwI+eQ2hgLxhmQpx33nmm4LQnAnE6CMRSiNoWTRGo+5uz9DP8wAMPuBV4R2DQtGWahkmDT3psd+YcQ9PMaoBZa9fYM9BBFQCC3+Cc8w8AgBdGO2vH+ltvvWVOxjUXrxaTdEQ7BrWzUTvjNZ1TKAceehsxYoQsXrxY1q1bJ0uXLpVPP/3UZjHI3he3mtdeZw3oBa4zF0f+om3TzkJNmxDsxaeV7qOfffaZuWnObH3ftNC65sh2hhbF1JRCF110kRx88MEyWGinms5mePHFF03ATIuB2tq3tXPppz/9qccv0s855xxTZ0SLuet7p4Xe9+7da2qxdHZ2evS14F26n2jh1zvvvDOgNrUGUx3VvbEXcNbZhHpst1cPQZ8zkKC1v2eJaHF5LSitwQj9HFZVVTlcTjvQ9ZxA06npNtYgTqim7NOUPTprQc8FtKbI+vXrZfPmzdLR0eHU8hrM1+OcFo7WzlBn6fbUjnbtiNfjtJ676etq8WYtruuoYLM/zp3+9a9/me/fp556ynxmdu7c6dQ5iRb01n1JA3m6P2mtB0fBnNtvv11OOOEEeeihh8z3vq39VItla8qxYBlUAfvXJVrDRj8TWpha6/NUVFSYAIMGKHSwle5Het1TUFDQs5x+Zu0JtOA5AN8L6w60b1UAAEKAfr1qAUXtcNQL2X379pkCgtqBpCPPdMqyXiRnZ2f7u6kBY8eOHeaiWjtHdXtp56iOkk9OTjYXtZqbNpCDDv3phZvm1LVGO5k/+OCDoJ6Srh1D2vGhF6bamaYBt5aWFrPvaxohvYjNyckxs3r0J8R0bGknTm1trTkeJCYmmuOABngcdQYBCB2VlZWyceNG06mn5wgaELQcN/X7QVPYaGdzqAYcnKEBbj0vKC8vNwMUGhsbTVBAZytYtpWeG+h3jAa5Byv9PtHOXz130n2poaHBfJ/ouaZ2GOv2GTVq1IDPN/R90CCaFiLW90G/vzSNlQYPPZFWE8FLz901TVhXV5fVv+u1jwZeg/mcF8DAEYAAAADwAp3hsHbtWqt/05QPOuoSAAAACIRBEhr8dCUlmQa8dMavvRkQmnpTZ9EAGNwG75AKAAAAL9AZAFqc3FbwQdMa2Mu7DAAAAPjSb3/7WzM7S9MradpMTZlpq/6XzlB699135b777pPt27fbXa/OjgAAAhAAAABu0vQZGmxQOvVc0x989913UlZWZnckGHmSAQAAEEg0teYjjzxibprqTGdE5OXlmTRbmhJOU6FpOq7i4mKTOs6RI4880pz3AgABCAAAADfp1PMnn3zS6efrxdvPfvYztjcAAAACltb6+P77783NHVqDRGdVAICyPp8KAAAAHnfVVVeZosMAAABAKNIi5f/85z855wXQgwAEAACAD2g+3euvv55tDQAAgJB07LHHyooVK2TSpEn+bgqAAEIKJgAAAC8KCwuTiy++WG699VaJiIhgWwMAACCgnHzyyVJRUWG3jpktWqz6mGOOMee7p59+ulfaByC4hXV3d3f7uxEAAADBSC/S9IKt/0WYTj3XQtOHH364nH/++VJQUOC3NgIAAADOKC0tldWrV8vatWtl+/btpuh0fX29tLS0mHNcLUit57np6ekyYcIEmThxohx99NGSnZ3NBgZgEwEIAAAAAAAAAADgcdSAAAAAAAAAAAAAHkcAAgAAAAAAAAAAeBwBCAAAAAAAAAAA4HEEIAAAAAAAAAAAgMcRgAAAAAAAAAAAAB4X6flVAv7V2dkpdXV15v7QoUMlIiKCtwQAAAAAAAAAfIwZEAg5GnxYsGCBuVkCEQAAAAAAAAAA3yIAAQAAAAAAAAAAPI4ABAAAAAAAAAAA8DgCEAAAAAAAAAAAwOMIQAAAAAAAAAAAAI8jAAEAAAAAAAAAADwu0vOrBHxrxowZfR6HhYVJXFwcbwMAAAAAAAAA+BEzIAAAAAAAAAAAgMcxAwJBb9WqVX0e19bWyoIFC/zWHgAAAAAAgMGkrq5ONm/eLNu3b5ddu3bJvn37pK2tTaKjoyUpKUny8vJk8uTJMmrUKLdfo6mpSb7//nspLi42fT8NDQ0SGRkpCQkJkpKSIiNHjpTRo0dLTk6OeEN3d7f5/7Zs2SI7duyQ+vp606b29naJiYkx2TgyMjIkOztbxo8fb3468uKLL8p3331n9zlRUVE96x4zZowcfPDBMmTIEI++dw888IDNv2umEf3/dDvr/5Sfny8TJkyQiIgIt15vz5498uCDD/Y8jo2NlZtuusm8l8769ttv5aWXXurzu8LCQrn44oudXofup3/729/6/C45OVluuOEGp9cB5xCAAAAAAAAAAOCyyspKMzC0vLzc6t9bWlrMTTt7V69ebQIQ5513nunodaXj/+uvv5Z33nlHWltb+/yto6PDrL+mpkZKS0tNZ/2vfvUrj7+TGvj44IMPZPfu3Vb/3tzcbG7auV5UVCTvvfeeCYocffTRcuihh7rUud6fBjj0pgGPTZs2yfvvvy+nn366Wa8v6Pa3vI+6ndetWydDhw4176MGfVylwYPedL0aVJo4ceKA2qnvvwaE4uPjnXq+o8APPIcABAAAAAAAAACX6UyE/sGHtLQ0M1pfR+1rwGDnzp2m81xt27ZN/vnPf8oVV1xhOuid6fxeuXKlrFmzps+I+eHDh/fMAmhsbJSqqirZv3+/x99B7fh/+eWXZf369X1+r4EOnWmhswJ0doD+n9qOiooKM/ND7d27V1577TUzY2L27NkOX2vYsGFmBkf//1/Xp4EPDfYofazbpLOzU4444gjxtEMOOcTMXLHo6uoy/5vO+tAOfsusiaVLl8q8efPM7BZn6f+zdu1aq8GAgQYgdHvo+3TkkUc61Q4NpMA3CEAAAAAAAAAAcFtqaqpMnTrVpAfSlEv9O3t11Lt2xmuHvgYKli9fLldeeaVJ72PPW2+91RN80IDDqaeealI5hYcfWNZWR+dv2LDBY++idmgvWbLEBFAscnNz5YQTTpCxY8daTUGknfVlZWWmzdrBreuwBCQc0XWfddZZNv+us0heeOEF89OybQ466CCPpmNSJ554opnh0J/+L5999pmZiWKZffLKK6/Ij3/8Y4fvo4UGoDQ9l9JZIboOpTM7NKWWBnTc2fc0IKLbXoMbzgQgNChkCVilp6fbnNkCz6AINQAAAAAAAACXaYfxueeeKz/5yU/k+OOPPyD4oLRzWoMTs2bN6vmddtJrzQh7dMS9dnhbXkdnTejofGvBB8sMAg0OeIoGTHoHH6ZNmyZXXXWVqYFgq/6Btm3EiBFmm/zsZz8ztRI8RWeVXHLJJT2vrcEcTffkK/q6+h5rWimL6upq8146q3faIw0kWepk2JoZ4QydaaM1N5TOxtFAlCvt0KAZvIsABAAAAAAAAACXaU2HKVOm2AwK9Kad8TrKv3fOfkcBAIsZM2aY1E6+oiP1v/nmm57HRx11lJx00kkurSMxMVEuuugiM2vDU3RmgqafsvDHyH3dFr1pYW5n6EyQjRs39un47935P5CaDBqYcnY9mi7LErjJzMyUrKwst18XziEAAQAAAAAAAMDreneea9ocW3RUvdZ1sMxsKCws9Om78/HHH/fp9D/llFPcXpdllL+naGDDQmdB+JpuD63DYeFs7Q0NPljSUWkRci1g3TudlqaWstS5cJXOStGZEEpTX+mMCls0TZdlu/UOXMB7CEAAAAAAAAAA8Cl7ncS90/F4Mo2RM7R4dO/0UIcffripVxAotFaChTs1EzxBi3BbWOo4uJp+SVNzaf0Kradh7TmupoeyFLHWGhM6g8XRvqWvr+2A9wXOpwcAAAAAAABAyLIUUFbW6kVY9K69YJlBoPn9v/76a5PyR0fdaye4jqQfPXq0HHHEEZKSkuKRNvbvvJ40aZIEivr6+j7bRutN+JoWo25qaup57EwRbA0KbN26tedx75kHmobJko5LZy9oyipbNTbs0XXq/mEJMuh+0Z/OurG8vxr48FcAZ7AhAAEAAAAAAADAq/p3Qo8ZM8bq83REfe9AhQYZ3n77bfn000/7zJrQ5zU3N5tUTV988YVMnz7dI0Wotfi1hXZQ6+sHgtraWnn++ed7Zhzk5OTY3IbepB34GoSwyMvLc7hM75kN2m5Nq2VRUFAgMTExpjaDBjY2bdpkfucqbYfWCdHtpGmWzjrrrD4zNfrPrCH9ku8QgAAAAAAAAADgVW+88UZPAEE79TVvvzU6u6Grq6vnsQYevv/+e3NfO6p1ZHt8fLx5ngY0tENen//ee++ZTuyBFn3WQIlF745yb9MZHq+++mqf3+n20noFWmy6oqKiTy2N2bNnmzRCvqQ1HN56662ex/o+9k6hZEvvjv/ehaeVBgk0zda3337bE6xwJwBhCSq8++67PQWv+7+WpR26H/m6rshgRgACAAAAAAAAgNdo57J2CFucfPLJNusqtLS09HlsCT5MmTJFzjzzTImOju75W2Njo7z44otm1LwlWKEd4gOZGaCzKix6F1u2RUfc6wwMe7Qj3NFMgZqaGnOzRzvONcBy6KGH+iz4oMEd3c4a7Pnoo4962qjvw/nnn+8wXZKmjNJtpLTgtLWUVho4sAQgSkpKzHtgKSrtCt3OGoCwBBt6ByC0sLmlHQcddFBA1fUIdWxpBL0ZM2b0eawHYHcOUgAAAAAAAPAsHbm/atWqnsfaAW2v+K+O+O9PR8Sfe+65B/xe6w/oTIDHHntMqqurze8+/PDDAQUgdPS8Re9ghy06E+Orr76y+xxNO+RMqiJHdIbHK6+8IkVFRTJz5ky7dTTc9cADDzh8jv4/GgxyNf2SBoes1YwYOXKkmU2hs080vZPWgjjyyCNdbruuY9SoUSZN1JYtW8x7k5iYaP5mCXAo0i/5FgEIAAAAAAAAAB63d+9eefrpp3vqFmRmZsrZZ59tdxlrI9NPOeUUm8/XEfgnnXSSeR2lRaobGhrcLjDcO+jQOxjhbdopft5551kNyGjxaZ2B8Mknn5hCyjrj4/HHH5crr7xShg4dKr6kdRZ05oP+dETfd8sMFtU/JVLvwcQalPr44497ghbuBCAs21EDEJq+SmdBHHfccSaoYWmHbi9/FO8ezAhAIOj1jqIrnU61YMECv7UHAAAAAABgsNPR50uWLDHBAJWSkiJz5swxaYTs6T/rID093WEthnHjxpnAhSXQoWl/tK6AO3pn1eifDsoaHXH/m9/85oDf33///X3qSbhLayRoZ7/etJP+iSeeMLM9dPu+9NJLMm/ePPEk7cC3vAfaia+FoTXtkqUwuPa76YyTyy+/XHJzc+2uq7i4uGcb6vtur7aDvq4lAKGzZrTuhb73rtL0SlpLQwM3lgBE73ZoEMTXtTMGu3B/NwAAAAAAAABA6NBOaw0+6AwIpbMRtMPakg7Hnv5ptZ3phNbaAr1H5OuMAXdpGh8LRzUZfM1SA8JCR/r3Lk7tCSeeeKKcddZZ5qZpzy+88EK59tprZf78+T3bRlNBPfPMM33qZThKv6QBIQ2m2KJBJk3tZG1ZV2jwxBJ80qBJZWVlnyLYpF/yPWZAAAAAAAAAAPAI7ZxeunSpGcGu4uPjZe7cuWYGhDO0RoAGISyd287UYfBk6iStR2CpF6CzN3QWQ++ghL/pjAtNO6VphZTWOujdce8tul0uu+wyM/tBt6/OwHjzzTet1uawbDtLcXCl27R3HQZHNGigxcrdma2gsxwsQYfPPvtMSktLzf3hw4dLamqqy+vDwDADAgAAAAAAAMCAacf0k08+aUadW0bsa9olV1PpZGRk9Fmns69t4SjNk6MO/t7Wr18vgUSDDxrUsfBEmidn6SwFnSHRe5aCpfh3fxoA0BRO7tIAhwZX3KFFyC2zbbSgdVdXl7nP7Af/IAABAAAAAAAAYEC0/sKyZctM/QWl6XYuvfRSt0bnjx49uue+ZSaFPdrBrLUJLAYyY0GLFI8dO7bn8ddff91TWyJQaH0DC1/XM9Di0JbC1xpgeO+996w+r3cKJX2+1otw5ta7eLi7aZgsRa170xohEydOdGt9GBhSMAEAAAAAAABwm6YDevbZZ2Xr1q09o/Rnz54tI0aMcGt9msP//fff7wlAaHChd42H/jZv3twTJNDOZ3df1+L4448361R1dXXy9ttvyxlnnCGBQOtq9C6O7UxdDU/S93batGmycuVK81gLPFdVVUlWVlbPc3QGjKVotbroooskOzvbqfVv2LBBnnvuOXN/48aNJqWXOzNadLbDp59+2vM4Pz9fYmNjXV4PBo4ZEAAAAAAAAADcorMPli9f3pNnXwtCa+FiTYPjLk3B1HsWhAYA7AU/3n333Z7HBQUFBxSydicN02GHHdbz+IsvvujzGv70+eef93k8kO3sLu3c7z3L5KOPPurz994zFzRtk7PBB0ugwBJw0KDS999/7/Y+dM0118jVV19tbmeeeaZb68HAEYAAAAAAAAAA4DJNwfPyyy+bUeuW2QezZs0yQYCBOuWUU3rSCxUVFZkR9/3rQTQ2NsozzzxjRuBbRudPnz7dI++kdlhr0eLeneyPP/64lJSU9BSAtkZnbLzyyitSX18vnk67pLNCvvzyy57f6UwPTVvka7qddZaIhb7/llRZum207oJF/1RIjmiqpIMOOmjAaZiUBj40BZjeeqd2gm+RggkAAAAAAACAy7Q+Qu8O4pSUFNmxY4e5OeOss86y+TftNNYgwKuvvmoer1692nR06+wELcKsRYq3bdvWpx6CPr93KqCBdrJffvnlJsBiKURdXl4uTz/9tKlvoe3T9Ec6Wl873TUYop3wmrKpN90mzrRJ1235X3vTWQAazCgrKzPpiCyGDBki55xzjvjL1KlTTVDGEmj58MMP5Yc//KFs2rRJmpqa3A5AWJZZs2aNua/7kqad0u2I4EQAAgAAAAAAAIDLtNO9tz179pibs+wFINQRRxxhAgFvvPGGmf2gtQ90NkR/mtt/xowZMmnSJPEkDTRop7rO6NAOdssofw16bN++3e6yWrNC0zhp0Wb9HxypqakxN2fozIxzzz3Xbl0MX82CsARNNFXSiSee2Ccgpe10J3CgQaakpKSe4IauU9eN4EQAAgAAAAAAAEBAOvTQQ2XcuHHy7bffmvRHOsOgubnZ1HnQ+gL6N+3oH2jdB3s0sDFx4kQz42LLli1mVL52jms7NBihsyD09dPT083MCK3LkJeX55HX1poaGmAZOnSoWbe2QzvoA4FlFoTORtF0XG+++WZP8W53Zz8oTb2l29xSRFoDEJpay5KSy5s27GmQZZsq5fs9+6W9q1tOHT5Mrp6QJ7GRjoNIsC6sW/cOIITU1tbKggULzP1HH33Ur9FgAAAAAAAAAIFLu8dX19TLwqJy+XLXvgP+/oPcVLnzqHwJ90EAJBQxAwIAAAAAAAAAMOgCD59W1cnC4nJZW7vf5vPeK98jm+ubZHzyEJ+2L1QQgAAAAAAAAAAADAqd3d3yfvkeWVhUJiX7/lsw254t+whAuIsABAAAAAAAAAAgpHV0dckbO2tkUVG5bG9ocWnZCSkJXmtXqCMAAQAAAAAAAAAISa2dXbJy2y5ZWlIhlU2tLi9/Sl6ajEj0XpHzUEcAAgAAAAAAAAAQUpo6OmX5lmp5qrRCalvaXV4+OjxMLhmfIz+eONwr7RssCEAg6M2YMaPP47CwMImLIyoJAAAAAAAADDb72trluc1V8symKqlv63B5+fjIcPnhmCy5ZHy2pMVGe6WNgwkBCAAAAAAAAABAUKttaZOnSyvlhS1V0tTR5fLySVERcvG4bLlwXJYkR0d5pY2DEQEIBL1Vq1b1eVxbWysLFizwW3sAAAAAAAAA+EZVU6up7/Dy1mpp7ep2efnUmCi5dHy2nD8mS4ZERXiljYMZAQgAAAAAAAAAQFDZsb9ZFhWXy2s7aqSz2/XAQ1Z8tFyenytnj0qX2AgCD95CAAIAAAAAAAAAEBRK9zXKwqJyebesVlxPtCQyIiFW5hXkyhkjhklkeLgXWojeCEAAAAAAAAAAAALautr9ZsbDR5V73Vp+fHK8zC/MlR/kpklEWJjH2wfrCEAAAAAAAAAAAAJOd3e3fLO7Xp4oKpOvd9e7tY7JqQkyvzBPjssaKmEEHnyOAAQAAAAAAAAAIKACDx9X7pWFxeWyfk+DW+s4Ij3ZzHg4LD2JwIMfEYAAAAAAAAAAAPidFpPW2g4aeNi0r8mtdUzLTjGBh0mpiR5vH1wX1q3hJCCE1NbWyoIFC8z9Qw89VGJiYuw+Pzs7Wy655JI+v3v66aelsrLS4Wsdc8wx5mbR2toqDz/8sFPtnD17tuTk5PQ8LikpkVdeecXhctHR0XLdddf1+d2bb74p69evd7js+PHjZebMmX1+949//EMaGhxHkk899VSZPHlyz+OamhpZvHixOOPqq6+WxMT/HvS/+eYb+eCDDxwul5aWJvPmzevzu+XLl8u2bdscLqvv/Yknntjnd/fdd59T7T3//PNl1KhRPY/19fR1nfHzn/+8z+P3339fVq9e7XA5fT193d4WLVpk9mdHpk+fLocddljP4/3798tjjz3mVHvnzp0rw4YN63m8bt06eeuttxwul5CQINdcc02f361cuVJKS0sdLjtp0iQ57bTT+vzuoYcekra2NofLnn322ZKfn9/zuKKiQpYtWybO+MlPftLnePDZZ5+ZmyMcIzhG9McxgmMEx4j/4jziQBwjOEZwjOAYYQ/HCI4RHCM4RnCMsN4f8e1338mrb74lTR2d0tFlu7u6LTJGPhl3dJ/fTSrfIBn7d0tMRLjER0VIpI00S/RHPCzu9G0NFDMgENIaGxsddmomJycf8LumpibTieuIBhz6c2Y51dnZ2edxe3u7U8tqAKK/lpYWp5bV5/WnwQdnltX29dbV1eX0/9o/zqnviTPLxsbG+uW96ejoOOCxs8taa4czy+r/ZW3/dWbZ/vu4bm9n26vvozv7oTUD2Q91OWcCEP33Q/0cefu94RjBMaI/jhEcIzhG/BfnEQfiGMExgmMExwh7OEZwjOAYwTGCY0Tf/oiWzk5ZuXWXvPbtFhnZ1ChRIubmLC0mPSI2XGRvq0iHSPOB3UE96I/YL/5AAAIhbciQIQ5nQMTHx1v9Xe8R+7ZYW7czy6mIiIg+j6Oiopxa1loAQjvqnVnWWoe+jmR3hravt/DwcKf/1/4FfvR/cGZZff/88d5ERkYe8NjZZa21w5llre2H+v9b+3J0tE/o9na2vfo+urMfWttvBrIf6nLOBCD674f6OfLHe8MxwvF7wzHC+/shxwjH+yHHiANxHsF5hKN9gvMI53AecSCuNbjW6I9rDa41OEb8F/0RB2rp6pYlxeXyVGml7Gltl5yObsmItN+HZ5kBYb6Lw8Pk3NGZMic/R756Z7eUNux1uCz9EYniD6RgQkinYHr00UdNGh8AAAAAAAAA/rWvtV2e2Vwlz26qlP3tfbODOCM+MkIuGJMps8dnS1rsgYN0EXiYAQEAAAAAAAAA8Jqa5jZ5qrRClm+plubOvumgnZEUHSmzx2XLhWOzzH0ED94tAAAAAAAAAIDHVTS2yNKSClm5bZe02SkubUtabJTMGZ8js8ZkmtkPCD5+D0BoAaaKigopLy+XvXv3SnNzs/md5uTS3MbZ2dmSk5MjKSkp/m4qAAAAAAAAAMCBbfXNsrikXF7fUSOd3a4HHrLjY2RuQY7MGJkhMRF961ciuPg8ANHY2CifffaZfPrpp/Ldd99JSUmJCTg4kp6eLhMnTpQjjjhCjj/+eMnPz/dJewEAAAAAAAAAjpXUNcrConJ5t7xWXA87iIxKjJN5Bbly2vA0iQwn8BAKfBKAaG9vl3feeUdefvll+eijj3oCDt0uRL927dolu3fvlvfff1/uvvtuMyti5syZcs4558iYMWO82HoAAAAAAAAAgC1ra/fLE0Vl8mlVnVsbqWDoEBN4+EFuqoSHhbGhQ4hXAxB79uyRp556SpYtWya1tbV9gg5hYWHm5mwQQp9roctoyqa///3v5nbsscfK3LlzZfr06V76TwAAAAAAAAAAvftov9q1T54oLpfVu+vd2jAHpyXKFYW5ckzm0D79vwgdXglAaC2Hxx57TJ5++mlpaWkxO2P/YIPe17RKmkpJZzNkZmZKYmKiqf0QEREhra2tph6EznrQ2Q9bt26VzZs3m9kUvel6NJ2T3iZMmCA/+9nPCEQAAAAAAAAAgBd0dXfLR5V7ZVFRuXy/t8GtdRyVkSzzC3Nl6rAkAg8hzqMBCE2ttHDhQvnb3/4mTU1NBwQcRo4cKSeeeKKp43DYYYe5XFi6s7PT1Iz48ssv5fPPPzdBBw1UWF5n48aNsmDBAjn88MPl17/+tRQUFHjy3wMAAAAAAACAQUmLSb9dViuLispkc32zW+uYnpMi8wvy5KDUBI+3D4EprNuVQgwOnH766bJjxw5z37Jandkwa9YsmTFjhowfP148SWdIfPjhh7J8+XL5+OOPTYDCQmdR3HnnnXLuued69DUR+DTdlwai1KOPPippaWn+bhIAAAAAAAAQlNq7uuTV7btlcXGFlDW2uLy8lpI+dfgwU+NhbHK8V9qIQTIDYvv27T33dYbDj370IzPjIdxLFcvj4uJM0ENv1dXV8uSTT5p6E/X19dLV1SU7d+70yusCAAAAAAAAQChr6eiUl7btkqUlFbKruc3l5SPDwmTGyHS5vCBHhifEeaWNGIQ1II4++mi54YYbZMqUKeJLOtPi5z//uRn5vnTpUnn88cd9+voAAAAAAAAAEOwa2jvkhS3V8nRphext7XB5+ZiIcDlvdIbMGZ8jmfExXmkjBmkA4oknnpBjjjlG/Ck+Pl6uueYamT17NjMgAAAAAAAAAMAJda3tsmxTpTy3uUoa2v+b6t5ZQyIj5MKxWTJ7fLakxESxzeH5AIS/gw+9JSUlycSJE/3dDAAAAAAAAAAIWLub2+Sp0gpZvqVaWjq7XF4+OTpSLhmfLReMyZLEaI8n3EGQY49A0NMC572FhYWZ+iAAAAAAAAAArKtobDGFpV/Zvkvau7pd3kzpsVEyJz9HzhudKXGREWxmWEUAAgAAAAAAAAAGia31TbKouFze3Fkjna7HHSR3SIxcnp9rCkxHR4R7o4kIIQQgEPRWrVrV53Ftba0pRg4AAAAAAADg34r2NsjC4nJ5v3yPuBF3kNGJcTKvMFdOzRsmkeFhbFY4hQAEAAAAAAAAAISob2vqZWFRuXxWXefW8oVDh8gVhblyQk6qhIcReIBrCEAAAAAAAAAAQAjp7u6WL3btk4VFZbKmZr9b65gyLFGuKMiTozKTTc1VwO8BiFtuuUX8QT8Ad955p19eGwAAAAAAAAACQVd3t3xYscekWtq4t9GtdRyTOdSkWpo6LMnj7cPg49EAxIoVK3weDdNoHgEIAAAAAAAAAINVR1e3vF1WY4pLb6lvdmsdJ+akyvzCXJmQkuDx9mHw8msKJg0e9OYoeOHq8wEAAAAAAAAgVLV1dsmq7btlSUm5lDe2urx8RJjIacOHydyCXBmTFO+VNmJw83gAon+QwBFLEEGXc2ZZV58PAAAAAAAAAKGkuaNTXty6S54srZDdzW0uLx8VHiZnj8yQy/NzJDch1ittBDwegCgqKnL6ua+88orccccdUl9fbwIJU6ZMkbPOOksmT54sI0eOlISEf0/1aWhokO3bt8u6devk1VdflW+//dYEIZKTk+XXv/61zJw5k3cSAAAAAAAAQMjb39Yhz2+pkmWllVLX1uHy8rER4TJrdKZcmp8tGXExXmkj4PcUTI8//rjce++95n5OTo784Q9/kGOOOcbqc1NTU81t6tSpMnfuXPn000/ltttuk/LycvnFL34hlZWVcs011/j4PwAAAAAAAAAA39jb2m6CDs9trpLGjk6Xl0+IipALx2bJxeOyJSUmyittBAIiAPHZZ5/JPffcY+4PHz5cnnzyScnIyHB6+WOPPVaefvppmTNnjuzcuVP+8pe/yMSJE+W4447zYqsBAAAAAAAAwLeqm1rlqdJKeXFrtbR0drm8fEpMpMwelyMXjM2UhCi/lgPGIOXzve6RRx4xPzWN0u9//3uXgg8Wuszvfvc7ufLKK83jhx9+mAAEAAAAAAAAgJBQ1tAii4vLTYHpDjfq4GbERctl+Tly7qgMiY2M8EobgYALQOzYsUO++uorE3zIy8uzmXbJ2ZkQOoNCZ0GsWbPGrHvEiBEebS8AAAAAAAAA+MqW+iZZVFQub+6sEdfnO4jkDYmRuQW5cuaIdImOCPdCC4EADkBs2LCh535+fv6A16fr0ACEZd0EIAAAAAAAAAAEm417G2RhUbm8X7HHreXHJMXJ/IJcOTlvmESGh3m8fUBQBCCqqqp67kdHRw94fb3X0XvdAAAAAAAAABDo1uyulyeKyuSLXfvcWv6glCEyvzBPpmWnSHgYgQcEHr9VHrHMXPD3OgAAAAAAAADAV7q7u+Wz6joz4+G72v1urePQYUkyvzBXjsxINunugUDl0wCEpeC0fsg0ZVJZWZmpBeFu8OH777/v+YC5U8waAAAAAAAAAHyhq7vbpFjSwENxXaNb6zg2a6hJtXTIsCSPtw8I+gDEYYcd1hMw0CDEb37zG3n88cddjtJ1dXXJb3/7W7MOFR4ebtYNAAAAAAAAAIGko6tL3txZK4uKy2Xb/maXl9ee0x/kpppUSwVDh3iljYC3+LQUemZmphx99NE9jz/99FP56U9/KvX19U6vQ5+ry3zyyScmcKE3XaeuGwAAAAAAAAACQWtnlyzfUiUXvPGt/O7rTS4HHyLCRGaMTJdlp06Ru44uIPiAoOTzGhC/+tWv5Pzzz5eOjg4zg+Gdd96R0047TebMmSNnnHGGjB8/3upypaWl8tprr8lTTz0l+/btM4EHXV4LUd96662+/jcAAAAAAAAA4ADNHZ2yYku1PFlaITUt7S5voejwMJk5KkMuy8+RnCGxbGEEtbBuSx4jH3rxxRdNIMLy0vrTkoYpNjbW1IUYMuTf04kaGxtNrYiWlpY+z9WfmnrpzjvvlPPOO8/X/wICWG1trSxYsMDcf/TRRyUtLc3fTQIAAAAAAECI29/WIc9urpJnNlXKvrYOl5ePiwiX88dkyqXjc2RYXLRX2giE/AwIpQGDuLg4ue2220xKpd51IZqbm81sh96/s7CkXNLfJSUlyR133CGnn366P/4FAAAAAAAAAJA9Le3y9KYKeX5ztTR1dLq8RRKjIuSicdly8dgsSY6JYosipPglAKE0cKCFo++9915ZtWqVtLW1md/3L0jdOxBhSbk0Y8YMuemmm2TYsGF+aTsAAAAAAACAwa26qVWWllTIS1urpbXL9SQzKTFRMmd8tswakykJUX7rpgW8yq97tgYQ7rrrLvnlL38pb7zxhqxZs0Y2bNhgUuhYClPrTAdNoXPQQQfJ1KlTTeAiOTnZn80GAAAAAAAAMEjtbGiWxcUV8ur23dLhRnb7zLhouTw/R2aOzpDYiAivtBEIFAERWtOAwkUXXWRuCA0aQFq3bp2sXbvW3PT+7t27zd+OPPJIWbJkib+bCAAAAAAAADht075GWVRcLm/vrJUuN7bb8IRYmVuQK2eOGCZR4eFseQwKARGAQOjROh/l5eX+bgYAAAAAAAAwIN/v2S9PFJXLR5V73Vp+XHK8zC/IlZPy0iSiX/p5INQRgIBPUm1NnjxZ3nvvPbY2AAAAAAAAAp7Wol1dUy8Li8rly1373FrHxNQEuaIgV47PTjmg7i0wWBCAgFfMmTNH8vLy5OCDD5bs7Gzzu4KCArY2AAAAAAAAAjrw8ElVnUm1tLZ2v1vrODw9SeYX5pmfBB4w2BGAgFf86Ec/YssCAAAAAAAgKHR2d8v75XtkYVGZlOxrcmsdx2elyPzCXJmclujx9gHBKiACEC0tLVJSUiJ79uyRhoYG6ejocKvmANyn23zNmjWmbsOuXbskISFBsrKyZMqUKZKamsqmBQAAAAAAQMjp6OqS13fUyOLictne0OLy8ppY6eS8NJlXkCv5Q4d4pY1AMIv0Z4f3888/L8uXL5fvv/9eurrcqR0fugEI3R6bN2+WtWvXyrp168ytuLhY2tvbe55z1113yfnnnz+g12lubpZHHnnEvA81NTUH/D0qKkqmTZsmN9xwAymUAAAAAAAAEBJaO7tk5bZdsrSkQiqbWl1eXotJnzlimAk8jEiM80obgVDglwCEdqxrh/amTZt6cqu5Q3Oo6bKhlEvt9ddflyeffFLWr18vTU3uTfdyVmlpqVx//fWyZcsWm8/RgMe7774rn3zyidxyyy1yySWXeLVNAAAAAAAAgLc0dXTK8i3V8lRphdS2/Hegr7Oiw8PknFEZcll+rmQPifFKG4FQ4vMARFVVlcyfP9+Mth9o8MDdwEUg++abb+TLL7/0+utomiWt01BdXd3n9xMnTpThw4dLXV2dmXXR2Nhoft/a2iq//e1vTWqmmTNner19AAAAAAAAgKfsa2uX5zZXyTObqqS+zfX07/GR4fLDMVlyyfhsSYuN5o0BAjUAcffdd8vu3btN4MEyg0HrDEyfPl3GjBkjiYmJEhkZEKUpAopul/j4+AMCBu7Qba4zH3qvKz8/37w3hYWFPb+rr6+XBx54QJYuXdrzu1tvvdU8Z/z48QNuBwAAAAAAAOBNtS1t8nRppbywpUqaOlxPAZ8UFSEXj8uWC8dlSXJ0lFfaCIQyn/b066j61157rSfwkJycLPfee68cf/zxvmxGwIuNjZUJEybIpEmTZPLkyeY2evRoeeihh8xtoN58801TcNoiLy/PBBn0/egtKSlJbrvtNvN+LVmypGcmhAYlPNEOAAAAAAAAwBuqmlpNfYeXt1ZLa5frWVTSYqPk0vE5Mmt0pgyJivBKG4HBwKcBCE0tZCk2rZ3af/7znwk+9HPttdfKzTff7NVZIP2DB7fffvsBwYfebrrpJlMHory83Dx+6623ZOPGjSZIAgAAAAAAAASKHfubZVFxuby2o0Y63Ujfnh0fI5fl58jMURkSExHulTYCg4lPAxCWlD8afBgxYoRJu4S+UlNTvbpJiouLpaSkpOfx2LFjHb4PcXFxMnv2bDNbxWLlypUEIAAAAAAAABAQSvc1ysKicnm3rFZcT7QkMiIhVuYV5MoZI4ZJZDiBB8BTfPppamlp6blfUFDgy5fGf7z33nt9toWzBaX7P09nRAAAAAAAAAD+tK52v9z0SZFc9vZaeduN4MP45Hi586h8WXbaFDl7VAbBByCYZ0BkZmb+94UpNO0Xn3zySZ/Hhx9+uFPLZWdnS25ubk8apq1bt0plZaX5PQAAAAAAAOArWlv2m9318kRRmXy9u96tdUxOTZD5hXlyXNZQk60FQAgEIMaMGdNzv6qqypcvjf/YtGlTz7YIDw83ha6ddcghh/QEIFRpaSkBCAAAAAAAAPgs8PBJVZ0JPKzf0+DWOo7MSDaplg5LTyLwAIRaAEI7u7XmwObNm2XdunXS0NAgCQkJvmzCoLZv3z7Zs2dPz+O0tDRT38FZeXl5fR7rLIgTTjjBo20EAAAAAAAAetNi0lrbYWFxuWza1+TWxpmWnSLzC3NlUmoiGxfwIZ9XVLn66qvNz46ODnn88cd9/fKD2o4dO/o8djV9UlZWlt31AQAAAAAAAJ7S0dUlK7ftktlvfiu//rLU5eCDdnyempcmT55ysNxzbCHBByDUZ0Co8847Tz799FN5+eWX5bHHHpMJEybI6aef7utmDEo646S31NRUl5ZPSUnp83j//v02n7tx40Zzs2b37t2yfPnyPr+bNm2apKenu9QeAAAAAAAAhJ6Wzk5ZuXWXLC2tkKqmNpeXjwwLk7NGpsvl+TkyItH57B8AQiAAof7v//5PYmJi5LnnnpMbb7xRLrroIvnRj34kw4cP90dzBo3GxsY+j/U9cEVsbGyfx01NtqPOb7/9tjz00ENW/6apm2655ZY+v1u8eLHTAQhH9UP27t3r1HoAAAAAAAAQOBrbO2X5lip5qrRS9rS2u7x8THiYnDs6Uy7Lz5HMeNf6vQCESACid8fz0KFDpa6uTp555hlzGzlypKkRkZiY6FIRGH3unXfe6aUWh47m5uY+j6Ojo11avn/Aov/6fGX69Ol2/x4VFWVm1gAAAAAAACDw7Wtrl2c3Vcmzmyqlvr3T5eXjIyPkgrGZcsm4HEmNjfJKGwEESQBixYoVfYILel8r2Ktt27bJ9u3bXVqfLksAwj2uBHmsPd/yvlnz05/+1NwAAAAAAAAAa2pb2sxsB5310NTR5fJGSoqOlNnjsuXCsVnmPoDAExmMHeFwT1xc35x3ra2tLi3f0tLS53F8fLxf3ooPPvjAYQqm22+/3WftAQAAAAAAgPMqG1tlSUm5KTDd1mV7gKstabFRMmd8jswak2lmPwAIXH4JQNgbOQ/v6R8wcDUA0f/5/gpAZGVlOUzBBAAAAAAAgMCyfX+zLCoul9d31EinG/2D2fExMrcgR2aMzJCYiHCvtBFAkAcgioqKfP2S+I+EhIQBFWves2dPn8daqwMAAAAAAACwp6SuURYWlcu75bXizrDkUYlxMq8gV04bniaR4QQegGASECmY4BsjRozo87iystKl5auqqvo8Hj58uEfaBQAAAAAAgNCzrna/PFFUJp9U1bm1fMHQISbwcGJuqkSQwh0ISgQgBpGhQ4dKampqz0yGmpoaaW5uPqA2hC1lZWV9Ho8ZM0YCwYwZMw6oKeLs/wQAAAAAAADPpl7/ene9CTx8s7verXUcnJYoVxTmyjGZQ6kdCwQ5AhCDzLhx4+TLL78097u6umT9+vVyxBFHOLXsd999d8C6AAAAAAAAAA08fFS5VxYWl8v3exrc2iBHZiSbwMPUYUkEHoAQQQBikDn22GN7AhDq66+/dioAoemaysvLex6PHj1acnJyJBCsWrWqz+Pa2lpZsGCB39oDAAAAAAAwWGgx6XfKamVRUblsqm9yax3Tc1JkfkGeHJTat34pgOBH1ZZB5qSTTurzeOXKlU4t1/95/dcDAAAAAACAwaO9q0te3lotF73xrdz2ZanLwQftlDx9+DB56pRD5M/HFBJ8AEIUMyAGmYKCAsnPz5eSkhLzePPmzfLBBx/I9OnTbS7T0tIiy5Yt6/O7mTNner2tAAAAAAAACCwtnZ3y8tZdsrSkQqqb21xePjIsTGaMTJfLC3JkeAI1PIFQ5/MAxNy5cz2+Ti06vGjRIo+vN1Rdd911cv311/c8vuOOO2TKlCmSnJxs9fn33ntvn/RLp5xyikyYMMEnbQUAAAAAAID/NbR3yPIt1fJUaaXsbW13efmYiHA5b3SGzBmfI5nxMV5pI4DA4/MAhNYf0ICBJwvceHJ9gaCsrMzq7+vr6/s83rt3r9XnxsTESHp6us31n3baaTJ16lRZs2aNebxz50657LLL5J577jEzJCz2798v999/vyxdurTPum+44Qa3/i8AAAAAAAAEl32t7bJsU6U8t7lK9rd3urz8kMgIuWBslswely2psVFeaSOAwBXWrT34PlRYWDjggEH/Juv6Nm7cKKGidxDAHUceeaQsWbLE7nOqq6vlggsukF27dvXZjhMnTpThw4dLXV2drF27VhobG/ssd/fdd8s555wjgax3EepHH31U0tLS/N0kAAAAAACAoFLT3CZPllbIii3V0tzZ5fLyydGRJuhw4dgsSYwmCzwwWPn805+Tk+NysEFH/ls6wrWTXG+xsbGSkpLipVaGvszMTPnnP/9pUjFt3bq1Z1uvX7/e3PrTmQ+//OUvAzL4MGPGjD6Pdf+IiyOHIAAAAAAAgKsqGltMfYeV23ZJW5fr45bTY6NkTn6OnDc6U+IiI3gDgEHO5wGId999163ldMT+hx9+aGo9bNq0STo6OuRHP/qRzJkzx+NtHCy0GPWKFSvk4YcfluXLl5uZA/1FRUXJtGnTTNqlgc7MAAAAAAAAQGDaVt8si4rL5Y2du6XTjXwpOfExcnlBrpw9Ml2iI8K90UQAQcjnKZgGqr29Xf7whz/IM888Y0a633jjjXLNNdf4u1lBTwM6q1evNjUlampqZMiQIZKVlWVqRaSmpkowIQUTAAAAAACAc4rrGmVhUZm8V75H3OkkHJUYJ/MLc+XUvGESGR5adVoBDFzQJWDTEfm/+93vTO2C9957zxRJnjJliql7APdFRkaabch2BAAAAAAACH3f1dTLE0Xl8ll1nVvLFwwdIlcU5sr0nFQJH2C9VwChK+gCEBY333yzCUDoBI4HH3zQYdFlAAAAAAAAYDDTfrQvd+2ThUXlsrqm3q11HJKWaAIPR2cONdlJACAkAxCjRo0yNQxKSkrk66+/lp07d8rw4cP93SwAAAAAAAAgoHR1d8tHFXtlYXGZbNjb6NY6js5MlvmFeTJ1WJLH2wcgdAVtAMIShNAAhFq/fj0BCAAAAAAAAOA/Orq65Z2yGlNcenN9s1vb5cScVFPjYUJKAtsVwOAKQGg9CIvq6mq/tgUAAAAAAAAIBG2dXfLqjt2ypLhcyhpbXV4+Ikzk1OHDZF5BroxJivdKGwEMDkEdgNC0S71z2AEAAAAAAACDVUtHp7y4dZcsLa2Q3c1tLi8fFR4mM0amy9z8XMlNiPVKGwEMLkEbgNiyZYtJu2QpdpOWlubvJsFPZsyY0eex7hNxcXG8HwAAAAAAYFBoaO+Q5zdXydOllVLX1uHy8rER4TJrdKZcmp8tGXExXmkjgMEpKAMQe/fulZtuukm6urp6OpwPPfRQfzcLAAAAAAAA8Jm61nZZtqlSnttcJQ3tnS4vnxAVIReOzZKLx2VLSsx/U50DwKAKQGigob6+3sx6+Oijj+Tpp5+Wffv29cx+mDJliuTl5fm7mfCTVatW9XlcW1srCxYs4P0AAAAAAAAhaVdzqzxVUikrtlZLS+e/B+i6IiUmUmaPy5YLxmZJQlRQdA8CCFI+P8JMmDBhwOvQeg8afNCfWoj6lltu8UjbAAAAAAAAgEBV3tAii0vKZdX23dLe5Xo91PS4aLlsfI6cNzpDYiMjvNJGAPBrAGKgxaI18GAJPsTGxsqf//xnOfjggz3WPgAAAAAAACCQbKlvkkXF5fLWzhrpdKNrLXdIjMwtyJWzRqRLdES4N5oIAFb5ZY6VJYDgDl0uOjpaTj31VLnhhhtk+PDhHm8fAAAAAAAA4G8b9zbIwqJyeb9ij1vLj0mKk/kFuXJy3jCJDP93KnMACOkAxKxZs1xeJiIiQhISEiQ1NVUKCgpk6tSpkpSU5JX2AQAAAAAAAP60pqZeFhaVyefV+9xafkLKELmiIE+m5aRI+H9qqALAoAhA3HXXXb5+SQAAAAAAACCgadaPz6vr5Imicvmudr9b6zh0WJLML8yVIzOSTQYSAPA3ytwDAAAAAAAAftLV3S0fVOwxqZaK6hrdWscxmUPlisJcOWQYGUMABBYCEAAAAAAAAICPdXR1y5s7a0xx6W37m11eXuc3/CA3VeYX5knB0CFeaSMADBQBCAS9GTNm9HmsUwzj4uL81h4AAAAAAABb2jq75JXtu2VJcblUNLW6vKEiwkROH54ucwtyZHRSPBsaQEAjAAEAAAAAAAB4WXNHp6zYUi1PlVbI7pZ2l5ePDg+TmaMy5LL8HMkZEuuVNgJAyAYgioqKZM2aNfL999/Lnj17pL6+3vw+KSlJUlNTZeLEiTJ16lQpLCz0d1MRYFatWtXncW1trSxYsMBv7QEAAAAAALDY39Yhz22ukmWbKmVfW4fLGyYuIlxmjcmUOeNzZFhcNBsWQFDxawCiu7tbnn/+eVmyZImUlpbafe4LL7xgfo4bN07mzp0rF1xwgUm1AwAAAAAAAASaPS3tJujw/OYqaezodHn5xKgIuXBslswely3JMVFeaSMAhGwAoqKiQv73f/9XVq9ebQIRFraCCpbnaKDi9ttvlxUrVsif//xnycvL81mbAQAAAAAAAHuqm1plaUmFvLRtl7R2drm8sVJiouTS8dly/phMSYgKmOQlAOAWvxzFKisrZc6cOVJVVWUCC5agg97vHYzor/fzNHBx2WWXydNPPy3Z2dk+azsAAAAAAADQ386GZllSXCGrtu+WDjv9W7ZkxkWb+g7njMqQ2MgINjCAkODzAERnZ6dcc801JgihAQW9aUBhwoQJcvbZZ8shhxwiY8aMkYSEBPP8hoYG2bp1q3z33XfyyiuvyIYNG3qW0QCGruvFF1+UiAgOzAAAAAAAAPCtzfuaZFFxuby1s0Zcn+8gkjckVuYV5sqZI4ZJVHi4F1oIAIMoALFs2TKTRskSRBg2bJhJqXTaaadZfb4WoNbbYYcdJldeeaW8/fbb8tvf/tYUGtblN23aZGZB6GwIAAAAAAAAwBc27GmQJ4rK5MPKvW4tPy4p3gQeTs5LkwjqnAIIUT4PQCxcuLAn+JCRkSFLly6VESNGOL38KaecIgUFBSaF0+7du816Fi1aRAACAAAAAAAAXqX9UGtq6uWJonL5ctc+t9YxMSVB5hfmyvHZKRJO4AFAiPNpAEJTKe3cubMn9dJvfvMbl4IPFsOHDzezJq677jrzuKysTLZs2WJSNwEAAAAAAACeDjx8WlUnC4vLZW3tfrfWcVh6klxRmCeHpyf11DkFgFDn0wCE1m+wHLSzsrLk5JNPdntdOhNC16F1INTGjRsJQAAAAAAAAMBjurq75b3yPabGQ3Fdo1vrOC5rqAk8TE5L5J0BMOj4NAChdRuURnkLCwsHvD4tXG0JQFjWjcFnxowZfR7r/hUXF+e39gAAAAAAgODW0dUlb+6sNYGHbfubXV5e5zeclJtmUi3lDx3ilTYCQDDwaQCipaWl5/6QIQM/+MbHx1tdNwAAAAAAAOCq1s4ueWXbLllSUiGVTa0uL6/FpM8YMUzmFeTKyEQGRwKATwMQKSkpPffLy8sHvL6Kigqr68bgsmrVqj6PdTbMggUL/NYeAAAAAAAQXJo6OmXFlmp5qrRCalraXV4+OjxMzhmVIZfl50r2kBivtBEAgpFPAxA5OTk9NSDWr18vNTU1MmzYMLfWpZ3Ma9euPWDdAAAAAAAAgDPq2zrk2c2V8symKnPfVfGR4XL+mCy5dHy2pMVGs9EBwJ8BiMMPP1xiY2OltbVVOjs75U9/+pPcfffdbq1Ll9N1qJiYGLNuAAAAAAAAwJHaljZZVlopz2+pNrMfXJUUFSEXjcuWi8ZlSXJ0FBscAAIhAKGBgunTp8sbb7xhHr/yyiuSmpoqN998s4SHhzu9nnvuuUdefPFFU2xYnXjiiWbdAAAAAAAAgC1VTa2ytKRCXt5aLa1d3S5vqNSYKDPbQWc9DImKYEMDQCAFINSNN94o77zzjpm9oKmYFi9eLJ999pn85Cc/kR/84AcSHW19ulp7e7u8//778re//U02btxogg+6fGRkpPzsZz/z9b8BAAAAAACAILFjf7MsLimXV7fXSGe364GHrPhoU99h5qh0iY0g8AAAARuAGDVqlPzP//yP/PGPf+wJIpSUlMgNN9xg0jMVFhaa5wwZMsQ8v7GxUbZv3y5FRUXS3Nxsnq90Wb3pukaPHu3rfwMAAAAAAAABrnRfoywsKpd3y2qly43lRyTEytyCXDljxDCJciF7BwDATwEINX/+fGlqapIHH3ywJ42SBhY0wPDtt9+aW3+9Aw+W+zprQtcFAAAAAAAAWKyr3S+Lisvlo8q9bm2U8cnxMr8wV36QmyYR/+m7AgAESQBC/b//9//k0EMPld/85jdmhoOyBCOssQQe9DZixAj53e9+J8ccc4wPWwwAAAAAAIBApX1G3+yulyeKyuTr3fVurWNSaoJcUZgnx2UNtdtPBQAI8ACEOvroo+X111+XDz/8UF5++WVZs2aNVFRUWH1uTk6OTJ06VWbOnGkKWfMlAAAAAAAAAA08fFJVZwIP6/c0uLVBjkhPNjMeDktPos8JAEIlAKE0kKABBb2puro62bNnj+zfv988TkxMlNTUVBk6dKifWwoAAAAAAIBAocWktbbDwuJy2bSvya11TMtOMYGHSamJHm8fACAAAhD9aaCBYAMAAAAAAACs6ejqktd21Mji4nLZ0dDi8kbSUtIn56XJvMJcGZ88hI0MAIMpAAEAAAAAAAD019LZKSu37pKlpRVS1dTm8gbSYtJnjRwmc/NzZURiHBsYAHyAAAQAAAAAAAACVmN7pyzfUiVPlVbKntZ2l5ePCQ+Tc0dnypz8HMmKj/FKGwEAQRKA0NoPtbW1Ul9fbx4nJSVJWlqaqQUBWDNjxowD6orExTGSAQAAAACAYLavtV2e3Vwlz26qlPr2TpeXj4+MkAvGZMrs8dmSFhvtlTYCAIIgAPHhhx/KypUr5dtvv5WysjKrz8nLy5MpU6bIzJkz5YQTTvB5GwEAAAAAAOB9Nc1t8lRphSzfUi3NnV0uL58UHSmzx2XLhWOzzH0AgP+EdXd3d/vrxT/55BP5/e9/Lzt27DCPHTVFR7arESNGyG9+8xs59thjfdJOBBedQbNgwQJz/9FHHzUzaAAAAAAAQGCrbGyVJSXlsnLbLmnrcr27alhslMwZnyPnjck0sx8AAP7ntzDwXXfdJYsXL+4JOmhwQW+2ghCW4IP+ffv27fKjH/1I5s2bJ7/85S992m4AAAAAAAB4zvb9zbKouFxe31EjnW6Mk82Oj5HL83Pk7FEZEhMRzlsDAIM9AHHHHXfIU089ZYIJlqCD3hISEqSwsNDMcLDUfNCaEDpDoqioSBoaGvoEIhYtWiTt7e1y2223+ePfAAAAAAAAgJtK6hplYVG5vFteK+6k5xiVGCfzCnLltOFpEhlO4AEAApHPAxDvvfeePPnkk31mPEyePFmuvfZaU9shMtJ6kzo6OuSDDz6Qv/3tb7J+/fqeZTWQoctNnz7d1/8KAAAAAAAAXLSudr88UVQmn1TVubXtCoYOkfmFuXJiTqqE/2egKgAgMPk8AHHvvfean5bZD9ddd525OaKBiZNPPtncHnzwQXn44Yd7ghD33HMPAQgAAAAAAIAApf03X+3aJ08Ul8vq3fVurePgtES5ojBXjskc2pMhAwAQ2HwagCgpKZFNmzb1fEnMnj3bqeBDfz/96U9NoeFly5aZx7rO4uJiKSgo8HibAQAAAAAA4J6u7m75uHKvSbX0/d4Gt9ZxVEaymfEwdVgSgQcACDI+DUBs2LChJ+odHR0tN954o9vr+vnPfy4vvPCCSc2kNm7cSAACAAAAAAAgAGgx6bfLamVxUblsqm9yax3Tc1JkfkGeHJSa4PH2AQBCMACxe/du81NnQBxyyCGSlJTk9rp02SlTpshXX31l1mdZNwAAAAAAAPyjvatLXtu+WxYVV0hZY4vLy2sp6VOHDzPFpccmx3uljQCAEA1ADBkypOd+RkbGgNfXex291w0AAAAAAADfaenolJe27ZKlJRWyq7nN5eUjw8Jkxsh0ubwgR4YnxHmljQCAEA9AZGVl9dzfv3//gNfXex291w0AAAAAAADva2jvkBe2VMvTpZWyt7Xd5eVjIsLlvNEZMmd8jmTGx3iljQCAQRKAOPTQQyUqKkra29tl3bp10tnZKREREW6tS2s/rF271tyPjIw06wYAAAAAAID37Wttl2WbKuW5zVWyv73T5eWHREbIBWOzZPa4bEmNjfJKGwEAgywAMXToUDnllFPktddek7q6Olm+fLlceOGFbq1rxYoVZh1a/+Gkk04y6wYAAAAAAID37G5uk6dKK2TFlmpp7uxyefnk6Ei5ZHy2XDAmSxKjfdotBQDwA58f6W+88Ub54IMPpLm5Wf70pz/J2LFjXZ69sHr1avnjH/9o7sfFxZl1AgAAAAAAwDsqGltkSXGFrNy+S9q7ul1ePj02Subk58h5ozMlLtK9bBgAgODj8wDEiBEj5OGHH5brrrtOGhoa5Morr5Srr75arrjiComPj7e7bFNTk/zrX/+Sxx9/XFpaWkzw4aGHHpJRo0b5rP0AAAAAAACDxbb6ZllUXC5v7Nwtna7HHSQnPkYuL8iVs0emS3REuDeaCAAIYGHd3d1ufH1Y99VXXzn93M2bN8t9990n9fX1Jo1SbGysTJs2TSZPnmyCFAkJCeZ5GqTYsWOHqRnx0UcfmcCDNjk5OdnMfNAZFOqII47w1L+BIFdbWysLFiww9x999FFJS0vzd5MAAAAAAAgqxXWN8kRRmbxfvkfc6TganRgn8wpz5dS8YRIZHuaFFgIABl0AorCw0AQTXGVpgqNlbT1PH2/YsMHl10VoIgABAAAAAIB7vquplyeKyuWz6jq3li8YOkSuKMyV6TmpEu5GHxEAILR4JQWTszENSyChf0Ch//LOPg+D04wZM/o81v1E03MBAAAAAADHtH/ly137ZGFRuayuqXdrk00ZlihXFOTJUZnJbg1OBQCEJo8HIFwJCjj7XAINAAAAAAAAntXV3S0fVeyVhcVlsmFvo1vrODozWeYX5snUYUm8PQAA7wYg7rrrLk+uDnDKqlWrbKZgAgAAAAAAfXV0dcs7ZTWysLhcttQ3u7V5TsxJlfmFuTIh5d81PAEA8HoAYtasWZ5cHQAAAAAAADykrbNLXtuxWxYXl0tZY6vLy0eEiZw6fJjMK8iVMUnxvC8AAP/UgAAAAAAAAEBgaOnolBe37pKlpRWyu7nN5eWjwsNkxsh0mZufK7kJsV5pIwAgNBGAAAAAAAAACEEN7R3y/OYqebq0UuraOlxePjYiXGaNzpRL87MlIy7GK20EAIQ2AhAAAAAAAAAhpK61XZZtqpTnNldJQ3uny8sPiYyQi8ZlycXjsiUlJsorbQQADA4EIAAAAAAAAELAruZWeaqkUlZsrZaWzi6Xlx8aHSmXjM+WC8ZmSUIUXUYAgIHj2wQAAAAAACCIlTe0yOKSclm1fbe0d3W7vHx6XLRcNj5HzhudIbGREV5pIwBgcCIAAQAAAAAAEIS21DfJouJyeWtnjXS6HneQ3CExMrcgV84akS7REeHeaCIAYJAjAAEAAAAAABBENu5tkIVF5fJ+xR63lh+TFCfzC3Ll5LxhEhke5vH2AQBgQQACAAAAAAAgCKypqZeFRWXyefU+t5afkDJErijIk2k5KRIeRuABAOB9BCAAAAAAAAACVHd3t3xeXSdPFJXLd7X73VrHocOSZH5hrhyZkSxhBB4AAD5EAAIAAAAAACDAdHV3ywcVe0yqpaK6RrfWcUzmULmiMFcOGZbk8fYBAOAMAhAAAAAAAAABoqOrW94qq5FFReWydX+zy8trYqUf5KbK/MI8KRg6xCttBADAWQQgAAAAAAAA/Kyts0te2b5blhSXS0VTq8vLR4SJnD48XeYW5MjopHivtBEAAFcRgAAAAAAAAPCT5o5OWbGlWp4qrZDdLe0uLx8dHiYzR2XIZfk5kjMk1ittBADAXQQgAAAAAAAAfGx/W4c8t7lKlm2qlH1tHS4vHxcRLrPGZMqc8TkyLC7aK20EACCoAhANDQ1SVlbW83jEiBESH8+0QAAAAAAAMDjsaWk3QYfnN1dJY0eny8snRkXIReOy5eKxWZIcE+WVNgIAEJQBiFdeeUV+97vfmfvR0dHy4Ycf+vLlAQAAAAAA/KK6qVWWllTIS9t2SWtnl8vLp8REyaXjs+X8MZmSEEVCCwBAcPDpN1ZdXZ10d3eb+wcffLAkJyf78uUBAAAAAAB8amdDsywprpBV23dLx3/6RFyRGRdt6jucMzpDYiMivNJGAABCIgBhCTiEhYVJRkaGL18aAAAAAADAZzbva5JFxeXy1s4acX2+g0jekFiZV5grZ44YJlHh4V5oIQAAIRaASE9P77nf3Nzsy5cGAAAAAADwug17GuSJojL5sHKvW8uPS4o3gYeT89IkIizM4+0DACBkAxCTJ0+W8PBwk4Zp8+bNvnxpAAAAAAAAr9B+jjU19fJEUbl8uWufW+uYmJogVxTkyvHZKSZzBAAAocCnAYjMzEw54ogj5IsvvpAdO3bI+vXrZdKkSb5sAgAAAAAAgMcCD59W1cnC4nJZW7vfrXUclp4kVxTmyeHpSQQeAAAhx+dJBK+//nqJ+E/RpD/84Q/S1tbm6yYAAAAAAAC4rbO7W94pq5V5766Tn39a5Fbw4fisFHn8xEnyyAkT5YiMZIIPAICQ5PMAxGGHHSY33nijGSXw3XffyY9//GOprq72dTMAAAAAAABc0tHVJau275JL3/pOfvVFiRTXNbq0vCZWOiUvTZacfLDce1yhTE5L5B0AAIS0sG6NBPjBq6++Kr/+9a9NMerY2Fg588wz5aSTTpIJEyZIamqqxMXF+aNZCEIzZszo81hzZVr2n0cffVTS0tL81DIAAAAAQCho7eySV7btkiUlFVLZ1Ory8lpM+owRw2ReQa6MTKS/AwAwePi0BoTSAENvGv/QIMSKFSvMzR3a4bxhwwYPtRAAAAAAAECkqaNTVmyplqdKK6Smpd3lTRIdHibnjMqQy/JzJXtIDJsUADDo+DwA0XvChQYO9Gbtb4CzVq1a1edxbW2tLFiwgA0IAAAAAHBLfVuHPLu5Up7ZVGXuuyo+MlzOH5Mll47PlrTYaN4FAMCg5fMAhNKggwYbCDgAAAAAAIBAUdvSJstKK+X5LdVm9oOrkqIi5OJx2XLhuCxJjo7yShsBAAgmPg9AzJo1y9cvCQAAAAAAYFNVU6ssLamQl7dWS2uX69kZUmOizGwHnfUwJCqCLQ0AgL8CEHfddZevXxIAAAAAAOAAO/Y3y+KScnl1e410upEWOis+2tR3mDkqXWIjCDwAABAQKZgAAAAAAAD8ZdO+RllYVC7vlNVKlxvLj0iIlbkFuXLmiGESGR7uhRYCABAaCEAAAAAAAIBBYf2e/Sbw8FHlXreWH58cL/MLc+UHuWkSERbm8fYBABBqCEAAAAAAAICQ1d3dLd/srjeBh69273NrHZNSE+SKwjw5LmuohBF4AADAaQQgAAAAAABASAYePqmqk4VFZbJuT4Nb6zgiPdnMeDgsPYnAAwAAwRyA2L59u6xevVrKy8ulrq5OmpqazO/vvPNOfzcNAAAAAAAECS0m/V55rZnxULrv330LrpqWnSLzCnJlclqix9sHAMBg4tcARFtbmzz//POyePFiE4DoP1JBpzVaC0C8+eab8q9//cvcT05Olr///e8+azMAAAAAAAg8HV1d8tqOGllcXC47GlpcXl4rOpycl2ZmPIxPHuKVNgIAMNj4LQCxadMmuemmm6SkpMQEG5Qlj6LlsS3HHHOM3HzzzdLc3GyW+eyzz8zvAAAAAADA4NLS2Skrt+2WpSXlUtXU5vLyWkz6rJHDZG5+roxIjPNKGwEAGKz8EoDQoMNll10m+/fv7xNs6B2IsBeESExMlDPPPFOWL19uHr/66qsEIAAAAAAAGEQa2ztl+ZYqeaq0Uva0tru8fEx4mJwzOlMuy8+RrPgYr7QRAIDBzucBiMbGRrnmmmukvr6+Z8bD5MmTTUDi8MMPl9bWVjnrrLMcrue0007rCUB88sknXm83AAAAAADwv31t7fLspip5dlOl1Ld3urx8fGS4XDAmS2aPz5a02GivtBEAAPgpAPHYY49JVVVVT/DhuuuuMzcLLULtjKOOOkrCw8Olq6tLKisrpbq6WjIzM73WbgAAAAAA4D+1LW1mtoPOemjq6HJ5+aToSLl4XJZcNDbb3AcAAN7n029cTav0zDPP9AQffvjDH/YJPrgiLi5ORowYIdu2beupKUEAAgAAAACA0FLZ2CpLSspl5bZd0tZlv2akNWmxUXLp+Bw5f0ymxEdGeKWNAAAgAAIQa9eulb179/77hSMj5ec///mA1peXl9cTgHB25gQAAAAAAAh82/c3y6Licnl9R4102qkTaUt2fIxcnp8jZ4/KkJiIcK+0EQAABFAAYsuWLeanzoCYNGmSpKamDmh9WozaoqGhYcDtAwAAAAAA/lVS12gCD++U1YrrYQeRkQmxMq8wV04fPkwiwwk8AAAwaAIQtbW1Pfdzc3MHvD6tAWHR0dEx4PUBAAAAAAD/WFe7X54oKpNPqurcWj4/OV7mF+bJibmpEvGf1M8AAGAQBSA07ZInAwb79u3ruZ+UlDTg9QEAAAAAAN/Wivx6d70JPHyzu96tdRyclijzC3Ll2KyhPTUnAQDAIAxApKSk9Nyvqqoa8PqKi4t77g80nRMAAAAAAPBd4OGjyr2ysLhcvt/jXkrlIzOSZX5hrhw6LInAAwAAAcqnAYhRo0b1nGhs2LDB1G1ISEhwO/iwe/funseTJ0/2WDsBAAAAAIDnaTFpre2wqKhcNtU3ubWOE7JTTOBhYup/60ICAIDA5NMAhAYJkpOTpb6+3qRgevrpp+Xqq692a12PP/54z/3hw4dLdna2B1sKAAAAAAA8pb2rS17bUSOLi8tlZ0OLy8trBchThqfJvIJcGZc8hDcGAIAg4dMAhBaNPv300+XZZ581jx955BGZNm2aFBYWurSeV199VVauXNkzxfL888/3SnsxMJpma8mSJfLee+9JZWWlRERESF5enpxyyily+eWXm2AUAAAAACB0tXR2ystbd8nSkgqpbm5zefnIsDCZMTJdLi/IkeEJcV5pIwAACJEAhPrJT34iL7/8srS2tkpzc7PMnTtX7rjjDhOYcERnTSxevFjuu+8+E3zQVE7aia2d2QgsH374odx0001mtktvGzduNDcNQmkAatKkSX5rIwAAAADAOxraO2T5lmp5qrRS9ra2u7x8TES4nDsqQy7Lz5HM+BivtBEAAIRgACIzM1Nuvvlm+d3vfmeCCNpBfcMNN0h+fr6cdtppMmzYsD7P//jjj2XPnj2ydu1aeeedd8yoeg08WGZU/OEPf5AhQ5h+GUiKiorkZz/7mTQ1NUlcXJxcddVVcswxx0hnZ6d5D3VWRHV1tSxYsEBeeOEFs08AAAAAAILfvtZ2eWZzlTy7qVL2t3e6vHx8ZIRcMDZTLhmXI6mxUV5pIwAACOEAhLrkkktMIOHvf/97z0wGLSpdUlLS53n6+941IiyBB8sy2sl96qmn+rz9sO/OO+80wQdNufTYY4/JEUcc0fO3I488Ug466CD5xS9+YYqI33///XLXXXexSQEAAAAgiNU0t8mTpRWyYku1NHd2ubx8cnSkzB6XLReOzZLEaL90VQAAAC/QOk5+ceONN8q9997bM3vBUs9BAwt633LTx70DDyo2Nlbuuece+fGPf+yv5sOG9evXyxdffGHuz5o1q0/wweLcc8+Vo48+2tx/6aWXpLa2lu0JAAAAAEGoorFF/rxmi8x6fbVJt+Rq8CE9NkpuOHikvHjmoXLlhDyCDwAAhBi/DiuYMWOGSc2zdOlSUxOgpqbG/N4ScOhNf5eQkCAXXHCBSenTP1UTAsNbb73Vc1/fK1t++MMfyueff27SMr377rty4YUX+qiFAAAAAICB2lbfLIuKy+WNnbul88BLeIdy4mPk8oJcOXtkukRH+G1sJAAA8DK/z2tMTU2V66+/3tw2bdokq1evNumZ9u3bJ+3t7TJ06FDznClTpsjkyZNNWh94lhb3XrNmjZSXl8uuXbtMoCcrK8tsc932rvjmm2/MT639oO+XLUcddVSfZQhAAAAAAEDgK65rlIVFZfJe+R5xI+4goxLjZF5Brpw2fJhEhv87ywEAAAhdfg9A9DZu3DhzG+y6urpk8+bNpvD2unXrzE1rZGhAxkLrJpx//vkDep3m5mZ55JFHZPny5T2zT3qLioqSadOmmSLhBQUFTq1Tg0hq5MiREhlpe/fSwtOafquxsbFnGQAAAABAYFpbu1+eKCqTT6vq3Fq+YOgQuaIwV6bnpEr4f9IrAwCA0BdQAYjB7vXXX5cnn3zS1FHQIs7eVFpaamadbNmyxeZzNOCh6ZE++eQTueWWW0zxcHva2tpk79695r7OoHAkOzvbBB90xgsAAAAAILBoKuQvd+2ThUXlsrqm3q11HJKWaAIPR2cO7anrCAAABg8CEAFEUxF9+eWXXn8dTbP0ox/9SKqrq/v8fuLEiTJ8+HCpq6szsy50doJqbW2V3/72tyY108yZM22ut6Ghoed+fHy8w3ZYnmN5HQAAAACA/3V1d8tHlXtNqqUNe927Xjs6M1nmF+bJ1GFJHm8fAAAIHj4PQHz66ady2GGHSUxMjK9fOmglJiaazvr+AQN3R7DozIfe68rPz5e7775bCgsLe35XX18vDzzwgCkQbnHrrbea54wfP97qujVQ0Tt9kyPR0dEHLAcAAAAA8I+Orm55p6zGFJfeXN/s1jpOzEmV+YW5MiElwePtAwAAwcfnAYgrr7zSdDwfcsghcswxx8jRRx9t7lNc+t9iY2NlwoQJMmnSJFPEWW+jR4+Whx56yNwG6s033zQFpy3y8vJMkCE5ObnP85KSkuS2224zU2SXLFnSEyjQoIStdvQOKvWuV2EvZVP/5QAAAAAAvtXe1SWvbt8ti4vLpazR9QFiEWEipw4fZopLj0lyPBseAAAMHn5JwaSd019//bW5PfjggxIXFyeHH354T0BCO+AHo2uvvVZuvvlmu8WbB6p/8OD2228/IPjQ20033WTqQJSXl5vHb731lmzcuNHqe6QpmiycqWFheY4WowYAAAAA+FZLR6e8tG2XLC2pkF3N/x4g5oqo8DCZMTJd5ubnSm5CrFfaCAAAgptfAhCaBkjp6Hq9rx3RH330kbkp7RA/8sgjTUBCb6NGjZLBIDU11avrLy4ulpKSkp7HY8eOlenTp9tdRoNDs2fPlnvvvbfndytXrrQagNCZLSkpKaYQtTOFpS3PcaZgNQAAAADAMxraO+T5zdWybFOF7G3tcHn5mIhwmTU6U+bkZ0tGHDPaAQBAAAUgdAT+Z599Zm5btmw54O8akNAiyDrSXm8qMzPTzIzQmwYk9DFc99577/V5bK+gdP/n9Q5A6IyIX/ziF1afO27cOPnqq69k+/bt0tHRYXM2h9agsBSt1mUAAAAAAN5V19ouyzZVynObq6ShvdPl5YdERshF47Lk4nHZkhLjuO4fAACAzwMQp5xyirmp3bt39wQjvvjiC6moqLAakNCR8i+99JK5qREjRvSkazrqqKPMqHs49sknn/R5rGmvnJGdnS25ubk9aZi2bt0qlZWV5vf9aYFxDUA0NzfLunXrZOrUqVbX+eWXX/ZZBgAAAADgHbub2+TJkgpZsbVaWjq7XF5+aHSkXDI+Wy4YmyUJUX5JpAAAAIKUX88c0tPT5ZxzzjE3tWPHDvn88897AhJ79uyxGpDQ0fX63GeeecakcSooKJAVK1b44T8ILps2beq5Hx4ebgpdO0sLhVsCEKq0tNRqAOLUU0+VRx991Nx//vnnbQYgXnjhBfNTi4+fdNJJLv0fAAAAAADHKhpbZHFxhbyyfZe0d/07FbIr0uOiZc74HDlvdIbERUawyQEAgMsCauiCzmzQ20UXXdRTs0CDERqU0FH1jY2NB9SR0J9FRUV+a3Ow2LdvX5+ATlpamqnv4Ky8vLw+j3UWxAknnHDA8zSoofU7dIaDBoVmzZp1wEyLl19+2byv6txzzzVtAQAAAAB4xpb6JllcXC5v7qyRTtfjDpI7JEbmFuTKWSPSJToinLcFAACERgCiP53ZoLf58+dLZ2enSemzZMkSefXVVw8IRMA+nTHSm7XZC/b0LxTdf3293XrrrXLJJZeY4uJXXXWVXH311SZllr6H77zzjixevLhnBswNN9zAWwcAAAAAHrBxb4MsLCqXDyr2iDtXyqMT42R+Ya6ckjdMIsPDeE8AAEBoByCU1hqwzILQm9aNgOssBZ8tUlNTXVq+f52N/fv323xuYWGhPPDAA3LTTTdJfX29/PWvfzW33rSQ+COPPOJWQXGtCWLP3r17XV4nAAAAAASrNTX1srCoTD6v3ufW8hNShsj8glw5ISdVwsMIPAAAgBAOQNTV1fXUgdCf1kba9571kJycbApRw77e6atUTEyMS5ssNja2z2Od3WCPpmdauXKlme3w/vvvm0CS1p3QVE5aJ+Lyyy837507pk+fbvfvUVFRMmHCBLfWDQAAAADBQK+Lv6jeJ08Ul8m3NbYHiNkzdViizC/Mk6Mykk19RQAAgJALQGhHttZ3sMxw0LoPvQMM/VMsad2CQw891KT00dtBBx3EiZITmpub+zyOjo526X3qH7Dovz5baZt+8YtfmBsAAAAAYOC6urvlw4o98kRRuRTV9R1o5qxjMoeaVEtThiXxlgAAgNAKQHR0dMi3335rZjjobe3ataY2gK2AQ2RkpBx88ME9AYdDDjnEjHDHwLg6uqX/8/1Ze+ODDz5wmILp9ttv91l7AAAAAMDbOrq65a2yGllUVC5b9zseENafXtGdmJtqUi0VpiR4pY0AAAB+D0AcccQR0tLSYrMjWzu6tYbA0UcfbQIO+nyd9YCB6b8NW1tbXVq+93um4uPj/faW9C+I3R8BKgAAAAChoq2zS1Zt3y1LSsqlvNG16zgVESZy+vB0mVuQI6OT/HcdBwAABiefByA0dY8GGXoHHkaOHNkzw0HrOQwdOtTXzQp5/QMGrgYg+j/fnwEIAAAAAAh1zR2d8uLWanmypEJ2t7S7vHxUeJjMHJkhlxfkSM6QvjX9AAAABkUNCA1EHH/88TJr1iwz4yE1NdWfzQlpCQkJB6QpcsWePXv6PE5MTPRIuwAAAAAA/7W/rUOe31Ily0orpa6tw+VNExsRLuePyZRLx+dIepxrtf8AAABCJgBhqSnw8ccfm5s+Hj9+fM9MCE29xCh7zxkxYkSfx5WVlS4tX1VV1efx8OHDPdIuAAAAAIDI3tZ2ebq0Up7fXCWNHf+tk+isxKgIuXBsllw8LluGxlA3EQAADNIAxI9//GP5/PPPZf369QcUny4pKTG3RYsWSURERJ/i01OmTDEFqeEeTWulM0wsMxlqampMOixn62uUlZX1eTxmzJiAeStmzJjR57EGs6gbAgAAACAYVDe1ypOlFfLi1l3S2tnl8vIpMZFyyfgc+eGYTEmI4poZAAAEFp+fndx4443mZ0NDg3z11Vfy2WefmdumTZv61IXo6OiQNWvWmNsjjzwisbGxcvjhh/cEJCZMmODrpge9cePGyZdffmnud3V1mSCQzjRxxnfffXfAugAAAAAA7tnZ0CxLiitMgemOXtfCzsqIi5bL83PknFEZEhsZwdsAAAACUqQ/axL84Ac/MDelI/N1ZoTeNCCxc+fOPgEJHa1vSddkGdGvBastAYn+KYZwoGOPPbYnAKG+/vprpwIQmq6pvLy85/Ho0aMlJycnYDbxqlWr+jyura2VBQsW+K09AAAAAGDL5n1Nsqi4XN7aWSOuz3cQyRsSK/MKcuTMkekSFR7OhgYAAAEtYOZnanqgs846y9xURUVFz+wIDUpoyqDeAQktovzGG2+Ym6bc2bBhgx9bHxxOOukkuf/++3ser1y5Uq699lqHy+nz+q8HAAAAAOC8DXsaZGFxmXxQsdetzTY2KU7mFebJyblpEhn+75qKAAAAgS5gAhD96Qj7H/7wh+amNm/ebIIROgPigw8+ML/rHZCAYwUFBZKfn2/qbFi2qW7L6dOn21ympaVFli1b1ud3M2fOZHMDAAAAgAN6zbqmpl6eKCqXL3ftc2t7TUxJkPmFuXJ8doqEhxF4AAAAwSVgAxC9aYe5JTWTpg2y0JkPBCFcc91118n111/f8/iOO+4wBb6Tk5OtPv/ee+/tk37plFNOof4GAAAAANih16mfVtXJwuJyWVu7361tdWh6klxRkCtHZCSba18AAIBgFJABiLKysp7US3rT+hCDhf7v1tTX1/d5rCmorD03JiZG0tPTba7/tNNOk6lTp5ri3kprbVx22WVyzz33mBkSFvv37zfpmpYuXdpn3TfccINb/xcAAAAAhLqu7m55r3yPqfFQXNfo1jqOyxoq8wvz5OC0RI+3DwAAwNfCugNgCoEWDe5dgLr3iHtL8ywjPno3Nzc3t6cI9YwZMyQU9A4CuOPII4+UJUuW2H1OdXW1XHDBBbJr166e3+n2nThxogwfPlzq6upk7dq10tjY94T57rvvlnPOOUcCXe8i1I8++qikpaX5u0kAAAAAQlhHV5e8ubPWBB627W92eXm92j0pN82kWsofOsQrbQQAABg0MyAaGhrkq6++6ikyvWnTJpsBh96/HzZsmBx11FFy9NFHm6BDXl6ez9seCjIzM+Wf//ynScW0devWnu27fv16c+tPZz788pe/DNjgQ//gk+47cXFxfmsPAAAAgMGhtbNLXtm2S5aUVEhlU6vLy0eEhckZI4bJvIJcGZnINQwAAAg9Pg9AXHzxxfL9999LZ2fnATMaegcd9PeJiYlmRL8l4DBu3DhfNzdkaTHqFStWyMMPPyzLly83swb6i4qKkmnTppm0SwOdmQEAAAAAoaKpo1NWbKmWp0orpKal3eXlo8PDZOaoDLk8P1eyh8R4pY0AAACDMgVTYWFhT/Ho/kWkY2Nj5dBDD+0JOGhKoPDwcF82b1Dq6OiQ1atXm5oSNTU1MmTIEMnKyjK1IlJTUyXYkIIJAAAAgDfUt3XIs5sr5ZlNVea+q+Ijw+X8MVlyybhsGRYX7ZU2AgAAyGBPwWQJOkRERMjkyZNNsEGDDlOmTJHoaE7CfC0yMtLMNNEbAAAAAKCv2pY2WVZaKc9vqTazH1yVFBUhF43LlovGZUlydBSbFwAADBo+D0BMmDDBBBv0dsQRR0h8fLyvmwAAAAAAgEPVTa2ytKRCXtpaLa1dricPSI2JkkvHZ5tZD0OiItjiAABg0PF5AELrDgAAAAAAEKh27G+WxSXl8tr2GulwI2txVny0XJafKzNHpUtsBIEHAAAwePklBRMAAAAAAIFm075GWVhULu+U1UqXG8uPSIiVuQW5csaIYRJFPUMAAAACEAAAAACAwW39nv0m8PBR5V63lh+fHC/zC3PlB7lpEhEW5vH2AQAABKuAmwGhBar3798v+/btM4+Tk5MlMTFRwjiJAwAAAAB48Nrzm931JvDw1e5/X3+6alJqglxRmCfHZQ3lmhUAACBQAxDff/+9vPzyy7JmzRrZuHGjdHR09Pl7ZGSkKV49ZcoUOeecc2TSpEl+aysCz4wZM/o81mBVXFyc39oDAAAAILADD59U1cnCojJZt6fBrXUckZ5sZjwclp5E4AEAACBQAxAlJSXy29/+1gQeLCeC1rS3t8u6devMbcmSJSYQocsVFBT4uMUAAAAAgGDU2d0t75XXmhkPpfua3FrHtOwUmVeQK5PTEj3ePgAAgFAU1m2r19/LnnvuOfnDH/4gbW1tJvDQO8VS/yZZ+1t0dLTceuutcvHFF/uw1QgGtbW1smDBAnP/0UcflbS0NH83CQAAAICfdHR1yes7amRxcblsb2hxeflwETk5L03mFebK+OQhXmkjAABAqPLLDIiVK1fK7bff3hN40Jvlfl5enowePdrUfVBaD2Lbtm2yc+fOPoEKDVz87ne/M6l2NC0TAAAAAAAWLZ2dsnLbbllaUi5VTW0ubxgtJn3WyGEyNz9XRiSS4hUAACAoAhDV1dVy2223mfuWwENGRoZcddVVcvbZZ0tqaqrV5fbs2SOvvPKK/Otf/5KqqiqzbFdXlwlkHHnkkZKVleXj/wQAAAAAEGga2ztl+ZYqeaq0Uva0tru8fEx4mJw7OlPm5OdIVnyMV9oIAAAwWOhsUp+6//77paXlv9NezzjjDHnttddk7ty5NoMPSv+mz3n11VflrLPO6pkN0draKn/961991HoAAAAAQCDa19Yuj2/YKee99o08tH6Hy8GH+MgImZufIyvOPFRumjKa4AMAAECwzYDQYtJvvPFGTxqladOmmYCEK+Lj4+Xee++VhoYG+fDDD83vdJ2///3vJTLSrzW1AQAAAAA+VtvSJk+XVsoLW6qkqaPL5eWToiPl4nFZctHYbHMfAAAAnuPTs6s1a9ZIU1OTuR8eHm7SJ7lDAxiaxum0004zMyF0natXrzapmAAAAAAAoa+ysdXUd1i5bZe0dnW7vHxabJTMGZ8js8ZkmtkPAAAACPIARFlZWU8A4aCDDjIFp901fPhwmTRpkqxbt8481iLVBCAAAAAAILRt398si4rL5fUdNdLZ7XrgITs+Ri7Pz5GzR2VITITPsxIDAAAMKj4NQNTW1vYJIAyUBjAsAQgtUo3BacaMGX0ea4ArLi7Ob+0BAAAA4HkldY0m8PBOWa24HnYQGZkQK/MKc+X04cMkMpzAAwAAQMgFIKKionrua/HogWpra7O6bgAAAABAaFhXu18WFpXLx1V73Vo+Pzle5hfmyYm5qRLxn3qEAAAACMEARFpaWs/90tLSAa+v9zpSU1MHvD4Ep1WrVh0w02bBggV+aw8AAACAgdFaf1/vrpcnisrkm931bq1jcmqCXFGYJ8dmDTWzpAEAABDiAYjx48f3nExqPYgvvvhCjjrqKLfW9dVXX8mOHTt6Hufn53usnQAAAAAA39NrxY8r98oTxeXy/Z4Gt9ZxZEayzC/MlUOHJRF4AAAAGEwBiMLCQsnKypLq6mpzYnn77bfLsmXLJCUlxaX11NXVmWUtMjIyzLoBAAAAAMFHi0lrbQet8bBpX5Nb6zghO8UEHiamJnq8fQAAAHCPzytvzZ492wQfdArs9u3b5ZJLLpG1a9c6vfyGDRvksssuk23btpnHuh5dBwAAAAAguLR3dcnL23bJxW9+K7d9Wepy8EEvaE8bniZPnnKw3H1sIcEHAACAwTwDQl1xxRXy7LPPSmVlpQkeaCBBgxLTpk2TGTNmyCGHHCIjR47ss4ymWvruu+/k1VdflQ8//FC6urp6/padnW3WCQAAAAAIDi2dnfLy1l2ytKRCqpvbXF4+MixMZoxMl8sLcmR4QpxX2ggAAIAgDEDExMTIP/7xD5kzZ47U19ebIIQGFDSwoDelv4uL+/dJZHNzs5kxYWGZPaE/k5OT5e9//7tZJwAAAAAgsDW0d8jyLdXyVGml7G1td3n5mIhwOXdUhlyWnyOZ8VwHAgAABDqfByDUuHHjZNGiRfLzn/9ctmzZ0lMYzBJo0J+NjY0HLKfPswQfRo8eLffdd19PYWsAAAAAQGDa19ouz2yukmc3Vcr+9k6Xl4+PjJALxmbKJeNyJDU2yittBAAAQIgEIJQWjV6xYoX885//NIWod+3a1efv/YMSlvvp6ekmZdNVV13FzAcAAAAACGA1zW3yZGmFrNhSLc2d/02l66zk6EiZPS5bLhybJYnRfrt8BQAAgJv8eganqZP+3//7f3LNNdfIF198IWvWrJH169fL3r17TXomlZSUJCkpKTJp0iSZOnWqHHXUURIZyYknAAAAAASqisYWU99h5bZd0tb130FlzkqPjZI5+Tly7uhMM/sBAAAAwSkgevI1oHDccceZGwAAAAAgOG2rb5ZFxeXyxs7d0ul63EFy4mPk8oJcOXtkukRHhHujiQAAABhsAQhgIGbMmNHnce8i5gAAAAC8r7iuURYWlcl75XvEjbiDjEqMk3kFuXLa8GESGf7vdLwAAAAIfgQgAAAAAABu+a6mXp4oKpfPquvcWr5g6BC5ojBXpuekSvh/6gACAAAgdBCAQNBbtWpVn8e1tbWyYMECv7UHAAAACGXd3d3y5a59srCoXFbX/Lt2n6sOSUs0gYejM4eaGcwAAAAITQQgAAAAAAAOdXV3y0eVe02qpQ17G93aYkdnJsv8gjyZmp7EFgcAABgEPBqAqKioEH/Jycnx22sDAAAAQKjq6OqWd8pqTHHpzfXNbq3jxJxUmV+YKxNSEjzePgAAAAySAMRJJ53kl+mz+pobNmzw+esCAAAAQKhq7+qSV7fvlsXF5VLW2Ory8uEipqj0vMJcGZMU75U2AgAAYBCmYNKcoAAAAACA4NPS0Skvbt0lT5ZWyK7mNpeXjwoPkxkj0+Xy/FzJS4j1ShsBAAAwSAMQBB8AAAAAIPg0tHfI85urZdmmCtnb2uHy8jER4TJrdKbMyc+WjLgYr7QRAAAAgzgAcd1113lydQAAAAAAL6trbZdlmyrluc1V0tDe6fLyQyIj5MKxWTJ7fLakxER5pY0AAAAITgQgAAAAAGAQ2t3cJk+WVMiKrdXS0tnl8vJDoyPlkvHZcsHYLEmI8kp2XwAAAAQ5zhIBAAAAYBApb2iRJSUV8sr2XdLe5Xr9vvS4aJkzPkfOG50hcZERXmkjAAAAQgMBCAAAAAAYBLbUN8mi4nJ5a2eNdLoed5DcITEytyBXzhqRLtER4d5oIgAAAEIMAQgAAAAACGEb9zbIwqJy+aBij7gRd5DRiXEyvzBXTskbJpHhYV5oIQAAAEIVAQgAAAAACEFrauplYVGZfF69z63lJ6QMkSsK8mRaToqEhxF4AAAAgOsIQAAAAABAiOju7pYvqvfJE8Vl8m3NfrfWMXVYoswvzJOjMpIljMADAAAABoAABILejBkz+jzWi6S4uDi/tQcAAADwta7ubpNiSVMtFdU1urWOYzKHmlRLU4Ylebx9AAAAGJwIQAAAAABAkOro6pa3ympkUVG5bN3f7PLymljpxNxUmV+QK4UpCV5pIwAAAAYvAhAIeqtWrerzuLa2VhYsWOC39gAAAADe1tbZJau275YlJeVS3tjq8vIRYSKnD0+XuQU5Mjop3ittBAAAAAhAAAAAAECQaO7olBe3VsuTJRWyu6Xd5eWjwsNk5sgMubwgR3KGxHqljQAAAIAFAQgAAAAACHD72zrk+S1Vsqy0UuraOlxePjYiXM4fkymXjs+R9Lhor7QRAAAA6I8ABAAAAAAEqL2t7fJ0aaU8v7lKGjs6XV4+MSpCLhybJRePy5ahMVFeaSMAAABgCwEIAAAAAAgw1U2t8mRphby4dZe0dna5vHxKTKRcMj5HfjgmUxKiuOwDAACAf3AmCgAAAAABoqyhRRYXl5sC0x3d3S4vnxEXLZfn58g5ozIkNjLCK20EAAAAnEUAAgAAAAD8bPO+JllUXC5v7awR1+c7iOQNiZV5BTly5sh0iQoP90ILAQAAANcRgAAAAAAAP9mwp0EWFpfJBxV73Vp+bFKczCvMk5Nz0yQyPMzj7QMAAAAGggAEAAAAAPhQd3e3rKmplyeKyuXLXfvcWsfElASZX5grx2enSHgYgQcAAAAEJgIQAAAAAOCjwMNn1XUm8LC2dr9b6zg0PUmuKMiVIzKSJYzAAwAAAAIcAQgAAAAA8KKu7m55r3yPqfFQXNfo1jqOyxoq8wvz5OC0RI+3DwAAAPCWoAxAVFRU9NyPj4+XoUOH+rU9AAAAANBfR1eXvLmz1gQetu1vdnkDaWKlk3LTTKql/KFD2MAAAAAIOkEZgDjppJN6phtHRETIzJkzZcGCBTJy5Eh/Nw0AAADAINfa2SWrtu+SxcUVUtnU6vLyEWEiZ4xIl7n5uTIqKc4rbQQAAAB8ISgDEJb8qaqjo0NefPFFWblypZx11lkmEDFmzBh/Nw8AAADAINPU0SkrtlTLU6UVUtPS7vLy0eFhMnNUhlyWnyM5Q2K90kYAAADAl4I2ANG/4JoGIjQIsWrVKjnjjDPk2muvlXHjxvmtfQAAAAAGh/q2Dnluc5Us21Rp7rsqLiJczh+TKZeOz5FhcdFeaSPw/9m7D+g4quvx41e9S7Z6cS+SjLHpBkLHtNh0THHDNhDiJED4JdQQCBAChJKQhOIQCO7UmIAxvRkwxXRsjOTe1GzJRbKtLv3PffxnM1qr7K62afX9nLPHu/LO7NvZ2dmZd9+7FwAAIBB6/AwI5781NzfLq6++Kq+//rp8//33AWkbAAAAgNC3o65Rnl5bKi+sqzCzH9yVHBUhFw3LkYuGZktKTJRP2ggAAAAEUo8MQBQVFTnuNzQ0yLfffivLly+XL774Qr755hupra2VlpaWgLYR/jN+/Pj9ZsfExZErFwAAAL5Rsa9e5q8ulZc2VEh9y/4Do7rSNyZKJg/PkfOHZEtCVIRP2ggAAAAEgx4ZgLCLjo6WI444wtysVEwrVqyQzz//PNBNAwAAABBCNtfUytzVJfLapkppamdGdley4qJlan6unDU4U2IjCDwAAAAg9IW1tpfLCOjBqqqqTDFyNWvWLElLSwt0kwAAANCDrd29V2YXlcg7W6vEk3nW/RNjZVpBnpwxIF2iwsN90EIAAAAgOPX4GRAAAAAA4Avf76iRp4pK5MOynR4tPywlXqYX5MnJ/dIkIizM6+0DAAAAgh0BCAAAAAD4/3SC+Jfbq2VOcYks37bbo+0yMjVRZhTkybE5fU19MgAAAKC3IgABAAAAoNfTwMOy8l0yu2irrNixx6PtcXhGsswo7CeHZSQTeAAAAAAIQAAAAADozZpbW+W9kipT42HN7n0erePY7L4yvTBPRqUleb19AAAAQE8W1DMgioqKZOXKlbJz506JjY2VzMxMOeKIIyQ1NTXQTQMAAADQgzW1tMjrmytlbnGJbNpT5/bymlhpbL80U1w6v0+CT9oIAAAA9HR+CUBUVVXJ6tWrTSAhOjpa+vXrJ/n5+RIeHt7u899++2156KGHZN26dfv9ny5zwgknyE033SQDBgzwQ+sBAAAAhIq65mZ5ZeN2mb+6VMr21bu9vBaT/umAdBN4GJAU55M2AgAAAKHCpwGIb775Rh544AH56quvTE5Vu759+8rkyZPlyiuvlKioKMff9flPPvmkue+8jGpubpZ3331XPv30U3n88cfl8MMP9+VbAAAAABAC9jY2y6L15bJwTZnsqG90e/no8DA5e1CmTMnPk5yEGJ+0EQAAAAg1PgtAPP300/LHP/7RBBHaCyTs2LFDHn74Yfnss89MwEGDEAsXLpQnnnjC/H9YWJi52ZfVx0r/tm/fPhO8ePnll82MCgAAAABwtruhUZ5fWy7Pri2T6sZmtzdQfGS4XDAkWyYOz5G02Gg2MAAAABDoAMSyZcvkjjvuaBNIUFYwwf74888/l0cffVQuu+wyefDBB9v8X05OjhQWFkpCQoLs3r3bUQ/CCkxoEOKuu+6SWbNm+eJtAAAAAOihquoa5Ok1ZfKf9eWyr6nF7eWToyLk4mE5cuGwbEmJ/t+MbQAAAAABDEBoYOD22283961Agf6rgYSBAweav2/evFl++OEHx/8vWLBAUlJSZO/eveZvWmz67rvvlmOOOWa/dS9evFjuvPNO81x9vHTpUtmyZYv079/f228FAAAAQA9Tvq9e5hWXyOKN26S+Zf+Z2F1Ji42SScNz5bzBWZIQFeGTNgIAAAC9hdcDEB988IEJCFgzGYYPH25mNmjRabs1a9bIddddJ8XFxVJTU+NIvRQbGyvz5s1rt8C0rvPss8+W7OxsmT59uuPvS5YskZkzZ3r7rQAAAADoITbX1Mqc4hJ5bXOlNLeTArYr2fHRMjU/T84alCkxEeE+aSMAAADQ23g9APHee++Zf3V2gs5qeOqppyQ9PX2/52lgQms/jB8/Xqqrq6WystIEGC6++OJ2gw92Y8aMkZNOOkneeecds8x3333n7bcBAAAAoAdYvWuvCTy8s7VK3A87iAxIjJVpBXlyxoB0iQwn8AAAAAAEdQDi+++/N/9qYOD8889vN/hg0f+74IIL5N///rfjb2PHjnXpdU455RQTgFBr167tdrsBAAAA9BwrqmpkdlGJfFS+06Plh6fEy4zCfnJiXqpE/P/Z2wAAAACCPABRVlbmuH/UUUd1+fwjjzyyTQBi6NChLr2OzqCwZlpogWoAAAAAoU3P/b/YXi1PFW2VL7dXe7SOUamJMr2wnxyT3ceRNhYAAABADwlAaD0HS15eXpfPd36Opm1yRXJysuO+FqQGAAAAELqBh4/KdspTxSXy/Y49Hq1jTGaKSbV0WEYygQcAAACgpwYg6uvrHffj4+O7fH5cXFybxxERES69jv15zc3NbrURAAAAQPDTYtJa20FrPKzdvc+jdRyX01emF+bJgalJXm8fAAAAAD8HIAAAAACgOxpbWuS1zZUyt7hEtuypc3t5LSU9tl+aCTwMS0ngwwAAAAAChAAEAAAAgKBQ19wsizdsk3mrS6WitsHt5SPDwmTcwAyZmp8rA5LazrQGAAAA4H8EIAAAAAAE1J7GJlm0vkIWrimTnfWNbi8fEx4m5wzOkin5uZIVH+OTNgIAAABwHwEIAAAAAAGxu75Rnl1XLs+tLZOaRvfrusVHRsiEoVkycViupMZG+aSNAAAAAII0ABEWFubL1QMAAADogSprG2ThmlIz66G2ucXt5ZOjI+WSYTly4dBscx8AAABAcIr0ZeBh4sSJEhER0elzm5vbjnQaO3asS6/hvBwAAACA4Fa6t07mry6VxRu3SUNLq9vLp8dGyeThuXLukCwz+wEAAABAcPPZcKHW1lYpLy93e5mSkhK3Ah26DAAAAIDgtbG6VuYUl8gbW7ZLswen7znxMXJpQa6MH5gpMRHhvmgiAAAAgJ4UgCD9Evxl/Pjx++17cXFxfAAAAAABVrxrr8wu2irvlewQT4YNDUqKk2kFeXJa/zSJDCfwAAAAAPQ0PglAMCsBAAAA6L2+q6qRp4q2ysfluzxavqBPgkwvzJMTc1MlnLpyAAAAQI/l9QDEO++84+1VAp1asmRJm8dVVVUyc+ZMthoAAIAf6SCk5dt2y+yiEvmqstqjdRyUliQzCvPkqKw+zKgGAAAAQoDXAxB5eXneXiUAAACAINXS2ioflu2UOUUl8v3OPR6t48jMFJlR2E8OyUj2evsAAAAAhGANCAAAAAChq6mlVd7ZWmmKS6+rrvVoHSfk9pXpBf3kgNREr7cPAAAAQOARgAAAAADgssaWFnl103aZW1wiW/fWu73ltJT0af3TZVphngxJjmfLAwAAACGMAAQAAACALtU1NctLG7fJ/NWlsq22we0tFhUeJuMHZsjU/DzplxjLFgcAAAB6AQIQAAAAADq0p7FJXlhXIc+sLZWd9U1ub6mYiHA5b3CWTM7Pkcy4GLY0AAAA0IsQgAAAAACwn131jfLM2jJ5fl257GlsdnsLJURGyIVDs+WS4TnSNyaKLQwAAAD0QkETgLj00kvdXubMM8+Uiy66yCftAQAAAHqj7bUNsnBNqSxaXyF1zS1uL98nOlImDs+RCUOzJTEqaC43AAAAAASAT64IKioq5LTTTpOGhh9zw6akpMhLL70kWVlZHS6zfPlyCQsLc/k1WltbZdWqVTJ27FhJS0vzSrsBAACA3qp0b53MLS6VVzZtk8aWVreXz4iLlsnDc+XcwZkSFxnhkzYCAAAA6Fl8EoB49NFHpb6+3tzXoMJdd93VafDBObDgaiBi79698ve//13uuOOObrUXAAAA6K02VO+TOcUl8uaWSml2P+4geQkxcmlBnowbkCHREeG+aCIAAACAHsrrAYgdO3bIf/7zH0cQYfz48XLKKae4vLwup0EIV5/38ssvy/XXXy+JiYndajcAAADQmxTt3COzi0vk/ZId4kHcQQYnxcn0wjw5pV+6RIa7PpMZAAAAQO/h9QDEW2+9JU1NTeZ+RESE/PKXv3R5WWv2w4IFCyQ7O7vD5+3evVsmTZokdXV15vbaa6/JhRde6JX2AwAAAKHsm8pqmV1UIp9U7PJo+RF9E2RGQT85LrevhLuRQhUAAABA7+P1AIQGA5QGEnTmw5AhQ9xeR05OjuTm5nb4/3l5eSbgMG/ePPP41VdfJQABAAAAdDLQ57OK3fJU8Vb5prLGo+10SHqSTC/sJ0dmprhVuw0AAABA7xXp7Qubb7/91vH49NNPF185++yzTQBCX3PFihU+ex0AAACgp2ppbZUPSneYVEs/7Nzr0TqOzupjUi0dnJ7s9fYBAAAACG1eDUBs2LBBamtrzf3w8HA55phjxFdGjRpl6j7s2bPHFKNet26dDB061GevBwAAAPQUTS2t8vbWSlNcen31j+fn7tD5DSfmpcr0gjwp7EutNQAAAABBEIBYu3at+VenZA8aNEhSUlLcWt7dqdwacLBmXGjwgwAEAAAAerOG5hZZsmm7zFtdIiV7691ePiJM5PT+GXJpQa4MTo73SRsBAAAA9B5eDUBocWhLenq628tbRahdZX+NXbs8K6IHAAAA9HS1Tc3y3w3bZMGaUtle2+D28lHhYXLWwEyZWpAruQmxPmkjAAAAgN7HqwGI6upqjwMQEyZMcNxPSEhwaZn4+Ph2XxsAAADoDWoamuSF9eXyzJoy2dXQ5PbysRHhcv6QLJk0PFcy4qJ90kYAAAAAvZdXAxANDf8bbdXY2OjWsnfddZfbr1dfX9/uawMAAAChbGd9owk6PL+uXPY2Nbu9fGJUhFw4NFsuGZYjfWKifNJGAAAAAPBqACIuLs5xf+fOnT7fuva0S7GxTBUHAABAaKvYV2/SLGm6pfrmFreX7xsTKROH58oFQ7IkMcqrlwIAAAAAsB+vXnXY0y5VVFSIr5WVlTnup6Wl+fz1AAAAgEDYuqdO5haXmALTTa2tbi+fGRctU/Jz5ZxBmRIbGeGTNgIAAACATwMQAwYMcBST3rJli5SXl0t2drb4Kvigr+H82gAAAECoWF+9T+YUlcibWyrF/fkOIv0SYmVaQa78dGCGRIWH+6CFAAAAAOCnAERhYaFERUVJU9OPBfCWLl0qF198sfjC+++/77gfGRkpI0aM8MnrAAAAAP72w8498lTRVlla6lla06HJcTKtsJ+MzUuTyPAwr7cPAAAAAPwegIiOjpYxY8bIsmXLzOMnnnhCJkyYIBER3p3mrQWudd1hYT9eTOlr6msjOFRXV8uKFSvku+++Mze9v337dsdnNW/evEA3EQAAICh9vb3aBB4+27bbo+UP6Jsg0wv7yXE5fSX8/58rAwAAAECgeL3y3FlnnWUCEBoc2Lp1qzz55JNy5ZVXevU1NPhQUlJi7uvr6GsieJx77rmOzwcAAACd0/Sln1TsktlFJfJtVY1Hm+vQjGSZUZAnR2SmOAbpAAAAAECgeT0R7Pjx4x11H/Ri6qGHHpJXX33Va+t/+eWX5e9//7vjwiozM9O8JoKTFiY/6aSTAt0MAACAoNPS2irvllTJtHdXyP8tK/Io+PCT7D7yrxMPlMeOHyljsvoQfAAAAAAQ2jMgtAbEddddZ24aJGhpaZHrr79eVq1aJddcc43HqZIaGhrkr3/9q8ydO9cENpSuX19HXxPBY/LkydKvXz8ZPXq05OTkmL8VFBQEulkAAABBoamlRd7cUiVziktkY02t28vrMJyT8lJNqqWCPgk+aSMAAAAABGUAQp155pny3nvvyZIlS0yQoLm52aRieu2110xRak3RozMXXLFt2zZZtGiRPPfcc1JWVmaCD9bsh3HjxpF+KQhdfvnlgW4CAABA0KlvbpElm7bJ3OJSKdtX7/byEWEiZwzIkEvz82RQcpxP2ggAAAAAQR+AUPfcc49UVlbKZ599ZgIGGjjQugA6i0Fv/fv3l5EjR8rAgQMlKSnJ3PR5NTU15rZx40ZZuXKlqSOh7LMe9P5RRx1lXqO3aGpqkq+//tpsQw3KJCYmmlRXBx98sKSmpga6eQAAAOhAbVOzvLi+QhasKZXKuka3t1N0eJicNShTpuTnSm5CLNsZAAAAQI/hswCEplrSYtH33nuvLFiwwDFrwQokbN68WbZs2dLpOqzn2gMPepsyZYrceOONAU29pKml1q1bJ999952sWLHC3IqLi6Wx8X8XlRogOf/887v1OrW1tfLoo4+aWSAa0HGm2+C4446Ta6+9ljRHAAAAQaSmoUmeW1cuz64tk90NTW4vHxcRLucPyZJJw3MlPc6zNKYAAAAAEJIBCKtz/NZbb5UxY8bI/fff75jNYAUjnIMMdvoce9BCb1pX4IYbbpDTTjtNAuX11183ARWdnbFv3z6fvtaaNWtM3Yz169d3+BwNeLz77ruybNkyufnmm2XixIk+bRMAAAA6t6OuUZ5eWyovrKuQfU3Nbm+upKgIuWhYjlw8NFtSYqh1BgAAAKDn8mkAwnL66afLqaee6ui8//bbb01Koc5YgYnIyEg56KCDzKwHXU94eLgE0pdffinLly/3+etomiWtpVBRUdHm75q2StNX7dq1y8y62Lt3r/l7fX293H777SY101lnneXz9gEAAKCtin31Mn91qby0cZup9+CuvjFRMnl4jpw3JEsSo/xymg4AAAAAPuW3KxsNHGjRaL1pWqFvvvnGBCK2b99uOtOrq6vN85KTk6VPnz6Snp5u6hvoLS4u+IvsaQ2L+Pj4/QIGntDgi858sK8rPz/fzCIpLCx0/E232d/+9jeZP3++42+33HKLec7w4cO73Q4AAAB0bcueWlNY+tVN26Wpg9m9ncmKi5ap+bly1uBMiY2IYJMDAAAACBkBGVqlAYWjjz7a3Hqi2NhYGTFihBx44IEyatQocxs8eLA8/PDD5tZdb775pik4bdHUUxpkSElJafM8DdZoiitNVTVv3jzHTAgNSnijHQAAAOjYut37ZHbxVnl7S5W4P99BpH9irEwryJMzBqRLVIBn+QIAAACALzC3202/+MUvTAFsTQ3lK87Bg9tuu22/4IPdb3/7W1MHoqSkxDx+66235IcffjBBkvZokGL06NHdbudVV10lV199dbfXAwAA0JN8v6NGZheVyAdlOz1aflhKvEwvyJOT+6VJhK02GgAAAACEGgIQbkpNTRVfKi4ultWrVzseDx06VE444YQuZ5Rccskl8uCDDzr+tnjx4g4DEAAAAHA/ReZXldUm8LB8226PNt/I1ESZUZAnx+b0NTNYAQAAACDUEYAIMu+9916bx64WlNbn2QMQOiPihhtuaPe50dHR8uqrr3azpSJ9+/bt9joAAACCPfCwrHyXzCkuke+qajxax+EZyTKjsJ8clpFM4AEAAABAr0IAIsgsW7aszePDDz/cpeVycnIkLy/PkYZpw4YNUlZWZv7uTEfc6cwKAAAAtK+5tVXeL9khs4u2yurd+zzaTMdm95XphXkyKi2JzQwAAACgV/JqAOLee++VmTNnSp8+fSTQPvjgA9mzZ4+MGzdOepK1a9c67oeHh5tC16466KCDHAEItWbNmnYDEAAAAGhfU0uLvL65UuYWl8imPXVubyZNrDS2X5opLp3fJ4HNDAAAAKBXC/fmymbPni2nnnqqKaK8a9cuCYRPPvlEpk6dKj//+c9l/fr10pPs3r1bduzY4XiclpZm6ju4ql+/fm0e6ywIAAAAdK2+uUVeWFcuE974Rv745Tq3gw9aTPrMgRny3GkHy5+OzCf4AAAAAAC+SMGksw4eeeQRefLJJ+WCCy6Qiy66SPLz8326sevr6+W1116TefPmyapVq6Sn2rx5c5vH7s5eyM7O7nR9AAAAaGtvY7O8uKFCFq4plaq6Rrc3T0x4mJw9OEum5OdKdnwMmxcAAAAAfBWAuPjii+X555+XlpYWqa2tlQULFpjbqFGjZPz48WZ2RG5urldeq6mpycx2ePPNN+X11183gQ+rUKAaMGCAHHfccdKTWO/Bkpqa2q2i0DU1nhVK7K4ffvjB3Nqzfft2WbRoUZu/6eeUkZHhp9YBAACI7G5olOfXlsuza8ukurHZ7U0SHxkuFwzJlonDcyQtNppNCgAAAAC+DkDccccdJghxzz33yOeff+4IBqxYscLctEbEkCFD5IgjjpBDDz1UCgoKzOOoqKgu160d11rTYOXKlWbdX331lezb92NBQOt1VFJSkvzsZz+T6dOnS3R0z7oY3Lt3b5vHMTHujaKLjY1t89jaPv729ttvmzRc7dG0UDfffHObv82dO9etAER5eXmn/79z506X1wUAAHqXqroGeXpNmfxnfbnsa2pxe/nk6Ei5eFi2XDg0W1Kiuz6HBQAAAIDezOspmA444ACTCmnZsmXy97//Xb799lsTIAgLCzP/rlu3ztRmePbZZx2FltPT0yUzM1OSk5NNp3tkZKRJq6Qd6JWVlbJt27b9OtPtQQeVkJAgU6ZMkcsvv9yspyfSWSN27gZQnAMWzusLFSeccEKn/68BrREjRvitPQAAIPiV76uXecUlsnjjNqlvaXse6Yq02CiZNDxXzhucJQlRET5pIwAAAACEGq8HICzHHHOMuX355ZcyZ84ceffdd03aJOcAQnNzs1RUVJibBimcOQcanJ+jqZYmTZokEyZMkMTERAkl7W0Pd57vvO385eqrrzY3AACAQNtcUytzikvktc2V0uzBuVFOfIyp73DWoEyJiQj3SRsBAAAAIFT5LABhOeyww8xtx44dplC01mv4+uuv2wQjOutot2ZOWPS+Fls+6aST5KyzzjKpnEJFXFxcm8c6C8QddXV1bR7Hx8dLKFq6dGmXKZhuu+02v7UHAAAEn9W79prAwztbq8STIRkDE2NlWmGenN4/XSLDCTwAAAAAQFAGIOwFlSdPnmxuWmx5+fLl8t1338mqVatk06ZNUlZWJg0NDfstl5aWJnl5eZKfny8HHnigHH744TJs2DAJRc4BA3cDEM7PD9UAhAagOuNKTREAABCaVlTVyOyiEvmo3LOaUPkp8TK9sJ+cmJcqEW7ORgUAAAAABCgAYaepkk4++WRzs9PAhNYt0LRMWlBZ6zr0ps5k5xRS7hZT1lkmdlqQGwAAINTpDNkvtlfL7KKt5l9PjE5LkukFefKT7D5up8EEAAAAAARRAKKzDvhQq+PgDq1nYaezQtxRXl7e5nH//v290i4AAIBgDTx8VLZTniouke937PFoHWMyU2R6YZ4cmp5M4AEAAAAAQjkA0dv16dPHpKqyZjJUVlaaGSHOtSE6snXr1jaPhwwZ4pN2AgAABJIWk353a5XMLi6Rtbv3ebSO43P6msDDyFRmjAIAAACArxCACDJa30LrY6iWlhZZuXKlHHHEES4t++233+63rt5g/PjxbR5r2gRXgzYAAKDnaGxpkdc2V8rc4hLZsqfO7eW1lPQp/dNkWkGeDEtJ8EkbAQAAAAD/QwAiyPzkJz9xBCDUF1984VIAQtM1lZSUOB4PHjxYcnNzfdZOAAAAf6lrbpbFG7bJvNWlUlHb4PbykWFhMm5ghkzNz5UBSQxSAAAAAAB/IQARZLQw90MPPeR4vHjxYvnFL37R5XL6POf19BZLlixp87iqqkpmzpwZsPYAAADv2NPYJC+ur5CFa8pkR32j28vHRITLOYMyZUp+rmTFx/CxAAAAAICfEYAIMgUFBZKfny+rV682j9etWydLly6VE044ocNl6urq5Jlnnmnzt7POOsvnbQUAAPCF3fWN8uy6cnlubZnUNDa7vXx8ZIRMGJolE4flSmpslE/aCAAAAADoGgGIIHTVVVfJNddc43j8xz/+UQ4++GBJSUlp9/kPPvhgm/RLp5xyiowYMcIvbQUAAPCWytoGWbimVBatr5Da5ha3l0+OjpRLhuXIhUOzzX0AAAAAQGBxZeaBrVu3tvv36urqNo937tzZ7nNjYmIkIyOjw/Wfdtppcsghh8jXX39tHm/ZskWmTJkiDzzwgJkhYampqTHpmubPn99m3ddee60nbwsAACAgSvfWyfzVpbJ44zZpaGl1e/n02CiZPDxXzh2SZWY/AAAAAACCQ1hra6v7V3m9nD0I4IkxY8bIvHnzOn1ORUWFTJgwQbZt2+b4W1hYmIwcOVL69+8vu3btku+++0727t3bZrn7779fzj77bOnN7DUgZs2aJWlpaYFuEgAAaMemmlqZU1wir2+ulGYPTklz4mPk0oJcGT8w09R7AAAAAAAEF2ZABKmsrCx58sknTSqmDRs2mL9prGjlypXm5kxnPtx00029PvgAAACC3+pde2V2UYm8W1IlnoyEGZQUJ9MK8uS0/mkSGU7gAQAAAACCFQGIIKbFqF988UV55JFHZNGiRWZkv7OoqCg57rjjTNql7s7M6KnGjx/f5rHOFImLiwtYewAAQPu+q6qRp4q2ysfluzzaRAV9EmR6YZ6cmJsq4WFhbGYAAAAACHKkYOohmpqa5KuvvjI1JSorKyUhIUGys7NNrYjU1FTpzToLQJCCCQCAwNIZnJ9v2y1PFZfIV9vb1sty1ei0JJlRmCdHZ/Uxv/MAAAAAgJ6BAARCDjUgAAAIvJbWVvmwbKfMKSqR73fu8WgdR2amyIzCfnJIRrLX2wcAAAAA8D1SMAEAAMBrtJj021urZE7RVllXXevROk7I7SvTC/rJAamJfDIAAAAA0IMRgAAAAEC3Nba0yKubtsvc4lLZurfO7eW1lPRp/dPl0oI8GZoSzycCAAAAACGAAAQAAAA8VtfULC9t3CbzV5fKttoG909Gw8Jk/MAMmVqQK/0Tf6zhBAAAAAAIDX4PQFx66aXmXy0g+Je//EXS0tI8Wo8WYv7Nb37jWNecOXO82k4AAAB0bE9jk/xnfYU8vaZUdtY3ub2pYiLC5dzBmTJ5eK5kxcewqQEAAAAgBPk9ALF8+XITMFD19fUer0eX1XUpa30AAADwrV31jfLM2jJ5fl257Glsdnv5hMgIuXBotlwyPEf6xkT5pI0AAAAAgF6cgqm1tZWgAbxm/PjxbR5rQCoujhQOAAB40/baBlm4plQWra+QuuYWt5fvEx1pgg4ThmRLUjRZQAEAAACgN+DqDwAAAB0q3VtnCku/smmbNLa0ur2lMmKjZHJ+nkm3FBcZwZYGAAAAgF6kxwYgmpr+l2s4Korp+73ZkiVL2jyuqqqSmTNnBqw9AACEgg3V+2ROcYm8uaVSmt2PO0heQoxMzc8zBaajI8J90UQAAAAAQJDrsQGI0tJSx/2EhISAtgUAACBUFO3cI7OLS+T9kh3iQdxBBifFyfTCPDmlX7pEhlOnCwAAAAB6s8iePupd8/33798/0M0BAADo0b6prJbZRSXyScUuj5Yv7JMgMwrz5PjcVAkPI/AAAAAAAPBRAOK///2vS8974403pG/fvi6vt7GxUbZv3y6fffaZLF++3PH30aNHe9ROAACA3qy1tVU+27ZbZhdtla8razxaxyHpSTK9sJ8cmZliBoYAAAAAAODTAMRNN93U5QWoXvDed999Hr+GLm8555xzPF4PAABAb9PS2ioflO4wqZZ+2LnXo3UcndXHpFo6OD3Z6+0DAAAAAIQGn6ZgsgcJPPl/Z+0FNaZPny6jRo1yu20AAAC9TVNLq7y9tdIUl15fXev28nomdmJeqkwvyJPCvok+aSMAAAAAIHT4LADhbnDBnXVGRkaatEtTpkyRcePGef11AAAAQklDc4ss2bRd5q0ukZK99W4vHxEmclr/dLm0IE+GJMf7pI0AAAAAgNDjkwDE3LlzOwwgTJs2zTGb4cEHH5T09HSX1qnPj46OlsTERFN0Wu8DAACgY7VNzfLfDdtkwZpS2V7b4PamigoPkzMHZsrU/FzJS4xlUwMAAAAAAh+AGDNmjEuplA4++GDJzc31RRPQi4wfP36//SsuLi5g7QEAINBqGprkhfXl8syaMtnV0OT28rER4XL+kCyZNDxXMuIY9AEAAAAACMIaEO2xBxwiIiL8/fIAAAAha2d9owk6PL+uXPY2Nbu9fGJUhFw4NFsuGZYjfWKifNJGAAAAAEDvEdbqi2INQABVVVXJzJkzzf1Zs2ZJWloanwcAIKRV7KuXhWvK5L8bKqSuucXt5fvGRMrE4blywZAsSYzy+/gUAAAAAECI4goTAACgh9q6p07mFpeYAtNNHowpyYyLlin5uXLOoEyJjWRmKgAAAADAuwhAAAAA9DDrq/fJnKISeXNLpbg/30GkX0KMXFqQJ+MGZkhUeLgPWggAAAAAAAEIAACAHuOHnXtkdlGJvF+6w6PlhybHybTCfjI2L00iw8O83j4AAAAAAIJqBkRNTY188803UlRUJDt27JC9e/dKY2OjW+sICwuTu+++22dtBAAACKSvt1fLU0Vb5bNtuz1a/oC+CTK9sJ8cl9NXwsMIPAAAAAAAQjwAsXnzZnnooYfknXfekYaGBo/XozW0CUAAAIBQo+c4n1TsMjMevq2q8Wgdh2Yky4yCPDkiM8WcLwEAAAAAEPIBiNdff11uvvlmqaurMxfXiotiAAAAkZbWVpNiSQMPxbv2erRJfpLdR2YU9pPRaUlsUgAAAABA7wlAaLql6667TpqamhyBBw1CWIEIAACA3qippdUUlZ5TXCIba2rdXl7nN5yUl2pSLRX0SfBJGwEAAAAACOoAxD333GOCD1bgISkpSSZOnCjHH3+8DBkyxDyOjAx4aQoAAAC/qG9ukSWbtsu84hIp3Vfv9vIRYSJnDMiQS/PzZFBynE/aCAAAAACAJ/za079lyxb59ttvHemWCgsL5cknn5S0tDR/NgMAACDgapua5cX1FbJgTalU1jW6vXx0eJicNShTpuTnSm5CrE/aCAAAAABAjwlAfPXVV+ZfnfkQHh4uf/nLXwg+oNvGjx/f5rEGuOLiGAEKAAhONQ1N8vy6cnlmbZnsbvgxJaU74iLC5fwhWTJpeK6kx0X7pI0AAAAAAPS4AERVVZWjg3jkyJEm5RIAAEBvsKOuUZ5eWyovrKuQfU3Nbi+fFBUhFw3LkYuHZktKTJRP2ggAAAAAQI8NQOisB8uAAQP8+dIIYUuWLNkv0DVz5syAtQcAALuKffUyf3WpvLRxm6n34K6+MVEyaXiOmfWQGEWdLAAAAABAz+HXq9icnBzH/YaGBn++NAAAgF9t2VMr84pLTYHpptZWt5fPiouWqfm5ctbgTImNiPBJGwEAAAAACJkAxAEHHOAoQL1p0yZ/vjQAAIBfrNu9T2YXb5W3t1SJ+/MdRPonxsqlBXny0wHpEmWbPQoAAAAAQE/j1wBE//79ZcyYMfLZZ5/JmjVrZMuWLeZvAAAAPd2qHXvkqaKt8kHZTo+WH5YSL9ML8uTkfmkS8f8HbAAAAAAA0JP5fVjdb37zG4mM/DHucf/99/v75QEAALymtbVVvty+W67+cJXMeG+FR8GHkamJ8sDRBTJ/7Gg5tX86wQcAAAAAQMjwewDioIMOkhtvvNFcsL/11lty1113SXNzs7+bAQAA4DE9j1lWtlOuXPq9/PKDVbJ8226313F4RrI8fNwB8uSJB8pxuamONJUAAAAAAIQKv6ZgskydOlVSUlLk1ltvlQULFsjy5ctlxowZcuyxx0pGRkYgmgQAANCl5tZWeb9kh8wu2iqrd+/zaIsdm91Xphfmyai0JLY4AAAAACCk+T0AMXbsWMf98PBwM4Jw9erV8rvf/c78LTExUZKSktwaBajPffvtt33SXgAAgKaWFnljS6XMLS6VjTW1bm8QPasZ2y9NphXkSX6fBDYoAAAAAKBX8HsAoqSkxAQMNPCg/1qBBn2sampqzM0dpCwAAAC+UN/cIq9s3CbzVpdK2b56t5fXYtI/HZAulxbkycCkOJ+0EQAAAACAYBWQFEwWK+gAAAAQTPY1Ncui9RWycE2pVNU1ur18dHiYnD0oU6bk50lOQoxP2ggAAAAAQLDzewDiiCOO8PdLAgAAuGR3Q6M8v65cnl1bLtUNTW5vtfjIcLlgSLZMHJ4jabHRbHUAAAAAQK/m9wDEvHnz/P2SAAAAnaqqa5Cn15TJf9aXy76mFre3VnJUhFw8LEcuHJYtKdFRbG0AAAAAAAKdggkAACCQyvfVy7ziElm8cZvUt7ifGjI1JkomDc+R84dkS0JUhE/aCAAAAABAT0UAAj3e+PHj9ytKHhdHoU8AQMc219TKnOISeW1zpTR7UJMqOz5apubnyZmDMiQ2gsADAAAAAADtIQABAAB6jTW798rsohJ5d2uVuJ9oSWRAYqxMK8iTMwakS2R4uA9aCAAAAABA6CAAgR5vyZIlbR5XVVXJzJkzA9YeAEDwWVFVYwIPH5Xv9Gj54SnxMqOwn5yYlyoRYWFebx8AAAAAAKGIAAQAAAhJra2t8uX2anmqaKt8sb3ao3WMSk2U6YX95JjsPibFHwAAAAAAcB0BCAAAEHKBh4/Kdsrs4hJZuWOPR+sYk5liUi0dlpFM4AEAAAAAgGAIQIwdO7bNYx0p+Pbbb3f6HG9o73UAAEDvosWktbaDBh7W7t7n0TqOy+kr0wvz5MDUJK+3DwAAAACA3sarAYiSkhITDNCRh6q9VAXOz/EGUiIAANB7Nba0yGubK2VucYls2VPn9vJaSnpsvzSZVpgnw1MSfNJGAAAAAAB6I1IwAQCAHqmuuVkWb9gm89eUSvm+BreXjwwLk3EDM2Rqfq4MSIrzSRsBAAAAAOjNvBqAOOKII7zyHAAAgI7sbWyWRevLZeGaMtlR3+j2hooJD5NzBmfJlPxcyYqPYUMDAAAAANATAhDz5s3zynMAAACc7a5vlOfWlctza8ukurHZ7Q0UHxkhE4ZkySXDcyQtNpoNDAAAAACAj5GCCQAABLWqugZZsLpUFq2vkNrmFreXT46OlEuG5ciFQ7PNfQAAAAAA4B9chQMAgKBUtrde5q0ukcUbt0lDS6vby6fFRsnk4bly3pAsM/sBAAAAAACEcABi48aN8sEHHzgejxs3TtLT0/3ZBAAAEOQ21dTKnOISeX1zpTS3uh94yImPkUsLcmX8wEyJiQj3SRsBAAAAAECQBSA0+HDPPfeY+3369JFJkyb58+UBAEAQW71rr8wuKpF3S6rE/bCDyKCkOJlWkCen9U+TyHACDwAAAAAA9KoARF1dnbS2tkpYWJgccMABEhlJBigAAHq7FVU18lTRVllWvsuj5Qv6JJjAw0l5qRIeFub19gEAAAAAAM/4NQKQmprquN+3b19/vjQAAAgiOiDhi+3VJvDw5fZqj9YxOi1JZhTmydFZfczgBgAAAAAA0IsDEFlZWY77u3fv9udLAwCAINDS2iofle00qZa+37nHo3UcmZki0wvz5JD0ZAIPAAAAAAAEMb8GIA477DCJi4uT2tpaWblypSMdEwAACG1aTPqdrVUyp6hE1lbv82gdJ+T2lekF/eSA1ESvtw8AAAAAAPTwAER8fLyMHTtWXnnlFdm1a5e8+eabcvrpp/uzCQhB48ePb/NYg1oa6AIABF5jS4u8tmm7zCkula1769xeXktJn9o/3dR4GJoS75M2AgAAAAAA3/B7Fejrr79ePvroI5OC6U9/+pOMGjVKcnNz/d0MAADgQ3XNzfLyhm0yf3WpVNQ2uL18ZFiYjB+YIVMLcqV/IkFlAAAAAAB6orBWzYPkZx9//LFcc801smfPHsnMzJTbbrtNTjnlFH83AyGqqqpKZs6cae7PmjVL0tLSAt0kAOg19jQ2yX/WV8jTa8pkZ32j28vHRITLuYMzZfLwXMmKj/FJGwEAAAAAQIjOgPj8888lKipKbrzxRrnnnntk27ZtcvXVV0v//v3lxBNPlBEjRkhqaqpJ1+SOI444wmdtBgAAndtV3yjPri2T59eVS01js9ubKyEyQiYMzZZLhuVIamwUmxsAAAAAgBDg9wDE1KlT2xSe1vs6CWPz5s0yb948j9ap61i1apUXWwkAAFyxvbZBFq4plRfXV0htc4vbGy0lOlImDs+RCUOyJSna76clAAAAAADAhwJ2pa9BBysQYQ9IBCAjFAAAcFPp3jqZV1wqizdtk8YW93+7M2KjZHJ+rpw7OEviIiPY/gAAAAAAhKCABCCsIAPBBgAAepYN1ftkbnGpvLFluzR7MGYgNz5GLi3IMwWmoyPCfdFEAAAAAADQWwMQWvcBAAD0LEU798js4hJ5v2SHeDJXcXBSnEwrzJNT+6VLZPj/Zj4CAAAAAIDQ5fcAxHnnnefvlwQAAB76prJaZheVyCcVuzxavrBPgswozJPjc1Ml3JZyEQAAAAAAhD6qPQIAgDY0ReJn23bL7KKt8nVljUdb5+D0JJlR0E+OzEppU+sJAAAAAAD0HgQgAACA0dLaKh+W7pSnirfKDzv3erRVjs7qY1ItHZKezFYFAAAAAKCXIwABAEAv19TSKm9vrZQ5xSWyvrrWo3WcmJsq0wvzZETfRK+3DwAAAAAA9EwEIAAA6KUamlvk1c3bZW5xiZTsrXd7+YgwkdP6p8ulBXkyJDneJ20EAAAAAAA9V9AEIDZv3ixff/21fP/991JVVSU1NT/mnE5KSpK0tDQZOXKkHHLIITJgwIBANxUAgB6ttqlZ/rthmyxYUyrbaxvcXj4qPEzOHJgpU/NzJS8x1idtBAAAAAAAPV/AAxD//e9/ZeHChbJixQqXnj9q1CiZPHmynHPOOT5vGwAAoWRPY5O8sK5cnl5TJrsamtxePjYiXM4bnCWT8nMkMy7GJ20EAAAAAAChIzKQMx5uuukmM+tBtba2mn/DwsLafb71/999950JVjz33HNyzz33MCMCAIAu7KxvlGfWlMnz68plb1Oz29srMSpCLhyaLRcPy5G+MVFsbwAAAAAAELwBiKKiIrn88stlx44dJrCgQQe96X0r0NAeKzihz/nyyy9l0qRJ8u9//1vy8/P92HoAAHqGbbX1smB1mfx3Q4XUNbe4vXzfmEi5ZFiuTBiaJYlRAZ80CQAAAAAAehi/9yZUV1fLFVdcYeo82AMPBxxwgPz0pz81KZYGDhxoaj8orQWxadMmM+vhtddek1WrVjkCEZWVlSaQsWTJEklOTvb3WwEAICiV7KmTuatLZMmm7dLY0nFgvyMZcdEyZXiunDs4U2IjI3zSRgAAAAAAEPr8HoB44IEHTODACjxkZ2fL7bffLieeeGK7z09MTJScnBw56qij5Gc/+5m89957cscdd0hFRYX5f12XrvPOO+/08zsBACC4rK/eJ3OKS+StLZXS7H7cQfolxMilBXny0wEZEh0R7osmAgAAAACAXsSvAYh9+/bJyy+/7Ag+5OXlyYIFC0wQwlUnnXSSjBgxwqRfKisrM+tZvHix3HzzzRIXF+fT9gMAEIx+2LlHZheVyPulOzxafkhynEwvyJOx/dIlMrz9WkwAAAAAAABBHYD47LPPpK6uztzXIITOWnAn+GDRZXRZTeWkdJ2ffvqpCU4AANBbfL29WmYXb5VPK3Z7tPwBfRNkemE/OS6nr4T///SGAAAAAAAAPTIAUVpa6gg+6OyHY445xuN1HXvssdKvXz/ZunVrm3UDABDKdObfpxW75KmiEvm2qsajdRyanizTC/NkTGaKo64SAAAAAABAjw5A1NbWOu4XFBR0e326DisAYV83AAChpqW1VZaW7jCplop27fVoHUdn9ZEZhXlyUHqy19sHAAAAAAAQ0ABEZmam4350dHS312dfh33dAACEiqaWVnlzS6UpLr2xxv1gu85vOCkv1aRaKuiT4JM2AgAAAAAABDwAMWjQIMf9LVu2dHt99nXY1w0AQE/X0Nwir2zaLvOKS6R0X73by0eEiZwxIEOm5ufK4OR4n7QRAAAAAAAgaAIQo0ePloEDB8qmTZtk1apVJn2S1nHwNPjw/fffm9zVAwYMMOsGAKCnq21qlhfXV8jCNaWyva7R7eWjw8PkrEGZMiU/V3ITYn3SRgAAAAAAgKALQKhp06bJnXfeaYpo/uEPf5AnnnjC7QKYLS0tcvvtt5t16LJTp071WXsR/MaPH9/mse4TcXFxAWsPAHiipqFJnl9XLs+sLZPdDU1uLx8XES7nDcmSycNzJT2u+2kOAQAAAAAAuitc/GzixIly8sknm+DBxx9/LFdffbVUV1e7vLw+V5dZtmyZeXzCCSfIlClTfNhiAAB8Z0ddozyycpOc/dpX8s9VW9wOPiRFRcjlI/rJSz89VH49ehDBBwAAAAAAEDTCWjUS4GeNjY1mFsTzzz9vRqunpKTI5MmT5YwzzpDhw4e3u8yaNWvktddek4ULF8ru3bvN3y644AIziyIqKsrP7wDBrKqqSmbOnGnuz5o1S9LS0gLdJADYT8W+epm/ulRe2rhN6ptb3N5CfWOiZNLwHDl/SJYkRvl9QiMAAAAAAEDwBSAuvfRSx/2ioiLH7AcrDVNsbKypC5GQkGAe792719SKqKurM4+t5mrQoqCgwOXX1fXPmTPHq+8FwYkABIBgtmVPrcwrLpUlm7ZLkwc/wVlx0aa+w9mDMyU2IsInbQQAAAAAAPAGvw+ZXL58eZuaD9Z9K7BQW1trZjs4/935+Rq4+Pzzz116TatWBAAAgbJu9z6ZXbxV3t5SJe7PdxDpnxgrlxbkyU8HpEtUuN8zKAIAAAAAALgtYDkbnIMCHQUIOgscEFgAAAS7VTv2yFNFW+WDsp0eLT8sOV6mFebJ2H5pEkEwHQAAAAAA9CB+D0Dk5ub6+yUBAPArDZB/VVkts4tKZPm2H+sWuWtkaqLMKMiTY3P6MosPAAAAAAD0SH4PQLz77rv+fkkAAPwWePi4fJfMLi6R76pqPFrHYRnJMqOwnxyekUzgAQAAAAAA9GgBS8EEAECoaG5tlfdLdsic4hIp3rXXo3Ucm91Xphfmyai0JK+3DwAAAAAAIBAIQAAA4KGmlhZ5Y0ulzC0ulY01tW4vr1WOtLbDtII8ye+TwOcAAAAAAABCCgEIAADcVN/cIq9s3CbzVpdK2b56t7efFpP+6YB0ubQgTwYmxbH9AQAAAABASPJ7AOLhhx923J8+fbokJiZ6tJ6amhqZM2eO4/FVV13llfYBANCRfU3Nsmh9hSxcUypVdY1ub6jo8DA5e1CmTMnPk5yEGDY0AAAAAAAIaQEJQISFadIJkfPPP9/jAER1dXWbdRGAAAD4SnVDkzy3rkyeXVtu7rsrPjJcLhiSLROH50habLRP2ggAAAAAABBsApKCqbW11RE4CKZ1AQBgV1XXIE+vKZP/rC+XfU0tbm+c5KgIuXhYjlw4LFtSoqPYuAAAAAAAoFehBgQAAE7K99XL/NWl8vKGCqlvaXV7+6TGRMmk4Tly/pBsSYiKYPsCAAAAAIBeqccGIFpa/jcSNSKCzh0AQPdtrqmVOcUl8trmSmludT/wkB0fLVPz8+TMQRkSy28TAAAAAADo5XpsAGLHjh2O+/Hx8QFtCwCgZ1uze6/MLiqRd7dWifuJlkQGJMbKtII8OWNAukSGh/ughQAAAAAAAD1Pjw1AfPLJJ+Zfrf+QlZUV6OYAAHqgFVU1ZsbDh2U7PVp+eEq8zCjsJyfmpUoE9YgAAAAAAAB8H4AoLS116XkVFRVurbexsVG2b98un332mTzxxBOOvxcWFrrdRgBA79Ta2ipfbq+Wp4q2yhfbqz1ax6jURJle2E+Oye5jAuEAAAAAAADwUwDi5JNP7rJDRjuAJk2a5PFr6PKW008/3eP1AAB6B/3dWFa+ywQeVu7Y49E6xmSmmFRLh2UkE3gAAAAAAAAIZAome5DAk//viAY39KbLH3744XLKKad42EIAQKjTYtJa22F2cYms3b3Po3Ucl9NXphfmyYGpSV5vHwAAAAAAQKjyWQDC0+CCq+uOjY2Vc889V2644QafvQ4AoOdqammR1zZXytziEtm8p87t5bWU9Nh+aSbwMCwlwSdtBAAAAAAACGU+CUBcddVVHf7fww8/bP7VGQxTp06V5ORkl9cbHR0tSUlJMnjwYBk1apQkJNAhBABoq665WRZv2Cbz15RK+b4GtzdPZFiYjBuYIVPzc2VAUhybFwAAAAAAoCcFIKz6ENOnT5fc3FxfNAEA0MvsbWyWRevLZeGaMtlR3+j28jHhYXLO4CyZkp8rWfExPmkjAAAAAABAb+LTGhCdpVDqqkg1AACu2F3fKM+tK5fn1pZJdWOz2xstPjJCJgzNkonDciU1NoqNDgAAAAAA0FMDEO+8847jflZWlr9fHn6ycuVK+eCDD+TLL7+UtWvXSlVVlURGRkp6erocdNBBcs4558jxxx/P5wHAY1V1DbJgdaksWl8htc0tbi+fHB0plwzLkQuHZpv7AAAAAAAA8C6/97jk5eX5+yXhZ1OmTJHPP/98v783NjbKli1bzO2VV16RE088UR544AFT1wMAXFW2t17mrS6RxRu3SUNLq9sbLj02SiYPz5Vzh2SZ2Q8AAAAAAADwDYZ8wusqKirMvzrb4fTTT5fDDz/cBJ407daKFStk7ty5snHjRnn//fflF7/4hXkcHh7OJwGgU5tqamVOcYm8vrlSmlvdDzzkxMfIpQW5Mn5gpsREcMwBAAAAAADoVQGIdevWyffffy87duyQ6upq87fk5GRJTU2VkSNHytChQwPdRLhgyJAhcu2115rgg6Zdshs9erScd955ctlll8nXX39tZkosXrzYpGQCgPas3rVXZheVyLslVeJ+2EFkUFKcTCvIk9P6p0kkwU4AAAAAAIDeE4DYsGGDLFiwwHRCW0GHjmgw4qyzzpJJkyaZTm4Ep3/+85+d/n98fLzccccdcvbZZ5vHr7/+OgEIAPtZUVUjTxVtlWXluzzaOgV9EmR6YZ6cmJsq4WFhbGEAAAAAAIDeEoBobm6Wv//97/LEE09IS0uLtNrSaWiqHjvr/3bv3m2CFU8//bRcfvnlcs011+w3wh49Q0FBgfTp00d27dolmzdvDnRzAAQJPd5/vm23PFVcIl9t7zwo3ZHRaUkyozBPjs7qs9/vCQAAAAAAAPwnIL33DQ0NcuWVV8pnn33mCC5oJ5F13x6MsLM6kjR48a9//Uu++eYbE8CIjo6W3qCpqcmkLSopKZFt27ZJYmKiZGdny8EHH2zSVPXE96Oo/wCgpbVVPirbaVItfb9zj0cb5MjMFDPj4ZD0ZAIPAAAAAAAAvTUAccMNN8inn37aJvAQEREhxxxzjBx00EEyePBgSUpKMv9fU1NjChZ/++238tFHH5lOa2sZrR9w/fXXy9/+9rdAvA0zc0PrVnz33XemuLLeiouLpbGx0fGce+65R84///xuvU5tba08+uijsmjRIqmsrNzv/6OiouS4444zdRd0ZkFPsGrVKtmz58dORmp7AL2XFpN+Z2uVzCkqkbXV+zxaxwm5fWV6QT85IDXR6+0DAAAAAABADwpAfPjhhybnvzWbQUe/T5w4UX71q19J3759O11W0/U88sgjsnDhQkfapjfffFM++OADOf744/30Dn6sWaCpoFauXCn79nnWYeaqNWvWmFRT69ev7/A5GvB49913ZdmyZXLzzTeb7RnsHnvsMcf9cePGBbQtAPyvsaVFXtu0XeYUl8rWvXVuLx8uIqf2TzfFpYemxPukjQAAAAAAAOhhAQgNICgNHmjqpL/+9a8yduxYl5bVmgG33HKLHH300Wa0v3a863p0doA/AxBffvmlLF++3Oevo2mWtNZFRUVFm7+PHDlS+vfvbwIyOuti79695u/19fVy++23m9RMWqw7WC1ZssQEjtSBBx4op556aqCbBMBP6pqb5eUN22T+6lKpqG1we/nIsDAZPzBDphbkSv/EOJ+0EQAAAAAAAD0wALFjxw6TSsma/aAj+10NPtidfPLJZtkHHnjAPNYUSFVVVZKWliaBpGmj4uPj9wsYeEIDK/oe7evKz8+X+++/XwoLCx1/q66uNimo5s+f7/ibBmn0OcOHD5dgU1RUJL///e/N/bi4OLnvvvvI1Q70Ansam+Q/6yvk6TVlsrP+f2nqXBUTES7nDs6UycNzJSs+xidtBAAAAAAAQA8OQOjMAavAtI7Sv/TSSz1ely6raXx09L+u86uvvvLrSPrY2FgZMWKEGcE/atQoc9PaFQ8//LC5dZfOENCC05Z+/fqZIENKSkqb5yUnJ8utt95qOvHnzZvnmAmhQYnO2qHPGT16dLfbedVVV8nVV1/t0nO3bt1qio9r2ipNvXXvvfdS/wEIcbvrG+WZtWXy/LpyqWlsdnv5hMgImTA0Wy4ZliOpsVE+aSMAAAAAAABCIACxfft28692lmuxaU3B5Cld9uCDDzZ1D+zr9odf/OIXcuONN0pkpO82n3Pw4Lbbbtsv+GD329/+1tSBKCkpMY/feust+eGHH0yQJBhoOqnLLrvMMaPjzjvvlDPOOCPQzQLgI9trG2ThmlJ5cX2F1Da3uL18SnSkTByeIxOGZEtStN+zBQIAAAAAAMAL/NqrU1NT47jfVcFpV2tCWPbs2SP+kpqa6tP1FxcXy+rVqx2Phw4dKieccEKny2g6o0suuUQefPBBx98WL17cYQBCAzivvvpqt9vqyueoqbc0+LBp0ybzWAtlX3jhhd1+bQDBp3RvncwrLpXFm7ZJY8uPM97ckREbJZPzc+XcwVkSFxnhkzYCAAAAAAAgBAMQmi7IsnPnzm6vT4sw2+svhIr33nuvzWNXC0rr8+wBCJ0RccMNN7T7XJ2FooENfwSdtJD2mjVrzONf//rXMn36dJ+/LgD/2lhdK3OKS+SNLdul2f24g+TGx8ilBXmmwHR0RLgvmggAAAAAAIBQDkBkZGSYf7Vmg9Y30DoEMTGeFRPVZe01Eqx1hwIrrZTl8MMPd2m5nJwcycvLc6Rh2rBhg5SVlZm/B4LW57jiiitk1apV5vHPfvYz+eUvfxmQtgDwjeJde2V20VZ5r2SHeBB3kMFJcTKtME9O7ZcukeFhPmghAAAAAAAAekUA4tBDDzXFhzUAUVtbK3PmzDFFiT0xd+5cU8zYGs2v6w4Va9euddzX7aWFrl2ltTWsAITSmQeBCEDU1dXJzJkz5ZtvvjGPp06dKtddd53f2wHAN76trJanikrkk4r/zURzR0GfBJlRmCcn5KZKeBiBBwAAAAAAgFDk1wCE1k7QDnLtlNYghBZaHjJkiJxyyiluref999+Xf/zjHybwoHSdvq7L4C+7d+82NRMsaWlppr6Dq/r169fmsc6COP7448WfGhoa5Oqrr5bly5ebxxMmTJBbbrnFr20A4H163F6+bbfMLiqRryqrPVrHwelJMqOgnxyZleI4hgMAAAAAACA0+TUAobRjWgsSa8eTdlRrTYCLL77YpOZJT0/vdFntmJ81a5YsXLhQmpubTWeYrudXv/qVhIrNmze3eezu7IXs7OxO1+cPOtPhgw8+MPcPOeQQM/vBqgHRkfz8fD+1DoC7Wlpb5cPSnTK7eKus2rnXow14VFaKTC/sJ4ek/68WEAAAAAAAAEKb3wMQP/nJT+TMM8+UV155xQQPNJDw9NNPy3PPPSdHHHGEjB49WgYNGiSJiYnm+Xv27JFNmzbJd999Z0bU2wMPehs3bpwce+yxEir0/dq5O7Ojb9+++xWB9rc33njDcV/rdJxzzjldLlNcXOzy+svLyzv9f28UOAcg0tTSKu9srTTFpddV13q0SU7MTZXphXkyou+Px3QAAAAAAAD0Hn4PQKh77rnHdBJrsWUrBUdTU5N8+umn5tYRe+BB7x9zzDFy7733SijRws127hbpjo2NbfPYqpMRSk444YRO/z8qKkpGjBjht/YAoaahuUVe3bxd5hWXyNa99W4vHxEmcmr/dJlWkCdDkuN90kYAAAAAAAAEv4AEILSD+PHHH5fHHnvMpFTS4IM9F7gGF+ys/7MCDxEREfLzn//cpF7S+6FEi3PbRUdHu7W8c8DCeX3+4M5sBgDBo66pWf67YZvMX1Mq22sb3F4+KjxMxg/MkEvz8yQvsW0wFAAAAAAAAL1PQAIQSgMHV111lZx77rmyYMECk5Jp+/bt7T7XCkhkZGSY9E2TJk2S/v37S2/gbpFW5+c7B3NCwdKlSzv9f51dc9ttt/mtPUBPt6exSV5YVy5PrymTXQ1Nbi8fGxEu5w3Okkn5OZIZ596sLQAAAAAAAISugAUgLP369ZMbb7zR3EpKSuT77783xaarq6vN/ycnJ5s6CCNHjpS8vDwJdXFxcW0e19e7l/6krq6uzeP4+NBLf+JcaLu9GTYAurarvlGeWVsmz68rlz2NzW5vsoTICLloWLZcPCxH+sbwvQMAAAAAAECQBSDsNMDQG4IMnXEOGLgbgHB+figGIAB0z7baelm4ukxe3FAhdc0tbi/fJzpSJg7PkQlDsyUxKqh+RgAAAAAAABBE6DkKMomJifulE3KHzh6xS0pK8kq7APR8JXvqZO7qElmyabs0trifni0jLlqmDM+VcwdnSmxkaNXfAQAAAAAAgPcRgAgyAwYMaPO4rKzMreXLy8vbPO4ttTIAdGx99T6ZU1wib22plGYPysLkJcTIpQV5Mm5AhkRHhLOpAQAAAAAA0LMDEC0tP6YFCQ/vXZ1dffr0MTUvrJkMlZWVUltbu19tiI5s3bq1zeMhQ4ZIqBs/fvx+hbhd3V5AKPth5x6ZXVQi75e2nRnlqiHJcTK9IE/G9kuXyPC2Be4BAAAAAACAHhGA0E72V199Vb7++mtHEeq9e/ea/0tISHAUoT744INl3LhxkpGRIaFs2LBhsnz5ckcgZuXKlXLEEUe4tOy3336737oA9C5fV1bL7KKt8mnFbo+WH9E3QWYU9JPjcvtKeBiBBwAAAAAAAPTAAMT27dvlvvvuk9dff12amprM31pb2+YH2bNnj7lt2bLFPO/++++XM844Q2644QbJzMyUUPSTn/zEEYBQX3zxhUsBCE3XVFJS4ng8ePBgyc3NlVC3ZMmSNo+rqqpk5syZAWsPEAh67Py0Ypc8VVQi31bVeLSOQ9OTZXphnozJTDEziQAAAAAAAIDuCFh+o6VLl8pZZ50lr7zyijQ2NjoCD9rpZe/4sj/W52igQjucddn3339fQtHJJ5/c5vHixYtdWs75ec7rARB6Wlpb5b2SKpn+7gq5dlmRR8GHo7P6yOMnjJTHThgpR2b1IfgAAAAAAACAnjsD4pNPPpGrrrrKBB6UBhg0uKC36OhoUzg5KSnJ/F9NTY2pa1BfX+94rtq9e7dcffXV8vjjj8vRRx8toaSgoEDy8/Nl9erV5vG6detMwOaEE07ocJm6ujp55pln2vxNgzQAQlNTS6u8uaXSFJfeWFPr9vJ6JD0pL1WmF/aTgj4JPmkjAAAAAAAAeje/ByA0cHDttdea4IMVeIiNjZWLLrpIzjzzTDnggAMkIiKizTLNzc3yww8/mBH+zz//vOzbt88sq+vQdb3xxhumeHMo0QDNNddc43j8xz/+0dTASElJaff5Dz74YJv0S6eccoqMGDHCL20F4D8NzS3yyqbtMq+4REr3/RiYdUdEmMgZAzJkan6uDE6O90kbAQAAAAAAgIAEIB599FEThLCCD4cddpg88MADkpOT0+EyGpA48MADze2yyy6T6667Tj7//HPzf9XV1fLYY4/JzTff7Md3IWZWRnu0PXY7d+5s97kxMTGdFtM+7bTT5JBDDjGFuZXWwJgyZYrZVjpDwqIzRB566CGZP39+m3VrYAZA6KhtapYX11fIwjWlsr3ux9lj7ogOD5OzBmXKlPxcyU2I9UkbAQAAAAAAALuwVueqzz6kL6XpkjQAoUaNGmU6zjXtkjsaGhpMZ/yKFSvMOnX2g6Z18mfRVHsQwBNjxoyRefPmdfqciooKmTBhgmzbts3xN32PI0eONGmqdu3aJd99953s3bu3zXJaqPvss8+W3spehHrWrFmSlpYW6CYBHqtpaJLn15XLM2vLZHdDk9vLx0WEy3lDsmTy8FxJj3PvWAsAAAAAAAD0mBkQ3377rek0tzrSNa2Qu8EHpcvceeedcu6555rHGtDQdWuKolCSlZUlTz75pEnFtGHDBvM3DbisXLnS3JzpzIebbrqp1wUfxo8f3+ax7ltxcXEBaw/gDTvqGk3Q4YV15bK3qdnt5ZOiIuSiYTly8dBsSYmJ4kMBAAAAAABAaAcgNm7c6OggHjZsWLdmERQWFrYp1KzrDrUAhNL3+OKLL8ojjzwiixYtMqP7nUVFRclxxx1n0i51d2YGgMCq2Fcv81eXyksbt0l9c4vby/eNiZJJw3Pk/CFZkhjl9yx7AAAAAAAAgINfe6fsneeDBw/u9voGDRrkCEDs2LFD/Km4uNhvr6Wj+bXuhQYYvvrqK1NTorKyUhISEiQ7O9vUikhNTZXeasmSJR2mYAJ6ii17amVecaks2bRdmjzIjJcVF20KS2udh9jICJ+0EQAAAAAAAAjaAIQ/azSEosjISFM7Qm8AQsO63ftkdvFWeXtLlbg/30Gkf2KsXFqQJz8dkC5R4eE+aCEAAAAAAADQAwIQ9lH669ev7/b6rLoIzusGgGC3asceeapoq3xQttOj5Yclx8u0wjwZ2y9NIgjuAgAAAAAAoLcHIIYMGeIopLxu3Tr54YcfZMSIER6tq6ioSNasWePVlE4A4Et67PuqslpmF5XI8m27PVrHyNREmVGQJ8fm9GVWGQAAAAAAAIKaXwMQo0aNkr59+8quXbtMR9xtt90m8+fPl5iYGLfW09DQIH/4wx8cj/v06SMHHXSQD1oMAN2nx7uPy3fJ7OIS+a6qxqN1HJaRLDMK+8nhGckEHgAAAAAAANAjhPu7BsS5555rOuP0/sqVK2XGjBlSVlbm8joqKirk8ssvl2+//bbNOgEg2DS3tso7W6tk2rsr5DcfF3kUfDgmu488ceKB8ujxI+WIzBSCDwAAAAAAAOgxwlo1GuBHNTU1cuqpp8ru3T+mH9GXj4uLk/PPP1/GjRtnZklER0fvN+NBgxWvvvqqLFq0SGprax3L6uyHN998U5KTk/35NhBExo8f3+axBqV0n1KzZs2StLS0ALUMvVVTS4u8saVS5haXysaaH49X7ggTMbUdphXkSX6fBJ+0EQAAAAAAAAi5AIT6/PPP5YorrjCBBWXNiFARERGSmZkpiYmJ5vGePXtk27Zt0tzc3Oa5+q+mbvrXv/4lY8aM8fdbQBAhAIFgUd/cIq9s3CbzVpdK2b56t5fXYtJnDEg3gYeBST8G0QAAAAAAAICeKiABCLVs2TK54YYbpKqqyhF86Kop9uelpqbKn//8ZznuuOP80l70HLpPzZw509xnBgT8YV9Ts7y4vkIWrimVyrpGt5ePDg+TswdlypT8PMlJcK8mDgAAAAAAABCs/FqE2u6YY46RxYsXy4MPPiivvPKK1NfXtwkyONOggzXrQUe8//a3vyW1DoCAqm5okufWlcmza8vNfXfFR4bL+UOyZdLwHEmLbZt6DgAAAAAAAOjpAhaAUDqL4U9/+pNcf/31po7D119/bWo97Ny5U6qrq81ztLZD37595cADD5RDDjnE1I/QxwAQKFV1DfLMmjJ5YX2Fmf3gruSoCLloWI5cNCxbUqKjfNJGAAAAAAAAoFcHICxaSPqiiy4yNwAIVuX76mX+6lJ5eUOF1Le4n70uNSbKzHbQWQ8JURE+aSMAAAAAAAAQkgGIuXPnOu5fcMEFkpCQ4M3VA0BAbK6plbmrS+TVTZXS7EHZnOz4aJkyPFfOGpwpsREEHgAAAAAAANA7eDUAcffddztqOJxyyintBiAefvhhx/3p06dLYmKiN5sAAF6zZvdemV1UIu9urZIWD5YfkBgrlxbkyRkD0iUqPJxPBgAAAAAAAL2K11MwaaHojgpJWwEI6//PP/98AhAAgs7KHTUm8PBh2U6Plh+eEi/TC/PkpLw0iejkeAgAAAAAAACEsoDUgOgqSAEAgTgufbm92gQePt++26N1HJiaKDMK+8kx2X04xgEAAAAAAKDX82oAIiYmRurr6839hoaGXr9x4R/jx49v81iDW3FxcWx+uBx4WFa+S54q2iord+zxaKsdkZFiZjwclpFM4AEAAAAAAADwRQAiJSVFtm3bZu5v2LBBBg0a5M3VA4DXaDHp90qqzIyHNbv3ebSO43L6yrSCPBmVlsQnAwAAAAAAAPgyAFFYWOgIQPz73/+WY445RqKjo735EsB+lixZ0uZxVVWVzJw5ky2FdjW1tMhrmytlbnGJbN5T5/ZW0lLSY/ulybTCPBmeksBWBgAAAAAAAPwRgDjhhBPkgw8+MPe/+OILOeWUU+TYY4+VzMxMiYzc/6XmzJkjSUneGTl81VVXeWU9AEJTXXOzLN64XeavLpHyfe6niNNi0uMGpsul+XkyIIkUXwAAAAAAAEBXwlo1AbqX1NbWmnz8ZWVlHRabtr+cNwtR//DDD15bF3o2+wyIWbNmSVpaWqCbhADa29gsi9aXy8I1ZbKjvtHt5WPCw+TswVkyJT9XsuNjfNJGAAAAAAAAIBR5dQaEFv595JFHTOdvRUWF+EN7QQ4A2N3QKM+tLZfn1pZJdWOz2xskPjJCJgzJkkuG50haLKnkAAAAAAAAgIAGINSIESPklVdekaefflreeustKSoqksbG9kcde3HyBQAYVXUNZraDznrY19Ti9lZJjo6US4blyIVDs819AAAAAAAAAJ7xSe+a1nW48sorza2lpUWqq6tl37595r7WhVA6a2H+/PmSnZ3tiyYA6GXK9tbLvNUlsnjjNmlocT+4mRYbJZOH58p5Q7LM7AcAAAAAAAAA3ePz4b3h4eHSp08fc7NYKZNycnIkNzfX100AEMI21dTKnOISeX1zpTR7MKsqJz5GpubnypmDMiUmItwnbQQAAAAAAAB6o4DkF6FuA4DuWr1rrwk8vLO1SjxJ5jYoKU6mFeTJaf3TJDKcwAMAAAAAAADQ4wMQV111leN+cnKyv18eQA+3oqpGniraKsvKd3m0fH5KvEwv7Ccn5qVKBAXsAQAAAAAAgNAMQACAq7OmvthebQIPX26v9mijjU5LkukFefKT7D6ONHAAAAAAAAAAQiwFE+BN48ePb/NYO5fj4uLYyCESePiwbKfMLi6R73fs8WgdYzJTZEZhnhySnkzgAQAAAAAAAPAjAhAAgo4Wk9baDnOKSmRt9T6P1nF8Tl+ZXpgnI1OTvN4+AAAAAAAAAF0jAIEeb8mSJW0eV1VVycyZMwPWHniusaVFXttcKXOLS2TLnjq3l9dS0qf2T5dLC3JlWEoCHwUAAAAAAAAQQAQgAARcXXOzvLxhm8xfXSoVtQ1uLx8ZFibjB2bI1IJc6Z9I+i0AAAAAAAAgGBCAABAwexqbZNH6Clm4pkx21je6vXxMRLicMyhTpuTnSlZ8jE/aCAAAAAAAAMAzBCAA+N3u+kZ5dl25PLe2TGoam91ePj4yQi4cmi2XDMuR1Ngon7QRAAAAAAAAQPcQgADgN5W1DbJgTam8uL5Captb3F4+JTrSBB00+JAUzeELAAAAAAAACGb04AHwudK9daa+w+KN26ShpdXt5TNio2Ryfq6cMzjLzH4AAAAAAAAAEPwIQADwmY3VtTKnuETe2LJdmt2PO0hufIxMLciTMwdmSHREuC+aCAAAAAAAAMBHCEAA8LriXXtldtFWea9kh3gQd5BBSXEyrSBPTuufLpHhYXxCAAAAAAAAQA9EAAKA13xbWS1PFZXIJxW7PFq+oE+CzCjMkxNyUyU8jMADAAAAAAAA0JMRgADQLa2trbJ8226ZXVQiX1VWe7SOg9KSTODhqKw+EkbgAQAAAAAAAAgJBCAAeKSltVU+LNtpUi2t2rnXo3UclZUi0wv6ySEZyXwKAAAAAAAAQIjxewDisccek/PPP1+ysrL8/dIAvKCppVXe2Vppikuvq671aB0n5qbK9MI8GdE3kc8EAAAAAAAACFFhrZo/xY8KCwslMjJSjj32WLnooovkxBNPlPDwcH82ASFm/PjxbR5rCp+4uDhzf9asWZKWlhagloWWxpYWeXXTdplbXCJb99a7vbx+y7Wo9LTCPBmSHO+TNgIAAAAAAADo5SmYmpubZenSpeaWnp5uZkRccMEFMmDAgEA0B0An6pqa5b8btsmCNaWyrbbB7W0VFR4m4wdmyNT8POmXGMu2BgAAAAAAAHqJgMyAsIrMWi9tPR4zZoyZFXHqqadKdHS0P5uFEFJVVSUzZ84095kB4bk9jU3ywroKeWZtqeysb3J7+ZiIcDlvcJZMzs+RzLiYbrQEAAAAAAAAQE/k9xkQjzzyiLzwwgvywQcfmJkQ9mDE8uXLzS05OVnOOeccmTBhguTn5/u7iUCvtqu+UZ5ZWybPryuXPY3Nbi+fEBkhFw7NlkuG50jfmCiftBEAAAAAAABA8PP7DAjL9u3b5T//+Y8sWrRINm/e/GNjwsL2mxUxatQoufDCC02e//h48saja8yA8PA7WdsgC1aXyosbKqSuucXt5ftER8rE4TkyYWi2JEYFJLsbAAAAAAAAgCASsACE3aeffirPP/+8vP3221JfX99mVoR1X4sKaxBCZ0UcdNBBAW4xghkBCPeU7KmTeatL5ZVN26Sxxf3DQUZctEwenivnDs6UuMgIt5cHAAAAAAAAEJqCIgBhqa6ulpdeesnMjCgqKupwVsSwYcNMrYizzz5bUlJSAtpmBB8CEK5ZX71P5hSXyFtbKqXZg6NAXkKMXFqQJ+MGZEh0RLj7KwAAAAAAAAAQ0oIqAGG3YsUKMyvi1VdflT179rQbjNBC1aeccopJ0XTUUUcFuMUIFgQgOvfDzj0yu6hElpbuEE++/IOT4mR6YZ6c0i9dIsN/DAoCAAAAAAAAQI8JQFhqa2vltddeM4Wrv/rqq/3+35oV0b9/f5Oe6bzzzpOMjIwAtBTBggBE+76urJbZRVvl04rdHm3XEX0TZEZBPzkut6+E///vHQAAAAAAAAD02ACE3fr1682siP/+97+ya9cu8zd78zUYERERIWPHjpUpU6bIEUccEcDWIlAIQPyPfj8+q9gtTxVvlW8qazzanoekJ8n0wn5yZGaKI+AHAAAAAAAAAF2JlB5kx44dpnNZZ0VYnDtEm5qa5M033zS3MWPGyK233mpqRgC9SUtrq0mxpKmWinbt9WgdR2f1MamWDk5P9nr7AAAAAAAAAIS+oA9AVFZWyosvvmgKU2/atKnN/1mzH5KSkiQmJsY81/5/n332mUnJdNddd8k555zj97YD/tbU0ipvba2UOUUlsqHmf4E6V2k478S8VJlekCeFfRN90kYAAAAAAAAAvUNQBiA0eLB06VKTbkn/bW5ublN82rp/8MEHy0UXXSTjxo0zBan1uc8995z5V5+jz21sbJRbbrlFhg8fLgcccECA3xngO69v3i6Pr9oiJXvr3V42Ikzk9P4ZcmlBrgxOjvdJ+wAAAAAAAAD0LkEVgNiyZYspNq01HrZt22b+ZgUSLImJiXL22WebwENBQUGb5U866SRz27hxo9x7773y/vvvm2U1gDF79my57777/P6eAH/4x4pNMn91qdvLRYWHyVkDM2VqQa7kJsT6pG0AAAAAAAAAeqeAByAaGhpMvQad7fD555+bgIN9toPSxwcddJAJOowfP15iYzvvKB00aJDMmjVLZs6caYIQavny5X54N4D/ba6pdTv4EBsRLucPyZJJw3MlIy7aZ20DAAAAAAAA0HsFLABRVFRkZjssXrxYqqur253tkJCQ4JjtUFhY6PZrXHrppY4AhL0+BBBKPq3Y5fJzk6Ii5MKh2XLxsBzpExPl03YB+B+tSaS/SZZ77rlHzj//fDYRgDYzerV2mc7iDaR//OMf8vDDDzsev/POO9KvX79uPxfumzp1qmMQVV5enrz77rshtRn/8Ic/yDPPPGPuT5s2TX73u995vK4NGzbI6tWrpaysTPbt2yeRkZHSp08fGThwoBx44IHmujJY2+4L1113nbnOVr/85S/l17/+tcvL7ty5U0455RTZs2ePREVFycsvvyxDhgzxYWsBAECo83sAQk/UdLbDqlWrzOP2ZjuMHj3aMdshLi7O49eyXwBpGiYgFOXEx3T5nL4xkTJxeK5cMCRLEqMCPvEJQA+2detWGTt2bJfPi4+PN2kTtdNs5MiRctxxx5lbRESEV9oxffp0+eSTTxyPw8PDTedcTk5Otzv67J588kk59thj3V5fRUWFSQvpfP6hHWIakEJgLVq0SG6++eb9/n7mmWfKgw8+6NE6r7jiCvnwww/3+7vOytV9AcD/6LWg1u5T+luhM9fdVV5eLnPnzpUlS5aY+x3RYMSRRx4pF154oZxxxhltBrz5uu0d/WbedtttMnnyZLde9+STT5aSkhKXfks04PD666+beoj6O3bBBRe4HBzs27evXHbZZfL3v//dLH/33XfLE0884VZbAQAA7MLFz26//XZz0mYFHiw6KuWSSy4x9R/0hG7ChAndCj4ob3VyAMHs4PRkyUtoPwiRGRctvz1okPz3jENlWkEewQcAfqMjULWe09dffy3z58+Xn//852ZE5XvvvdftdesIV+eOl5aWFjNK09v0vMQT2hYGP/Q8b7/9thn16y7d1z/++GOftAn+CUjpTBTrRpDQ93QmoB63rRkEqampLi+rneI6++bUU081neudBR9UU1OTLFu2TK699loz+9AaCBeIttsDk/X19eIr/fv3d8y01NdxN7CqQX4NcigNrH7wwQc+aScAAOgd/B6AsNMgxKhRo+SPf/yjObHR4IQnqZY6G32pU9n1du6553ptvUAwSYqOlL/8ZISM7Jvo+NugpDi55dAhsuiMQ+SiYTkSG0kwDkDglZaWmpGijz/+eLfW89JLLzk6f+xefPFF8UWH9N69e91ezhfBEPheXV2dGTXsLk11QsAJcI0GA6wZZzrgbMqUKS5vul27dsmMGTNM6i+tJWiXlpYmhx9+uIwbN05OPPFEGTFixH4D0jT4cPHFF3t8jO5O252DlgsXLhRf0lkMOjtQvfbaa1JcXOzysjo4cOLEiY7Hf/vb33zSRgAA0DsEJBeLBgZ0irvOeNATQ1/R6aM6QgWhTVN12em06u7OnulpBiXHyb9PHiXl++olPjJCkqNJswTAP7KystrtRNEZEJoq4qOPPjKji/Wx5S9/+YsZZXzCCSd49JodBRo0B/i3334rBx10kHSX5r3WUba1tbXyxhtvuFWz4/vvvze5yJ3XheBl/4w0wKUzcd2hy7S3LgD709Q+Fh0o5uoMAp2dpJ3qeoy1O/74400KtCOOOMLR4W6pqqqSV155RR555BHZvXu3+ZsGLm644QYzGO6cc87xS9vbo8F4TTvs7foUlkGDBpn0T2+99ZZ5r841W7qitat0holur5UrV5oaL66kYAQAAAj4DAhrtsMdd9zh0+AD0Btlx8cQfACCkOae1pGH1i2UClBrbm3NK+18y8/PN3nvb731VtM5q+kgLNoRoukgnNMxuuKbb76RjRs3tul48sUsCK1X4WkaJvvzdWZnZmamV9oE37F/3p9//rkjz7orfvjhhzYji533ya7Yjw2BLkCtrr766jZtoqg0vOmLL74wx3GLzkZw5zrSHnzQ3x+tT/Cvf/3L/M46Bx+sWRGaJknrRGiAwqK/P1qHYd26dX5pe3t27Nhhalj4kgY4LBpA2Lx5s8vLanBFUydannrqKa+3DwAA9A5+D0Bo8S+dAQF4i15Q2G9z5sxh4wJAEBkwYICZ9WCnHZv2WQKucg4wXH/99TJ8+HDH41dffXW/tByesKdu1HQbWnfCFZprXEfbtrceBK+jjz7azOaxOibdSc9i3yc18MYAG6Bj9g53TcXravpdTSFkD+5qsOG+++4zxZVdkZGRIf/85z9l9OjRbVKuXXfddS7PWPK07XZjxoxpM+NBO/VramrEV4499ljJzc019zV1obsBD712twdnu1s/AwAA9E4BrQEBAAB6B+300Q4bO+3McIcGFrQTyqKdP9rhe/bZZzv+pik2vFHo+ic/+YnpsHK3Q1oLdeqoVqW5xzXlJIKfdmaeddZZ7aZU6irgpIMfLAScgI7psVFH4VtcPT7q9+zPf/5zm79Nnjx5vzSsXdGOf61lYB8Mpx3qrsyc87Tt7aUI1gLP9t8sTXPky2PbGWec4Xisv2XuBOmPOuooM4vE8txzz3m9jQAAIPSRKB4AgB5ER2pqGoitW7eaDpGkpCQZOHCgKbwZExMjwUwDECtWrHA8dnVWgUU7f6wc3soKPGjHsc6wsFI6aWfS6aef7pUO6X//+9+ODumf//znXS5nH6FrD2IEks42Wbt2rZSXl5uUJdnZ2WYUrnaEeYvW+NCAkr7Gzp07JSUlRQ4++OAuZwPU19fLl19+adqnxb61TVofRJfVmk7+pLngn3jiCbfqiWiNk8rKyjYBp0B20On3a/369VJRUSGxsbFmVoe3P2t3aDu+++47s1/oZ63pyHSb6jHLU/o91/eoN12v7jfR0dFmn9Oc93qc0cfBoLq6Wr7++mtTcFi/F6pPnz4yZMgQE0BNTEzs1vq3bNliUhLpdtAi6NpRfNhhh7VJeRdMdIaaBhMs9vQ+nXnzzTfb/F6kp6fLtdde61EbdDbAL37xC5MG0KKzAuypirzZ9vZoEe358+c7fs909rbWW+hOPYnOnHrqqY7fMn3N999/X0477TSXfwu17oN1XNNBALfccoupdQMAAOAqAhAAAPjYZ599ZjoXLPfcc0+HdSA6eq6OWHz00Ufl2WefdYywt9MRnZrneubMmabjMRglJyfv1znnDuf0G9YI1JycHJPbW1MlKa01pYVH7aM2Pe2QtjptNE+4dqTa03c40/djn33hr9HwU6dOdbz3vLw8effddx0dZpobvb2UGdp5dPLJJ8vvfvc7E5DoSkf7pe6LOqJ48eLFpiPYmXYGa92vkSNHtvm7pj557LHHTCecFpZ1pnUHND+7p4XKPaGzaQ444ADH9tL9rasAhH2ftKdxcocGXOxFbTurA9HRc7UdWtC2vXz2+l356U9/atKV6XelK86FajXw524dCC1Yq8V6NUCjHePOdLvedNNNcuihh7q0Pt1fdL/WgvCffvqp7Nq1q8Pn6vFv3LhxJmCoAYnO2LennX1fd9bVZ6QBEi36q8cOPWa09/6t7ztR9ywAAIvjSURBVOAhhxxivkc6kt+doIkGxzToqt/L9mrp6Pa9+eabzfqDyeuvv+64r0EYV/erhQsXtnmsReK7E7yZOHGi2cc1KKbWrFlj9isd7e/ttrdHBw5cfvnljtSEGsDVY/WNN94ovqD7gQa+rO+NBhFcDUAoPQ5bAQhdxyeffOJ2rRsAANC7BUUAorS01JxI60gmvXjXC1HNUekqHSGnBcgAAAhFOrr1V7/6lenU64h2YGiHrnZEa0dkd0fW+oJzB7U7MzZ0lLl2Zlq04Ki9s1eDBVYnvI5S1Q5xe5oLT+gIZb0VFRWZxzoLorMAhL3+hG7/7oyQ7S7t9HfutHOeSaOdubpNtcP5mGOOcfs1tLP7sssuM/tnZyPyNVWK7pvaQa+2b99uOt/shZud6Qwf7UDWQId2+PqL7kdWAEI/T+3E7ahjWPO2W8Eea1l/0/3t97//facpo/ScWtNE6fdj9uzZMmzYMJ+2SVO8aGCrs7z6et6v+4WOYndlZpEGo1xNi6XBikWLFpnP709/+pNf06DpjI9f//rXZtZDV3T76GeiNz3OuFo7ZN68eSYdUVfbVwOTWiNBgzHBQK/v7NvFXhC6M/rb5rw9uxvc1QCAHp/t6dM+/vjjDgMQnra9M/r56MwHDZYrPV7rzAidJeRteq2sM2OsFFL6XjUwprO2XKEzLHUdVrBLUw0SgAAAAD0mAKEXvtboIE/piRABCABAqKqtrTUddFYnuHZsaye4pmrQDv1vvvnGkdpDaTob7bTVjrdg88MPP7R5rKP1XaUBBXv6C3vdB6U5ru+8807HiFYdEd7dAITV0WWNdtbOKh213VHqCftoeE0BFaiZKNrZbw8+aJHuwYMHO1LX2EfJ6z6kwS09H3N1NLo1CvaKK65wBB801dSBBx5ocqzrwBI9t7M+L92HtaNZRxDrNrEHHzRtjs6S0NG52hH31VdfOT5Dbe+tt95q0jFp+/1B027df//9pu36HpcuXWrSl7RHO7ittur77uh5vqT7vNUxr/ulfgY6o0Xbr0XeN23a5HiuBn60c1xTlPkqRZGm4dKAiPXZa5BQO9Z1hpa1X1iDjPRfHQGugciuvqvOA5N0f9FAiqaW0n1K92NNm7Vx40ZHJ6lVYFg7m/0xk0a/V/o+NN2SnW5rnQGkHcva4avHa93/25vJ1hX9rO+66642s3YGDBhgXmPz5s0mHZP1/jVAoQE0ndXT1UwQf9DZGvZjuKvHGw2m2JfTWTzeOB5oQNQegNBjj7fb3hn9Tlx55ZXm99o+K+wPf/iD+II9AKHHNt1XOguot/d905kiVgADAAAg6AMQerGmI6P0ws0eRLDuW5xz/zpPMfZ3bmAAAPxN05hoZ4F2AGgKFe0Q1zz+Fu0U0c5jew2EF154wYyk9PVIZ3dox7dzB4+OqnSVvUiodjg6p4/QwIymFLKKVGuwQzv5Okqx4m6HtI4W1Y5DHfmp+bCdaUevfYRsoIoRaxsfeeQRc187PbWzUjsg7bQTWHN4awe1FSDQ1B8a5HE1aKIdZTprVTtVdXS6jia2n5dpbnrt7NZOLqX7sBZa1ZHE+rlop7C+ps5usO/P2imr+7k120U7UR966CGT5skfNG3Xsccea3KkWx2+HQUWnANOcXFx4k+a7ku3q6ZY+tnPfmYCQs5pzvR96Pa00p1prQ09PkyaNMknbdIZyXpM0u2oHam67bR99hkCuk9qTn/LAw88YOqlaGd6Z/T/NV3RSSed1GGHuu53//znP+X55583j/WYqEFD7Xi1Fx62WB2yOihKZwtY9HjaUfqt9taj+7UG8uzBB90GV199tTkWtLdv6PdAO8A1rZ6r320NyCn9vt1www371dLQIMj//d//OQJ82qmt3x+9BZrzgDOd9eFqAMLO+XjmKef16GwtDXTZ99futt2VVFD6+63fC6X7rX6P3QnOu8r5t1C3q6sBCGt5KwChv+e6zwfjTEsAABCc9j/D8gMNPugJtz2goPetQIR1s/5m3ZT1f9YyAACEMiv4oCPaNe+1vbNW6WMdRak3O01BEkzvQUci20cx66hoHfnuCiuYYNEOyPY6PpxnRdg7iD2lxU7t6Yk6SgNjfy3tPPJGig5PaLoS7bTXjiWtr9BeZ53+n+5P9g4pHT1tFWB2hXZo66yHp59+2nQyOw8K0SK4mgpMZwZY9Lna2aoduJpG5sILL9xvf9aZPZqb3V6rQNMcuVsvpDvsqZS0A7+9egPa0W0PqAUi/ZK2S7e7FtP9zW9+s1/wQZ144omOPPP+ODbo56THK/18NSjj3JmrMyI05Ze9Bo7ur3/84x87Xa++Pw2QaWC1s9H8ut9pgEODDvagVkfHAs3jrzfnIt36vbf+z/nWXqFg3cY6A8OiI/Q10KMdzB0FpvT7p+9LA0k6i8GV77YO4tLUVfodaa+Q99ChQ02Htn1f0CCLVew4kOy1aHS/0DoKrrA65y3eCqzrzDA7DdZ0tJ08bXtXdPaPFsS2fxesALK3Ob9fKzjsyfJ6De7u8gAAoHfz+wyIt99+2wQfrAtVnXp/1VVXmdGEetJljWjU/9d0EzqlWvM+a4oJvfBYtmyZ+X89+ddRVnphBQCh5Ittu2XBmlL5trJGGtyoh4O2osPD5aD0JJk8PFcOz0zp0ZtHR71qx1JndAS05ne3UsJoKpRA0s6ykpISUxBa22XvRNJUMdZIXndnP7QXaLAcd9xxpiPRSkml5w0a+HA1z3VHdASzznxQ2lmonVR6/mLvjNG89/b2BXKWpm5fHc3d3khti85A0OdoR7BVJFeDA1rE3Dko0BGd+dBZIVbtxNUZJM8884x5bBWb1nRMneW71w5bHaGvHetWnQOdXeKvgtR6TqoduNqZruemet6qnb529g7t3NxcU5MkEC666KIuc/zr90KL0FozdLTjUL+fne0f3aEzW7o6Xum+o0V/NS2T0joIOjujo85l3cbu0ECFfv+tTlKdGeWrWR96naLBBovOItJOZFfb7M7noLNANK1SZ8cX/d5p4ENngljfH72O8mdB9/bY04FpG11NA+YcfGwv0OYJfX0NAFi/mUqP7c7BqO603RU6sECDv1r3xgpy64ACb6fN0tlq+ttg1Q7RdGXucN6fdZsE6rgHAAB6Hr/PgNDRcNbFuo6Q0ovSKVOmmJFu7XUQ6Mg5HeGjI8v05EyLdenoKe1c0KnO3hjdCADBFHz4v2U/yMflu2RvU7M0trRy83Ab6PbT7ajbU7drT6Wj6V0pIqodyva81DpjwB8zBTXIoCN5nW/a4anFX7VYqj34oJ3LOlpY81G7QtO5vPLKK47H2jmkHart0c4V+7bSnPf2wtWe0nQn1owL7cyzUkhavvjiC0fnUSDTL1l++tOfupQjXdOIaNoqi6aP0VznrtCR5q7UPHAu6qrndTrzwd3lOitY7W3aKak1RTqa9aLfK/vfAhlw0nQtrrAXjNWZSFb6LW/TTkpXiobrcUCDBHYaMPAm+769cuVKR6DN23RfsHdiX3LJJV0GYDw1bdq0DmvQ2DkXCPbn96c9+p0pKytzPHan0LLzrARvpv3R30279mZadaftrtDPUwfj2X/zdJaQt+kxyt52K/jnKr3+dv7tBwAACMoAhE6B1hyaVholnXbs7uiOMWPGmCCEzoDQCwkdQRnok2oA8Bad+dDQQno5b9LtuXCNexfawUTT/7SXk7o99rQQ2iGmswiDhY7y1dHaOprcuX5DZ7QIsBYntneud9YB5zw7wnn2hC86pO2voQWTA13wVbeRq5yDW/Y6Fp3RnP2udLo7p5bRwJQrI76dl/OkYG932FMqaa50e3odLfSuKZjae64/6X7mSuoe5ZwyxlfbU9MuuRqM8XTfs9NrAe2g1g5iDQLab/ZR6jrjwyqY7m3OQTtXAmye6ij4GqjP21X6W6TBW4t9BllXfFkDsKt6g91tu6v0d8v+mWmQ2xdBQnvAxd19wnnmiTXTEAAAIOhSMOn0X+vkTkc+dZRCoSs6I0IL6ml+Vx0lotOctUgnAPR0RTuDp8M4lPzQg7erOyNpnUdzaseJ82hRHeHpaj79ztLruEsDIvr7r4VZ3eE807GrcwcNAOh5gpUyQ+sH1NTU7Ldt3KWdzFaaFe0o1fXr62jecC1g62r7/MHV2hrtPdfVvN6u5kB33v9cmZnR3nJW+iZ/0Rk6uv/b06Jo6ijnfVK3n7fywbvLndd13v99tT3d2fc0lY3OgLZGl7uy7+kx7a233jJ1DTRVqwaCXJ3ppcc9XxT3tRdJ1tnd3qpR0N53wnkUeqA/b1dpAMjOnRRGzh3+ejz3FuffwvaCC91pu6s0C4AWLNcC4tYsJb221Vof3qTpwSyaikkDeK6mKNRAvF1tba1X2wYAAEKbXwMQOrXfGm2iuX+dT2Sc6WiTjk7yNK2D1oDQE0ctEOiNzgUACLTCvgkmbRC8a0Tf/xXC7WncSTfhnLvfyvVsp7MIXe3UcGWGoXaIaUFj544JTe+gRXr1//S3Wjs6nn/+eZN3+sknn+zyHMAqsqs1Fyw62ltH0HdFgwBWCgsNfOho0osvvli6Q4tKa+ellXZCO6SvueYaU9vK6tzTmRnjx4+XQNIBHloc2lXaya4zbKwC4a6OinV1v3Tu3HL1XM15X9YBJ/6k56oadLIKwmqNj1//+tfm3PT1118PinRb7pz3+mt7amoud+h32gpA6Peos3N/LZ6tdUs8HXnti054Pcbai5T7MhjlzuftPEvM398fZ87pr1ytM9NeUMBbBel1X7OnzuqovkR32u7uzDWt26GBNaWBNk0dduCBB3rtNZyPx7pfuBqACPQxGQAA9Gx+TcFkP2FsL3+m8wWH80mh80mQNcpKT/61kwMAejotmBwdHrjitaFIt+ek4e4VMA0mrqZfChT9PdZObPtt+PDhpuCpjubUGQjaeW/R4th/+tOfXFq3pmuyB1G0oLErfJGGSTuk7evVDmkdeW0fDX/iiSeaEdCB5G5+dH1f9mVc7dzzdL8M9v3Zzh5c0MCT7rs68t4agR3ogFMwbkt39z9XcvArHQ2uxZe7k/bFCrJ5k3N9Am8VSO4pn7cnI++7usZz5hxQXbdunVfapEXP7TQo3t4MiO603d1jsQa17R566CGvvobO2LPvT64MBGhv2fa2CwAAQNDMgLCfOLeXv9n5okVnTHQ22kcLUdqfCwA93eGZKfLXY0aYmgXfVNZIgw86THqL6PBwOTg9yQQfdLsiMPR3XGdcaGFaq+jls88+a0Z7Hn300Z0u6xw40NHo1oh0d2jKJJ150d3aDNoh/dhjj5n7mvrltddek48//rjN/yN0WDNurNoEGmzSwub2Qr/2c1H4xvLly/f73muqNT2G6Ojw7Oxs8znoQCb79YXOmNCghT8Fqhh5sNPi856m79HP2m7VqlVeaZNzyi8d2NbebIDutN1dY8eOldGjR5uaierDDz80NWc0JZw32IMIrtTi6WjZ9rYLAABA0AQg7KOC2svfqWkDdDSFdYKzefPmTnNf24trUggLQKjQznI6zOErmmdab/6kswJuueUW+dWvfuX4m86C0FkEHY3q1VGuK1as8FobtPPYyuHvKQ1gaGeYVdPqD3/4gyM9h75HnfURaO6mmdFZHPZlfDmCuyfSoJIVgNCAk330MwGn7u9/ztcD7e1/jz76aJvHv//972Xq1Kldrtt+neArvkoPFGo0OKT1f6qqqsxjeyCvKwcddJAJDFjHWg1ka1F4V+vJdOTTTz9t8/jQQw/1ets9ob9Tl112WZtZEPPmzfPKuu1t1/or7nAe7Ofu8gAAoHfz61xeLdZoqaioaPc59sJtOtW9s2nU9o4Jd0dxAAAA/znllFPa1G9Ys2aNLF68uMPneyNtkp2VMqm7tC5Ae52Nmoqnvdmd/qajc93pINMiy/bUNKmpqT5qWc80btw4R4pQLUZrDzhpyi20pTOD3KGDjewzoZ3TsWoQ4YsvvnA8/slPfuJS8EFVVlb6/OPR77w97dr69et9/po9lb0AuDsz13WkvfMsCHvqO09o4EtrLNjpvuXttnvimGOOaZO2UGcA2WfaeUoH+Nl/s3Jz3UtN6fy+3V0eAAD0bn4NQGhOaKUdADqy0bmol7JOMPU52lnQ0eglPfG0X1joNHkAABC8fvnLX7Z5rOmM2svLrn/TcwD7IIM33njD5N9356Zpn+w5/LUjp7s6CjQE02h4d2aOOD935MiRPmhRz6Uj8k866aT9/q7pfzoqltybubPvaaDMKkDd0b6no93tdWCOPfZYl9dvzVTydeoke+e4zsj2Vo2CUJOfn++4r8G8jgajtWfy5MltHr/wwgvdKir+9NNPt5nNpAPgOksJ2J22e8J5tt7f/va3bq9TZ43YFRQU+HV5AADQu/k1AKGj6qwZDnoxYR/RZB9pZl0I6FRXTdfgfJL30ksvyZ133um4WNC0TYcffrhf3gMAAPCM5sw/4IAD2nRoaFobZzra0/7br6mNNP2Rc7Hrrm722QreGDVrpVxx7pDWVCCatztYtLdNO/Lqq6+2eWyfpYKOg0vBFHAKJhoodHWmkfN+2t6+55yiydUi1+Xl5e1eZ3TEOZhkD3p05cgjj2zz+Pnnn3d52d5E63XY6Sw4V51++umSlZXleKyD0Dwt0KxBLauWj6WrWTXdabsn9LrWHmzTYNp7773XrXUWFxd3+p66snr16jbfw+6mwAIAAL2LXwMQ6rjjjnPc19GJzjT/pr3Q1meffSYnn3yynHXWWTJx4kQzLfWmm24y00j1AkeDEBMmTCAFEwAAPcCVV17Z5vGsWbP267B0Tr9kDU5w15gxYyQ9Pd3x+PXXXzejV7tLc9AvWLDAcdP3EEy0Y9d5tGp7ioqK5N1333U8zszM3K8zFT8Gzuyf98KFC/dLCYP/de66kj5Nz+OfeuqpNn/Tc31nzjUhtJi8K3TEeFNTk8sfi3Ngw530TRrojImJcTx+5plnXPr+9TbOg8VWrlzp8rKRkZFyww03tPmbfheXLFniVht0Zr3OLrD/DhQWFsr555/vs7Z7cxZEd9II2tus18/uDN7TWYn24t96rU7BdQAAENQBCE1doPQESkciaq5iZzq7wX7BoamadKSJjv7QWRFW4EENGTJE/u///s+P7wAAAHhKR7LqbAb7qMq3337b8VjTatgfa/olT4s7a+HS0047zfFYO53efPNN6S4diaudN9bN/n6CgY7e1s66zoItOrJcn2NPh3nxxRebjr6eSEcwa0oQ67Zo0SKvrVu3if3ztg+Uwf7+/Oc/d1kL4Y477jDBCnuw0F4Hzp5iNS4uzvFYrx00zVFX6XXc/fz1esJOB0C5SgsUX3jhhY7Hem2jM7h1FoYr9Hvqj4LZgaapeLOzs12q9deeM888s02QSjvF9Rjm6metQaWZM2fKt99+6/ibBo4eeOCBLtOpdbftnhg1apSMHTvW8fiHH35o851xl73NI0aMMPutO7Mn7PUj3EmFBgAAEJAAhE73nD17thn1pCM52pviPHToUPn3v/9tLjo6Gumhf9cCXXPnzmX2AwAAPUR4eLj87Gc/a/M3ezoMHb2vo6MtOgvSPrrYXZqr3zmNYyjTgI3WqPjuu+9kypQpbUatWvT/NKe6PSVH//795YorrvBzaxFqdADRrl27TEBI0zE513jR1GrXXHNNm05j3V9vvfXWdtenHcP2Yt87duyQyy67rE06GHsH8x/+8Ae5/fbbzeO+ffu63G7d/+0dzBro0ECKpnHatGmTKdZu3bQNzn7zm9+0SUmjdSAuuOACefbZZ9sdbKX0Pfz1r381Kd3sxbhDmT193VdffSUNDQ1uLa+frc5YsOgsl5tvvtn8pmiNn/ZqCunnpdeLOpPOXgdIB7NpIMyqUejrtnvi17/+tVdmGug2sH9n2qtr05lPP/20zWN3lwcAAAjIMLejjjqqy+doIbpXXnnFdEQsXbrUnPzraD29sNFCYDqCUqfEAwCAnkVTljz88MOOArTff/+9vP/++6aj0blOg3MAwV06Yj0jI8MUvLU6UnRksr2zMZRop6t2fP79738321ULcet5k3aO6uAN7Rh1LpKrI8zvu+8+U1ML6A5Nk6rBBA0GaKBBZwtp3RcNjOnobR197txJfN1117Up8uvsqquuMqnCrKLBGlTTkfA6ilv3a12frltTzFjrHjhwoAmy3X333S63XZ//4IMPmvu6Hh0MpTdn+p2699572/wtISHBHNNmzJgh27ZtM3/TbXDbbbfJXXfdZUaz63FIZ2VpZ7AG/9oLZIQ6/dx0hoo18+OTTz5xa4abpsrSQWy//OUv5euvv3b8/YMPPjA3Tbmns1l0W1vForXj3Tkdlwa9/vjHP5rP0l9t94TO5tLAibupppzp98f+vWsv3Vln7LMStVaLBuwAAADcEdTz7PXk8OyzzzY3AAAQGvT3XTvq7J2Djz76qJkB+eWXX7bpbOruYAOdcXHGGWfIvHnzzGPthNFZED//+c8lVGnnnHa86ehrpR1w7Y0YtzpO//GPf5gaXEB3aQ2RP/3pT3LLLbeY9F66H9oLytvpyG5Nozp9+vRO16mpmXQ2wvXXX99m5rSmpNGbM02J9sQTT7idJufyyy83QTutFeMJbacWoNb0S/Z8+zpS3n5c6830OKMz3K0ZH5oSz91O/NTUVDOj4ZFHHpEnn3yyzT6hQZ+u6ndop77uoxoU8nfbPXH11VebfdKeLs9db731luP+QQcd5FYBad2eOuPDPoAAAAAg6FMwAb6oK2K/TZs2jY0MAEHuoosuMh1JFh0ZrR2M9tSLmn6pq9zcrnCeReFKkdyeTDt2tZ7W/fffbzrbOgoC6WxSHVl7zDHH+L2NCF06qlwLMWueeA0Atmf06NGmiLCrgUD9Dmvx784K52oRdc3xr+mdPBmhrbMTND2sjrDXWUSa6iclJcV8V1ylM6teeOEFU1dAZ3N3lj5H16vfvb/85S8upwHq6XR76EwTi3asd5SiqjP6u6DBK+1Y15Rc+tl39dnqDHxNeaWz7NwNPniz7e7SYEF3Ov11Rs5HH33keKyp+dyh28uaPaGZCAhAAAAAT4S1dlRkwQd0JNKcOXMcj3V0VE5Ojr9eHiHKKmxuv0CwChbOmjXLrSJrAAD0NJpv38ptnpeXZ9Jt2BUVFcmaNWvMSHQtqKydpDpS3Z0c+YAnNN2ZBhf1X50JoKlxDj744G4Vbt+yZYuZUWCNdNd1asBB19tRwCNQtI2aKkj/1doYGnTQoIZ2KlupqXqbPXv2mJltVuHte+65R84///xur1dTy+lMLz3OaYokPdbpttZZCxpw0Bl1wdp2X3r88ccdqcX0u/Lee++5FVTT4J9VVF6DPTfeeKPP2goAAEKXX1MwadFDzSGpHcT9+vUj+ACvcM6LWlVVZUbAAQAAMSO57YVbAX/RYJe3661osKGn5KDXmgSnnnpqoJsRVDQQoEFTHSSk5s+f75VOfE3hp7ee2HZf0doXVt0KdcUVV7gVfNCZE1bwQWsEaepEAAAATwRsmJCvTxABAAAAAMFFR9InJSWZ+1p3Y9myZdJT9KS26yAtLdCuNE3VxIkT3Vpea6lYdNmuUl0BAAAERQBCp33aix4CAAAAAHoPTY105ZVXOh4/9thj0lP0lLZr3YZ//etfjsfXXHONxMTEuLz8N998I5988omj9oP9PQMAAAR1AELzElu2b9/uz5cGAAAAAASB6dOnm/oMVp3ApUuXSk/RE9quxaO19o868MADTWF1d1h1I9TVV18tqampXm8jAADoPfxaA+KQQw4xuVC1EJvWg6irqzP5JAEAAAAAvUN0dLTcd999ps6AVeC5p+gJbW9ubparrrrK3Nc6JO4UaN+5c6eMGTPG3LRmxKRJk3zYUgAA0Bv4NQChJz5aqOvxxx+X+vp6mTt3LtM5AQAAAKCX0cFpeuuJgr3tF154ocfL9u3b18x6AAAA6LFFqHUkRn5+vrS2tsrDDz8sH374ob+bAAAAAAAAAAAAQi0AoVNWZ82aJQcccIA0NDTIzJkz5e6775bS0lJ/NwUAAAAAAAAAAPhIWKtORfAjnfWgNAXTggULZN++fRIWFmZuOjNixIgRZtpnfHy8W+u1clwCVVVVJrClNNiVlpbGRgEAAAAAAACAUK4BYQUgNNhg0fsaA9FbUVGRFBcXe7ReAhAAAAAAAAAAAPTiAER77AEJd2ngojvLAwAAAAAAAACAEAlA+DnrEwAAAAAAAAAACPUAxDvvvOPvlwQAAAAAAAAAAKEegMjLy/P3SwIAAAAAAAAAAD8L9/cLAgAAAAAAAACA0OfXAMQrr7wiY8aMMbcjjzxSSktL/fnyAAAAAAAAAAAgFAMQJSUlUl1dbW6aiik3N9efLw8AAAAAAAAAAEIxABEXF2f+DQsLk/79+/vzpQEAAAAAAAAAQKgGIDIyMhz3W1tb/fnSAAAAAAAAAAAgVAMQw4YNa5OOCQAAAAAAAAAAhCa/BiCGDx9ubjr74YcffpCKigp/vjwAAAAAAAAAAAjFAISaPn26+VeDEH/961/9/fIAAAAAAAAAAMAPIsXPLrjgAvnggw/kjTfekJdeekmysrLk2muvNYWp0fPV1tbK0qVL5bvvvpOVK1dKeXm57Ny5U/bt2yeJiYkyePBgOeaYY+Siiy4ynz0AAAAAAAAAIDSFtQagGnRDQ4Pcfffd8swzz5jAw4gRI2TatGly0kknSXJysr+bAy9asWKFTJgwocvnxcfHy2233SbnnXee17d/VVWVzJw509yfNWuWpKWlef01AAAAAAAAAABBNgPi0ksvbdMJrSPjV61aJTfddJP5W05OjqSmppr/c5UGMebMmeOT9sJ9GRkZcuSRR8qBBx4oubm55nFERISp+fH+++/LK6+8Yj73m2++2XzWJ5xwApsZAAAAAAAAAEKM32dAFBYWtptuyd4Md9Ix6XL6fC1qjcBrbm42wYbOaHqmSZMmSWNjoxxwwAHy4osverUNzIAAAAAAAAAAgF5YhNriHPfQIIJ1Q8/VVfBBjR492syQUDr7Ze/evX5oGQAAAAAAAAAgpFMwaUoeICEhoU1NEPtjAAAAAAAAAEDP5/cAxLvvvuvvlwwJTU1N8vXXX0tJSYls27ZNEhMTJTs7Ww4++GBTR6En2bFjh3z66afmft++fc0NAAD8T0FBgeP+eeedJ/feey+bB+gFpk6dKsuXLzf38/LyuHby0J49e+T000+XyspKM8NeU76OGDHCo3XpYKmVK1fKxo0bzXWMPo6NjZWsrCwZNmyY5Ofne3UWvzttZ3/xLX6LvUvrftrTLxcXF0so0PqW+r603mVRUZHs3r1b6uvrHf+vtS+nT5++33KaknrJkiXy1ltvmcwQu3btMuuy10+95ZZb/PY+EHiLFi0y+4tl7ty5juwh2N9nn33Wps7wPffcI+eff75XNtXHH38sM2bMMPcHDhxoatlGR0e7vPx1110nixcvNvd/+ctfyq9//Wvp7fwegAgVLS0tsm7dOlPPYMWKFeamP6D6I+LNnb+2tlYeffRRcyDSk1BnUVFRctxxx8m1117b5gQp2OgPsAZO9Ev8xBNPmB9lNW3atEA3DQAA9FBbt26VsWPHdvqcyMhIc8GggzfS09OlX79+ptNQB3Ecdthh5u/wTmeSNy/8gO74xz/+4bh2Gj9+vEfBB71uefrpp2Xp0qVtOhOd9enTR8444wzTCTJ06FAJhrYD8A+tRfqrX/3KDBR1hz7/F7/4RcgEYYBQ85Of/ESOOuooM3h606ZNMnv2bLnyyitdXl4DDq+//rrpI37yySflggsuMNcgvRkBCDfpDrRgwQIzCsYenfaFNWvWyDXXXCPr16/v8Dm6M+uskmXLlplI6cSJEyVYvPfeezJz5swO/18vUC+//HK/tgkAQgUjZADXZ5HqTc/bdDCEjjJ88803zf/FxMTICSecYEYRjxkzJug2KaNvAffpTIX58+c76tPp9ZQ7tmzZIrfeeqt88sknLj1fRy0/88wz8txzz8nFF18s119/vcfpZbvb9p4qVEfGI7RVV1eb/o7y8nK3ltM+HA1asJ8DwU0Hel9yySXm/mOPPSYTJkxwOQNN//79TZ/ns88+awYxPPjgg/LXv/5VejMCEG768ssvHdOifUkvkLVzvqKios3fR44caXZkPdHVWRdWAWfdoW+//XYziu+ss86SYKbTl+644w45+uijA90UAADQi+n5kwYj9HbMMceY8xM9zwLQc+kMAg06qtNOO81ce7hKB3Vph4N2LDrTmVOaAjclJcXMUNBARWlpaZsZ8jpj4osvvpB//vOfJoWWP9sOwL808GgPPhx++OFy2WWXyeDBg9ukatFZUnZ6zqEzJyyaxk1nQ+i/mt7NwgxNILAOOeQQM1ta+4F1INO//vUvufHGG11eXo8Hzz//vDk/eO2110zAMpgz1/gaAQgvSUpKkvj4+P0CBp5obW01o13s69Ifo/vvv18KCwsdf9MT47/97W+OUTJKcwTqc4YPHy6BdsQRRzhynmmuVD1B19kaL7/8stxwww3yf//3f0zTBwAAXqM52RcuXLjfeZUO2NDzpqqqKjOAQ+tqffPNN+aCwN7xqPU29NxKgxEAeh5Nkfvqq6+2ufh3lc540M4BvW6x6CypK664wlyztJc6QWfFz5s3T/773/+2mcWus6o0GKHHJH+0HYD/WbMpVUZGhkmzYg8guLKcBiqeeuopkyISQPDRgeEagFB6jaGPXf2+Dho0yKSK1Tovra2tZpDBww8/LL0VAQgP6I+K5uI88MADZdSoUeamUW7dkbyxM+kPkl4YW/RkV4MMOtrGLjk52UwP1uJkeuJrjeTTC+fO2qHPGT16dLfbedVVV8nVV1/d4f9rxF4DJxbdXjqS55xzzjG50zRllAYldD0AAOB/mJbvGa330FV+1Z/+9KfmX83nOmfOHDOCsbm52fytpqbGjELUTgQdSAGgZ/n3v//tCCwecMABLl/z6IyG3/zmN22CDzrjQa+p9DqvI3p98+c//1nOPPNMk+/Zmp2u+d11wJXmjHa1QLWnbQdcTdlJYXvv0fMGLTht0bqcrgQf1Pfff++4r/WoCD54VvfrnXfe6fU59eF7J554ohlMoAPE6+rqTBDCnfSIF110kQlAWPvs5s2bZcCAAdIbhQe6AT2NXpRq9EsvVn//+9+bzvQhQ4a4fGLpCufgwW233bZf8MHut7/9bZspvrpz26f0BRtNvWRVqn/kkUfMaB8AAAB/0tQmeo6lnX5paWltBmroLM2dO3fygQA9yI4dOxyzr5XmanbV7373O7O8RTsHdFRyZ8EHO+18nDVrVpsOSC1cqccXX7cdgP/prEqt5WDJzMx0eVmdjenJcgD8T+sx6Qxpi/YF2wcrdOXYY4+V3Nxcc18HGcydO1d6q8hgOHBrWp5vv/3WFFvWx3v27GkzJb4r2vn/9ttviz+4WnCkOyMeV69e7Xg8dOhQUxyxM3FxcaYwihY1segJrM7SaI9O87NP7/VU3759PV5WI9dPPPGE+Zw1YKLvEwAAwN+OOuooM+NBRyhZFxTbt283A0J0pimAnkHTvGoA0eowsGY7deXDDz+UpUuXOh6Hh4fLX/7yF7c7BrWQvc6C0BkRFj2OXHjhhWbmui/aDiAwNB+88wxMT5aNioryarsAeN/48ePNIAMrgKj9z+PGjXNpWT2nOOOMMxwDEqyU9PY6Mb1FwAIQtbW18sADD5gpgTqNxaJ5sdzlzdkHgfbee++1eexqQWl9nj0AoUEd3ak72l6B7vC3B3LsxdsAoDfQgLtO29ZOTr0I0c6GhIQEU9xSc0V6e2adJ3R6qM6m0zbqwAAdIX7uued2eqGkv+H6vtauXWtOzrQzV4/3OpJUi3h15yJLByhogH7jxo2ye/duM+pM6y/ptPWDDjrIbDtv0gD5V199ZYqM6jbQPOCaVlDT8nR2kanbQGsMrFq1Snbt2mXqQ+lIe+3k1nUEgrZH9zmdOqyjc3UasXaUdWcggZViRFNGbtu2zZxc63p18IPuwz2Ntvu6666Tu+++2/G35557Tn71q1+5PPhERzDrPqqpnXR/1X1IOx21E1P3f18PYumKHms0N/2GDRvM7A49/9bvkLZL08gE+3Rw/e5r2gpNlaMdtbr/6vdei37qYJzu0O+5Hrf0nFSPd3pM1tnHms5L02N0d/1K1/vZZ59JWVmZuQ7S/UJH1/sizY5+N/V7r8dhTSum78XaVvqZe4se43VmuL4n/Vz0GK/HlpEjR0og6EW9RYtGuvqdcx6NqMFITbHrCZ3lrde3+l2zvnd6LNE6Er5ouyuD23Tf1kK5+tul+0F3j//6uev7098V3cf0Peo5jK5Tj6WauirQ5zAW/U7rd0GPz3ruoB0+um21jVo30Z1OY3/+vvZW+vup50+6PfV8Srenfh/cqaXiL570W/mCfv/0fFW/47qf6z6u5+z6u+7qDC5X6TmOHk/0vFh/xzQLh6v9Vb0pNZeeq+g5i34eup30+KjbSo859swlntD1ff755+Z3V68ztMC59u3puUp3j2f+ptcPOhBd34ue1+mxWc+JgqGerTO9BtTrG/1crd9sVwMQ6tRTT3UEIPS36P333zfp6XubyECdCGhRLf2B0QO3nqDYD+DWCYv1t/ZOYILlgO9tWgDRTi8UXJGTk2MOZnrBofTiUr/I+vdgZC+wrZ0zABDqtMNc6/VoUUrt3O+MdhBpujqd3eZcDLegoKDdZazUdu3RaaP33ntvm79pgczly5fvlxNY/9URHnpC6Oz0009vN4igJ1KPP/64vPTSS+aipD168q1pC7V2kKsdK5qi75VXXjEjUzUY0tnsSO1M0HMLfQ1XT8Dt29LaRvo5ac7uBQsWmN9RZ9ppqLmU2zvpfPHFF01xMeu32Pn9a82j6dOnm856T9rm7nO1KKp+Lu2lOtQ26Ajb66+/3u1zBd037rnnnjb1qpzPXXQUsHbCuPteAmnixIlm37fSsGgnm15g6GfWWefTkiVLzAhqvSjvjF5UaQ0svQjpzE033WT2JWf6t/b+3lndEN0XtX06wEXbak8X4UyPA3oc0eOOq3ms/XEhr523mgpHrxvao4E9nS187bXXujzARreDnnO/9tpr8vHHH5uL4I7oMe+kk04yxYk96VjXDgIddKWzk+2Drix6QTtt2jSz/3Wn07apqUmef/55Uzeuo31Rj42aCkBTjGlHSFc0YGL/bdHvvRZj1gCbzhDQfUvv2+nzAxGA0CCSPa+6pkRyhf4e6/fXrrPvfFd0G0+ZMkX+8Ic/OP6m+aI7C0B42vbO6Gz3f/3rX6Yzt719+uSTTzZpp1wN3mugQb8vOupTj/3t7csW/Y2fPHmy2Rc6m/nhvH+5cq5j3w87om3Tcy1Nk2F1FnVUr1C/D3rM03Mud/jq97W30uPwo48+ajpW29ue+vno/qrneu7UvLACjEceeaRL7XD+/XX+Xe2sgGxH9UD13E9/X+1tcuW3vavzpe+++86ks9Zt11E6GP19+fnPf24GELly7qnHBescVs/hrDqj//nPf8zvsBVYtV+vEID4kZ6jPPbYY+YYqYH/juigJB0Nr7/57hwfdBCD/u7qtZbed6aBCE0Xr9d3OojCk992Vzh/Bzqrw9HR90mDtvfff7988MEH5tzFmQ7Eu/HGG03tBV/55z//KX/9618d/coatNNrgM4Ghei5pvWb8tFHH5nPQX9HXKEDkfr06WPOCZX+nvbGAITfa0BoZOuXv/yl+eCsD1v/1Z1Wo7TWY6WjDLUAV0ZGhvmbPeigowr0//Xmaid9T2C/YNAfCWubuEJHgNo5/0AEk9dff91x316oGgBCkXZo6ohKTcvQVfBB6Ynrm2++aTrB/UV/Y//0pz+Zk9f2gg8d0RPtU045xaTV6yj4oLQwp3bC6MmW5sXuinbIaCe/XpBqx0xXqRn191MvTi+//HLHyZ27tI0axNDOwvaCD0o7KrXzTjseLHryrPWY9ES7veCDtW79/LV+lK8HUeiFqM6C1JP3juos6fbUzkNNDdJVx7mdpirSzpqOgg/qiy++MB14GsjpSXTUoHPedb046sgnn3xinq8X5a5sQ+0s0I4I3Vc667zzZue9przUGbI6OrKz4IPSfVcvQi+++OIO92N/D1a54IIL5Pbbb+8w+GBdW+jx8uyzzzbHIVfocUU7ZbQTsbPgg9LtpuvX74qr+fwtOiNMp+1rYKCjz1yvie644w4z28adnMJ2ejGvnUC6rTrbF/VYpaPutGPL3fdi0WCwdmZpB69z8CGQnIMIVgC0K9p5Zz8maydBd0cN62dun/Gm3ycdHObttndE9yf9nWov+GDt02+88Yb5jXUe/NYR/S7+8Y9/NMe9ro5fes6jnVS6n9iL9PqD/v5okFc7bjsLPijtPNJrUh0YEQy/r72Vnk/NmDGj3eCDtT11P9X9SY/ZvZ1+fzU9pO5fejzv7HdDvwMa+NAgtyfHa123DijR8+tg7lsKJP39eOihh8zxVDvbOws+KD2f0c5vd36DdeaJ/m7rdWF7wQel1z56DqfH/vY69YOFHnP190QHvHXUTj2n0XM0X1xH6LmxDhDQYI71269BIQ1YdzUj1f7brN9D/T10lQ4wOeyww9qce2hbehu/z4DQD1ZPRKwRPtppridJOspBq9prJ4bFirgqnVKmo9DmzJljRmDoaEvdQXRkQajQ92QvfqZROHemfDtHHvVE9/jjjxd/0pMCPenTkZ6djch59tlnHVFzjbQDQKjSkxvtdHTuCNDUQRqA1YC6/ibqCaUGJ/QWiBNHHSlpT0Oh01/1hExHc2pn/MqVK9v9TdffcHtwQIPn+puuo720A8aaXmtdIOmJ+c9+9jNzwdnZiEPngIOO5tH29O/f3/x26Embng9oZ5j9ZF+DGzrQQc8huhoB5Pw5/eY3v3HMCtHZeTqdWUdy6km9drhrEMGiJ656DqOj6vREVmdqWCNL9e+aLkCnxX/zzTdtihnrKDIdOOHqSCNP3HnnnWaElNUeHcygo1x1v7JSBFk0aKQXl3rR1FUuUh1Zfd9997X5m37eOtpZz0H0M9YOGb3g1c9HL4SCPa2PM02VZQ8u6ednzdZ15hxI0m2tI/B1W+uIKL040f1fz3t1yrxF9xVdVvchX3IevKPvQT8n/R7pqGR9rPumfofs+6i2VwNxOoq0s/M5X9Lz/kmTJu0XCNFBSTo4SdulxyUN6lgXcLp/64g6PR7oBXhnnI8vuj49Huv3Xb/72sGq3xPdn631678aRNRzcx252BU9D9dONfu5vdLvhB5fdX/R473126CjCLWD1126DfSY6hx41c9aX0ffm15j6AwY6zn6/vW96PvU46WrdD/RILUVoNV16/WYbjd9DXsdO3/Ti3mLHstcnYWhwTk7d0fCt0d/o/T17evW+x0FNjxte3t0BK4G+y26D+jr6rFAO3Xsneb6m6aBL+0IO/TQQ936zujIXV2vHkv0PEE/fx3dag/o6fdXOz71Gt4fKXR0RKn2DTgHWzUFmR43dD/V/9PfPefjcqB/X3sr3TfsaaQ7Oj4q/ey0I1yPwTojtzfSgLvOxrMfM6xjse6Lem2h52LO33U9t9WZWdoH407/kg5MsgaO6jmDfo/0/F7v60Ah7b/rzfS8QM83NKDrTI+Per6l54N6rNVzY/sgbFdpekMNeup5kdLjmP5G6HGtvesTbYuex7rz2+4vOvNCB+FY17lW2mE979IsOXo+Y78G1kCy7tfeGnCu5zz6eVmz/pWew2hAyJXZ+c6/kxoY7WpWs91hhx1mzvWUfnY6wM4XaTiDmd8DEBpAsC7i9MdFo1rWNO/Oph3rD7tOW9eRZrrT6M6rJ0v6ZdPRPqHAeVSsu1M2nafQujLK1tt0FKBePOkXUQ8UelDRg652wugPoR4QrdGE+nnfcsstZioSAIQqTR+kObItejKqnfba0dne754eL/XCQkfPtTcywjpx0eOpvTPY6hBvT1ep7vTkVkfvWKkfdLSUcyoTHY1sX4+OMtQLcatTQjv7ddS7dlzqBZCdBlf05E6DHHrirRdH2kmgF56dnfBpAENHkuqsCe0Uai8tjF6QvvXWW6bj0aoppNtbf4+6yrltpyejOjpML3q1w0CnJttHsOr5hv6+WVOJ9X3oRbO+5xdeeMF8ljr7Qi8M7TnWtX1/+9vfzHu3f1aaKsqdAImrNNWOntRqYEA7JXUbOKfA0BFzuv2t0XA6QlPfg3b4dkQvZO2FVZV2ALSXwkMvhm677TbTGdNRyoFgpXnf7alB9XPXC2wNfLVHP2sdlakzDfS8p70UZdrJpfu6fu5WB7B+v/Vcqb1Cs3qhqUFLpeu1b++O6nt1RDsGdR063V+/2+3l/7dGl+rxxOpA1otknQlkTyPjL7rtdXSxPfigxxQd8anHAnsaCT12abutDkGlU+j1Iq+rQTjaQa+j8HQgjAZN2zseaweiXrvo8cS6KNbAmk7Dz83N7XDduk1137cHH7RjTY+Zzh3cGqjQ7azXNppuqqtixc4BAd1X7MEHfT96DHNOsaS/J7qdtM6JFbTVUeo6os/Vi3tNu6C/UXpurzVTdPvZO1b1NbqaUeIr2nFh0d8vV2sOOc/481b6KOcAhL6Obi9vtr29/UFTslivf9ddd5kOQ+fX0usv67uuxyf9vmmKsM5Sr+n3To8h+pus+39Hv916/Ndjh54jKN039btrD+xaNMhvndPo99jeiWf9vT3t1VfQgILOQrQHH/R7rZ1dmmbJ+fdW91XtlNVjs/4mBvL3tTezgq4dHR+1/0DPJ6zZEfq56f6kx3jnc01/0ICajkRX2iGsqcYset6o/+9M9xHd/+yjp135bW/v3F0DAvbgg/YX6bFYf+Od04/qd0KvN6zjkAYI9fjvaqBbBx5Zg3J0hqF+l5zP93p7AOLvf/97m+OWnkfoACMN1Ld33qjHD71msQbjukK/F7qcnnPoObfuO/bzID2G6/WIffC2ngfpvqlBimByzTXXmHMpPefQ9+L8e6uDG/T3SM+HrPNBvfbQWaTdpedjeo1m/83XFE96/etqUE5/93QgjDXj3/7b7YoCp9SC2hYCED6kUVKrc0C/nLpzuZtjVj907cTQkUc6Yku/vHpSYZ850VM5T6dyt/iY88lYV9O/fPk+uspTrEEHTUNBzkAAoU4vWC16caApbDrqzLQuOPQ3TW860qmj2W7Ox3y9EOsoB2dXrNc588wzTUd+e3li7aMXrRQEVoBEO0s0pUlHHX7aWaUXLjoayOqQ1pM37ZTXc4H2aK0FvZDv6rdQX1unPetMBD03sEYf6mwODQ64Wg9COwv0fWvnjXauONPRZdrxqJ2FOireOnG0Omj1gk6nw7fXPr041EEB1kWKvnfNHdre63SXXqToOZZejHRUHE1PuLUz3B6g0dHunXWQ6AWAfZq/dqTZCzbbaQoTHYGro+38nX6ju7SjQL9L9nRi2snQ3ndWO/Y0wNjVLAG9sNG0Rtr5oOmrrM5iHUjTXgBC9/n29ns9NrjzHdcOD73Q7ayj3N6xqB06OmLf2r91n9CObH8PFNEOQXuaNt0WOvumvdHj+llpx6XOGranM9Dvpb73jr7/eqzQ99ZVTmy90NTvrwam9MLZOl5qGoTOZmFrR789TZnuP5qyqL0OM31f2nYdraj7kztpMvS4Y6+rpuvQ99XR/qAdI3rBr/uhBhI0UKLHNZ2Z5QpdRvdDDcq0lyZWXyMQOe91Npx9O7iSJ95iX065WkekK85t6Cgw0522t/f5KO3Q0M+ovQ5M/T89PmvnlJWPW3+fNH2ZFfhsj+7zrhRP1eO//v7qfqjfQaX7tQaxnbetBvmtY5pzW9051mknlX4f7amhNMCrv3MdzTzQfVU7u/9fe3cCPtX4/3/8biVCpBBFSYSvkC1byl4i5IvIvpQk2fpmy14SWULW7EUpWyp7RLYQ2RMqIlIiJan/9br/15nfPadZzpk5s53P83Fdn8tnMjOfM2fO3HPO/X7f73ebNm1sILOU369Vmca7TOOjsqM1PqocizfxrtU2eg80dpXiHCFdkFj/num4TXeuEPS7XaXa3IlrBZn1OU/3He39f02G65xTFORWj4AgJbC98URJwDqHTyXX64440OSz5iU9Ot9Qxn6m+S29V7pW0E+YcUcBOo3BujZKdY6peTUFoL3V2DpP0e9ugKwc6LXo+laT/qkC7Tp/ULBa16OaO/b2swK5+Xw36jtOQWO3LJ9KIyuRPWwymD473jWCAvm6Ngq6wm0LX3Ntt/dTVVHUFRBa+uvRxIm/sWZQClooc0MXtqIvpTgEINylU+JmXgbhD+Z4XxrFpMwoTbYp0q4PuAZWDTQaYPSe6wOri1wNzGEyvFze8rN03DICQCX696uZZvkrU8yKmbOM+ad8aziWvVo1TfVmTUzN9m1MjRbNSrYZbi8BXQxkCj74hf0eyIdOapUpFaRJnTIl3exkTTQEKfmnyS8FFbyJeGUFalIvVeaJ/i3MMnFNQCr7URdaogkdrYQI2nhQdKKeKSigiQdlt3kTtN7Fs85BUgUfXFoZ4mZJKausEAEI76Q63eSIR9/FmijyJkl1EuxNLvopw82tUa5JKGUkZqJMewUt9J5XWo1TbbsbgEh3XhF2Yl4TzZrgU1ayd1GlfVuoC3gdr9mCDy699xoDtDpHNJmnlUGFLBeWipvFJ8p2zVaTX0EC1eJVcpIo4Un9aZQVmkrQxrtuhqpWX6gXhFvqJR0FTPxZq5mydb2JC/2doAEIXVBrOzwag9MFH/wZeNpfyqr0slx13p6tBI9HY3aYHnXF4K/1H2SiXDRx4C/DE1XGqP86J11voly3PR1dcykol2nlo8Y43ccdnzWxqezQdEG7MNulCR0dXwo8eIFrleAN02shDK2WcMt/KdChVRhBJ4XCZNJH/f2K7OOj3kedT2g89+ZLtIpQCSxVqZKCO9mteR8lzGR7/dp3+qwra94b6/QdG3QVRMuWLc25556b55bH9/1wyylp/AyTXBtm3NF7mCr44FI5PS8AIVo1VG4BCF2r6Vwn0yo/HdsKFrjXGbpmyjUAoXNtvTcK9nv0XZQp4J6Je16tFXc631TVlyAaNmxoX7u3Ui9bn6I4KmoTau8CThdEGsz8/EufU2V+erRsR2++PvT6ci+HZnn58p8Ah60V6Z+oyqWuZb40sablh4pqqh+Eou26sNF7pItYbzlYrsEH0YRNpp9sk0BAuQcflg171Kz47Gtjlv6ttcb85LoPlv5t96P2p/ZrOSjnAKkytLVSIQhlqnl08ZNqyXk6+o7waKLNLU+VL130u9+dYZppi1ZMZKPSWX7KGs9G5bHcyYdCrgwIWnrKDRopEzpd/XZNurr1v5UAEmQFqwJuek8qjf8cJcqG0W7ZhVyWbxea3jM3IBL2M5QvrTByE5Z0wZltss+b8PRfTGolRZTc907XHemyF/Ua3J45WlkSJBCqlR5Beku4PXjcz2WQ4INHJW3dAG/QEjQaw/RdUW68FfaebBM1nlTBnqDfg9n4nyddYCnXbU9Hq6qCNNHWZ93tw6cVGl7ZiyjoeNYkfDHGEn95DmVrh62yUKrv16ou6Pioz4V7ja95Ii8gXBUoA9xt0q0xPGjygiZ93e/RoOO96Py+EOVCK50ms90ycQomaJVOIWje0x1L09FqIbfvWjmuQNb5Q6pSoH7+BK1cX4uOdR3DXvBBAXYFPHMNPoi/n5H/OzyTatWqJX3Hh3lsXBR1BYRbEihVtNZ/oqAId6bsT2XweG+aGhTlmzFSbjL1xAhy/7ANbgCUnlY+mOWVlS1c9pb/a5a/OqVkqyA0EaAsQG/SSktoyy0jRdyJiGxl9twlo3pcmIC5yj+42R/KvFUpxTAUYNc5gi5A/d91yl71stc1ERiUsleCXMxpUkXL6L0sPE3IBbkw8BoAe5MQhQpG6XUEbfysixWXv1mux13xIcoED0r3DXOxWw78zVbD0jGpY1SfFbdsVarnVn3rUtBnR9un4Ir/M6RzdK+uc7G3zy1bJEGCD+4Fq4JH3kSv/7mC0Pujz7Z+3EaI4l8dpn2TKoPR/3lJVWYrHb1eN8s1E3eyWNdAYVYl6PpKZaW8+t7+RsyZArDlmMXtH0+DJjoV8lop6HVZrtueTtjjzSuT5H1mwlYo0BinsURjnv81uhNNhRpLtILD6zfhlU1r165dxXy/5jOGuyv1MnGPMY1rQev2a5zQ/iyksMer+oe6x6tWpFQF/uBg2CbcmsT2Su0p2KjyNkFWZBfqs+QX9Jj0V8LIVhnDFeVqU313uuOdVo4WqtF8mEQejTteH9hyTHoL+lq0SlXnGl5Fl1xei1b1aVWvt8pPz6ck6XxXn/sDKGHH9rWcx0f9vVAJihqAyJaJ4M8U0eCYqfaze/9SNTyLkr/URKYVIKn4M/TK8QIhCt5EXjoaoLKVhgDK1YrZVS8SXgwrZpVuv6pho3vBpLIEKg2i0gfKkCuHBmHKbvLXpUxHk2vuJKoyOcI2odP3t3cyGWQFozJfVPZJqyU0ge8vWZiO6gQH5Z8syLb93jboAi5odph73uLv+xSVMK/DfxKdbpu8OuHe5HSYi7ioGroWk7+HVpBMWh2bKjGiz4cyFYOumghzjOZDy7y1NF+TGPoMpSsH4xemH0EU/PVwtXIoKAU21ZfD6x+hCzuVwMvUk0CTp15ZOCUzqYdM0JJh6faN/zVooj8ojcOa9Mt2DaDJXq/clGhSNOw47NYjD7qSPNUK9nLgX/UdtHxhqu9fjYNh+/AFGUfSfdfnuu3phDne/PcNUo9a5Q01lqhxvb4bgtYxL9RYot4S7neXEgKClJIsl+/XXOm7xl1RGpTeP/9KvHTUt8hfEi9qYY5XjT/KYPaCw1Wpfro/SKzjK8yY7w+IaszPFoBQtZFilbgKekz6hUnocs9l8+VfvaoAT6GE6UtUjOuMYr4WLwAR9rXccssttjehuwpIyR1hxpt0/NcEYavOrO48Xsl4OuesSquMihqAcAewVCchihrq5My7GPv666/t8tB03DpeQScjypk/YBA2AOG/f1wDENnq9maqKQeUu+qNG/3/8kuIdr82CV4HPWrK+FfzQTV99KiRnn50kazeOKq9vfPOO9tl6DpJKjZdyAQdO/3ZRiqtp59cZZqA1d9SnVoFbHIR5oQ1yJJgj1sjO9fH+bOro5Lr9mTaJney2r/0OJuwtfbLgX/iMNNkpCa/lPTgZuCGUegLRJ1vq3a2Mh9zyfYu9gWsPxts0003Db3izG1gredLF4BQ4EEB4VzLuKbbN+71iQTNmBZdhCrAp+MqE036uoES9b/IdQJHggakopiYLwT/2BX0Yl7XnkoAcycQ9J0Uxev0X+umC0Dkuu2p6LWEyVjXsabzEC+pIFM2pq4zb731VtvU1lvBWA5jiT8AEiZIUA7fr1VdmPFRn1edU3gT71Upe9h/7t25c+e8ni9I8oP6dyI1/+qjMBPrYYUpC+hey5XjmBNmDM31tag3ihuc08q1e++9N1QPxkz839Fh93ONFI+vSgGIovaAcE8IlGGUiiZiPJqcSUcBB3eJc77LVcuBf3AJu9TI/yUc5gMOoDyoYbKpWXW+hIqiZg1Ts12bkm6CJij79Omzyko3XfQru1/BCf1/lSLq1q2bmTBhQlHL6LmZsMXO2E6XQKCTRwVucg0+SJh9mGvGZKEyLXNViO1xJ47CHCtR1lMvFh3f/gnkdEEUZV+qZn+uwQcp5Odcr0W1b9XsPde/U+xynv5J27DHj//cN13WtfaJmhLm00MuXamuqF9DPgGDoLwsw2zKNbkpn1Xk/gn7bMGfoLQSKkhvh3xXwOdzrCkr2n1Mus+LVgqdddZZdhInl+BDIccS/2ehkHMC5fZ9Hwf5jI/FXqFXSlGfewcZ88Oe71Ul/nGnkPNucRp3ivFa/CuDTj311MiCD6m+o/3f4dksdVZIa3/ku+qx0hR1BYSWFXvL5nRg6ILW/6Wj7E812NFJyvjx4+3FQarsp9tuuy1p4FQ/iLhlAGjZeD6R8Sg/aACKQ30Kanc/zvYsWPHNLGP+Kb/shYpRq6apvnkTG3woVf8H9yJf32dqGjd27FjbuEyNVv1ZE5rQUl1R/Wg575AhQ/JuSBm1YmXUXHzxxUmTgzpf6NSpk61Brixn7Red9OnEzV1arp4U+UwqInUWkve+h33//T0Qyp2W1bsTZTruUvUY0+s677zzkiYFtHpJdYD12dW5qyY2tdTaf3FRrHPWgQMH2rJCHm2H6m3vvvvuNuFHnyFNKOvf3YtCBUG93gBxpHJUV1xxRdL7rGsUjS9asab3WxnwyrZ1azqrfFUuJU8KoRwzG0vJP1EWpiSCynx5NbNFn5n99tsv721yP3ve34l624vl7rvvNpMnT07cVramVtyojKSqFWhlnMZKjXfuWPK///3PnvOUcw/FSqU5k6AlZcaMGWP69etnf9f49sorrxR46xC1XAN/6ZRbr9Cgx7LmEN3VfrqeirK3Q66qyrhTCZTMp0R2L0lEfSB0rnvIIYdE8vz+7+iwgbqlTgCiXJM6YhOA0ESBTuxVw06Dnk5kDjrooKT76ORfy2a8N0cn+hdddJFtiKUTG100DB8+3H6R6oOu51EDOD1vpVOJKl3weCsZtKRUB3jQqJo/2lfIJagACkeT5aWeMEdh6Pvq9NNPtz/K/Ndkp2rHa2JL343upJKyqk877TSbqVuoxma58JeR6N+/v12pECW9drfhniZL77vvvkDBmDiUZCw3yij1TrjDZhxWWoaiW75Htt9++5QXlkqS0TmpWyv7zjvvzJrNWaySRkpicSf+9NlR6ZQg54al/Az5s5e1v8KUgfCXz0qVDX3PPfckTeaccsop9loj2wRC0P2S6jWEKenjfw1BxmEFltRcsaryr1IK2pdAVAJRPYY8mrg455xz8toevYfTp09f5e9Eve35ji+6jnYfk+rzomxPff+6Eyb333+/7bVQDmOJ/7NQad85VV0+42MhVrukW9lWav5eDNOmTQvUnwrFGXd0XFaVklXl+hlxey+qRFnfvn1tqUpdW1944YU2cUj9F6MuvxW21OwvzuMz9SiLq6Kv52nXrl3idzV881PGmA4MnRDpQkCNki644ALTpk0b2zREAQoFH/T/vfuceeaZsamb1bx586QPt//kNRN9EaV7rjjTIOP+qNwAAJQ7ZUzou+3ss8+2Tf7eeOMNc+655yZdUCgjSAGIcuLvURG2XGAQkyZNSrqtbOUgwQdNlDD5ED01IvTMnj07cINlr59XpdDxo14JrrZt22Y9RpXxqz4LQUpJ5DO5GIa2z81w1MVX0MQU/8VVMfknotzM9CDcoFCq5/O/d6oNrH0TJHsx6HvnHyPDvAZdLAdpLFqMcbiS+DNgdf0YlJLc3Pf/ww8/XOU4CmvcuHFJZRo0hqb7/OWz7X4KFIf5/OpYcyeTUn1etBrKrTpwxhlnBAo+FGssUWKHa+bMmQX/m4hOmPFRE4huxYd0gQv/vJDbLyebcj2H9L/Wqj7ml1qhSvcVSz6fkSBJEqWmOWNVEfD6SOh7TivrR44cmfdzz5s3L+l2qlXS6SxdujRpjHGvr6qKogcgNEEsuih68cUXVykb5C3XbNmyZSLA4AUbvB/xThQPPPBAc/zxx5u40LJ4V9C6wsp0c0tOqDxFVTygAaBS6eKiR48etumy69VXXy2r5b7+FYdaxRE1t0+Usi1bt24d6HEqa1XumTmVSAkg7kWKVusElU9/hGIbMWJE0kW9Vh6lW7LtHqNqPhj0nMvtX1ZI/l5rWpIe9HzSf3FVTNtss03G5JpMlOWmvhzumOrPLtNEqjspqsnnoDWJg753/tegcSlMwC5IDwBlXyp44tHrDjOBEDdKYHNLnYUJIOixOg5cDzzwQM7bouPw0UcfTfq3TKsE89n2VMIcb/77+o/dVBPEQccSTRb7y1Blkus5jRLu3OCvvp84D6gcYY7Xzz//PGmlcKrjVfzJAGGCCuU6kazVmLl+N6Lw70clnevm+xnx9zcqV5onVtl+r4qA5pG1av+hhx7K63m//fbbpEBUmBVc3zqPjUsbgbIPQOgCTScjOlHWwJlqyYqauGiZ+OGHH57yOXTwKENUzbBuuukmEyeqXe1ylwRn4r+f/3niTFlG7o+OHQCoVB06dEgquZSun4G/LFPU9WHTUa1nd4WdSiV5pQNL3fT4mWeeiXQ78H+1pl3+VQLp6JislPdE56Y33nhj0r/997//TXth4R6jYZpoBj2vS/U5D/MZ95diCbqNYbcvav7MapW6CkorG9yLaP8EQarMvaD7RZnlL730UqD7+v9umNfw/PPP55S0pNfl1uivatRjUL0IPF999VWox/t7ezz++OOhJkZdmtxw/75K6WosKdS2++VzvKVa2ZDrZ0aJhmEaanuZqmH7BymTd+edd07cVoDxtddeC/x3UVpRH6/i/94OuipGgfuwq+6KRSumc91viJ7KbrrJC08//XTRrsOikOtnROeWxUqkiar6jsqjutUFrr32WluKMxcKgLpBym233TavXifbhnx8HJSkpbo+rDpZyFQ2STX9BgwYYDM/1ThE9Vl18qYu5lrmruZJqs8Zp67wXhRMta49OsD9pShSLeXxLyfSsiMAQOXRZITblMp/UZ5uEqBYpV3k6KOPTpqYU4ZJlJSI4FFwI0hmjk6edQGA6O2zzz5Jy80V7FeZkmxUJ/zHH38s+7dEQTT1W3EnvPR6VR4tyDGqSYsgGbcqZfLmm2+G2jb374QpZ+I+LmhWtT5r+WR+R0EreN2sVk3GaiIzG+1/r4ec59BDD13lfv6a4UGzzVUD3204nu01uBeVXp+fIPtfq3DCjMNu1rh6QFRa0/couSvl9FkJU8pIpdb22muvpONJTebDrgZ67733zC233JL0b0qY89cLj3Lb/TQx6c+yTOWLL75IakasMof+YHOuY4kCD3fccUfgbU71d8Kc0/gDPEpQDBP8QOkEHR/1WRw1alTitlYNHXDAASnvq7kUd45IJU6DyHVSslirj93vxhdeeIFVECWewN9vv/2Sxu27777bVAqN924pRyUwBGlMrpLB/ibM5U6r9vTZdq+tBw8evMo5YxA6J3UDTTvttFOox093yuvr/C3s4+Og7GfvtUJCJ9hqDnfVVVfZOq2HHXZYrJu8+C94VY4j00WPMvbcDFkNhiphBQAovcceeyzUCgEFnRcuXJg0mZWKv550kAu4qOh72e3JoNeoyeYwdAKbLtvaDcSrrEi25bLav7179w7VmwDhgmJKAPHoIqVnz562HEI6Tz311CoTceVGmY46xzr55JPN/PnzkyY2NJmb6VzTPUZ1/GVb6aEghc5hg1zgudzPvzKygzZ1dbdPhg8fnvXz2KdPn6T9UCr+bHQlIqn3SCY6F3bLL6n00v7777/K/ZSN3rhx48RtJTplm1DVfZRBF8Zxxx2XdPuSSy7JOKGqrLp+/fqFKoOgrHl3AkSreHSMhRkHdTzq9ZXD+56vvffeO6+SGNddd13SZ17jg8aGIJP53gRO9+7dk/a/smQV3Cz0trs0OaLrZrdvQ6pVDbqPW7ZL3+sa6/38JSK00jzTOKbn1PEetkyH/1wnzDmNslzd7VQpM6/paBDFTODAqrKNjzqmVaLb/f5TWW9/Y2Z35awbBFZZrilTpmTc9TofdQMc5UjnXW6QVLf9GdXZ6FykmNcLcaZ+OG4SgIKuYVYxlnrc0feTR2Xxs61uVhJNLpP25UCvVYkkbvLerbfeavtEhKF9kEtJQjdJwaP5Wn8/r6qg7AMQ5UgNu1L9+C8aVEc41f2yZbApmu8uKdRFl/pc+L9gdPKoC2d3YkYXzWpiCgAoD/fee6/NID///PNtCQ9/aRR3AkqTtrpftixe0SSaW8ZQj9UKQU1c6ALD/d6JukSSvmuUYeiuztDfPv300202W7rJCU0G6CJQ313aJwMHDkz7Peie1OuEV9kr/skEXYC9/PLLNvtRWSnaLjfDBdFOCru9IDRhedRRR5nLL7/cvP7663bFpjJqn3vuOTvh1rdvXzsRddBBBxX1bdDnyH/epfMobZtO/CdMmGBuuOEGW5Nd9WEfeeSRpEk4ZeHqIjJbVpIe61JdWV28+evwa+JEn81jjz3WXuCFTaBxt0MTimeeeaZdEaD97X+d/glNTbZ7xowZY1cWpxp/NGZo+95++237uUs3qRNEunPfbD/uObTGPPfCWOfNer/0uv1ji45DTdxrnPUHLVJNpvrfO40pWmWdasJX59kKRCkxSMdVmPeuc+fOq5zL6zVoH/spAKIAn1c2xr9KIxO9TrfPhY7vI4880k6CpJt81bipz8PQoUPt51OT5m7Qu1JpFYG771Lt60wUVFcgy/1e0yS6Et+0yi9dOUQFvjQ5qvfQ/Xyp7LDGmiCr9fPddo++/7T96s2ka8dUPRj0/xQgc68rdT6RLlCikmIqvejRd/gFF1yQ8rxCx5WCNl5yQZjPjH/M1fmBrnGVMarPjzte+AOxGre0r90yGxMnTrRBFWW/p1qhprFa+1njR7rzLBSejvtM46NWt2qMdlcPakWRVihlonHQpSSVVFUllOSpbGgF5LztKVf77rtv0gpkfTfq/FfJHplWa+l7UucAOn/QmB90RQgy0zmx9qlH5wk6Li+77LJVzsk8+q4dPXq06dKli7nrrrtKuov9nxGdT+h81X+epaC6zrH0Hadz2nL+jGSy44472lW+7orEYcOG2XPjoNwxSs2n3fKJ2eg78yunxKIC51VR6jNzZB38gxg0aJD98dNFlZYvpaOTKH2RaGDyvkx0sOoEWEvvdJKowUsnkP4TsGuuucZsscUWvIMAUEZUCkATs/rRGK/moZtsskniJEhjvbLJ/bWWld2aqaePJhG8uvW6wNYqhFQrEdRTKd1kf65Uc1mBBJ1oe8tRNRGtH01gKrNDkw96vXpdmnzVhaTbQHD99ddP+dz6HtOEgFdSSa9NF4gKQrRq1cruN01YauLHzSDSpLcyXDJlfiI3KpupQFC3bt0SDY71vqtWun5SadKkiT0v0aRosZqnq3RJ0PM0PzWiveKKK+x2BynZos+Al82kC7SLL77YZlMp81IZmN65mje5rolIZVmr2XxQCvJo9YJXTkR/z82gcrkTiioPoIlAtwyKLryeeOIJO6GorCtNluoxbpksPUYTfv4sr6DSnftmo0n+Xr16JfaTApqakFJTbG+M1H00Ebr11lvbiVb9P+1fd0wRBUL1/qSji2hNAHiT7ppY1liqVWUae3Ss6zjSc3tjm8YyjS+aaA5Cr0EXtQrseI3N9bk58cQT7fGlFSoKkGjizV25oXN/Zd4H3f96H/Ue6zV7Y6EmzbWaRZOxGodVTkzBWb3fmojS/4/jGKl+KQoueVnMWtmhiZQwY47GAO1P7T8vmKDPnoI1+tHxocC/voO0L/VepQpM6FhSOY5UvQ4Lte3ecapJJWV26rjSd7+ONa0u0PMpcOlvsqtApT6z7uS9S8epyh4rS92jcxmVgNH3sT6TWkGl73d3tYhejz6nY8eODbTtOi9SGSxvclSfT9XqTkWfrSOOOCLp37QCQv+uiWTvc6sAjAIrOifRuKFxUf9P44mCJV4pEX/5JxSPjiuNrUHGR9H4rHNPtyxkKjo+lGCg1TBeoEEZ6zrOdKzoefQdou8773gJe8yWwqWXXmqPX40R3rmHxiz9aNxRU3tleSsArdesz2XYUnIITmOjxj0FPEXjrM6z9LP55pvb41nvh75PdIwr4cALiKbrYVIsGm+Vwe/1j9Ixo8+idx6r7wQdOzoX8lb2aZs1l1nq4Ek+QSMFtnWu6wXRdW6s166EqkzfuTpveuuttxK3DznkkFB/WyUPVzjB8KpaMp8ARJnSyZwmUbxBzRvQ9CXp1g7z6MJCF0VVMYNDSzBdGjjcrD8AKCcayzWuZyvroGbUmoTLRBNpujhzJ3iLSZMbCoor48etWa2Jg2zL3SVTFo1XdkVL5z26mFKAIxVlt2kSUd+dKNy5iRIodAGc7n3w6KJGQSN/D5NyW6Gi8ydNVitbOFX980yUHa+AjNu8TxmJ3sSAS/tBpUQzBRRTUaBSwUNl6YYtMaYJe002ehfGqS6gXMqsVNkSTQKVmrLH1Q9BE0ZuxpjGmXT18TWhpHHAzUhMRZOQCqbpfm7Gut7HVI0YFSjVxXbQ8lceTfoqeKQJUDdQqknrVI1OlQ2nlTRuubMgNLGq1TdaAe32ZtHxEqRXiz4D+okDrTzxJvH1WdSKvLA1lrV6SNnCGuf8gSBNZnoTmukCTwoaahI8THP6qLbd6zmhz4gXGNbnJ11jawVKtbpDmaGZKDCm1+32iNGETbpgqI5lrUjQ8RyGAg76vOTaiFvnTfq8KoDkfuZ0TpJu3ENp6bhXIEDfp5nGR+97VEkN/hWI6YJ6Sug86aSTkibgNQGcquyeEhd0zqJJyHKm16Vgg16bstLdAHy67zC/Ss1gL0c679Cxq2QwJYC5E8ypAr7lRueXmox3v9eUMKYfPwUftGIgUyJ1JdCqBb0GjQ1eZRqVEtZ3moKb6VYt6rrH7S2k5PAw3H5mrVq1SltiOe4owVTGFP1XBF5ZTenqg+mLWBezOmFVphgAoLzoQkETgco6TtdQ2qOTnjZt2tjMSWWg6EIj24mvLkI0yaWsR51UKTMz29+JkiZIdFKllRD+uvOp6PtMWSPKJs1UN1+BZGWpqMZtuuxEvX5lrOp+mnRBcYIQWomiCy0FoJRNqKCCJrKU7aWMHk1S6cJY2bj+8pRhJ+WioM+VMrmUMamMcPUGUNkZfc40KaUJuLDBB9FEl86/FIRIlz2sz6L+njLu/Rm7YSbVVFJH44gyz/Q60v29VOODMkwzZYvqolL7QAGSIOViikWlhXQerAmhTKtSNE5qtZhWTGULPrjjlibtNVGaLuNNWdMKaqqcjFtPPAwdb2rargncdO+ZXpsmu/VdkW3MT0eZ9iNHjrS9KnSMZPsO0Piq7EdNECv7UYGuOND76pZEyDWTWVnEmqBQQFsTk9neF33vKoCnY0Wfo1zGuai2XceztkEBAH//Bo+OD03i6tjUd2gQCoJqgjbTZ1F/TxPEOg5zCWrp+0WfSyVfaPv0/aJ9GWZc0vGvcxIF5FQiI9tErL6zNKmG0tGqQJ1XpMsI1/u/++6728+EAhZB6ZxE39E659T3YSo6xpTwoqB0ruNvsWl/KMim5CONO9nKJmpM0GdTwT2NUQrsI9r3Q8kb2rc6X8uWaKOJZ53PaY6v1HRu+Oijj9oknHTnDRqXlWim78R8SnSWk+bNm9sVUu4qRZ2ne+VjU1FigkfjkcaXoBQE9VaaiPZ3VVVtZdhueCgJRbeVBap6csro0IW+PjD6olYmF/6PlkRrYkF0QlkVm7sAKE/KnFBmn7K7lHWhbGQtM9cEuyY8lMla6Sd3el3Tpk2zY7GyDnXhowkETSbqhE8TXWHLSniZvMokUjknTfaoXrcySLItw0dpqV6qm1GviVZN2MeNMumVsawl9vpcK/iiizadp7n1ZktFJSa0jF4llxQU0mdSnx2NOW5T5nKmzE6VVNF5sMZSjZUaVzRxm8/KGmWL671Txp/Ot7Vf9Lyqyx9lMFfHiD4PyvZV6ReNYZqI2G677SIvTaZjUGOm/pbGYb0uXTsoaKYyHfq7lTLZFpYmCTRZLjoudNGv154PHW9aga6ViyrboM+TAkp6D1WWScH3KIJ3hdh2lRpSdquOc51v6PpRAdew/Wg8yvDV51A/Ki/mBXe1D/QdX268/kR633T+oOCbPgfaVm1zuolplIZWJ+jY0vGqaSp9xlTu0O1DkguNg1rRpDFRn2dv/NV5ZKXTftIxrvJ6+kzqu0bHuQJsurbQsV7p1xaVRJn0+v5VeT6NO/r+1TmXrn8UZA5amq/YtMpTnxHNN+ocQuOkAs5aIVfVx0mNR0pY8YITmmMM08NBCU9eyeQGDRrYldLFTBYsJwQgEDsEIAAAKA/KKlQ9co8yrXIpKwIAQSg4oOx5rzeDVgCdcMIJFbHzKnnbAQCIo5tuuinR90IrW7UaK2jiiAJQWgnt9Vvr16+fLf9UVZXPOmsAAADEhjKF3CXLyvbRiTsAFIrGGbckn0rC+RuVl6tK3nYAAOJGK0PUk8yjHr1hVq2q1KEXfGjYsKE59thjTVVGAAIAAACRU7aQlnJ71LMq33IiAJCN6sSrNJJoNYH6g1SKSt52AADiRCu3vX52Ks2pa5kwJQvV38YNXqyWQ3+kOCEAAQAAgKwmTZpka5iqxnCQE2639JJ07dqVvQyg4NTrQOWLPEOHDrV1uStBJW87AABx8fvvvycCCOrzdNlll4V6/FNPPWV7MMm2225rjjzySFPV1Sz1BgAAAKD8qXGuGqmpLIiar6mRqRrqqZmpgg7qwaTGe8rYVcNg1xFHHGF22223km07gKqlTZs2pn///nZcktmzZ5vNN9/cVIJK3nYAAOJA371eH6bGjRuHLiOrUrRnn322/V19IKpXJ/+fJtSIHZpQAwAQvQkTJpjevXuHfpwCD2pGXbduXd4WAAAAAKhiWAGBitexY8ek22oKU6dOnZJtDwAAcaQAgsqDBG2Kqu/i448/3gYt1FwVAAAAAFD1EIAAYuLfr2aa5a9MMStmzjLmn2CTQwAABNXaGPPSPoebyb/8YD74bZ6Z8cdCM3fJYvPHP8vMPytXmDVr1jLr1FrNbLFWPbNz/Q3MgRttZtb7YYlZftFAw7cSAAAAgIpTq6ap3qyJqdm+janRolmpt6ZiUYIJsVMVSzAp+LBs2KPGLP+31JsCAAAAAAAAxEfNGqZ29+MIQuSILhhADGjlA8EHAAAAAAAAIOqJt3/N8lensFtzRAACiIEVs38s9SYAAAAAAAAAsbRiFnNvuSIAAcRA9caNSr0JAAAAAAAAQCxVb8LcW64IQAAxoGY4qkcHAAAAAAAAIMqJtxqmZrs27NIc1cz1gQDKR40WzWwzHNWjW/HNLGP+WV7qTQIAAAAAAAAqV62apvrmTWzwQXNvyA0BCCAmNBAyGAIAAAAAAAAoF5RgAgAAAAAAAAAAkWMFBCpex44dk25Xq1bN1KlTp2TbAwAAAAAAAABgBQQAAAAAAAAAACgAVkCg4o0bNy7p9vz580337t1Ltj0AAAAAAAAAAFZAAAAAAAAAAACAAqAJNQAAAAAAAAAAiBwBCAAAAAAAAAAAEDkCEAAAAAAAAAAAIHIEIAAAAAAAAAAAQOQIQAAAAAAAAAAAgMgRgAAAAAAAAAAAAJEjAAEAAAAAAAAAACJHAAIAAAAAAAAAAESuZvRPCRRXx44dk25Xq1bN1KlTh7cBAAAAAAAAAEqIFRAAAAAAAAAAACByrIBAxRs3blzS7fnz55vu3buXbHsAAAAAAAAAAKyAAAAAAAAAAAAABUAJJgAAAAAAAAAAEDkCEAAAAAAAAAAAIHIEIAAAAAAAAAAAQOQIQAAAAAAAAAAAgMgRgAAAAAAAAAAAAJEjAAEAAAAAAAAAACJHAAIAAAAAAAAAAESOAAQAAAAAAAAAAIgcAQgAAAAAAAAAABA5AhAAAAAAAAAAACByNaN/SqC4OnbsmHS7WrVqpk6dOvb3BQsW8HYAAAAAAAAAQED16tUzNWrUMFGotnLlypWRPBNQhgEIAAAAAAAAAEBww4YNM/Xr1zdRYAUEKt64ceOSbs+YMcP069evZNsDAAAAAAAAAGAFBGJo2bJlZtasWfb3tddeO9RyoRNPPNH+98EHH8xrG/J9nlweP2/ePHPUUUfZ30eNGmUaNmyY099GOFEdM+WonF9bqbat0H+3EM/PuIZyORbLQTm/rriOa4X4G4xrKIfjsFyU8+tiXCvN/uJatOoo589/XF9bKbeLa1Hm2Cq5BBMrIBA7tWvXNs2bN8/psV5FsnyXGOX7PLk8/p9//rE/su6660a2TArFOWbKUTm/tlJtW6H/biGen3EN5XIsloNyfl1xHdcK8TcY11AOx2G5KOfXxbhWmv3FtWjVUc6f/7i+tlJuF9eizLFVsuql3gAAAAAAAAAAABA/BCAAAAAAAAAAAAABCAAAAAAAAAAAUP6qrfSKiAGoaD/99JNp27at/X3SpElmww03LPUmAUBeGNcAxA3jGoA4YmwDEDeMa9GiBBMAAAAAAAAAAIgcAQgAAAAAAAAAABA5AhAAAAAAAAAAACBy9IAAAAAAAAAAAACRYwUEAAAAAAAAAACIHAEIAAAAAAAAAAAQOQIQAAAAAAAAAAAgcjWjf0oAAIDKsWjRIvPkk0+al156yXz//fdm4cKFZp111jENGzY0O+64o2nXrp3Zc889S72ZAAAAVc4///xjnnnmGTNhwgTz+eef2/O0GjVqmAYNGphWrVqZI444wuyxxx6l3kwACGzZsmVm7NixZuLEiXZc++OPP+z15xZbbGEOOeQQc/jhh9txLk5oQg0gqy233DLQXtp4443NK6+8wh4FUDEUdLj88svN/Pnz095nq622Mk8//XRRtwsAwujWrZt59913Qz1mwIABduIOAMrV3LlzzRlnnGG++uqrjPc7+OCDzaBBg0zt2rWLtm0AkIvvvvvO9OzZ08yYMSPtfRRcHTZsmFlvvfVis5NZAQEAAKqkZ5991vTt29f8+++/pn79+uaYY44xrVu3Nuuuu67566+/zMyZM82rr76aMTgBAJWqWbNmpd4EAEhr+fLlScEHZQafdNJJduz6+++/zSeffGLuu+8+uyJi/Pjxpl69euaKK65gjwIoW7/99ps5+eSTzY8//mhv77///na1w0YbbWR+/fVXO5ZpZcS0adPs+PfYY4/FJrDKCggAgVdAHHvssaZr165p71erVi3TtGlT9iiAsqfgQufOne0F7K677mruuOMOU7du3bRLZONy4gcgnmbPnm2WLFmS8T4KpmryTjbbbDO77B8AypVKLvXu3TuRDayJuJo1k3No58yZY8/nVL6kevXqZvLkyTapBADK0dVXX20eeeQR+3uPHj3Mueeeu8p9RowYkQim9uvXL3HuVulYAQEgMJ3MtWjRgj0GoOJdc801Nvig+sFDhw5NG3wQgg8Ayl3jxo2z3mf48OGJ35VtBwDl7MMPP0z83r1791WCD7LJJpvYUnIPPvigWbFihc0abt++fZG3FACy+/fff20/G2nUqJHp1atXyvsp8Vf9CbXK65577jEnnHCCDbBWusp/BQAAACFXP7z55pv29+OPP96svfba7D8Asacl/aKL2MMOO6zUmwMAWZtPBwmybrrppikfAwDl1vth0aJF9vc99tgjY5Pptm3b2v+qLNP7779v4oAVEEBM62UqY+SHH34w8+bNs5m9G264odl+++1j1cQGQNUR5bim2poeN0vuzz//tCd5a621Fsv3AcTqfO2LL74wX375pf19t912s7WGAaCcxzW3tK/KzKkHRCqzZs1K+RgAKKdxbeHChYnfs5WKc///e++9Z3bZZRdT6QhAAEWg5aDffPON+fjjj+0yKv3oItDN0BgwYIBdPpoP1f5VHfMxY8bYSbRUPRr22msvW2fO6+sAAFVtXNPyfO+xamSo1RAqw/TBBx8k7qPSTAcffLCtzUngFqgaKnlcC7r6QVQvHUDVUMnjWseOHc3NN99sE0TuvvtumxHszxhWI1f9Tdl5550pFwxUAZU6rq255pqJ3zWuZaK+Np4ZM2aYOCAAARS4cdajjz5qpk+fbv7666+C7uuvv/7anHPOOba0SDoakF955RU72aZmNqotF/b16EeR32rVqtmorBqCderUyeyzzz4RvAoA5S4O45p3EqeVDmoCNnDgQLNy5cqk+/zyyy/moYcesk1aVXuToC0QX3EY17Jl7j377LOJi98DDjggr+cDUP7iMK4pAWTQoEHm/PPPt9nH6l2jZqxa5bB06VL72u677z5b0qRJkybm2muvjfiVASgnlT6uNWnSxAYt9Lh3330343216sEzd+5cEwcEIIACmjp1ataBJQpaBnbqqaean3/+Oenft9lmG1svU0u9FBVevHix/Xc1Xr3iiivs0jEFD4LyR1416Gs57HPPPWd23XVXc9NNN5n1118/olcFoBzFYVz7/fffE5klCj7UqVPH9O7d23To0MHUq1fPfPvttzbooAk7/f2ePXuap556KmOjagCVKw7jWiZvvPGGmT9/vv39wAMPtGMegHiLy7i277772uzjBx54wDzxxBN2ks+l51H2cdeuXc0666wT8asDUE4qfVxbY401TJs2bczrr79uvvrqK/P000+n7Mk1ZcoUM3ny5MRt7+9UOgIQQAko61aDj39Ay4WydhWZdZ+rRYsW5oYbbjBbbbVV4t+UGXLLLbfYbF/PJZdcYu+Trp6mRxeq7dq1s4OlypUoe06DrjJRRo4caf/2O++8Y04++WQzYsQIJumAKqiSxjUtpxVln2g1l8ovqRGYR6sdBg8ebFZbbTUzevRoG2jVWHfaaafl/doAVI5KGtcyUQDVk285AgCVrdLGNZ2raZLu5ZdfXmW1qlfGRAkjDRs2NEceeWTerwlA5amkca1Xr17mrbfesqtTL774YtvDRqu71FNCySLqVajnrVmzZqKklFZ8xUH1Um8AEHerr7662WGHHUy3bt3sElINKFpOddRRR0Xy/C+88IINBHg22WQTOwi6g6Osvfba5rLLLrPb4VGkVoNbNorQDhkyxPz3v/81O+20k2nZsqUNRpx11lnm+eeft7+Lori33357JK8LQPmq9HFNgQWP6gm7wQeXlvxrmaxorAMQX5U+rmVa8aXyAN7f1HkcgKqh0sc1rbZXgtuwYcPMggUL7O9aea/MY/Xt0t9SGWDVgtdEHiWYgPir9HFtu+22s2OVrjEVhFAinFZ6aXXF3nvvbXtX6Hn69++fsndEJWMFBFBAal7at29fG70sFA1Yrssvvzzj8lNNqOlCVH0c5MUXXzSff/65DSqko8E1HS0z0yCresJaFaEs4T59+pjatWvn9HoAlLc4jGs6ifPqhupEL1Pt4W233daehH7xxRc2C8ULSACIjziMa+koeLps2bJE82mt+gIQf3EY12677bZEHfSrrrrKdOnSJfH/dK2pptP60fMqMKHeXUqMa9++fUSvEEA5icO45p2PKaBx55132jKZXoml6tWrm912282Wldt4441NkPm4SsIKCKCANHlVyMHxyy+/tKsOPJtvvrnN5s1WTumYY45J+jevMWGuNCAffPDB9ndN6n366ad5PR+A8hWHcW2jjTZK/K7lrpk0atTI/vfff/9N9I4AEC9xGNeylV9S4EEXvACqhkof11QGRb0fZLPNNksKPvhdcMEFid+ffPLJwK8BQGWp9HHNpQCEEnkVZJ00aZJdeaGVXcOHDzetWrUy3333XeK+uZbgLDcEIIAK9uqrrybdDtqg0H8/b2l+PtxB8aeffsr7+QBUTcUY15o3b574fcWKFRmfV4EHj7JSAKBSzte+/fZb89FHH9nfVXpJjRMBoBLGtV9//dWurpett94643MqsaR+/fr295kzZwbaDgAoh/O1GjVq2IS4TTfd1AYzPNOnT08q2xQHXEkDFezNN99Muh20rq9O0twlXbpAnTt3buTbBwDlOK5pub5Hjb8yUQNqr95ovXr1Am0LAJTD+ZrbfJrVDwAqaVzTpJxHddKz8e5TyOxoAPFWTvNrEyZMsP9VUCIuZeUIQAAVbMaMGUmZuapVHpSWdbm+/vrryLalYcOGeT0XgKqrGOOaGn15vRy03DVT8EE1PGXHHXdkBQSAijlfU/mSZ555JnHxetBBBwX+mwBQ6nFt3XXXNWuttZb9XSu5MgUhVDbFK5OphrEAUMnza5MmTUo0wj700EMTY2GlIwABVCidZP3222+J21p26i7ZysZ/cqYoba4WLVpkmxyKtuE///lPzs8FoOoq1rimvjVerU5d1I4YMWKV+6jhdP/+/RMlmvy1PQGgnM/X3n77bfPjjz/a3/fbbz9Tt25d3jAAFTOuqW+NV3t93rx5qzSG9SxdutRcc801idtxyRQGEN/ztZ9//jnt//vss8/MRRddlOh50adPHxMXrE8DKpS/bIjbVDUIf+PVdGVIVL9u7733Truc9c8//zS9e/dO1OhUg7DatWuH2hYAKOa4Jj179rR1PufMmWOuvPJKW2ezY8eONjihpl9qAPbJJ5/Y+7Zr184ccMABvEkAynpccz399NOJ34844ohQfxMAymFcO/vss+216F9//WXuvPNO8+mnn5rDDz/c1krXigjdfuihhxITfepJSLk5AOV+vnbIIYeY1q1bm3322ceOWyr1q6DEa6+9ZsaMGWMT4RT8GDJkiF0NFhcEIIAKpYl/l6KjYfgHsj/++CPl/ZRRsmzZMjv5tv3229vIrgZDRYinTp1qHn/88UQEt1mzZqZXr16hXwsAFHNc8+577733mh49etgL19GjR9sfP2XSDR482GbiAUA5j2seTdZNnDgxcUG82267hfqbAFAO41rTpk1t4OG8884z8+fPN6+//rr9SWWbbbYxt99+e6LEJgCU6/na8uXLbSKcv+m1R3Nu119/feAeFJWCAARQoRYvXpx0e7XVVgv1eEVZ/Rer6fzyyy/m0UcftT/p6OJ20KBBNnsYAMp9XPMubNWkdeTIkbbRlwIR2gadQKqOp7KGWcoPoJLGNa+3jXe/ww47jP41ACp2XNM15vjx482TTz5p66KrrrrK/6o+u0qkbL311rbHzcEHH0wDagAVMa5de+21ZvLkyXa1vUrMedefzZs3t4m/Wunlf744IAABVKglS5Yk3Q5b9sg/oPqfzzNw4EDz7rvvmo8//tguI1uwYIGNDmsVxAYbbGAn6Tp16mTatGmTw6sAgOKPay6d3J100kn2BwDiMK6NHTs28TvlSABU+rimBLdTTjnF/gBApY9rHTp0sD9VDQEIICbClgfx33/lypUp77fLLrvYHwCIy7gGAHEe1x588MHQ2wUAueJ8DUDcMK5Fr3oBnhNAEWgFguvvv/8O9filS5cm3V5jjTUi2S4AyBXjGoC4YVwDEDeMawDihnGt8AhAABXKHzAIG4Dw358ABIBSY1wDEDeMawDihnENQNwwrhUeAQigQtWtWzfptnozhPHbb78l3V5rrbUi2S4AyBXjGoC4YVwDEDeMawDihnGt8AhAABWqSZMmSbfnzp0b6vE//fRT0u3GjRtHsl0AkCvGNQBxw7gGIG4Y1wDEDeNa4RGAACpUvXr1zHrrrZe4/euvv5olS5YEfvycOXOSbjdr1izS7QOAsBjXAMQN4xqAuGFcAxA3jGuFRwACqGDNmzdP/L5ixQozffr0wI+dNm1a2ucCgFJhXAMQN4xrAOKGcQ1A3DCuFRYBCKCC7b777km333///UCPU7mmH374IXG7adOmplGjRpFvHwCExbgGIG4Y1wDEDeMagLhhXCssAhBABWvfvn3S7WeffTbQ4/z38z8PAJQK4xqAuGFcAxA3jGsA4oZxrbAIQAAVbMsttzQtWrRI3P7mm2/MpEmTMj5m6dKlZuTIkUn/1qlTp4JtIwCEwbgGIG4Y1wDEDeMagLhhXCssAhBAhTv77LOTbl999dXm999/T3v/G2+8Man80n777WdatmxZ0G0EgDAY1wDEDeMagLhhXAMQN4xrhVOzgM8NwBgzZ86clPth0aJFSbcXLFiQ8r6rrbaaadCgQdp9ecABB5gddtjBfPjhh/b27NmzzfHHH28GDx5sI7ieP/74w9x8883mkUceSXruc889l/cJQCiMawDihnENQNwwrgGIG8a1ylVt5cqVK0u9EUCcuUGAXOyyyy7m4Ycfznifn3/+2XTp0sXMmzcv8W/VqlUz22yzjWncuLFZuHCh+fjjj83ixYuTHnfDDTeYQw89NK/tA1D1MK4BiBvGNQBxw7gGIG4Y1yoXJZiAGNhggw3MfffdZ5o2bZr4N8UWp0+fbsaPH2+mTJmSFHzQyof+/fsTfABQthjXAMQN4xqAuGFcAxA3jGuFQQACiAk1ox47dqw5/fTTTf369VPep1atWqZ9+/Zm1KhRpmvXrkXfRgAIg3ENQNwwrgGIG8Y1AHHDuBY9SjABMbR8+XLzwQcf2Pp4v/76q1lzzTXNhhtuaHtFrLfeeqXePAAIjXENQNwwrgGIG8Y1AHHDuBYNAhAAAAAAAAAAACBylGACAAAAAAAAAACRIwABAAAAAAAAAAAiRwACAAAAAAAAAABEjgAEAAAAAAAAAACIHAEIAAAAAAAAAAAQOQIQAAAAAAAAAAAgcgQgAAAAAAAAAABA5AhAAAAAAAAAAACAyBGAAAAAAAAAAAAAkSMAAQAAAAAAAAAAIkcAAgAAAAAAAAAARI4ABAAAAAAAAAAAiBwBCAAAAAAAAAAAEDkCEAAAAAAAAAAAIHIEIAAAAAAAAAAAQOQIQAAAAAAAAAAAgMgRgAAAAAAAAAAAAJEjAAEAAAAAAAAAACJHAAIAAAAAAAAAAESOAAQAAAAAAAAAAIgcAQgAAACgxLp162a23HJL+9O+fftSbw6Q0pw5cxLHqX5uu+029hQAAAAyIgABAAAAAECOxowZkxSY0U/Lli3NV199lVeA55xzzuE9AQAAFY8ABAAAAICEd955J2kSVJOryJ9WC7j7VZPNiK8VK1aYW2+9tdSbAQAAUHIEIAAAAAAAiNiLL75oPv30U/YrAACo0ghAAAAAAABQADfffDP7FQAAVGk1S70BAAAAAIDyt8kmm5gvv/yy1JtRUV5//XUzdepU07p161JvCgAAQEmwAgIAAAAAgIi0a9cu6fYtt9zCvgUAAFUWAQgAAAAAACKy5557mp133jmpsfuUKVPYvwAAoEqiBBMAAABQIH/++aedfJw7d65ZsmSJadiwoWnatKnZbrvtIvsbs2fPNjNmzDA//vij/Xs1atQw66yzji2Xs/3225s6deqYUvrtt9/MV199Zb7//nuzaNEis2LFCrP22mvbfbHDDjuY9dZbL6/n//nnn81nn32WeP2y+uqrmwYNGpjGjRubLbfc0tSuXTun5/7333/NJ598Yr777juzYMECs3z5clO/fn3TrFkz+x5Wr145+VyF3E/lYObMmeaLL74w8+fPN3/99ZdZd911TaNGjcxOO+1kX2exnXvuuea4445L6gXRpk2bom8HAABAqRGAAAAAACK2cOFCM3jwYPPss8+apUuXrvL/N9tsM3PiiSeaY4891lSrVi3Uc//zzz/mzTffNOPHjzdvvfWWmTdvXtr71qpVy5aD6d69u9lmm20yPm/79u3NDz/8sMq/9+vXz/6ksssuu5iHH354lX/XpP24cePMG2+8YYMjmWgi/4wzzjD777+/CeOll14y9957r/nwww8z3k/7QIGYDh06mK5duwaerL/99tvNxIkT7XuZiia4jz76aHP66aebunXrprzPnDlzzL777pvy/6X7dzn77LNNr169TBSi3E/+15NuO2+77TYzdOjQvLb78MMPNwMHDsx4H322HnzwQfPEE0/YbUtltdVWs8dWnz59bFCuWBT40EqIyZMn29sfffSRefXVV1cpzwQAABB3lZOyAwAAAFQAZWF37NjRjBo1KmXwQZRRf+WVV5qePXuaZcuWhXr+O+64w5x55pnmqaeeyhh88IIVL7zwgjnqqKPM/fffb4pBpWa6dOlihg8fnjX4IB9//LGdyD7//PPT7i//qoS+ffvafZdtUt3bB++995656aabAm3/iBEjzAEHHGAef/zxtMEH0YqIYcOG2fe6HBszF3o/ldoHH3xgAwva3nTBB/n777/Nc889Zw4++GD732Lq3bv3Kr0gVq5cWdRtAAAAKDVWQAAAAAAR+fbbb83JJ59syw65mjRpYrbYYgubZT5r1ixbCkdefvllc/XVV4f6Gyph5FpzzTVNixYtbCmjNdZYw07iq9zRN998YyehRf+9/vrrbTkmrbooJP8Eq17z5ptvbjbccEO7UkAT3QqcKFCjslQeTQ7rsdkmwG+99VYbfHHpdbds2dKWE9LfW7x4sf0b2gfu38hmyJAhNqjg334990YbbWTLW6mE0fTp0205Jvnpp59sqZ3HHnvMvg/lopD7qdReeeUVW+JIwQWXSmNpdZFep0oxTZs2zZZjEgX6LrjgAvu+de7cuSjbqdU9WjGiz7l8/vnnZsKECTYYAgAAUFUQgAAAAAAioMCAShW5wQcFHq666qpVar8rUNG/f3/bH0LlY9QTIQyVkjnyyCNt2STV7k9VxumXX36x5Wm0EsGbLB8wYIBp27atrY3vpwl03U+Ttuedd17i3y+66CJz4IEHpi1vk8paa61lJ3k1+apSNJrs9tOE9zPPPGMDDt5KA5VtUlZ7ugla3e++++5L3NZEs/a5/laq/gUKvCj7/8UXX7Q/mehvu8EH9Q0466yzbDkivR7Xr7/+arf7ySeftLf/+OMPu4Jj9OjRSftEQRdv8lnvxUMPPZT4f48++qj9/6mEPR6KuZ+CUHkxlVAKSkGpc845x/YK8ahHSLrVQ9rXbvBBn4UePXrYXhYuBR0eeeQR+17pbyjApZVHrVq1sr1YirUKQgETLzCn8lT6PFVS/xAAAIB8EIAAAAAAIvD0008nlbrRZKjK+ay//vqr3FeTnyqJpAnuSZMm2ebMQWkFgyY1s01gKstdGd//+c9/7OSuaNJWE98XXnjhKvf3JsP9fSDU6yBM7fytt97aviatzMhEqzHUQ0F9JI455phEEEL7JV0AQj0vNJHsueKKK8xhhx2W9m9oxYICIPpxgyp+ChpdfvnlidsKOChgkK5vht7T6667zq6K8HodaPJcAQi38XDNmjUT+84fVND+LlRPgkLtp6D0WoMGUTQxr+PUDT7suuuutixXKior5a1qkGuuucaWGEtFwZZTTjnFbLXVVua0006zgRY9VitdtEKkGBQg1PH8/PPP29tabaLAW7FWYQAAAJQaaRcAAABABJRp7br22mtTBh/cyWk12Q2b7a6J6zDZ08q2Vk8Dj5pXF1K9evWyBh/8wRh3slk9IdLV9Ff5I1eYxtXpVmt4792ff/6ZuK1VK9madou2WwEej7vCoZQKtZ8K4YYbbkjqzaAyVmoAnmqlxttvv22bObvBuHTBB9fuu+9uV2W4jbn9+6iQ1KhbQR6PXp+3KgkAACDuCEAAAAAAeVJJJfUF8LRu3dpmcWejvg2F7skgKoXk0QoHlRAqJ+72eUGIIPy9NnKhDHw1nHYnwDt06BDosSp9dfzxxyeVB9KxUG6i2E+FoNU4bqkoBdfuueeeVUpeebSiyA3gpVslkUq3bt0Sv2slxBtvvGGKRb0p3BUo6gPjle8CAACIOwIQAAAAQJ7crGwJ02Q26GR3kB4U6kWgpshaQeD++FdMzJw50xSbJvpV/kZNj/3b588GT7d9/rr9N954Y6LRdq5mzJiRFJBJ1+8iHZUtcn3wwQem1Aqxn6KmXhMqn+RR0OHuu+9O2xdD3n333cTvO+64Y8YVRn7qe7LxxhuX7H3q2bNnUi+UO++80/aoAAAAiDt6QAAAAAB5+vTTT5Nuu2V5stliiy1s2Ru3qW4Qmrx87bXXzMSJE81nn31mvv/++8CTzGF6TuRj6tSptva9AjSa6F+6dGmgx/3+++8p/13NvNWTYsGCBfa2nvuLL76wvST222+/nHoq+Cei69evn7YEVCr+SWR/D41SKMR+ipJ6pajvg4Jmool59dJQv4R0tLLEXcmhYEKY90lU7sx7f4r9Pmmfd+nSJbGKY+7cuWbkyJHmhBNOKOp2AAAAFBsBCAAAACBP8+fPT7rdpEmTwI9VbXhNTqo5bVAKPKhPQa6TqG6/g0LQa1FT5/fffz/S7VtjjTXs86pRslZUeKslBgwYYH80Ka3yV1qVoObW/pUAqWjFiEsNm/ORLnhSTIXYT1FRIKF79+6JYJTKWGmbdtttt1Dv09ixY+1PJb1PPXr0MGPGjEkEG++66y7bw0IN2QEAAOKKAAQAAACQJ/+Kgrp164Z6fLqa96mMHj3aXHrppYmJ5Vx4meeFWg1y8skn5zXBm+m1qWTV6quvbq688spVJqUVkNHPM888Y283b97cZp0fc8wxaSd5o56IVpmpchD1fooqUHf66aebhQsXJv7t/PPPN506dcr62Kjfp8WLF5ti22CDDWzPlwceeMDeVumvhx9+2JxxxhlF3xYAAIBiIQABAAAAVAg1OVaGvjtBrxJOmsDdbrvtbGa7GlvXrl3b/njeeeedopR6UTkiZd27k8UqaaQGvMq233TTTU2DBg3sxLjKTrkyld/xa9++vdljjz3M+PHjbXmh9957L+XEv8o+DRw40Dz44INmyJAhZocddljlPv7+E/nKJzAUtSj3U770dzXRPnv27MS/HXfccTYgEUTU71OpnHnmmeaJJ55IvA9qwt21a9fQQUsAAIBKQQACAAAAyJNqy/tLCCkQEJSaRwdxzz33mH/++Sdx+5RTTjEXXXSRLWNTDtnemuhWkMSj8j5qtpttcjWXklAKYHTu3Nn+aHL6888/t/0c1Kj4rbfeSppoV7390047zYwaNco0a9Ys6XnWWWedpNtqhNy2bVsTF1Htp3zo75577rlm+vTpiX9TLwqt5AnK/z5pOy+88EJTaTQuKBg4bNgwe1urQYYPH2569epV6k0DAAAoiOqFeVoAAACg6lCWv2vWrFmBH6vG0UGb6U6aNCnx+2abbWYnYLMFH7xSL8Xgbl/16tXN9ddfHyizO9/tq1mzpm38feKJJ5rbb7/drvgYNGiQ2WijjZKCHLfeemvW985r3BxH+eynfGjVjntsaIXFjTfeaI+RoOL0Pilw6JZdU0kmtywVAABAnBCAAAAAAPK0zTbbJN3+5JNPAj/266+/TjSlzUSZ6r/88kvitkrrBJ3A/eijjwJvT5CARjrff/994vfNN9/cNGrUKPLtC0Llp1T2SZnlbk8DTYIr4ONq1apV0u2PP/7YFEI++7VQwuynXA0dOtSuqHADZ3fccYctwxWGSo2puXah36di0GoO9Ulxgz733ntvSbcJAACgUAhAAAAAAHnafvvtVylFFJRq8+dSpilozfglS5aYl156KfD2uL0jxC35lI1bSilMTftnn33WFELTpk2T3hsFcfyZ5loR4G7ryy+/bHtZRC2f/VpoQfZTLp588klz2223Ja1i0ER7mPJknlq1apmdd945KXCnn0qlVSjrrrtu4vYjjzxim3QDAADEDQEIAAAAIIIJ3G233TZxe+rUqba8TTa//fabGTFiRE59JtxeC5moya3bFDobtzSMuKsuwjxWqyFWrFiR9THqRfDmm2+aQvEHQjSR7S9L1KVLl8Ttn376yTz00EMF345ilcWKaj+F9cYbb5jLL788cVurF+666y7TuHHjnJ/z6KOPTro9ePDgsmr6HXZ/uw24FSj0+kIAAADECQEIAAAAIALHHXdc0u1LLrkk4ySzGvP269fPLFq0KNDzq0SOO3n76quvZg1C6D5qAh2G/oYm5T1BAimeFi1aJAVXnnnmmYz3V5BCfSyCTiJPmDDBzJgxI/D2aP9PmTIlcXv99ddfJZDjNTR2SxDddNNNgVemePQ+Tpw4MWOQyhVmv4ZVqP0U1GeffWbOOecce4yLjqebb77ZrjbJx7777ptU7uy1114z1113XahyUdomrbjxtq3UY0aDBg0St1955ZWSbg8AAEAhEIAAAAAAItC5c2fbXNcze/Zs07VrV/P222+vcl8FDk499VQ7gSpBJ3sPPPDAxO8qE6Rmtu+//37Kck2a8D377LPtRKtb6iVIqaDtttsuaYWCgimaoNZ2q2G29+NfHeFun/Tv39+W4fFPEKv80FNPPWWOPfZYu+Ig6PZpfx1yyCHmpJNOMk888YSZN29e2vtqv6jMjVsWqlOnTinvq0ngq6++OnFb29unTx9zwQUXmC+++CLt31CpIgV5FEhq27atXW2Sjibf3b4H99xzjw0Oqf+Fmpa7+zVoUKrY+ymoM844w+4bN8CjniDua8z2owBWKjfccEPSag2tVtFE/uuvv542EKHPgPazHqsght7XqHpc5EPHQ48ePUq9GQAAAAX1f6lNAAAAAHKmhtADBgywk+oLFixIZPhrcrdJkyZ2dYAywRWY+PTTTxOPU/kfTUBroj8bBS1Gjx6dqM//ww8/2MnXZs2a2Sa9NWrUMD///LNt0Ov1GNDkft++fc3//ve/wK+lW7du5oMPPkjc1t/Uj98uu+xiHn744cRtTcKrTv97771nby9dutRcfPHFZsiQIbZE1Zprrmm3XdvnTbJrvymLPehErFZLKBjiZexvsMEG9vWrsa/KBqnc1Jdffmn3g2vjjTc2PXv2TPu8mnTXxPctt9ySWJGhTHn9KECx5ZZbmnr16tmJawV4dF+9b0HKTIkmzRWkGjlyZGLfKEikHz8Fjnr16mXyUaj9FIQ/MKXSQmHLCx1++OFm4MCBq/y7Ahl6j7R/vCDHhx9+aMsZ6fjSCgn1mNBnTUEVBV+0GqQQfT2icNRRR9nAlT7LAAAAcUQAAgAAAIiIyuwMHz7cZny75Zc0Ua0fv3bt2tlVAgosBKGJ1dtvv92ceeaZSRnrM2fOtD9+KqWjuvuLFy8O9To6dOhgpk2bZh544AETlibUFcBwt0cT0lop4KeJ8Kuuusq0b9/e5EoT6P5JdD8Ff7Qf/P0t/BQE0ST9pZdemrQKQdsfpBdGtpUsF110kfnmm28SAZpiinI/ldqee+5pV3b07t3b7k+PjvMggTwFg6pVq2bKgVYc6bjTMQcAABBHlGACAAAAItSyZUszbtw4u7LBLbnj0ooITTjecccddgIyjJ122smWNVLwIt0kqjL1tTJC2ftuc+wwVFZo1KhR9nlUPkjPGaQxsYIeepyCEOlev55n//33t6sqjjjiiMDbpLJIl112mdlrr71WaZqcbkJd9x87dqxp1KhRoL+hMlIKlmhyO0jDZD2v3uv777/f3H333Rnvqwx9lQwaOnSoLZHUvHlzO9nv9tyIQjH2U6lpxY+O70GDBtnjUytpsgWHdMxdf/31ZvLkyaE/d4Wk1R6bbbZZqTcDAACgIKqtDNrxDQAAAEAoWqWgHhBz5841S5YsMQ0bNrSrJNRjIYoMbGW0T5061fZRUJ17lQraaKONTOvWrQMFC4rx+rV9KkWlcjkqB6VSQOqVoVJA+VDpI62yUF8KvX5vlYcm+TfccEMbCFI5oXypNM4nn3xiexJoVYSCBZrU13MrgKD9Xc6KtZ9KTSWlVIpJK1VUAk2XuXqf9JnTqhZN8KtEGQAAAIqLAAQAAAAAAAAAAIgcJZgAAAAAAAAAAEDkCEAAAAAAAAAAAIDIEYAAAAAAAAAAAACRIwABAAAAAAAAAAAiRwACAAAAAAAAAABEjgAEAAAAAAAAAACIHAEIAAAAAAAAAAAQOQIQAAAAAAAAAAAgcgQgAAAAAAAAAABA5AhAAAAAAAAAAACAyBGAAAAAAAAAAAAAkSMAAQAAAAAAAAAAIkcAAgAAAAAAAAAARI4ABAAAAAAAAAAAiBwBCAAAAAAAAAAAEDkCEAAAAAAAAAAAIHIEIAAAAAAAAAAAQOQIQAAAAAAAAAAAgMgRgAAAAAAAAAAAAJEjAAEAAAAAAAAAACJHAAIAAAAAAAAAAESOAAQAAAAAAAAAAIgcAQgAAAAAAAAAABA5AhAAAAAAAAAAACByBCAAAAAAAAAAAEDkCEAAAAAAAAAAAAATtf8HVATGDT7TsfcAAAAASUVORK5CYII=",
+      "text/plain": [
+       "<Figure size 1600x1000 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "ncols = 4  # 3 features + observed\n",
+    "n_grid = np.logspace(5, 9, 50)\n",
+    "inram_gb = n_grid * ncols * 8 / 1e9  # whole dataset resident (array lower bound)\n",
+    "stream_rows = batch_size + 15_000 + 5_000  # one batch + shuffle buffer + one shard\n",
+    "stream_gb = np.full_like(n_grid, stream_rows * ncols * 8 / 1e9)\n",
+    "\n",
+    "fig, ax = plt.subplots(figsize=(8, 5), layout=\"tight\")\n",
+    "ax.loglog(n_grid, inram_gb, lw=2.5, label=\"in-RAM pm.Minibatch  (O(N))\")\n",
+    "ax.loglog(n_grid, stream_gb, lw=2.5, label=\"streaming DataLoader  (O(batch + buffer + chunk))\")\n",
+    "ax.axhline(26, color=\"0.5\", ls=\"--\", lw=1)\n",
+    "ax.text(n_grid[-1], 30, \"26 GB RAM\", color=\"0.5\", ha=\"right\", va=\"bottom\")\n",
+    "ax.set_xlabel(\"dataset size N\")\n",
+    "ax.set_ylabel(\"array footprint (GB, lower bound)\")\n",
+    "ax.set_title(\"Memory is flat in N when streaming\")\n",
+    "ax.legend(loc=\"lower right\", framealpha=0.95);"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d304c6c7",
+   "metadata": {},
+   "source": [
+    "Those lines are only the bare arrays. Actual peak RSS is higher, because of the\n",
+    "framework and PyTensor's resident copy, and it hits the RAM ceiling sooner. As a\n",
+    "real-data check, outside this notebook, we ran the same logistic model (13 numeric\n",
+    "features plus the click label) on the\n",
+    "[Criteo 1TB Click Logs](https://huggingface.co/datasets/criteo/CriteoClickLogs), a\n",
+    "standard, publicly available out-of-core learning benchmark. Peak memory for the\n",
+    "streaming `DataLoader` stayed flat at about 0.7 GB across a sweep from 1M to 150M\n",
+    "rows, while the in-RAM `pm.Minibatch` baseline rose linearly to 15.7 GB at 150M\n",
+    "rows, which extrapolates to out-of-memory around 240M rows on the same 26 GB\n",
+    "machine. On a 1M-row slice, where the in-RAM fit is cheap, the two posteriors\n",
+    "agreed on the large coefficients (the intercept near -3.6 and the strong slopes),\n",
+    "but the weakest slopes were noisier: the largest gap in posterior means was 0.12,\n",
+    "and two slopes the in-RAM fit placed near zero came out near +0.08 under streaming,\n",
+    "flipping sign. On coefficients near the noise floor the two stochastic fits are not\n",
+    "interchangeable at this slice size.\n",
+    "\n",
+    "## When to use it\n",
+    "\n",
+    "* Use `pm.Minibatch` when the data fits in RAM: it is simpler and its random\n",
+    "  index gives perfectly i.i.d. minibatches for free.\n",
+    "* Use the streaming `DataLoader` and `Trainer` when it does not: it keeps memory\n",
+    "  flat in `N` by streaming from disk, with no callbacks to wire up. The one thing to\n",
+    "  watch is shuffling. Pass `shuffle=True`, or pre-shuffle on disk and interleave\n",
+    "  shards for strongly ordered data, since a bounded buffer over strongly ordered\n",
+    "  data only block-shuffles it and can bias the posterior."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "918da33a",
+   "metadata": {},
+   "source": [
+    "## Authors\n",
+    "\n",
+    "* Authored by [Yicheng (Ethan) Yang](https://github.com/YichengYang-Ethan) in June 2026\n",
+    "  for the Google Summer of Code project Streaming Variational Inference for Large\n",
+    "  Datasets (PyMC / NumFOCUS)."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1a8d9f52",
+   "metadata": {},
+   "source": [
+    "## References\n",
+    "\n",
+    ":::{bibliography}\n",
+    ":filter: docname in docnames\n",
+    ":::\n",
+    "\n",
+    "## Watermark"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "91539613",
+   "metadata": {
+    "execution": {
+     "iopub.execute_input": "2026-06-11T15:48:54.489384Z",
+     "iopub.status.busy": "2026-06-11T15:48:54.489297Z",
+     "iopub.status.idle": "2026-06-11T15:48:54.498664Z",
+     "shell.execute_reply": "2026-06-11T15:48:54.498267Z"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Last updated: Thu, 11 Jun 2026\n",
+      "\n",
+      "Python implementation: CPython\n",
+      "Python version       : 3.12.12\n",
+      "IPython version      : 9.13.0\n",
+      "\n",
+      "pytensor: 3.0.3\n",
+      "pyarrow : 24.0.0\n",
+      "\n",
+      "arviz     : 1.1.0\n",
+      "matplotlib: 3.10.9\n",
+      "numpy     : 2.4.6\n",
+      "platform  : 1.0.8\n",
+      "pyarrow   : 24.0.0\n",
+      "pymc      : 6.0.1\n",
+      "\n",
+      "Watermark: 2.6.0\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "%load_ext watermark\n",
+    "%watermark -n -u -v -iv -w -p pytensor,pyarrow"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5d51af15",
+   "metadata": {},
+   "source": [
+    ":::{include} ../page_footer.md\n",
+    ":::"
+   ]
+  }
+ ],
+ "metadata": {
+  "jupytext": {
+   "default_lexer": "ipython3"
+  },
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.12"
+  },
+  "myst": {
+   "substitutions": {
+    "extra_dependencies": "pyarrow"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/examples/variational_inference/streaming_dataset.myst.md b/examples/variational_inference/streaming_dataset.myst.md
new file mode 100644
index 000000000..034023b72
--- /dev/null
+++ b/examples/variational_inference/streaming_dataset.myst.md
@@ -0,0 +1,267 @@
+---
+jupytext:
+  default_lexer: ipython3
+  text_representation:
+    extension: .md
+    format_name: myst
+    format_version: 0.13
+kernelspec:
+  display_name: Python 3 (ipykernel)
+  language: python
+  name: python3
+myst:
+  substitutions:
+    extra_dependencies: pyarrow
+---
+
+(streaming_dataset)=
+
+# Out-of-core minibatch variational inference with DataLoader and Trainer
+
+:::{post} June 2026
+:tags: variational inference, minibatch, out-of-core
+:category: intermediate, how-to
+:author: Yicheng (Ethan) Yang
+:::
+
++++
+
+`pm.Minibatch` random-indexes an array that must be fully resident in RAM, so
+minibatch variational inference {cite:p}`hoffman2013stochastic,kucukelbir2015automatic`
+only works on data that already fits in memory.
+This notebook uses the streaming API in `pymc.variational.streaming`, which follows
+the same structure as PyTorch's `torch.utils.data`:
+
+* a {class}`~pymc.variational.streaming.DataLoader` batches (and optionally
+  shuffles) an out-of-core source into fixed-size minibatches without loading
+  the whole dataset into memory. Here the source is a directory of Parquet shards
+  read by {func}`~pymc.variational.streaming.parquet_source`.
+* the model observes a `pm.Data` placeholder of one batch, not the whole array.
+* a {class}`~pymc.variational.streaming.Trainer` drives ADVI, writing each
+  minibatch into that placeholder with `set_data` every step. There are no
+  callbacks to write.
+
+The unbiased-gradient rescaling works exactly as in `pm.Minibatch`. The `DataLoader`
+is sized, so `total_size=len(loader)` passes the dataset size `N` to the observed
+distribution, and PyMC scales the minibatch log-likelihood by `N / batch_size`.
+Batches are exact-size, so each pass drops the rows that do not fill a final batch;
+with `shuffle=True` that remainder is re-drawn every epoch instead of being a fixed
+tail. The one extra requirement is shuffling itself. A streaming source is only as
+well mixed as the order it yields rows in, so pass `DataLoader(shuffle=True)` to
+shuffle through a bounded buffer.
+
+`N` is small here so the notebook runs in seconds. The streaming code is the same at
+any size, and the last section shows what changes at scale.
+
++++
+
+:::{include} ../extra_installs.md
+:::
+
+```{code-cell} ipython3
+import glob
+import tempfile
+
+import arviz as az
+import matplotlib.pyplot as plt
+import numpy as np
+import pyarrow as pa
+import pyarrow.parquet as pq
+import pymc as pm
+
+from pymc.variational.streaming import DataLoader, Trainer, parquet_source
+
+RANDOM_SEED = 8927
+rng = np.random.default_rng(RANDOM_SEED)
+az.style.use("arviz-variat")
+```
+
+## Write the dataset to disk
+
+First, we build a logistic-regression dataset, write it to Parquet shards, and
+delete the in-memory table. From here the streaming path reads only the disk copy.
+We keep `X` and `y` around only to build the in-RAM `pm.Minibatch` baseline later;
+the streaming fit never reads them.
+
+```{code-cell} ipython3
+N = 30_000
+b_true = np.array([0.4, -1.2, 0.8, -0.5])  # intercept + 3 slopes
+
+X = rng.normal(size=(N, 3))
+p = 1 / (1 + np.exp(-(b_true[0] + X @ b_true[1:])))
+y = (rng.random(N) < p).astype("float64")
+table = np.column_stack([X, y]).astype("float64")
+
+shard_dir = tempfile.mkdtemp(prefix="streaming_demo_")
+for i, s in enumerate(range(0, N, 5_000)):
+    block = table[s : s + 5_000]
+    pq.write_table(
+        pa.table({f"c{j}": block[:, j] for j in range(4)}),
+        f"{shard_dir}/part_{i:03d}.parquet",
+    )
+del table
+print(len(glob.glob(f"{shard_dir}/*.parquet")), "shards written")
+```
+
+## Stream minibatches off disk and fit with ADVI
+
+Next, the source is `parquet_source`, an out-of-core
+{class}`~pymc.variational.streaming.IterableDataset` that reads one row group at a
+time and gets `n_rows` from the Parquet metadata, so `total_size="auto"` resolves
+`N` without a data scan. The model observes a `pm.Data` placeholder of one batch and
+passes `total_size=len(loader)` for the `N / batch_size` rescaling; the `Trainer`
+streams each minibatch into the placeholder.
+
+```{code-cell} ipython3
+batch_size = 1024
+loader = DataLoader(
+    parquet_source(shard_dir),
+    batch_size=batch_size,
+    sample_shape=(4,),  # 3 features + 1 observed column
+    shuffle=True,  # the shards were written in order
+    buffer_size=15_000,
+    seed=0,
+    total_size="auto",  # N from the Parquet metadata
+)
+
+with pm.Model() as model:
+    b = pm.Normal("b", 0.0, 3.0, shape=4)
+    batch = pm.Data("batch", np.zeros((batch_size, 4)))  # placeholder for one minibatch
+    logit = b[0] + b[1] * batch[:, 0] + b[2] * batch[:, 1] + b[3] * batch[:, 2]
+    pm.Bernoulli("y", logit_p=logit, observed=batch[:, 3], total_size=len(loader))
+
+    approx = Trainer(
+        method="advi",
+        dataloader=loader,
+        data_name="batch",
+        obj_optimizer=pm.adam(learning_rate=0.008),
+    ).fit(30_000, random_seed=RANDOM_SEED)
+    idata_stream = approx.sample(1000, random_seed=RANDOM_SEED)
+```
+
+The negative-ELBO trace shows the fit converging on minibatches read off disk:
+
+```{code-cell} ipython3
+fig, ax = plt.subplots(figsize=(9, 3))
+ax.plot(approx.hist, alpha=0.6)
+ax.set(xlabel="iteration", ylabel="negative ELBO", title="Streaming ADVI convergence");
+```
+
+## Compare with in-RAM `pm.Minibatch`
+
+For comparison, here is the usual fit that keeps the whole dataset in memory. The
+prior, optimizer, iteration count, and seeds match the streaming fit; only the
+minibatch source differs. On this toy problem the two posteriors land on top of
+each other, with ADVI's usual mild bias relative to the dashed ground truth.
+
+```{code-cell} ipython3
+with pm.Model():
+    b = pm.Normal("b", 0.0, 3.0, shape=4)
+    xb, zb, sb, yb = pm.Minibatch(
+        X[:, 0].copy(), X[:, 1].copy(), X[:, 2].copy(), y, batch_size=batch_size
+    )
+    pm.Bernoulli("y", logit_p=b[0] + b[1] * xb + b[2] * zb + b[3] * sb, observed=yb, total_size=N)
+    approx_inram = pm.fit(
+        30_000,
+        method="advi",
+        obj_optimizer=pm.adam(learning_rate=0.008),
+        progressbar=False,
+        random_seed=RANDOM_SEED,
+    )
+    idata_inram = approx_inram.sample(1000, random_seed=RANDOM_SEED)
+```
+
+```{code-cell} ipython3
+bs_stream = idata_stream.posterior["b"].values.reshape(-1, 4)
+bs_inram = idata_inram.posterior["b"].values.reshape(-1, 4)
+names = ["intercept", "slope x1", "slope x2", "slope x3"]
+
+fig, axes = plt.subplots(1, 4, figsize=(13, 3), layout="tight")
+for k, ax in enumerate(axes):
+    ax.hist(bs_stream[:, k], bins=40, density=True, alpha=0.5, label="streaming")
+    ax.hist(bs_inram[:, k], bins=40, density=True, alpha=0.5, label="in-RAM")
+    ax.axvline(b_true[k], color="k", ls="--", lw=1)
+    ax.set(title=names[k], yticks=[])
+axes[0].legend(fontsize=8)
+fig.suptitle("Posterior of b: streaming vs in-RAM (dashed = ground truth)", y=1.04);
+```
+
+## Memory usage
+
+Both paths feed the same `batch_size` to ADVI, but `pm.Minibatch` keeps all `N`
+rows resident, so its array grows linearly in `N`. The streaming path keeps only a
+bounded number of rows resident — the model batch, a shuffle-buffer fill, and the
+source chunks used to build it — independent of `N`. The dense `float64` design
+matrix dominates the cost. The lines below are array lower bounds, not measurements;
+they also ignore framework overhead, PyTensor's resident copy, and the transient
+copy `np.concatenate` makes while filling the shuffle buffer:
+
+```{code-cell} ipython3
+ncols = 4  # 3 features + observed
+n_grid = np.logspace(5, 9, 50)
+inram_gb = n_grid * ncols * 8 / 1e9  # whole dataset resident (array lower bound)
+stream_rows = batch_size + 15_000 + 5_000  # one batch + shuffle buffer + one shard
+stream_gb = np.full_like(n_grid, stream_rows * ncols * 8 / 1e9)
+
+fig, ax = plt.subplots(figsize=(8, 5), layout="tight")
+ax.loglog(n_grid, inram_gb, lw=2.5, label="in-RAM pm.Minibatch  (O(N))")
+ax.loglog(n_grid, stream_gb, lw=2.5, label="streaming DataLoader  (O(batch + buffer + chunk))")
+ax.axhline(26, color="0.5", ls="--", lw=1)
+ax.text(n_grid[-1], 30, "26 GB RAM", color="0.5", ha="right", va="bottom")
+ax.set_xlabel("dataset size N")
+ax.set_ylabel("array footprint (GB, lower bound)")
+ax.set_title("Memory is flat in N when streaming")
+ax.legend(loc="lower right", framealpha=0.95);
+```
+
+Those lines are only the bare arrays. Actual peak RSS is higher, because of the
+framework and PyTensor's resident copy, and it hits the RAM ceiling sooner. As a
+real-data check, outside this notebook, we ran the same logistic model (13 numeric
+features plus the click label) on the
+[Criteo 1TB Click Logs](https://huggingface.co/datasets/criteo/CriteoClickLogs), a
+standard, publicly available out-of-core learning benchmark. Peak memory for the
+streaming `DataLoader` stayed flat at about 0.7 GB across a sweep from 1M to 150M
+rows, while the in-RAM `pm.Minibatch` baseline rose linearly to 15.7 GB at 150M
+rows, which extrapolates to out-of-memory around 240M rows on the same 26 GB
+machine. On a 1M-row slice, where the in-RAM fit is cheap, the two posteriors
+agreed on the large coefficients (the intercept near -3.6 and the strong slopes),
+but the weakest slopes were noisier: the largest gap in posterior means was 0.12,
+and two slopes the in-RAM fit placed near zero came out near +0.08 under streaming,
+flipping sign. On coefficients near the noise floor the two stochastic fits are not
+interchangeable at this slice size.
+
+## When to use it
+
+* Use `pm.Minibatch` when the data fits in RAM: it is simpler and its random
+  index gives perfectly i.i.d. minibatches for free.
+* Use the streaming `DataLoader` and `Trainer` when it does not: it keeps memory
+  flat in `N` by streaming from disk, with no callbacks to wire up. The one thing to
+  watch is shuffling. Pass `shuffle=True`, or pre-shuffle on disk and interleave
+  shards for strongly ordered data, since a bounded buffer over strongly ordered
+  data only block-shuffles it and can bias the posterior.
+
++++
+
+## Authors
+
+* Authored by [Yicheng (Ethan) Yang](https://github.com/YichengYang-Ethan) in June 2026
+  for the Google Summer of Code project Streaming Variational Inference for Large
+  Datasets (PyMC / NumFOCUS).
+
++++
+
+## References
+
+:::{bibliography}
+:filter: docname in docnames
+:::
+
+## Watermark
+
+```{code-cell} ipython3
+%load_ext watermark
+%watermark -n -u -v -iv -w -p pytensor,pyarrow
+```
+
+:::{include} ../page_footer.md
+:::