├── Burgers_torch.ipynb
└── heat_pde.ipynb


/Burgers_torch.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 16,
  6 |    "metadata": {},
  7 |    "outputs": [],
  8 |    "source": [
  9 |     "#code ref:\n",
 10 |     "#https://maziarraissi.github.io/PINNs/"
 11 |    ]
 12 |   },
 13 |   {
 14 |    "cell_type": "code",
 15 |    "execution_count": 1,
 16 |    "metadata": {
 17 |     "tags": []
 18 |    },
 19 |    "outputs": [],
 20 |    "source": [
 21 |     "import math\n",
 22 |     "import numpy as np\n",
 23 |     "import seaborn as sns\n",
 24 |     "import matplotlib.pyplot as plt\n",
 25 |     "from collections import OrderedDict\n",
 26 |     "from tqdm import tqdm\n",
 27 |     "\n",
 28 |     "import torch\n",
 29 |     "import torch.nn as nn"
 30 |    ]
 31 |   },
 32 |   {
 33 |    "cell_type": "code",
 34 |    "execution_count": 2,
 35 |    "metadata": {
 36 |     "tags": []
 37 |    },
 38 |    "outputs": [],
 39 |    "source": [
 40 |     "# Multi-layer Perceptron\n",
 41 |     "class NN(nn.Module):\n",
 42 |     "    def __init__(\n",
 43 |     "        self,\n",
 44 |     "        input_size,\n",
 45 |     "        hidden_size,\n",
 46 |     "        output_size,\n",
 47 |     "        depth,\n",
 48 |     "        act=torch.nn.Tanh,\n",
 49 |     "    ):\n",
 50 |     "        super(NN, self).__init__()\n",
 51 |     "        \n",
 52 |     "        layers = [('input', torch.nn.Linear(input_size, hidden_size))]\n",
 53 |     "        layers.append(('input_activation', act()))\n",
 54 |     "        for i in range(depth): \n",
 55 |     "            layers.append(\n",
 56 |     "                ('hidden_%d' % i, torch.nn.Linear(hidden_size, hidden_size))\n",
 57 |     "            )\n",
 58 |     "            layers.append(('activation_%d' % i, act()))\n",
 59 |     "        layers.append(('output', torch.nn.Linear(hidden_size, output_size)))\n",
 60 |     "\n",
 61 |     "        layerDict = OrderedDict(layers)\n",
 62 |     "        self.layers = torch.nn.Sequential(layerDict)\n",
 63 |     "\n",
 64 |     "    def forward(self, x):\n",
 65 |     "        out = self.layers(x)\n",
 66 |     "        return out"
 67 |    ]
 68 |   },
 69 |   {
 70 |    "cell_type": "code",
 71 |    "execution_count": 3,
 72 |    "metadata": {
 73 |     "tags": []
 74 |    },
 75 |    "outputs": [],
 76 |    "source": [
 77 |     "class Net:\n",
 78 |     "    def __init__(self):\n",
 79 |     "        device = torch.device(\"cuda\") if torch.cuda.is_available() else torch.device(\"cpu\")\n",
 80 |     "\n",
 81 |     "        self.model = NN(\n",
 82 |     "            input_size=2,\n",
 83 |     "            hidden_size=20,\n",
 84 |     "            output_size=1,\n",
 85 |     "            depth=4,\n",
 86 |     "            act=torch.nn.Tanh\n",
 87 |     "        ).to(device)\n",
 88 |     "        \n",
 89 |     "        self.h = 0.1\n",
 90 |     "        self.k = 0.1\n",
 91 |     "        x = torch.arange(-1, 1 + self.h, self.h)\n",
 92 |     "        t = torch.arange(0, 1 + self.k, self.k)\n",
 93 |     "\n",
 94 |     "        # exact solution\n",
 95 |     "        self.X = torch.stack(torch.meshgrid(x, t)).reshape(2, -1).T\n",
 96 |     "        \n",
 97 |     "        # training data\n",
 98 |     "        bc1 = torch.stack(torch.meshgrid(x[0], t)).reshape(2, -1).T\n",
 99 |     "        bc2 = torch.stack(torch.meshgrid(x[-1], t)).reshape(2, -1).T\n",
100 |     "        ic = torch.stack(torch.meshgrid(x, t[0])).reshape(2, -1).T\n",
101 |     "        self.X_train = torch.cat([bc1, bc2, ic])\n",
102 |     "        y_bc1 = torch.zeros(len(bc1))\n",
103 |     "        y_bc2 = torch.zeros(len(bc2))\n",
104 |     "        y_ic = -torch.sin(math.pi * ic[:, 0])\n",
105 |     "        self.y_train = torch.cat([y_bc1, y_bc2, y_ic])\n",
106 |     "        self.y_train = self.y_train.unsqueeze(1)\n",
107 |     "        \n",
108 |     "        self.X = self.X.to(device)\n",
109 |     "        self.X_train = self.X_train.to(device)\n",
110 |     "        self.y_train = self.y_train.to(device)\n",
111 |     "        self.X.requires_grad = True\n",
112 |     "        \n",
113 |     "        self.criterion = torch.nn.MSELoss()\n",
114 |     "        self.iter = 1\n",
115 |     "        \n",
116 |     "        self.optimizer = torch.optim.LBFGS(\n",
117 |     "            self.model.parameters(), \n",
118 |     "            lr=1.0, \n",
119 |     "            max_iter=50000, \n",
120 |     "            max_eval=50000, \n",
121 |     "            history_size=50,\n",
122 |     "            tolerance_grad=1e-7, \n",
123 |     "            tolerance_change=1.0 * np.finfo(float).eps,\n",
124 |     "            line_search_fn=\"strong_wolfe\",   # better numerical stability\n",
125 |     "        )\n",
126 |     "        \n",
127 |     "        self.adam = torch.optim.Adam(self.model.parameters())\n",
128 |     "        \n",
129 |     "    def loss_func(self):\n",
130 |     "        # this is more like a not so elegant hack to zero grad both optimizers\n",
131 |     "        self.adam.zero_grad()\n",
132 |     "        self.optimizer.zero_grad()\n",
133 |     "        \n",
134 |     "        y_pred = self.model(self.X_train)\n",
135 |     "        loss_data = self.criterion(y_pred, self.y_train)\n",
136 |     "        u = self.model(self.X)\n",
137 |     "\n",
138 |     "        du_dX = torch.autograd.grad(\n",
139 |     "            inputs=self.X, \n",
140 |     "            outputs=u, \n",
141 |     "            grad_outputs=torch.ones_like(u), \n",
142 |     "            retain_graph=True, \n",
143 |     "            create_graph=True\n",
144 |     "        )[0]\n",
145 |     "        \n",
146 |     "        du_dt = du_dX[:, 1]\n",
147 |     "        du_dx = du_dX[:, 0]\n",
148 |     "        du_dxx = torch.autograd.grad(\n",
149 |     "            inputs=self.X, \n",
150 |     "            outputs=du_dX, \n",
151 |     "            grad_outputs=torch.ones_like(du_dX), \n",
152 |     "            retain_graph=True, \n",
153 |     "            create_graph=True\n",
154 |     "        )[0][:, 0]\n",
155 |     "        \n",
156 |     "        loss_pde = self.criterion(du_dt + u.squeeze() * du_dx, 0.01 / math.pi * du_dxx)\n",
157 |     "\n",
158 |     "        loss = loss_pde + loss_data\n",
159 |     "        loss.backward()\n",
160 |     "        if self.iter % 100 == 0: \n",
161 |     "            print(self.iter, loss.item())\n",
162 |     "        self.iter = self.iter + 1\n",
163 |     "        return loss\n",
164 |     "    \n",
165 |     "    def train(self):\n",
166 |     "        self.model.train()\n",
167 |     "        for i in range(1000):\n",
168 |     "            self.adam.step(self.loss_func)\n",
169 |     "        self.optimizer.step(self.loss_func)\n",
170 |     "        \n",
171 |     "    def eval_(self):\n",
172 |     "        self.model.eval()"
173 |    ]
174 |   },
175 |   {
176 |    "cell_type": "markdown",
177 |    "metadata": {},
178 |    "source": [
179 |     "# Training"
180 |    ]
181 |   },
182 |   {
183 |    "cell_type": "code",
184 |    "execution_count": 4,
185 |    "metadata": {
186 |     "scrolled": true,
187 |     "tags": []
188 |    },
189 |    "outputs": [
190 |     {
191 |      "name": "stderr",
192 |      "output_type": "stream",
193 |      "text": [
194 |       "C:\\Users\\mecha\\anaconda3\\lib\\site-packages\\torch\\functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at C:\\actions-runner\\_work\\pytorch\\pytorch\\builder\\windows\\pytorch\\aten\\src\\ATen\\native\\TensorShape.cpp:3484.)\n",
195 |       "  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]\n"
196 |      ]
197 |     },
198 |     {
199 |      "name": "stdout",
200 |      "output_type": "stream",
201 |      "text": [
202 |       "100 0.18777918815612793\n",
203 |       "200 0.0967542976140976\n",
204 |       "300 0.08578500151634216\n",
205 |       "400 0.07077104598283768\n",
206 |       "500 0.06669513136148453\n",
207 |       "600 0.06396176666021347\n",
208 |       "700 0.060198429971933365\n",
209 |       "800 0.05186719447374344\n",
210 |       "900 0.03577256202697754\n",
211 |       "1000 0.026719417423009872\n",
212 |       "1100 0.009295769967138767\n",
213 |       "1200 0.005431105382740498\n",
214 |       "1300 0.0036388784646987915\n",
215 |       "1400 0.002566322684288025\n",
216 |       "1500 0.0019015477737411857\n",
217 |       "1600 0.0013643826823681593\n",
218 |       "1700 0.0010937071638181806\n",
219 |       "1800 0.0009133566054515541\n",
220 |       "1900 0.0007465980015695095\n",
221 |       "2000 0.0005781286163255572\n",
222 |       "2100 0.0004759069997817278\n",
223 |       "2200 0.0003760950348805636\n",
224 |       "2300 0.00032897235359996557\n",
225 |       "2400 0.00029087072471156716\n",
226 |       "2500 0.0002510045887902379\n",
227 |       "2600 0.00022016573348082602\n",
228 |       "2700 0.00020470678282435983\n",
229 |       "2800 0.00020292938279453665\n",
230 |       "2900 0.00017246608331333846\n",
231 |       "3000 0.00015391276974696666\n",
232 |       "3100 0.00013784482143819332\n",
233 |       "3200 0.00012046219490002841\n",
234 |       "3300 0.00011163597810082138\n",
235 |       "3400 0.00010503875091671944\n",
236 |       "3500 9.92854475043714e-05\n",
237 |       "3600 9.533743286738172e-05\n",
238 |       "3700 8.807270205579698e-05\n",
239 |       "3800 7.886957610026002e-05\n",
240 |       "3900 7.278687553480268e-05\n",
241 |       "4000 6.689371366519481e-05\n",
242 |       "4100 6.118026794865727e-05\n",
243 |       "4200 5.804469401482493e-05\n",
244 |       "4300 5.492745549418032e-05\n",
245 |       "4400 5.136712934472598e-05\n"
246 |      ]
247 |     }
248 |    ],
249 |    "source": [
250 |     "net = Net()\n",
251 |     "net.train()"
252 |    ]
253 |   },
254 |   {
255 |    "cell_type": "markdown",
256 |    "metadata": {},
257 |    "source": [
258 |     "# Evaluation"
259 |    ]
260 |   },
261 |   {
262 |    "cell_type": "code",
263 |    "execution_count": 17,
264 |    "metadata": {
265 |     "tags": []
266 |    },
267 |    "outputs": [],
268 |    "source": [
269 |     "# net.eval_()\n",
270 |     "net.model.eval()\n",
271 |     "h = 0.01\n",
272 |     "k = 0.01\n",
273 |     "x = torch.arange(-1, 1, h)\n",
274 |     "t = torch.arange(0, 1, k)\n",
275 |     "\n",
276 |     "# exact solution\n",
277 |     "X = torch.stack(torch.meshgrid(x, t)).reshape(2, -1).T\n",
278 |     "X = X.to(net.X.device)"
279 |    ]
280 |   },
281 |   {
282 |    "cell_type": "code",
283 |    "execution_count": 18,
284 |    "metadata": {
285 |     "tags": []
286 |    },
287 |    "outputs": [],
288 |    "source": [
289 |     "model = net.model\n",
290 |     "model.eval()\n",
291 |     "with torch.no_grad():\n",
292 |     "    y_pred = model(X).reshape(len(x), len(t)).cpu().numpy()"
293 |    ]
294 |   },
295 |   {
296 |    "cell_type": "code",
297 |    "execution_count": 19,
298 |    "metadata": {
299 |     "tags": []
300 |    },
301 |    "outputs": [
302 |     {
303 |      "data": {
304 |       "text/plain": [
305 |        "<Figure size 15000x9000 with 0 Axes>"
306 |       ]
307 |      },
308 |      "execution_count": 19,
309 |      "metadata": {},
310 |      "output_type": "execute_result"
311 |     },
312 |     {
313 |      "data": {
314 |       "text/plain": [
315 |        "<Figure size 15000x9000 with 0 Axes>"
316 |       ]
317 |      },
318 |      "metadata": {},
319 |      "output_type": "display_data"
320 |     }
321 |    ],
322 |    "source": [
323 |     "sns.set_style(\"white\")\n",
324 |     "\n",
325 |     "plt.figure(figsize=(5, 3), dpi=3000)"
326 |    ]
327 |   },
328 |   {
329 |    "cell_type": "code",
330 |    "execution_count": 20,
331 |    "metadata": {},
332 |    "outputs": [
333 |     {
334 |      "data": {
335 |       "text/plain": [
336 |        "array([[ 0.00204214,  0.00155723,  0.00115392, ..., -0.00150058,\n",
337 |        "        -0.00148293, -0.00146407],\n",
338 |        "       [ 0.03344315,  0.03200454,  0.03069592, ...,  0.00614461,\n",
339 |        "         0.00610626,  0.00606957],\n",
340 |        "       [ 0.0647999 ,  0.06241718,  0.0602116 , ...,  0.01379141,\n",
341 |        "         0.0136956 ,  0.01360312],\n",
342 |        "       ...,\n",
343 |        "       [-0.09171508, -0.08923785, -0.08691494, ..., -0.02510004,\n",
344 |        "        -0.02490883, -0.02471937],\n",
345 |        "       [-0.06069633, -0.05913445, -0.057676  , ..., -0.01701722,\n",
346 |        "        -0.01687989, -0.01674375],\n",
347 |        "       [-0.02961773, -0.02897303, -0.02838138, ..., -0.00889459,\n",
348 |        "        -0.00881118, -0.00872785]], dtype=float32)"
349 |       ]
350 |      },
351 |      "execution_count": 20,
352 |      "metadata": {},
353 |      "output_type": "execute_result"
354 |     }
355 |    ],
356 |    "source": [
357 |     "y_pred"
358 |    ]
359 |   },
360 |   {
361 |    "cell_type": "code",
362 |    "execution_count": 21,
363 |    "metadata": {
364 |     "scrolled": false
365 |    },
366 |    "outputs": [
367 |     {
368 |      "data": {
369 |       "text/plain": [
370 |        "<AxesSubplot:>"
371 |       ]
372 |      },
373 |      "execution_count": 21,
374 |      "metadata": {},
375 |      "output_type": "execute_result"
376 |     },
377 |     {
378 |      "data": {
379 |       "image/png": "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\n",
380 |       "text/plain": [
381 |        "<Figure size 640x480 with 2 Axes>"
382 |       ]
383 |      },
384 |      "metadata": {},
385 |      "output_type": "display_data"
386 |     }
387 |    ],
388 |    "source": [
389 |     "sns.heatmap(y_pred, cmap='jet')"
390 |    ]
391 |   },
392 |   {
393 |    "cell_type": "code",
394 |    "execution_count": null,
395 |    "metadata": {},
396 |    "outputs": [],
397 |    "source": []
398 |   },
399 |   {
400 |    "cell_type": "code",
401 |    "execution_count": null,
402 |    "metadata": {},
403 |    "outputs": [],
404 |    "source": []
405 |   },
406 |   {
407 |    "cell_type": "code",
408 |    "execution_count": null,
409 |    "metadata": {},
410 |    "outputs": [],
411 |    "source": []
412 |   },
413 |   {
414 |    "cell_type": "code",
415 |    "execution_count": null,
416 |    "metadata": {},
417 |    "outputs": [],
418 |    "source": []
419 |   },
420 |   {
421 |    "cell_type": "code",
422 |    "execution_count": null,
423 |    "metadata": {},
424 |    "outputs": [],
425 |    "source": []
426 |   },
427 |   {
428 |    "cell_type": "code",
429 |    "execution_count": null,
430 |    "metadata": {},
431 |    "outputs": [],
432 |    "source": []
433 |   },
434 |   {
435 |    "cell_type": "code",
436 |    "execution_count": null,
437 |    "metadata": {},
438 |    "outputs": [],
439 |    "source": []
440 |   },
441 |   {
442 |    "cell_type": "code",
443 |    "execution_count": null,
444 |    "metadata": {},
445 |    "outputs": [],
446 |    "source": []
447 |   }
448 |  ],
449 |  "metadata": {
450 |   "kernelspec": {
451 |    "display_name": "Python 3 (ipykernel)",
452 |    "language": "python",
453 |    "name": "python3"
454 |   },
455 |   "language_info": {
456 |    "codemirror_mode": {
457 |     "name": "ipython",
458 |     "version": 3
459 |    },
460 |    "file_extension": ".py",
461 |    "mimetype": "text/x-python",
462 |    "name": "python",
463 |    "nbconvert_exporter": "python",
464 |    "pygments_lexer": "ipython3",
465 |    "version": "3.9.13"
466 |   }
467 |  },
468 |  "nbformat": 4,
469 |  "nbformat_minor": 4
470 | }
471 | 


--------------------------------------------------------------------------------
/heat_pde.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": null,
  6 |    "id": "8c44ce4a",
  7 |    "metadata": {},
  8 |    "outputs": [],
  9 |    "source": [
 10 |     "# code ref: https://deepxde.readthedocs.io/en/latest/demos/pinn_forward/heat.html?highlight=heat%20equation#problem-setup"
 11 |    ]
 12 |   },
 13 |   {
 14 |    "cell_type": "code",
 15 |    "execution_count": 1,
 16 |    "id": "713dd6c7",
 17 |    "metadata": {},
 18 |    "outputs": [
 19 |     {
 20 |      "name": "stderr",
 21 |      "output_type": "stream",
 22 |      "text": [
 23 |       "Using backend: tensorflow.compat.v1\n",
 24 |       "Other supported backends: tensorflow, pytorch, jax, paddle.\n",
 25 |       "paddle supports more examples now and is recommended.\n"
 26 |      ]
 27 |     },
 28 |     {
 29 |      "name": "stdout",
 30 |      "output_type": "stream",
 31 |      "text": [
 32 |       "WARNING:tensorflow:From C:\\Users\\mecha\\AppData\\Local\\Programs\\Python\\Python39\\lib\\site-packages\\tensorflow\\python\\compat\\v2_compat.py:107: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.\n",
 33 |       "Instructions for updating:\n",
 34 |       "non-resource variables are not supported in the long term\n",
 35 |       "WARNING:tensorflow:From C:\\Users\\mecha\\AppData\\Local\\Programs\\Python\\Python39\\lib\\site-packages\\deepxde\\nn\\initializers.py:118: The name tf.keras.initializers.he_normal is deprecated. Please use tf.compat.v1.keras.initializers.he_normal instead.\n",
 36 |       "\n"
 37 |      ]
 38 |     }
 39 |    ],
 40 |    "source": [
 41 |     "import deepxde as dde\n",
 42 |     "from deepxde.backend import tf\n",
 43 |     "import numpy as np"
 44 |    ]
 45 |   },
 46 |   {
 47 |    "cell_type": "code",
 48 |    "execution_count": 2,
 49 |    "id": "3e691459",
 50 |    "metadata": {},
 51 |    "outputs": [],
 52 |    "source": [
 53 |     "a = 0.4\n",
 54 |     "L = 1\n",
 55 |     "n = 1"
 56 |    ]
 57 |   },
 58 |   {
 59 |    "cell_type": "code",
 60 |    "execution_count": 3,
 61 |    "id": "6dc08607",
 62 |    "metadata": {},
 63 |    "outputs": [],
 64 |    "source": [
 65 |     "geom = dde.geometry.Interval(0, L)\n",
 66 |     "timedomain = dde.geometry.TimeDomain(0, n)\n",
 67 |     "geomtime = dde.geometry.GeometryXTime(geom, timedomain)"
 68 |    ]
 69 |   },
 70 |   {
 71 |    "cell_type": "code",
 72 |    "execution_count": 4,
 73 |    "id": "9c82f7ed",
 74 |    "metadata": {},
 75 |    "outputs": [],
 76 |    "source": [
 77 |     "def pde(x, y):\n",
 78 |     "    dy_t = dde.grad.jacobian(y, x, i=0, j=1)\n",
 79 |     "    dy_xx = dde.grad.hessian(y, x, i=0, j=0)\n",
 80 |     "    return dy_t - a * dy_xx"
 81 |    ]
 82 |   },
 83 |   {
 84 |    "cell_type": "code",
 85 |    "execution_count": 5,
 86 |    "id": "acd7a9b2",
 87 |    "metadata": {},
 88 |    "outputs": [],
 89 |    "source": [
 90 |     "#bc = dde.icbc.DirichletBC(geomtime, lambda x: 0, lambda _, on_boundary: on_boundary)\n",
 91 |     "\n",
 92 |     "bc = dde.icbc.DirichletBC(geomtime, lambda x: 2*x[:, 0:1], lambda _, on_boundary: on_boundary)\n",
 93 |     "\n",
 94 |     "ic = dde.icbc.IC(\n",
 95 |     "    geomtime,\n",
 96 |     "    lambda x: np.sin(n * np.pi * x[:, 0:1] / L),\n",
 97 |     "    lambda _, on_initial: on_initial,\n",
 98 |     ")"
 99 |    ]
100 |   },
101 |   {
102 |    "cell_type": "code",
103 |    "execution_count": 6,
104 |    "id": "612d3c51",
105 |    "metadata": {},
106 |    "outputs": [
107 |     {
108 |      "name": "stdout",
109 |      "output_type": "stream",
110 |      "text": [
111 |       "Warning: 2540 points required, but 2550 points sampled.\n"
112 |      ]
113 |     }
114 |    ],
115 |    "source": [
116 |     "data = dde.data.TimePDE(\n",
117 |     "    geomtime,\n",
118 |     "    pde,\n",
119 |     "    [bc, ic],\n",
120 |     "    num_domain=2540,\n",
121 |     "    num_boundary=80,\n",
122 |     "    num_initial=160,\n",
123 |     "    num_test=2540,\n",
124 |     ")"
125 |    ]
126 |   },
127 |   {
128 |    "cell_type": "code",
129 |    "execution_count": 7,
130 |    "id": "0cb15a84",
131 |    "metadata": {},
132 |    "outputs": [],
133 |    "source": [
134 |     "net = dde.nn.FNN([2] + [20] * 3 + [1], \"tanh\", \"Glorot normal\")"
135 |    ]
136 |   },
137 |   {
138 |    "cell_type": "code",
139 |    "execution_count": 8,
140 |    "id": "ba30496a",
141 |    "metadata": {},
142 |    "outputs": [
143 |     {
144 |      "name": "stdout",
145 |      "output_type": "stream",
146 |      "text": [
147 |       "Compiling model...\n",
148 |       "Building feed-forward neural network...\n",
149 |       "'build' took 0.029545 s\n",
150 |       "\n"
151 |      ]
152 |     },
153 |     {
154 |      "name": "stderr",
155 |      "output_type": "stream",
156 |      "text": [
157 |       "C:\\Users\\mecha\\AppData\\Local\\Programs\\Python\\Python39\\lib\\site-packages\\deepxde\\nn\\tensorflow_compat_v1\\fnn.py:116: UserWarning: `tf.layers.dense` is deprecated and will be removed in a future version. Please use `tf.keras.layers.Dense` instead.\n",
158 |       "  return tf.layers.dense(\n"
159 |      ]
160 |     },
161 |     {
162 |      "name": "stdout",
163 |      "output_type": "stream",
164 |      "text": [
165 |       "'compile' took 0.203035 s\n",
166 |       "\n"
167 |      ]
168 |     }
169 |    ],
170 |    "source": [
171 |     "model = dde.Model(data, net)\n",
172 |     "model.compile(\"adam\", lr=1e-3)"
173 |    ]
174 |   },
175 |   {
176 |    "cell_type": "code",
177 |    "execution_count": 9,
178 |    "id": "c5910dad",
179 |    "metadata": {},
180 |    "outputs": [
181 |     {
182 |      "name": "stdout",
183 |      "output_type": "stream",
184 |      "text": [
185 |       "Training model...\n",
186 |       "\n",
187 |       "Step      Train loss                        Test loss                         Test metric\n",
188 |       "0         [7.15e-02, 2.40e+00, 6.60e-01]    [7.00e-02, 2.40e+00, 6.60e-01]    []  \n",
189 |       "1000      [9.36e-03, 5.94e-02, 1.38e-01]    [7.41e-03, 5.94e-02, 1.38e-01]    []  \n",
190 |       "2000      [6.58e-03, 3.64e-02, 2.77e-02]    [4.68e-03, 3.64e-02, 2.77e-02]    []  \n",
191 |       "3000      [3.79e-03, 2.53e-02, 1.78e-02]    [2.06e-03, 2.53e-02, 1.78e-02]    []  \n",
192 |       "4000      [3.49e-03, 2.08e-02, 1.33e-02]    [2.15e-03, 2.08e-02, 1.33e-02]    []  \n",
193 |       "5000      [2.70e-03, 1.75e-02, 1.18e-02]    [1.93e-03, 1.75e-02, 1.18e-02]    []  \n",
194 |       "6000      [2.45e-03, 1.54e-02, 1.07e-02]    [1.86e-03, 1.54e-02, 1.07e-02]    []  \n",
195 |       "7000      [2.13e-03, 1.37e-02, 1.00e-02]    [1.58e-03, 1.37e-02, 1.00e-02]    []  \n",
196 |       "8000      [1.75e-03, 1.25e-02, 9.50e-03]    [1.33e-03, 1.25e-02, 9.50e-03]    []  \n",
197 |       "9000      [2.18e-03, 1.11e-02, 8.76e-03]    [1.40e-03, 1.11e-02, 8.76e-03]    []  \n",
198 |       "10000     [1.78e-03, 1.02e-02, 8.00e-03]    [1.23e-03, 1.02e-02, 8.00e-03]    []  \n",
199 |       "11000     [1.76e-03, 9.34e-03, 7.41e-03]    [1.22e-03, 9.34e-03, 7.41e-03]    []  \n",
200 |       "12000     [1.69e-03, 8.63e-03, 6.90e-03]    [1.19e-03, 8.63e-03, 6.90e-03]    []  \n",
201 |       "13000     [3.47e-03, 8.43e-03, 6.29e-03]    [1.57e-03, 8.43e-03, 6.29e-03]    []  \n",
202 |       "14000     [1.59e-03, 7.52e-03, 6.02e-03]    [1.09e-03, 7.52e-03, 6.02e-03]    []  \n",
203 |       "15000     [1.54e-03, 7.03e-03, 5.67e-03]    [1.08e-03, 7.03e-03, 5.67e-03]    []  \n",
204 |       "\n",
205 |       "Best model at step 15000:\n",
206 |       "  train loss: 1.42e-02\n",
207 |       "  test loss: 1.38e-02\n",
208 |       "  test metric: []\n",
209 |       "\n",
210 |       "'train' took 36.574986 s\n",
211 |       "\n"
212 |      ]
213 |     }
214 |    ],
215 |    "source": [
216 |     "losshistory, train_state = model.train(iterations=15000)"
217 |    ]
218 |   },
219 |   {
220 |    "cell_type": "code",
221 |    "execution_count": 10,
222 |    "id": "53e5a951",
223 |    "metadata": {},
224 |    "outputs": [
225 |     {
226 |      "name": "stdout",
227 |      "output_type": "stream",
228 |      "text": [
229 |       "Compiling model...\n",
230 |       "'compile' took 0.105828 s\n",
231 |       "\n",
232 |       "Training model...\n",
233 |       "\n",
234 |       "Step      Train loss                        Test loss                         Test metric\n",
235 |       "15000     [1.54e-03, 7.03e-03, 5.67e-03]    [1.08e-03, 7.03e-03, 5.67e-03]    []  \n",
236 |       "16000     [8.87e-04, 6.46e-04, 6.10e-04]    [8.87e-04, 6.46e-04, 6.10e-04]        \n",
237 |       "17000     [5.07e-04, 3.01e-04, 2.71e-04]    [5.07e-04, 3.01e-04, 2.71e-04]        \n",
238 |       "INFO:tensorflow:Optimization terminated with:\n",
239 |       "  Message: CONVERGENCE: REL_REDUCTION_OF_F_<=_FACTR*EPSMCH\n",
240 |       "  Objective function value: 0.001020\n",
241 |       "  Number of iterations: 1990\n",
242 |       "  Number of functions evaluations: 2205\n",
243 |       "17205     [5.17e-04, 2.81e-04, 2.22e-04]    [3.76e-04, 2.81e-04, 2.22e-04]    []  \n",
244 |       "\n",
245 |       "Best model at step 17205:\n",
246 |       "  train loss: 1.02e-03\n",
247 |       "  test loss: 8.79e-04\n",
248 |       "  test metric: []\n",
249 |       "\n",
250 |       "'train' took 7.572932 s\n",
251 |       "\n"
252 |      ]
253 |     }
254 |    ],
255 |    "source": [
256 |     "model.compile(\"L-BFGS-B\")\n",
257 |     "losshistory, train_state = model.train()"
258 |    ]
259 |   },
260 |   {
261 |    "cell_type": "code",
262 |    "execution_count": 11,
263 |    "id": "07bade3c",
264 |    "metadata": {},
265 |    "outputs": [
266 |     {
267 |      "name": "stdout",
268 |      "output_type": "stream",
269 |      "text": [
270 |       "Saving loss history to C:\\Users\\mecha\\OneDrive\\デスクトップ\\main\\Youtube_Videos_eng\\2023_07_10_Video_2\\loss.dat ...\n",
271 |       "Saving training data to C:\\Users\\mecha\\OneDrive\\デスクトップ\\main\\Youtube_Videos_eng\\2023_07_10_Video_2\\train.dat ...\n",
272 |       "Saving test data to C:\\Users\\mecha\\OneDrive\\デスクトップ\\main\\Youtube_Videos_eng\\2023_07_10_Video_2\\test.dat ...\n"
273 |      ]
274 |     },
275 |     {
276 |      "data": {
277 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjYAAAGwCAYAAAC6ty9tAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABS50lEQVR4nO3deXhU5d3/8ffJvu87JASQLSxhCxEQlZoK1OJWlVqeulSxtai1sVbtgkt9qj+t1lajj6VV7CpqBW3FFQVckD0Ihp2QBMjGkoSEkG3u3x8DIxHCOsmZmXxe13WumTnrdw7HzMf7PotljDGIiIiI+AA/uwsQERERcRcFGxEREfEZCjYiIiLiMxRsRERExGco2IiIiIjPULARERERn6FgIyIiIj4jwO4C3M3hcLB7924iIyOxLMvuckREROQUGGM4cOAAaWlp+PmdebuLzwWb3bt3k56ebncZIiIicgbKysro2bPnGS/vc8EmMjIScO6YqKgom6sRERGRU1FXV0d6errrd/xM+VywOdL9FBUVpWAjIiLiZc72NBKdPCwiIiI+Q8FGREREfIaCjYiIiPgMnzvHRkREfJfD4aC5udnuMuQMBAYG4u/v3+nb8ZlgU1BQQEFBAW1tbXaXIiIinaC5uZni4mIcDofdpcgZiomJISUlpVPvM2cZY0ynrd0GdXV1REdHU1tbq6uiRER8hDGG0tJSWlpazvoGbtL1jDEcPHiQqqoqYmJiSE1NPWYed/1++0yLjYiI+K7W1lYOHjxIWloaYWFhdpcjZyA0NBSAqqoqkpKSOq1bSpFXREQ83pHTDIKCgmyuRM7GkVDa0tLSadtQsBEREa+hZwB6t67491OwEREREZ+hYCMiIiI+Q8FGRETEi2RmZvLUU0/Zvg5PpWBziowx7NqxhfKybXaXIiIiXsCyrBMODzzwwBmtd8WKFdxyyy3uLdaH6HLvU/T58zMZW/EPlqZMJ/VHz9pdjoiIeLjy8nLX+7lz5zJr1iw2bdrkGhcREeF6b4yhra2NgICT/ywnJia6t1AfoxabUxSYOhiA2D2rba5ERESMMRxsbrVlONX72qakpLiG6OhoLMtyfd64cSORkZG8/fbbjBo1iuDgYD755BO2bdvGZZddRnJyMhEREeTk5PDBBx+0W+/Xu5Esy+LPf/4zV1xxBWFhYfTr148333zztPZnaWkpl112GREREURFRXHNNddQWVnpmr527VomTpxIZGQkUVFRjBo1ipUrVwJQUlLC1KlTiY2NJTw8nMGDB7NgwYLT2r47qcXmFKUN+wasgT4tm2lsqCc0POLkC4mISKdobGkja9a7tmy76KFJhAW55+fz3nvv5Xe/+x19+vQhNjaWsrIyvvWtb/G///u/BAcH89e//pWpU6eyadMmMjIyOlzPgw8+yGOPPcbjjz/O008/zfTp0ykpKSEuLu6kNTgcDleoWbx4Ma2trcycOZNp06axaNEiAKZPn86IESN47rnn8Pf3p7CwkMDAQABmzpxJc3MzS5YsITw8nKKionatUV3NZ4JNZz8rKrXXAKqJJdHaz/q1HzNk3JRO2Y6IiHQfDz30EN/85jddn+Pi4sjOznZ9/s1vfsO8efN48803ue222zpczw033MC1114LwG9/+1v++Mc/snz5ciZPnnzSGhYuXMi6desoLi4mPT0dgL/+9a8MHjyYFStWkJOTQ2lpKXfffTcDBw4EoF+/fq7lS0tL+c53vsPQoUMB6NOnz2nsAffzmWAzc+ZMZs6c6XrWhLtZfn6URQwjsX4xBzYtAQUbERHbhAb6U/TQJNu27S6jR49u97m+vp4HHniAt956i/LyclpbW2lsbKS0tPSE6xk2bJjrfXh4OFFRUVRVVZ1SDRs2bCA9Pd0VagCysrKIiYlhw4YN5OTkkJ+fz80338zf/vY38vLyuPrqq+nbty8Ad9xxB7feeivvvfceeXl5fOc732lXT1fTOTanoaVHLgChlSttrkREpHuzLIuwoABbBnfePTc8PLzd55/97GfMmzeP3/72t3z88ccUFhYydOhQmpubT7ieI91CR+8fdz4F/YEHHuDLL7/kkksu4cMPPyQrK4t58+YBcPPNN7N9+3a+//3vs27dOkaPHs3TTz/ttm2fLgWb05Aw+AIA+jSu77QuLxER6b4+/fRTbrjhBq644gqGDh1KSkoKO3bs6NRtDho0iLKyMsrKylzjioqKqKmpISsryzWuf//+/PSnP+W9997jyiuv5MUXX3RNS09P50c/+hGvv/46d911F7Nnz+7Umk9EweY0ZGbl0mBCiLIOUrxBrTYiIuJe/fr14/XXX6ewsJC1a9fyve99z60tL8eTl5fH0KFDmT59OqtXr2b58uVcd911XHDBBYwePZrGxkZuu+02Fi1aRElJCZ9++ikrVqxg0KBBANx55528++67FBcXs3r1aj766CPXNDso2JwG/4BAtoc60+ueLxfbXI2IiPiaJ598ktjYWMaNG8fUqVOZNGkSI0eO7NRtWpbFG2+8QWxsLOeffz55eXn06dOHuXPnAuDv78/evXu57rrr6N+/P9dccw1TpkzhwQcfBJxPXp85cyaDBg1i8uTJ9O/fn2efte9+b5Y51QvyvcSRk4dra2uJiopy+/qXvfhzckueZ0XkReTc9brb1y8iIsc6dOgQxcXF9O7dm5CQELvLkTN0on9Hd/1+q8XmNEX2Pw+AngfWnvJNmkRERKRrKNicpt7ZF9Bq/EhlD+WlW+wuR0RERI6iYHOaQiOiKQ50Xru/64uPbK5GREREjqZgcwb2x48CoG3HUpsrERERkaMp2JyBoD7jAEjcX2hvISIiItKOgs0ZyBg+EYDebTvYv2+PzdWIiIjIEQo2ZyAuOYOdVip+lmHHGp1nIyIi4ikUbM5QRcxwAA5u+8TeQkRERMRFweZMZYwFILp6lc2FiIiIHGvHjh1YlkVhYaHdpXQpnwk2BQUFZGVlkZOT0yXbSx3ifCDmOc0bOXSosUu2KSIi3sOyrBMODzzwwFmte/78+W6r1Zf4TLCZOXMmRUVFrFixoku2l9Z3GDVEEmK1sO2Lz7pkmyIi4j3Ky8tdw1NPPUVUVFS7cT/72c/sLtEn+Uyw6WqWnx8l4cMAqNmoB2KKiEh7KSkpriE6OhrLstqNe/nllxk0aBAhISEMHDiw3YMjm5ubue2220hNTSUkJIRevXrxyCOPAJCZmQnAFVdcgWVZrs+nYvHixYwZM4bg4GBSU1O59957aW1tdU1/7bXXGDp0KKGhocTHx5OXl0dDQwMAixYtYsyYMYSHhxMTE8P48eMpKSk5+x3lZgF2F+DNmtPGwJZPCS3vmlYiERE5zBhoOWjPtgPDwLLOahX/+Mc/mDVrFs888wwjRoxgzZo1zJgxg/DwcK6//nr++Mc/8uabb/LKK6+QkZFBWVkZZWVlAKxYsYKkpCRefPFFJk+ejL+//yltc9euXXzrW9/ihhtu4K9//SsbN25kxowZhISE8MADD1BeXs61117LY489xhVXXMGBAwf4+OOPMcbQ2trK5ZdfzowZM/jXv/5Fc3Mzy5cvxzrL/dAZFGzOQtygC2DL78lsXIejzYGfvxrARES6RMtB+G2aPdv+xW4ICj+rVdx///088cQTXHnllQD07t2boqIinn/+ea6//npKS0vp168f5513HpZl0atXL9eyiYmJAMTExJCSknLK23z22WdJT0/nmWeewbIsBg4cyO7du7nnnnuYNWsW5eXltLa2cuWVV7q2N3ToUAD27dtHbW0t3/72t+nb1/lYoUGDBp3VPugs+iU+C72GjOWQCSSOA+zYstbuckRExAs0NDSwbds2brrpJiIiIlzDww8/zLZt2wC44YYbKCwsZMCAAdxxxx289957Z73dDRs2MHbs2HatLOPHj6e+vp6dO3eSnZ3NRRddxNChQ7n66quZPXs2+/fvByAuLo4bbriBSZMmMXXqVP7whz9QXl5+1jV1BrXYnIWAoBC2hAxiUNMXVK77iD4DR9hdkohI9xAY5mw5sWvbZ6G+vh6A2bNnk5ub227akW6lkSNHUlxczNtvv80HH3zANddcQ15eHq+99tpZbftE/P39ef/99/nss8947733ePrpp/nlL3/JsmXL6N27Ny+++CJ33HEH77zzDnPnzuVXv/oV77//Pueee26n1XQmFGzO0oGk0VD2BVbZMrtLERHpPizrrLuD7JKcnExaWhrbt29n+vTpHc4XFRXFtGnTmDZtGldddRWTJ09m3759xMXFERgYSFtb22ltd9CgQfz73//GGONqtfn000+JjIykZ8+egPMy8vHjxzN+/HhmzZpFr169mDdvHvn5+QCMGDGCESNGcN999zF27Fj++c9/Ktj4moh+50HZC/Q4oK4oERE5NQ8++CB33HEH0dHRTJ48maamJlauXMn+/fvJz8/nySefJDU1lREjRuDn58err75KSkoKMTExgPPKqIULFzJ+/HiCg4OJjY096TZ//OMf89RTT3H77bdz2223sWnTJu6//37y8/Px8/Nj2bJlLFy4kIsvvpikpCSWLVtGdXU1gwYNori4mD/96U9ceumlpKWlsWnTJrZs2cJ1113XyXvq9CnYnKXM4RNxLLRIp5zyXSWk9uh18oVERKRbu/nmmwkLC+Pxxx/n7rvvJjw8nKFDh3LnnXcCEBkZyWOPPcaWLVvw9/cnJyeHBQsW4OfnPDX2iSeeID8/n9mzZ9OjRw927Nhx0m326NGDBQsWcPfdd5OdnU1cXBw33XQTv/rVrwBnC9GSJUt46qmnqKuro1evXjzxxBNMmTKFyspKNm7cyEsvvcTevXtJTU1l5syZ/PCHP+ysXXTGLGOMsbsId6qrqyM6Opra2lqioqK6ZJvFvxlO77Ziluc8xZhLbuySbYqIdCeHDh2iuLiY3r17ExISYnc5coZO9O/ort9vXRXlBnvinCcNt+zQHYhFRETspGDjBkG9xwGQuG+1zZWIiIh0bwo2bpA+/CIA+rRup7amxt5iREREujEFGzeIS+tDpZVIgOVg+1o9N0pERMQuCjZusjsqG4CGLR/bXImIiO/ysetdup2u+PdTsHETR7rzBkWRVatsrkRExPccuSNvc3OzzZXI2Th40Png0sDAwE7bhu5j4ybJQybC+ofp21REU3MTwUHBdpckIuIzAgICCAsLo7q6msDAQNf9XMQ7GGM4ePAgVVVVxMTEnPITyc+Ego2b9Og/ggOEEWkdpGjdMrJGnW93SSIiPsOyLFJTUykuLqakpMTucuQMne4Tyc+Ego2bWH7+lIQNYcjB5ezbsAQUbERE3CooKIh+/fqpO8pLBQYGdmpLzREKNm7UmDoGti0neLceiCki0hn8/Px052E5IXVSulHsAGcrTa+D63C0OWyuRkREpPtRsHGjXsPOo9n4k8R+dmzfYHc5IiIi3Y5HBpv//ve/DBgwgH79+vHnP//Z7nJOWWBIODuC+wNQse4jm6sRERHpfjwu2LS2tpKfn8+HH37ImjVrePzxx9m7d6/dZZ2y2sTRzjeln9tbiIiISDfkccFm+fLlDB48mB49ehAREcGUKVN477337C7rlIWfcx4AqbWF9hYiIiLSDbk92CxZsoSpU6eSlpaGZVnMnz//mHkKCgrIzMwkJCSE3Nxcli9f7pq2e/duevTo4frco0cPdu3a5e4yO03G8IkA9DZlVFR4T90iIiK+wO3BpqGhgezsbAoKCo47fe7cueTn53P//fezevVqsrOzmTRpElVVVe4uxRYRscmU+fcEoLRwkb3FiIiIdDNuDzZTpkzh4Ycf5oorrjju9CeffJIZM2Zw4403kpWVxf/93/8RFhbGCy+8AEBaWlq7Fppdu3aRlpbW4faampqoq6trN9itOnYEAE3bP7W5EhERke6lS8+xaW5uZtWqVeTl5X1VgJ8feXl5LF26FIAxY8awfv16du3aRX19PW+//TaTJk3qcJ2PPPII0dHRriE9Pb3Tv8fJ+GeOAyBu72qbKxEREeleujTY7Nmzh7a2NpKTk9uNT05OpqKiAnA+6OyJJ55g4sSJDB8+nLvuuov4+PgO13nfffdRW1vrGsrKyjr1O5yKnsO+AcA5rVuoqz9gczUiIiLdh0c+UuHSSy/l0ksvPaV5g4ODCQ72rCdpx6cPYK8VSzz7KSr8mBHnfcvukkRERLqFLm2xSUhIwN/fn8rKynbjKysrO/1pn13KstgZkQ3Agc2f2FyMiIhI99GlwSYoKIhRo0axcOFC1ziHw8HChQsZO3bsWa27oKCArKwscnJyzrZMt2jrmQtAeOUKmysRERHpPtwebOrr6yksLKSwsBCA4uJiCgsLKS0tBSA/P5/Zs2fz0ksvsWHDBm699VYaGhq48cYbz2q7M2fOpKioiBUrPCNIJA6+EIBzDn1Jc0urvcWIiIh0E24/x2blypVMnDjR9Tk/Px+A66+/njlz5jBt2jSqq6uZNWsWFRUVDB8+nHfeeeeYE4q9Xc9BORwkmGirgaKilWRln2t3SSIiIj7PMsYYu4twp7q6OqKjo6mtrSUqKsrWWor+30SyGlfz8YBfMOHae2ytRURExJO56/fb454V5Usak53n+wTuWmZzJSIiIt2DzwQbTzt5GCB6wAQAMuq/wMcaxkRERDySuqI6UXNDLX6PZRJgOSi+bgW9+/S3tR4RERFPpa4oLxAUHk1JUF8Adn+xyN5iREREugEFm05WkzAKAEfJZzZXIiIi4vsUbDpZSN/xACTXFNpbiIiISDfgM8HGE08eBug13HlPn76OHVRVV9lcjYiIiG/zmWDjaXcePiIiIZ3dfqn4W4biwkV2lyMiIuLTfCbYeLKqGOcDMQ9t+9TmSkRERHybgk0XsDKcD/iM3bPK5kpERER8m4JNF0gb5jzPpl/LJuoPHrS5GhEREd+lYNMFEnsPo4ZIQq1mtq5Vd5SIiEhnUbDpCpbFzohhANRs+sTmYkRERHyXzwQbT73c+4iWHrkAhFV41lVbIiIivsRngo2nXu59RELWBQD0bVxHS2ubzdWIiIj4Jp8JNp6ux6BzOUQg8VYdWzeutbscERERn6Rg00X8gkIoCRkEQPWXi+wtRkRExEcp2HShhiTnAzH9dy6zuRIRERHfpGDThSL7TwCg54G1GGNsrkZERMT3KNh0oYzsC3EYi16UU1q6w+5yREREfI7PBBtPv9wbIDgynrLAXgDsXPuRzdWIiIj4Hp8JNp5+ufcR++Kd59m0liy1uRIRERHf4zPBxlsE9R4HQNL+NTZXIiIi4nsUbLpY+oiLADinbTvV+/bZXI2IiIhvUbDpYlHJvamyEgi02iguXGx3OSIiIj5FwcYGFdEjAGjYqid9i4iIuJOCjR0ynA/EjK5eZXMhIiIivkXBxgYpQycC0L+5iIbGQzZXIyIi4jsUbGyQ1HcE9YQRYR1iyzo9XkFERMRdfCbYeMMN+lz8/CkNHwLA/o1LbC5GRETEd/hMsPGWG/Qd0ZQ2BoDQcu+oV0RExBv4TLDxNnGDzgeg98EvaG1ts7kaERER36BgY5P0wefRgj/J1n62btlgdzkiIiI+QcHGJn7B4ZQE9Qegcv0ie4sRERHxEQo2NjqQNBoAv7LPba5ERETENyjY2Ci833kApNUVYoyxuRoRERHvp2Bjo4xs5436+lJG2a5dNlcjIiLi/RRsbBQSk8wu/54AlK79yOZqREREvJ+Cjc32xI0EoKV4qc2ViIiIeD8FG5sF9h4HQML+1TZXIiIi4v0UbGzW8/B5Nv1bt7C3ptbmakRERLybzwQbr3pW1FGi0gawz4oh2Gpl69pP7C5HRETEq/lMsPG2Z0W5WBa7o4YD0LBFwUZERORs+Eyw8WaO9FwAIqtW2lyJiIiId1Ow8QDJgy8EoF/TlzQ2tdhbjIiIiBdTsPEASf1G00gwMVYDm9ar1UZERORMKdh4ACsgiNLQLAD2bVhsczUiIiLeS8HGQzSmjgEgaPdymysRERHxXgo2HiJ24PkAZDZ8QWubw+ZqREREvJOCjYfoOfR82oxFT6uardu22F2OiIiIV1Kw8RD+oVGUBZ0DQMU6PRBTRETkTCjYeJDaxFHON2Wf21uIiIiIl1Kw8SCh55wHQEptIcYYm6sRERHxPgo2HiT98AMx+zl2sKuiyuZqREREvI+CjQcJje9JhV8K/pZhx9pFdpcjIiLidRRsPEx17AgAmrZ/ZnMlIiIi3kfBxsP4Z44FIKFqKfWH9NwoERGR06Fg42H6jrucZgLIZhNv/3uO3eWIiIh4FZ8JNgUFBWRlZZGTk2N3KWclOL4X5YNuAmDM5sfZWr7X5opERES8h2V87Lriuro6oqOjqa2tJSoqyu5yzkzTAWoeyyambS8vR93EtJ8+gWVZdlclIiLSadz1++0zLTY+JTiS1m88AMDU2r/z0Yq1tpYjIiLiLRRsPFTCuO+zO3Io4VYTze/MorG5ze6SREREPJ6CjaeyLOKu+gMOLCY7FjP/zdftrkhERMTjKdh4sJBeo9jV+yoAhnzxW3ZU1dlckYiIiGdTsPFwPb/zWxqscIb6bWfR3N/bXY6IiIhHU7DxcFZEEo3j7wZg6p7ZLF67xeaKREREPJeCjRdImHgbe0IyibcOUPmfBznUohOJRUREjkfBxhv4BxJ++e8AuLLlLV57+wObCxIREfFMCjZeInTgNylPvYgAy0GfVQ+xc1+D3SWJiIh4HAUbL5Jy9RM0E8g4az3/fWW23eWIiIh4HAUbL2LF9aZu5K0AXLL7GT7dUGZzRSIiIp5FwcbLJEy+l9rAJNL9qtk07xGaWx12lyQiIuIxFGy8TVA4gVMeBuDapld5ZeFSmwsSERHxHAo2XihsxDVUx40i1Gom7rOHqag9ZHdJIiIiHkHBxhtZFvFX/Z42/PiW9Rmv/PtluysSERHxCAo2XsovLZuaQd8D4Js7nmDZ1kqbKxIREbGfgo0Xi//2Qxz0j2SQXynL//17Wtt0IrGIiHRvCjbeLDweLvwlAP9z8G/M/fgLmwsSERGxl4KNlwsbN4P9kf2Iterx++i3VB9osrskERER2yjYeDv/AKIufwKAa3iPv81/y+aCRERE7KNg4wP8+17A/sxL8LcMYzc/zuqSfXaXJCIiYguPDDZXXHEFsbGxXHXVVXaX4jViL3+UFiuIsf5FvPfq87Q5jN0liYiIdDmPDDY/+clP+Otf/2p3Gd4lJoPmc38CwPcPzOa1zzfbXJCIiEjX88hgc+GFFxIZGWl3GV4n/Bt3cSAklR7WXmree4z9Dc12lyQiItKlTjvYLFmyhKlTp5KWloZlWcyfP/+YeQoKCsjMzCQkJITc3FyWL1/ujlrlZAJDCbvkEQCuN2/w5/8usrceERGRLnbawaahoYHs7GwKCgqOO33u3Lnk5+dz//33s3r1arKzs5k0aRJVVVWueYYPH86QIUOOGXbv3n3aX6CpqYm6urp2Q3fmP+RyalPGEmK1MGT946zfVWt3SSIiIl0m4HQXmDJlClOmTOlw+pNPPsmMGTO48cYbAfi///s/3nrrLV544QXuvfdeAAoLC8+s2uN45JFHePDBB922Pq9nWURf8QRtz53HFP/lzHr172Td8WP8/Cy7KxMREel0bj3Hprm5mVWrVpGXl/fVBvz8yMvLY+nSpe7clMt9991HbW2taygrK+uU7XiV5ME0DXcGy+n7nmXeqhKbCxIREekabg02e/bsoa2tjeTk5Hbjk5OTqaioOOX15OXlcfXVV7NgwQJ69ux5wlAUHBxMVFRUu0EgbNKvaQyMYYDfTra//QfqDrXYXZKIiEin88iroj744AOqq6s5ePAgO3fuZOzYsXaX5H1CYwnMmwXALW0v8/yCZTYXJCIi0vncGmwSEhLw9/ensrKy3fjKykpSUlLcuSk5BQE5N1AfO4ho6yA91zzJxorufWK1iIj4PrcGm6CgIEaNGsXChQtd4xwOBwsXLuz0VpeCggKysrLIycnp1O14FT9/Ii5/EoBpfh/y4mtvYIzuSCwiIr7rtINNfX09hYWFriubiouLKSwspLS0FID8/Hxmz57NSy+9xIYNG7j11ltpaGhwXSXVWWbOnElRURErVqzo1O14nV7jONj/Cvwsw1VVf+Q/a0//knoRERFvcdqXe69cuZKJEye6Pufn5wNw/fXXM2fOHKZNm0Z1dTWzZs2ioqKC4cOH88477xxzQrF0nbBL/peWrW+Tw2Z+/Z8/cdGgXxMefNr/9CIiIh7PMj7WN1FXV0d0dDS1tbW6QuooLR89RuDi/6XcxPHPnH9z17dH2l2SiIiIi7t+vz3yqihxv8Dz7qAxvCep1j5Cl/2BbdX1dpckIiLidj4TbHTy8EkEhhA69TEAbvL7L8++/r5OJBYREZ+jrqjuxBgaX7iU0LIlvNc2Cse0fzJ5iC7DFxER+6krSk6fZRF66e9os/y52H8V777xDxqb2+yuSkRExG0UbLqbxAGYnFsAuK1pNs9/uMHmgkRERNxHwaYbCph4L03BcfT1K+fAx89y3+vrqG3Us6RERMT7+Uyw0cnDpyE0hqCLHwDg5wFziVn1DJOf+JB31p/6g0pFREQ8kU4e7q4cDnj1OtjwHwDWOvrw85Zb6J01hocuG0xSVIjNBYqISHeik4fl7Pj5wTV/g8ufw4REk+23nf8E/ZL+G59l0pMLeXl5qS4HFxERr6MWG4G6cngrHzYtAGCDI4OftfyQqN6jeeTKoWQmhNtcoIiI+Dq12Ij7RKXCd/8J3/kLJjSOQX6lvBH0a8aXPsu3n1rIc4u20drmsLtKERGRk1KwESfLgqFXYc1cDlmXE2A5uC3gDeb53cu77/6Xywo+Zf2uWrurFBEROSGfCTa6KspNIhLhmpfgmr9iwpPo57eLfwffz2VVz3FNwUc8smCDbuonIiIeS+fYSMcO7oN37oMvXgZguyOFe1puoSpuJI9cMZRx5yTYXKCIiPgKnWMjnS8sDq58Hq6dC5Gp9PGrYG7wb7ih9jlu/vNifv7aWmoP6sZ+IiLiORRs5OQGTIYffw4jvo8fhhsD3uXdoHvYufodLnpyMW99Ua5Lw0VExCOoK0pOz7YP4c07oLYMgH+2foNHWr/HuVm9+c1lQ0iJ1o39RETk9KkrSuzR9xvw46WQczMA3wv4kPeCf07Lxnf55pOL+fvnJTgcPpWVRUTEi6jFRs7cjk/gjdtgfzEAr7Wdz0Mt/8PAzAwe+c5Q+iZG2FygiIh4C7XYiP0yz4NbP4NzZ2KwuMp/CQuDf0506XtMeepjnvlwC82turGfiIh0HZ8JNrqPjU2CwmDyb7Fueg8S+pNo1TA76El+5/cH/vLeSi595hMKy2rsrlJERLoJdUWJ+7QcgsWPYj79A5ZxsI8oft18A2+bXL5/bib/c24v+iVH2l2liIh4IHf9fivYiPvtWg1vzISqIgDebsthVsuNVBPDwJRIpmancWl2GulxYTYXKiIinkLBpgMKNh6itRk+/h18/AQ4Wmm0wni7bTTzW8fymWMwrQQwIiOGS7PTuGRYKkmRukxcRKQ7U7DpgIKNh6lY52y9KV/rGlXnF82bLaN5s3UcK8wALMuPc/vEc2l2GlOGpBIdFmhjwSIiYgcFmw4o2HgghwPKPod1r0HRfDi41zVpj18885vH8GbbOL4wfQj09+OC/olMzU4jb1Ay4cEB9tUtIiJdRsGmAwo2Hq6tFYoXw/rXYcN/oKnWNWm3XwqvN+fyZts4Npt0QgP9uWhQEpdmp3HBgESCA/xtLFxERDqTgk0HFGy8SGsTbP0A1v8bNr0NLQddk7ZbGbzenMt/HGMpMSlEhgQweXAKlw5PY2yfeAL8feZOBSIigoJNhxRsvFRzgzPcrH8dtr4Pbc2uSRusvvy7+VzeajuXcuJJiAjikqGpXDo8jRHpsfj5WTYWLiIi7qBg8zUFBQUUFBTQ1tbG5s2bFWy8WWMNbPyvsyVn+2Iwba5JqxnI6y1jebttDHuJpkdMKN/OTuXS7DSyUqOwLIUcERFvpGDTAbXY+Jj6aucJx+tfh9LPXKMd+LHUDGF+67m825ZDHeH0TQzn28PSuHBAIkN7RKu7SkTEiyjYdEDBxofV7oQv5zlbcnavcY1uIZDFjmG80TqWDxwjaSSEiOAAxvSOY1zfeMb2jWdQSpS6rEREPJiCTQcUbLqJvdvgy9dh3b+heoNrdLMVRKHpx9LWASx3DGC1ox+NhBAbFsi5feIPB50E+iaGq9tKRMSDKNh0QMGmG6oscrbirH8N9u9oN6kVf4pMbz5vG8AKh3OoIZKkyGDG9Y1nXN8ExvaN1+MdRERspmDTAQWbbswYqN7kPBenZCmUfAZ1O4+ZbYvpyfK2ASxzDGSFYyDlxNMzNrRd0EmO0iMeRES6koJNBxRspJ2aUmfIORJ29mw6ZpadJoHljoGscAxguWMg20wafRMjGNc3gXF94zm3Tzyx4UE2FC8i0n0o2HRAwUZOqGEPlH7ubM0p/QzKv2h3OTnAXhPJSofzHJ3ljkFsoBf9U2KdLTrnxJOTGUdkiJ5nJSLiTgo2HVCwkdPSVA87lx9u1VkKO1dA66F2s9SbEFY7+jnP0TED+YJ+9O+RyLCe0fRPjmRgSiT9kiOJDlXYERE5Uwo2HVCwkbPS2gzlhVDyqTPslH0Oh2rbzdJs/Fln+vClI5PNpiebHT3ZbHoSFp1I/5RIBqREMiDZ+do3MYKQQD3jSkTkZBRsOqBgI27lcEBVkbM1p+Qz5+uB8uPOWm2iXSFni+nJZkcPttKTuIRkBqZEulp3+idH0is+HH/dV0dExEXBpgMKNtKpjHFeUl62zBl4qjY676NTU9rhIhUmls2Ow2HH9GSLowc7/DNIS05iQHIUA1IiGJASxYDkSJKjgnV/HRHplhRsvkbPihJbNdU7LzWv3gBVG6B6I6ZqI9ZxLjc/YpeJZ4urhacHmx09qQzOJCMliQEpkc5ureRI+iVF6KosEfF5CjYdUIuNeJRDtc7AczjsULUBU70Rq4PuLHBefr7Jke7qztpierI/tBepSQn0TYygT2I4fRMj6JsYQc/YUD0TS0R8goJNBxRsxCs07v+qG+vwq6naiNVQ1eEiFSaWbY40tpmvhhKrByFx6fRNivwq8CQ5w0+ULkkXES+iYNMBBRvxag1723VnUbURs2fzCQNPgwlmu0llu0lrF3zqwzPomRRHn8OtO30PB58eMaF6IKiIeBwFmw4o2IhPatwPe7bCns3OYe9WzJ7NsG87lqP1uIs4jMVOk8A2k+YMPYeDT5l/D6Li0+ibFOkMO0kR9EmIoHdiOBHBAV38xUREnBRsOqBgI91KW4vzKq0jgedw+DHVm7CaajtcrNaEuYKOM/Skst2k0hjag8T4WDLiwugVH0Z6XJjrfXJkiFp6RKTTKNh0QMFGBOdl6Q17jgo8W2DvFmcrz/4SLDr+z77aRLHTJFFmEg8PzvcVVjJ+sen0jI8iI84ZenrFhx9+H0pYkFp7ROTMKdh0QMFG5CRaDsG+bV8Fnj1bYM8mzL5irKa6Ey7aZizKiWenSWSnSaTM8VX4ORjek9C4HqTHR5BxuKUnIz6MXnFhJEbq/jwicmIKNh1QsBE5C437YX8J1JS0ezU1pZj9Jfi1HTrh4k0mgF0mwRl6TNLh10Qq/ZMx0b2Ijk8hPT6cnrGhpMUcGUJICA9WN5dIN+eu32+1HYvIV0JjnUPa8HajLcAyBuqrjgo9O1zhx7GvBKtuJ8G00seqoA8Vx667HhoOBFNanMQOk0KxSeUjk0KxI4Wd/j0IiUoiLTaMHocDj+s1NpTU6BA9c0tETolabETEPdpa4cDuY1p8HPt34NhXQkDDccLOUepMGMUmxTk4Ug+/T2WHSeEAYSREBLnCTtpR4afH4VafuPAgdXeJeDF1RXVAwUbEQ7UcgtqdsG+78xyfvducl63v3Qq1O096QnOxSaXY4Qw62w+HnhKTTBPOx02EBPp91dIT7Wzp6Z0QTp/EcHonhOvkZhEPp64oEfEugSGQcI5zOIoFztCzvxj2bnUFHvZucwag+koSrToSrTrG+G1qt6wDi0ri2daW7Aw++1Ip3pvCMpPCTpNI61F/4tKiQ+hz+JEUfRLCXe/TonXDQhFfohYbEfFsh+qOauE5HHr2HX491PG9etrwp8o/ie1tSWxtTaLUJLPDJFNikikzSe1aejLjw13P4XIGH+f7SD2WQqTLqCuqAwo2It2EMXBw71EtPFvbB6DWxg4XdWCxx4pne1sSxY5kV+gpNcmUmCTqCQMgMTLY1brT96jQo4ePirifgk0HFGxEBIcDDpQ7u7f2bT88FH/12nzghIvvJZodjiRn2HEcaelJocQksZ9IAv39yIgLc3VnZcQ5r+Y6chm7zucROX0KNh1QsBGREzrS0nN02Nl/VOg5uOeEi9eZMEpMkivo7Dh8Ps9Ok0C5iaeVAOLCg0iLCTl81VYYPWJDXcGnR0woMWGBuoJL5GsUbL6moKCAgoIC2tra2Lx5s4KNiJyZQ7XOgHN02DkSgA7sPuGibcaiklh2mkR2mQTXzQqPvO428TQRRFiQvzP0HA467YNPGEmRumGhdD8KNh1Qi42IdJqWRudDR7/e0lNTCjVl0NZ00lVUmZjDQad96Nl5OAg1EkKgv0VKdPsWn56H792TGhNCWnQooUG6YaH4FgWbDijYiIgtHA5oqIbaMucNCmvKnIGntuyr4NPScNLV7DWRx7T0HN3iU0cYYBEbFkhqtPPmhM7Xr96nRoeQEh1CoAec4NzU2sbe+mb21jezp6Hp8Psm9jY00+YwjO0Tz7hz4nVekijYdETBRkQ8kjFwcB/Uln4VdNoFn1I4yUNIARpMCLtMPOUmnt1HXjnq/eHuLsuCpMjgduEn9XArUGpMKGnRISREnH6Xl8NhqG1sYW9DE3sOB5a9DU3sOdDEnobDoaW+mb0Nzeypb+LAodaTrjPI34/cPnFMHJDExIFJ9E4IP62axDco2HRAwUZEvFZjTfsWnprS9kGocd8prWa/iTwq/MQdfk1wva8kllYCCPS3SI5ydm2lxYS4Ak9wgP8xrSt76p1BZd/hlpbTEehvER8eTHxEEPERwSSEBxEfEcTB5jYWb65m5/72l+Znxodx4eGQk9s7Ts8J6yYUbDqgYCMiPqu5Aep2O8NP7S6o2+V8TEXtzsPvd51Sd1cbflSZGMpNXLuWniPv95hoGgniICE0Ecjh+0O3Ex0aSHxEEAmuwBJEfHgwCZFHgkuwa3pUaECHV4EZY9hWXc+iTdV8tKmK5cX7aGn76mcpJNCP8X0TuHBAIhcOSCI9LuyMd594NgWbDijYiEi3ZQwcqjkcdnZB3eFXV/DZ6QxGjpbTWm2rfyiOgFAIDMMvKAy/4HD8gsIg0DmOwDAIOvx69LjAUAgKPzwuFALDv5oeFAbBURAc0W5b9U2tfLp1D4s2VfHRxmoq6g61m35OUgQTByQycUASozPjCAqw/zwicQ8Fmw4o2IiInIDDAQ1Vxwk+R71v3AdtzV1TT1QPSBwISYOOeh0AwZEYY9hYcYCPNlWxaGM1q0r3t+sGCw/y57x+CUwckMSFA5JIiQ7pmpqlUyjYdEDBRkTEDdpaoeWg8xL3dq+H3zc3HDuu5SA0f33eI5+PXqbR2WXmOMGJxdHph4POQEgcBEkDqY3oyycljc6gs6maPfXtL68flBrFxMNdViMzYvTYCy+jYNMBBRsRES/RuB+qN0HVBqje+NVrfWXHy8RkQOIgTOJAdgb24pPaRObvCmf5zkMc/WsWFRLAhP7OLqvz+yeQFKnWHE+nYNMBBRsRES93cF/7oHPktaG6gwUs2qJ7URXam/UtaXy0L57CQylsM2mup7iHBfmTFBlM4uEhKTLE9T4xMpjEiGCSooKJDw/GX3d9toWCTQcUbEREfFTDXqjecFTg2ej8fHDvcWd34Ee5XwpftqRSbuKoNjFUE0O1iXa+N9HsJZoWvro5oJ8F8RHOoOMMQEe/hrQbFx6smwq6k4JNBxRsRES6mfrqw4FnY/vXxv2ntHitFcUeE02FI4oqE+MKPVWuIOT8XEMER1/6Hh7k364FKCkqmO/mZDAgJbKTvqhvU7DpgIKNiIhgDNRXOQPOni3O83bqK53jDlQ4XxuqTnwC89e0EsA+K5oqRzQVjsOtPhxp/YmhysRg0kYw7/YLO+97+TB3/X6rHU1ERHyPZUFksnPoc+Hx53E4nK06R4eeY95XQX0FNO4ngFaSzF6SrL0M6eBmyK9XnUfJ3hx6xeuxEHZRsBERke7Jzw/C451DctaJ521tcp68fEzoORyE6sph10ou8VvG39Zs5ua8EV3zHeQYCjYiIiInExAM0T2dw/EYQ+0TI4mu307tmjdAwcY2unuRiIjI2bIsgoZdBcDIug8p3nPyZ3ZJ51CwERERcYPQEVcDcJ7fehauKrK5mu5LwUZERMQdEvuzP2oggVYb9YXz7a6m21KwERERcZOQw602OfUfsbWq3uZquicFGxERETcJHe48z+ZcvyIWrVpnczXdk4KNiIiIu8RmsjdmGP6W4dDa1+2uplvyuGBTVlbGhRdeSFZWFsOGDePVV1+1uyQREZFTFjbyGgByDy5ic+UBm6vpfjwu2AQEBPDUU09RVFTEe++9x5133klDgy6bExER7xA6/Ds4sMjx28ySFWvsLqfb8bhgk5qayvDhwwFISUkhISGBffv22VuUiIjIqYpKY2/8KADa1r2Ojz2S0eOddrBZsmQJU6dOJS0tDcuymD9//jHzFBQUkJmZSUhICLm5uSxfvvyMilu1ahVtbW2kp6ef0fIiIiJ2iBzl7I46t3Exmyt1dVRXOu1g09DQQHZ2NgUFBcedPnfuXPLz87n//vtZvXo12dnZTJo0iaqqKtc8w4cPZ8iQIccMu3fvds2zb98+rrvuOv70pz+dsJ6mpibq6uraDSIiInYKGXYlbfiR7bedT5Yts7ucbsUyZ9FGZlkW8+bN4/LLL3eNy83NJScnh2eeeQYAh8NBeno6t99+O/fee+8prbepqYlvfvObzJgxg+9///snnPeBBx7gwQcfPGb82T72XERE5GxUFUwhqfoz/hI0nR/cV4BlWXaX5NHq6uqIjo4+699vt55j09zczKpVq8jLy/tqA35+5OXlsXTp0lNahzGGG264gW984xsnDTUA9913H7W1ta6hrKzsjOsXERFxl6ic7wIw7tASNpTr6qiu4tZgs2fPHtra2khOTm43Pjk5mYqKilNax6effsrcuXOZP38+w4cPZ/jw4axb1/FNjoKDg4mKimo3iIiI2C1k6KW0EsAgvzI+X/aJ3eV0GwF2F/B15513Hg6Hw+4yREREzk5oLHtSJpBS8RH+RfMwl09Rd1QXcGuLTUJCAv7+/lRWVrYbX1lZSUpKijs3dYyCggKysrLIycnp1O2IiIicqpicaQBMaFrCl7tqba6me3BrsAkKCmLUqFEsXLjQNc7hcLBw4ULGjh3rzk0dY+bMmRQVFbFixYpO3Y6IiMipChnybZqtIPr4VbDi88V2l9MtnHZXVH19PVu3bnV9Li4uprCwkLi4ODIyMsjPz+f6669n9OjRjBkzhqeeeoqGhgZuvPFGtxYuIiLi8YIj2Zs2kdRd7xK0cR7GXKruqE522sFm5cqVTJw40fU5Pz8fgOuvv545c+Ywbdo0qqurmTVrFhUVFQwfPpx33nnnmBOKRUREuoO4Md+Fee9yQcvHrNtZw7D0WLtL8mlndR8bT+Su6+BFRETcovkghx7pQ4hpZE7Wn7nhmqvtrsgjeeR9bOykk4dFRMQjBYWxL915f7ewTfP17KhOphYbERGRTtZctICgV66l0sSw+8ZVjMhMsLskj6MWGxERES8R1D+Pg34RJFs1rP/sHbvL8WkKNiIiIp0tIIh9GZMACN/6Jg6HT3WWeBQFGxERkS6QNPZ7AFzQ9hlrSqpsrsZ3KdiIiIh0gaBzLuSAfwzx1gE2fPpfu8vxWT4TbHRVlIiIeDT/AGoyvwVA9Pb/qjuqk+iqKBERkS7SvO1jgv72bepMGJuvW83ovql2l+QxdFWUiIiIlwnqPZ7agASirINs/nS+3eX4JAUbERGRruLnR12fbwMQV/xf2tQd5XYKNiIiIl0oZfx0ACY4VrBq6y6bq/E9CjYiIiJdKDAjh32BKYRbTWz/7HW7y/E5PhNsdFWUiIh4Bcuivt9lACSVqDvK3XRVlIiISBdr2bWWwNnnc8gEsnbaCnKzettdku10VZSIiIiXCkwbRlVwBiFWCyWf/9vucnyKgo2IiEhXsywa+zu7o1JL36K1zWFzQb5DwUZERMQGPQ5fHXWuWcvKDdttrsZ3KNiIiIjYICBlEOUhfQm02ti59BW7y/EZCjYiIiI2aR54BQA9dy2gRd1RbuEzwUaXe4uIiLfpcZ6zOyrHrGfF+k02V+MbfCbYzJw5k6KiIlasWGF3KSIiIqckIKEPO8MG4W8ZKj+fa3c5PsFngo2IiIg3as1ydkdllL9Dc6u6o86Wgo2IiIiN0g93R41iAyu/WGdzNd5PwUZERMRG/jE92RGeDUD1MnVHnS0FGxEREbsNuRKA3pXv0tTaZnMx3k3BRkRExGYZ46+lDT+GsZWVa1bbXY5XU7ARERGxmV9UMjsiRwFQs1zdUWdDwUZERMQD+A9zdkf1rXqPQy3qjjpTCjYiIiIeIGPcd2nFn4FWCatWfm53OV7LZ4KN7jwsIiLezC88ju3RuQDUrVR31JnymWCjOw+LiIi3C8q+GoABe97nUHOrzdV4J58JNiIiIt6u17jv0EQgfazdrFr+sd3leCUFGxEREQ9hhURTHDsegIZVr9hcjXdSsBEREfEgISOuASBr3/s0Nqk76nQp2IiIiHiQXudezkFC6GlVs/rzhXaX43UUbERERDyIFRROcfz5ABxao+6o06VgIyIi4mEiRjm7o4bsX0jDoWabq/EuCjYiIiIeJiNnKgcII9naT+Gnb9tdjldRsBEREfEwVmAIOxInAtCy9t82V+NdFGxEREQ8UOTo7wIwtPYj6hsP2VyN9/CZYKNHKoiIiC/pNWoyNUQRb9Xxxcf/sbscr+EzwUaPVBAREV9iBQSxIzkPgLZ16o46VT4TbERERHxN3Bhnd9SwuiUcaGiwuRrvoGAjIiLiodKHX8QeK5Zoq4H1S+bZXY5XULARERHxUJZ/AKUpk5zv179uczXeQcFGRETEgyWcey0AQ+o/pbauzuZqPJ+CjYiIiAfLGHYBFVYSEdYhiha/Znc5Hk/BRkRExJNZFjvTJgMQsEHdUSejYCMiIuLhksZNB2BIwzJq9++zuRrPpmAjIiLi4TKyctnpl0ao1cyGJXri94ko2IiIiHg6y2J3z28BELRRl32fiIKNiIiIF0gdf7g76uAKavZW2VyN51KwERER8QLpA0ZS7J9JkNXGpkX/srscj6VgIyIi4iUq053dUeGbXsc4HDZX45kUbERERLxExvnfp81YDGku5PM599hdjkdSsBEREfESaX2yWDnIGWjGlv6Jz//1vzZX5HkUbERERLxI7nfvY2mvHwFw7qbHWDH/WZsr8iw+E2wKCgrIysoiJyfH7lJEREQ61bnXP8LnSdMAGLHml6x5/x82V+Q5LGOMsbsId6qrqyM6Opra2lqioqLsLkdERKRTONraWPXH75FT+w5NJpAtF89hyPhv213WGXPX77fPtNiIiIh0J37+/oy47W+sCRtHsNVC5ns3s2XNErvLsp2CjYiIiJcKCAxi0O2v8WVQNhFWIwlvfI+SjavtLstWCjYiIiJeLCQ0nF63vcHmgP7EcoCQl6+ivGST3WXZRsFGRETEy0VExZL4wzcp8Usnmb20zrmcPZU77S7LFgo2IiIiPiA2MZWQH7xBOYmkm93U/mkqtTV77S6ryynYiIiI+Ijknn1p/Z957CWavm3b2VVwKY0NB+wuq0sp2IiIiPiQ9HOGUnPly9QRRlbLejY/cyXNTU12l9VlFGxERER8TN9h49g9ZQ6NJojsxuV88cy1tLW12V1Wl1CwERER8UEDcyex5cJnaTH+jD6wkJXP/qBbPBFcwUZERMRHDZt4NV+M+X84jEXu3vl8/pd8u0vqdAo2IiIiPmzUJTNYMeRXAIzd9SKf//0BewvqZAo2IiIiPi736p+xtPdtAJy79feseP0PNlfUeRRsREREuoFzv/8bPk+ZDsDItfez5t2XbK6ocyjYiIiIdAOWnx+5tzzD8thL8LcMgz/LZ92SN+wuy+0UbERERLoJy8+PUTP/yuqI8wmyWum7cAabVn5od1lupWAjIiLSjfgHBDD4trmsCx5JmNVEyn//h+KiFXaX5TYKNiIiIt1McEgYfW6bx6aAgUTTQMQrV7Nr+0a7y3ILBRsREZFuKDwyhpRb36TYrxeJ7Ie/Xcqe3SV2l3XWFGxERES6qej4ZCJv/g+7rGR6mErq/nwptfuq7C7rrCjYiIiIdGMJab3g+29QTSx9HDsof/ZSDtbX2l3WGVOwERER6eZ69BlE/TWvUks4A1s3sPWZK2g6dNDuss6IxwWbmpoaRo8ezfDhwxkyZAizZ8+2uyQRERGf1zsrh/Jv/42DJphhh1bx5TPTaGtttbus02YZY4zdRRytra2NpqYmwsLCaGhoYMiQIaxcuZL4+PhTWr6uro7o6Ghqa2uJiorq5GpFRER8y7ol8xiw8CaCrDaWxnybATe/QFxEcKdv112/3x7XYuPv709YWBgATU1NGGPwsOwlIiLis4aefwXrz32CNmMxtua/PPz/fsPPX1vLl7u947yb0w42S5YsYerUqaSlpWFZFvPnzz9mnoKCAjIzMwkJCSE3N5fly5ef1jZqamrIzs6mZ8+e3H333SQkJJxumSIiInKGRk65kZKhtwPwC7+XeG/lBi754ydc839LeeuLclrbHDZX2LHTDjYNDQ1kZ2dTUFBw3Olz584lPz+f+++/n9WrV5Odnc2kSZOoqvrq8rEj5898fdi9ezcAMTExrF27luLiYv75z39SWVl5hl9PREREzkSfy3+NSRxEglXH80mvE+BnsXzHPmb+czU/mVtod3kdOqtzbCzLYt68eVx++eWucbm5ueTk5PDMM88A4HA4SE9P5/bbb+fee+897W38+Mc/5hvf+AZXXXXVcac3NTXR1NTk+lxXV0d6errOsRERETlbZcvhLxcDhv3feYUXyzP55/JSHr58KJOHpLh1Ux55jk1zczOrVq0iLy/vqw34+ZGXl8fSpUtPaR2VlZUcOHAAgNraWpYsWcKAAQM6nP+RRx4hOjraNaSnp5/dlxARERGn9DEwZgYAsR/+nPwL0/n03m/wzaxkmwvrmFuDzZ49e2hrayM5uf0XTk5OpqKi4pTWUVJSwoQJE8jOzmbChAncfvvtDB06tMP577vvPmpra11DWVnZWX0HEREROcpFsyCqB+zfAYsfJTjAH38/y+6qOhRgdwFfN2bMGAoLC095/uDgYIKDO/8yNBERkW4pOBIueRL+NQ0+ewaGfAdSs+2uqkNubbFJSEjA39//mJN9KysrSUlxb1+ciIiIdJEBk2HwFWDa4M3boc1zb9zn1mATFBTEqFGjWLhwoWucw+Fg4cKFjB071p2bOkZBQQFZWVnk5OR06nZERES6pcn/D0KioXwtLHvO7mo6dNrBpr6+nsLCQld3UXFxMYWFhZSWlgKQn5/P7Nmzeemll9iwYQO33norDQ0N3HjjjW4t/OtmzpxJUVERK1as6NTtiIiIdEuRyXDx/4JfILQ1211Nh077HJuVK1cyceJE1+f8/HwArr/+eubMmcO0adOorq5m1qxZVFRUMHz4cN55551jTigWERERLzPif6DXOIjva3clHfK4Z0WdLT0rSkRExPt45H1sREREROzkM8FGJw+LiIiIuqJERETEduqKEhEREfkaBRsRERHxGQo2IiIi4jN8Jtjo5GERERHRycMiIiJiO508LCIiIvI1CjYiIiLiMxRsRERExGco2IiIiIjPULARERERnxFgdwHuUlBQQEFBAa2trYDz7GoRERHxDkd+t8/2Ym2fu9x7586dpKen212GiIiInIGysjJ69ux5xsv7XLBxOBzs3r2byMhILMty67rr6upIT0+nrKys294jR/vASftB++AI7QftgyO0H85uHxhjOHDgAGlpafj5nfmZMj7TFXWEn5/fWSW9UxEVFdVtD9ojtA+ctB+0D47QftA+OEL74cz3QXR09FlvWycPi4iIiM9QsBERERGfoWBzGoKDg7n//vsJDg62uxTbaB84aT9oHxyh/aB9cIT2g2fsA587eVhERES6L7XYiIiIiM9QsBERERGfoWAjIiIiPkPBRkRERHyGgs0pKigoIDMzk5CQEHJzc1m+fLndJZ2xRx55hJycHCIjI0lKSuLyyy9n06ZN7ea58MILsSyr3fCjH/2o3TylpaVccsklhIWFkZSUxN133+16VtcRixYtYuTIkQQHB3POOecwZ86czv56p+SBBx445vsNHDjQNf3QoUPMnDmT+Ph4IiIi+M53vkNlZWW7dXjz9z8iMzPzmP1gWRYzZ84EfPM4WLJkCVOnTiUtLQ3Lspg/f3676cYYZs2aRWpqKqGhoeTl5bFly5Z28+zbt4/p06cTFRVFTEwMN910E/X19e3m+eKLL5gwYQIhISGkp6fz2GOPHVPLq6++ysCBAwkJCWHo0KEsWLDA7d+3IyfaDy0tLdxzzz0MHTqU8PBw0tLSuO6669i9e3e7dRzv+Hn00UfbzePJ++Fkx8INN9xwzPebPHlyu3l8/VgAjvs3wrIsHn/8cdc8HnUsGDmpl19+2QQFBZkXXnjBfPnll2bGjBkmJibGVFZW2l3aGZk0aZJ58cUXzfr1601hYaH51re+ZTIyMkx9fb1rngsuuMDMmDHDlJeXu4ba2lrX9NbWVjNkyBCTl5dn1qxZYxYsWGASEhLMfffd55pn+/btJiwszOTn55uioiLz9NNPG39/f/POO+906fc9nvvvv98MHjy43ferrq52Tf/Rj35k0tPTzcKFC83KlSvNueeea8aNG+ea7u3f/4iqqqp2++D99983gPnoo4+MMb55HCxYsMD88pe/NK+//roBzLx589pNf/TRR010dLSZP3++Wbt2rbn00ktN7969TWNjo2ueyZMnm+zsbPP555+bjz/+2Jxzzjnm2muvdU2vra01ycnJZvr06Wb9+vXmX//6lwkNDTXPP/+8a55PP/3U+Pv7m8cee8wUFRWZX/3qVyYwMNCsW7eu0/eBMSfeDzU1NSYvL8/MnTvXbNy40SxdutSMGTPGjBo1qt06evXqZR566KF2x8fRf0c8fT+c7Fi4/vrrzeTJk9t9v3379rWbx9ePBWNMu+9fXl5uXnjhBWNZltm2bZtrHk86FhRsTsGYMWPMzJkzXZ/b2tpMWlqaeeSRR2ysyn2qqqoMYBYvXuwad8EFF5if/OQnHS6zYMEC4+fnZyoqKlzjnnvuORMVFWWampqMMcb8/Oc/N4MHD2633LRp08ykSZPc+wXOwP3332+ys7OPO62mpsYEBgaaV1991TVuw4YNBjBLly41xnj/9+/IT37yE9O3b1/jcDiMMb5/HHz9j7jD4TApKSnm8ccfd42rqakxwcHB5l//+pcxxpiioiIDmBUrVrjmefvtt41lWWbXrl3GGGOeffZZExsb69oHxhhzzz33mAEDBrg+X3PNNeaSSy5pV09ubq754Q9/6NbveCqO92P2dcuXLzeAKSkpcY3r1auX+f3vf9/hMt60HzoKNpdddlmHy3TXY+Gyyy4z3/jGN9qN86RjQV1RJ9Hc3MyqVavIy8tzjfPz8yMvL4+lS5faWJn71NbWAhAXF9du/D/+8Q8SEhIYMmQI9913HwcPHnRNW7p0KUOHDiU5Odk1btKkSdTV1fHll1+65jl6vx2Zx1P225YtW0hLS6NPnz5Mnz6d0tJSAFatWkVLS0u72gcOHEhGRoardl/4/l/X3NzM3//+d37wgx+0e4Csrx8HRysuLqaioqJdvdHR0eTm5rb7t4+JiWH06NGuefLy8vDz82PZsmWuec4//3yCgoJc80yaNIlNmzaxf/9+1zzesl/A+XfCsixiYmLajX/00UeJj49nxIgRPP744+26IX1hPyxatIikpCQGDBjArbfeyt69e13TuuOxUFlZyVtvvcVNN910zDRPORZ87iGY7rZnzx7a2tra/eEGSE5OZuPGjTZV5T4Oh4M777yT8ePHM2TIENf4733ve/Tq1Yu0tDS++OIL7rnnHjZt2sTrr78OQEVFxXH3yZFpJ5qnrq6OxsZGQkNDO/OrnVBubi5z5sxhwIABlJeX8+CDDzJhwgTWr19PRUUFQUFBx/wBT05OPul3OzLtRPN4wvc/nvnz51NTU8MNN9zgGufrx8HXHan5ePUe/X2SkpLaTQ8ICCAuLq7dPL179z5mHUemxcbGdrhfjqzDkxw6dIh77rmHa6+9tt2DDe+44w5GjhxJXFwcn332Gffddx/l5eU8+eSTgPfvh8mTJ3PllVfSu3dvtm3bxi9+8QumTJnC0qVL8ff375bHwksvvURkZCRXXnllu/GedCwo2HRzM2fOZP369XzyySftxt9yyy2u90OHDiU1NZWLLrqIbdu20bdv364u0+2mTJniej9s2DByc3Pp1asXr7zyikf90Halv/zlL0yZMoW0tDTXOF8/DuTkWlpauOaaazDG8Nxzz7Wblp+f73o/bNgwgoKC+OEPf8gjjzziE48V+O53v+t6P3ToUIYNG0bfvn1ZtGgRF110kY2V2eeFF15g+vTphISEtBvvSceCuqJOIiEhAX9//2OuiKmsrCQlJcWmqtzjtttu47///S8fffQRPXv2POG8ubm5AGzduhWAlJSU4+6TI9NONE9UVJTHhYeYmBj69+/P1q1bSUlJobm5mZqamnbzHP1v7mvfv6SkhA8++ICbb775hPP5+nFwpOYT/feekpJCVVVVu+mtra3s27fPLceHJ/1dORJqSkpKeP/999u11hxPbm4ura2t7NixA/Cd/XBEnz59SEhIaHf8d5djAeDjjz9m06ZNJ/07AfYeCwo2JxEUFMSoUaNYuHCha5zD4WDhwoWMHTvWxsrOnDGG2267jXnz5vHhhx8e0zx4PIWFhQCkpqYCMHbsWNatW9fuP+ojf/iysrJc8xy9347M44n7rb6+nm3btpGamsqoUaMIDAxsV/umTZsoLS111e5r3//FF18kKSmJSy655ITz+fpx0Lt3b1JSUtrVW1dXx7Jly9r929fU1LBq1SrXPB9++CEOh8MV/MaOHcuSJUtoaWlxzfP+++8zYMAAYmNjXfN48n45Emq2bNnCBx98QHx8/EmXKSwsxM/Pz9U94wv74Wg7d+5k79697Y7/7nAsHPGXv/yFUaNGkZ2dfdJ5bT0WTutU427q5ZdfNsHBwWbOnDmmqKjI3HLLLSYmJqbdlSDe5NZbbzXR0dFm0aJF7S7NO3jwoDHGmK1bt5qHHnrIrFy50hQXF5s33njD9OnTx5x//vmudRy5zPfiiy82hYWF5p133jGJiYnHvcz37rvvNhs2bDAFBQUec7nzXXfdZRYtWmSKi4vNp59+avLy8kxCQoKpqqoyxjgv987IyDAffvihWblypRk7dqwZO3asa3lv//5Ha2trMxkZGeaee+5pN95Xj4MDBw6YNWvWmDVr1hjAPPnkk2bNmjWuq30effRRExMTY9544w3zxRdfmMsuu+y4l3uPGDHCLFu2zHzyySemX79+7S7xrampMcnJyeb73/++Wb9+vXn55ZdNWFjYMZe2BgQEmN/97ndmw4YN5v777+/SS3xPtB+am5vNpZdeanr27GkKCwvb/Z04clXLZ599Zn7/+9+bwsJCs23bNvP3v//dJCYmmuuuu85r9sOJ9sGBAwfMz372M7N06VJTXFxsPvjgAzNy5EjTr18/c+jQIdc6fP1YOKK2ttaEhYWZ55577pjlPe1YULA5RU8//bTJyMgwQUFBZsyYMebzzz+3u6QzBhx3ePHFF40xxpSWlprzzz/fxMXFmeDgYHPOOeeYu+++u939S4wxZseOHWbKlCkmNDTUJCQkmLvuusu0tLS0m+ejjz4yw4cPN0FBQaZPnz6ubdht2rRpJjU11QQFBZkePXqYadOmma1bt7qmNzY2mh//+McmNjbWhIWFmSuuuMKUl5e3W4c3f/+jvfvuuwYwmzZtajfeV4+Djz766LjH//XXX2+McV7y/etf/9okJyeb4OBgc9FFFx2zb/bu3WuuvfZaExERYaKiosyNN95oDhw40G6etWvXmvPOO88EBwebHj16mEcfffSYWl555RXTv39/ExQUZAYPHmzeeuutTvveX3ei/VBcXNzh34kj9zhatWqVyc3NNdHR0SYkJMQMGjTI/Pa3v233o2+MZ++HE+2DgwcPmosvvtgkJiaawMBA06tXLzNjxoxj/ofW14+FI55//nkTGhpqampqjlne044FyxhjTq+NR0RERMQz6RwbERER8RkKNiIiIuIzFGxERETEZyjYiIiIiM9QsBERERGfoWAjIiIiPkPBRkRERHyGgo2IiIj4DAUbERER8RkKNiLidtXV1QQFBdHQ0EBLSwvh4eGUlpaecJmDBw9y33330bdvX0JCQkhMTOSCCy7gjTfecM2TmZnJU0891cnVi4g3C7C7ABHxPUuXLiU7O5vw8HCWLVtGXFwcGRkZJ1zmRz/6EcuWLePpp58mKyuLvXv38tlnn7F3794uqlpEfIFabETE7T777DPGjx8PwCeffOJ6fyJvvvkmv/jFL/jWt75FZmYmo0aN4vbbb+cHP/gBABdeeCElJSX89Kc/xbIsLMtyLfvJJ58wYcIEQkNDSU9P54477qChocE1PTMzk9/85jdce+21hIeH06NHDwoKClzTjTE88MADZGRkEBwcTFpaGnfccYe7doeIdCE9BFNE3KK0tJRhw4YBzm4lf39/goODaWxsxLIsQkJC+N73vsezzz573OUHDhxIdnY2f/7zn4mMjDxm+r59+8jOzuaWW25hxowZAKSkpLBt2zays7N5+OGHueSSS6iurua2224jOzubF198EXAGm3379vGLX/yCK6+8knfffZef/vSnvP3223zzm9/ktdde46abbuLll19m8ODBVFRUsHbtWtd2RMR7KNiIiFu0trayc+dO6urqGD16NCtXriQ8PJzhw4fz1ltvkZGRQUREBAkJCcddfsmSJUyfPp3Kykqys7M577zzuOqqq9q19mRmZnLnnXdy5513usbdfPPN+Pv78/zzz7vGffLJJ1xwwQU0NDQQEhJCZmYmgwYN4u2333bN893vfpe6ujoWLFjAk08+yfPPP8/69esJDAx0/84RkS6jrigRcYuAgAAyMzPZuHEjOTk5DBs2jIqKCpKTkzn//PPJzMzsMNQAnH/++Wzfvp2FCxdy1VVX8eWXXzJhwgR+85vfnHC7a9euZc6cOURERLiGSZMm4XA4KC4uds03duzYdsuNHTuWDRs2AHD11VfT2NhInz59mDFjBvPmzaO1tfUs9oaI2EUnD4uIWwwePJiSkhJaWlpwOBxERETQ2tpKa2srERER9OrViy+//PKE6wgMDGTChAlMmDCBe+65h4cffpiHHnqIe+65h6CgoOMuU19fzw9/+MPjnhNzshOWj0hPT2fTpk188MEHvP/++/z4xz/m8ccfZ/HixWrBEfEyCjYi4hYLFiygpaWFiy66iMcee4xRo0bx3e9+lxtuuIHJkyefUUDIysqitbWVQ4cOERQURFBQEG1tbe3mGTlyJEVFRZxzzjknXNfnn39+zOdBgwa5PoeGhjJ16lSmTp3KzJkzGThwIOvWrWPkyJGnXbeI2Efn2IiI21RUVJCZmUlNTQ2WZRETE8P27dtJTU096bIXXngh1157LaNHjyY+Pp6ioiLy8/Pp0aMHCxcuBODiiy8mNDSUZ599luDgYBISEvjiiy8499xz+cEPfsDNN99MeHg4RUVFvP/++zzzzDOA89yc/fv388tf/pLLL7+c999/n5/85Ce89dZbTJo0iTlz5tDW1kZubi5hYWG8+OKLPPHEE5SVlREfH9+p+0xE3Evn2IiI2yxatIicnBxCQkJYvnw5PXv2PKVQAzBp0iReeuklLr74YgYNGsTtt9/OpEmTeOWVV1zzPPTQQ+zYsYO+ffuSmJgIwLBhw1i8eDGbN29mwoQJjBgxglmzZpGWltZu/XfddRcrV65kxIgRPPzwwzz55JNMmjQJgJiYGGbPns348eMZNmwYH3zwAf/5z38UakS8kFpsRMTnHe9qKhHxTWqxEREREZ+hYCMiIiI+Q11RIiIi4jPUYiMiIiI+Q8FGREREfIaCjYiIiPgMBRsRERHxGQo2IiIi4jMUbERERMRnKNiIiIiIz1CwEREREZ/x/wENbcJMwGeyqgAAAABJRU5ErkJggg==",
278 |       "text/plain": [
279 |        "<Figure size 640x480 with 1 Axes>"
280 |       ]
281 |      },
282 |      "metadata": {},
283 |      "output_type": "display_data"
284 |     },
285 |     {
286 |      "data": {
287 |       "image/png": "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",
288 |       "text/plain": [
289 |        "<Figure size 640x480 with 1 Axes>"
290 |       ]
291 |      },
292 |      "metadata": {},
293 |      "output_type": "display_data"
294 |     }
295 |    ],
296 |    "source": [
297 |     "dde.saveplot(losshistory, train_state, issave=True, isplot=True)"
298 |    ]
299 |   },
300 |   {
301 |    "cell_type": "code",
302 |    "execution_count": null,
303 |    "id": "944ac26c",
304 |    "metadata": {},
305 |    "outputs": [],
306 |    "source": []
307 |   },
308 |   {
309 |    "cell_type": "code",
310 |    "execution_count": null,
311 |    "id": "51ff7e6e",
312 |    "metadata": {},
313 |    "outputs": [],
314 |    "source": []
315 |   },
316 |   {
317 |    "cell_type": "code",
318 |    "execution_count": null,
319 |    "id": "14d69ee8",
320 |    "metadata": {},
321 |    "outputs": [],
322 |    "source": []
323 |   }
324 |  ],
325 |  "metadata": {
326 |   "kernelspec": {
327 |    "display_name": "Python 3 (ipykernel)",
328 |    "language": "python",
329 |    "name": "python3"
330 |   },
331 |   "language_info": {
332 |    "codemirror_mode": {
333 |     "name": "ipython",
334 |     "version": 3
335 |    },
336 |    "file_extension": ".py",
337 |    "mimetype": "text/x-python",
338 |    "name": "python",
339 |    "nbconvert_exporter": "python",
340 |    "pygments_lexer": "ipython3",
341 |    "version": "3.9.13"
342 |   }
343 |  },
344 |  "nbformat": 4,
345 |  "nbformat_minor": 5
346 | }
347 | 


--------------------------------------------------------------------------------