Model training.ipynb 210 KB
Newer Older
Richard Weiss's avatar
Richard Weiss committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import requests\n",
    "from sklearn import decomposition\n",
    "from IPython import display\n",
    "import ignite\n",
    "from skorch import NeuralNetClassifier, NeuralNet\n",
    "from sklearn import model_selection\n",
    "from torch import nn\n",
    "import torch\n",
    "from sklearn import cluster, pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "if False:\n",
    "    display.clear_output(wait=True)\n",
    "    while True:\n",
    "        display.clear_output(wait=True)\n",
    "        res = requests.get(\"http://pizero.local\").json()\n",
    "        imshow(res['frame'], vmin=18, vmax=25)\n",
    "        colorbar()\n",
    "        plt.title(str(res['classif']))\n",
    "        plt.show()\n",
    "        display.display(plt.gcf())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "frames = []\n",
    "with open('./data.dat', 'r') as f:\n",
    "    for line in f:\n",
    "        try:\n",
    "            frame = json.loads(line)\n",
    "            frames.append(frame['frame'])\n",
    "        except json.JSONDecodeError:\n",
    "            print(f\"Failed at {len(frames)}\")\n",
    "            \n",
    "frames = frames[10:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x12e1ab5c0>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "imshow(frames[-100])\n",
    "colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_frames = np.array(frames).reshape(-1, 64)\n",
    "all_frames = all_frames.T\n",
    "all_frames -= all_frames.min(0)\n",
    "all_frames /= all_frames.max(0)\n",
    "all_frames = all_frames.T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x127c9d400>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "imshow(all_frames[-5].reshape(8,8))\n",
    "colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12e34d9b0>]"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "toob = pipeline.make_pipeline(\n",
    "    decomposition.PCA(4),\n",
    "    cluster.KMeans(5x)\n",
    ")\n",
    "whitened_frames = decomposition.PCA().fit_transform(all_frames)\n",
    "\n",
    "\n",
    "decomposed_frames = toob.fit_predict(all_frames)\n",
    "\n",
    "figure(figsize=(12, 3), dpi=100)\n",
    "plot(whitened_frames[:, 0], alpha=0.2)\n",
    "plot(whitened_frames[:, 1], alpha=0.2)\n",
    "plot(whitened_frames[:, 2], alpha=0.2)\n",
    "figure(figsize=(12, 3), dpi=100)\n",
    "plot(decomposed_frames)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[17.7500, 18.5000, 18.5000,  ..., 18.0000, 18.0000, 18.5000],\n",
       "        [17.7500, 18.0000, 18.2500,  ..., 17.7500, 17.7500, 18.2500],\n",
       "        [17.7500, 18.0000, 18.2500,  ..., 17.7500, 17.7500, 18.2500],\n",
       "        ...,\n",
       "        [18.0000, 19.0000, 18.7500,  ..., 18.5000, 18.7500, 18.5000],\n",
       "        [18.0000, 19.0000, 18.7500,  ..., 18.5000, 18.7500, 18.5000],\n",
       "        [18.2500, 19.2500, 19.2500,  ..., 17.7500, 18.5000, 19.2500]])"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torchdata = torch.tensor(frames).view(-1, 64).float()\n",
    "torchdata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = torchdata[:]\n",
    "y = torchdata[:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "decoder = nn.Sequential(\n",
    "#     nn.Linear(128, 64),\n",
    "#     nn.PReLU(),\n",
    "    nn.Linear(64, 32),\n",
    "    nn.PReLU(),\n",
    "    nn.Linear(32, 4),\n",
    ")\n",
    "\n",
    "encoder = nn.Sequential(\n",
    "    nn.Linear(4, 16),\n",
    "    nn.PReLU(),\n",
    "    nn.Linear(16, 64),\n",
    ")\n",
    "\n",
    "net = nn.Sequential(\n",
    "    decoder,\n",
    "    encoder\n",
    ")\n",
    "\n",
    "\n",
    "opt = torch.optim.Adam(net.parameters(), amsgrad=True)\n",
    "loss = torch.nn.MSELoss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([2480, 64]), torch.Size([2480, 64]))"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.shape, y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Results - Epoch: 50 {'accuracy': 23.514441229372608}\n",
      "Training Results - Epoch: 100 {'accuracy': 21.019813101632256}\n",
      "Training Results - Epoch: 150 {'accuracy': 17.924335308464205}\n",
      "Training Results - Epoch: 200 {'accuracy': 16.39731971585021}\n",
      "Training Results - Epoch: 250 {'accuracy': 15.417745146459463}\n",
      "Training Results - Epoch: 300 {'accuracy': 14.269157752212212}\n",
      "Training Results - Epoch: 350 {'accuracy': 13.364579632817483}\n",
      "Training Results - Epoch: 400 {'accuracy': 12.877606310163225}\n",
      "Training Results - Epoch: 450 {'accuracy': 12.451178507902185}\n",
      "Training Results - Epoch: 500 {'accuracy': 12.312315586635044}\n",
      "Training Results - Epoch: 550 {'accuracy': 12.101803962551818}\n",
      "Training Results - Epoch: 600 {'accuracy': 12.149661753128987}\n",
      "Training Results - Epoch: 650 {'accuracy': 11.908004542759487}\n",
      "Training Results - Epoch: 700 {'accuracy': 11.799759627361686}\n",
      "Training Results - Epoch: 750 {'accuracy': 11.873882605105031}\n",
      "Training Results - Epoch: 800 {'accuracy': 11.558896886085977}\n",
      "Training Results - Epoch: 850 {'accuracy': 11.604284854811064}\n",
      "Training Results - Epoch: 900 {'accuracy': 11.485453017877072}\n",
      "Training Results - Epoch: 950 {'accuracy': 11.37904052734375}\n",
      "Training Results - Epoch: 1000 {'accuracy': 11.296230767697704}\n",
      "Training Results - Epoch: 1050 {'accuracy': 11.313876794309031}\n",
      "Training Results - Epoch: 1100 {'accuracy': 11.210674083476164}\n",
      "Training Results - Epoch: 1150 {'accuracy': 11.105424468371332}\n",
      "Training Results - Epoch: 1200 {'accuracy': 11.056924656459264}\n",
      "Training Results - Epoch: 1250 {'accuracy': 11.073909993074379}\n",
      "Training Results - Epoch: 1300 {'accuracy': 11.00502604270468}\n",
      "Training Results - Epoch: 1350 {'accuracy': 10.971980410206074}\n",
      "Training Results - Epoch: 1400 {'accuracy': 10.919047951211734}\n",
      "Training Results - Epoch: 1450 {'accuracy': 10.885275517677774}\n",
      "Training Results - Epoch: 1500 {'accuracy': 10.826130069032008}\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "State:\n",
       "\titeration: 70500\n",
       "\tepoch: 1500\n",
       "\tepoch_length: 47\n",
       "\tmax_epochs: 1500\n",
       "\toutput: 0.11890707910060883\n",
       "\tbatch: <class 'list'>\n",
       "\tmetrics: <class 'dict'>\n",
       "\tdataloader: <class 'torch.utils.data.dataloader.DataLoader'>\n",
       "\tseed: <class 'NoneType'>\n",
       "\ttimes: <class 'dict'>"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_loader = torch.utils.data.DataLoader(\n",
    "    torch.utils.data.TensorDataset(x[:1500], y[:1500]),\n",
    "    batch_size=32,\n",
    "    shuffle=True,\n",
    ")\n",
    "\n",
    "test_loader = torch.utils.data.DataLoader(\n",
    "    torch.utils.data.TensorDataset(x[1500:], y[1500:]),\n",
    "    batch_size=32,\n",
    "    shuffle=True,\n",
    ")\n",
    "\n",
    "\n",
    "trainer = ignite.engine.create_supervised_trainer(\n",
    "    model=net,\n",
    "    optimizer=opt,\n",
    "    loss_fn=loss\n",
    ")\n",
    "\n",
    "\n",
    "val_metrics = {\n",
    "    \"accuracy\": ignite.metrics.MeanSquaredError()\n",
    "}\n",
    "evaluator = ignite.engine.create_supervised_evaluator(model=net, metrics=val_metrics)\n",
    "\n",
    "# @trainer.on(ignite.engine.Events.ITERATION_COMPLETED(every=50))\n",
    "# def log_training_loss(trainer):\n",
    "#     print(\"Epoch[{}] Loss: {:.2f}\".format(trainer.state.epoch, trainer.state.output))\n",
    "\n",
    "@trainer.on(ignite.engine.Events.EPOCH_COMPLETED(every=50))\n",
    "def log_training_results(trainer):\n",
    "    evaluator.run(test_loader)\n",
    "    metrics = evaluator.state.metrics\n",
    "    print(\"Training Results - Epoch: {} {}\"\n",
    "          .format(trainer.state.epoch, metrics))\n",
    "\n",
    "trainer.run(data_loader, max_epochs=1500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12e2aa908>]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cls = cluster.KMeans(3).fit(decoder(x).detach())\n",
    "c = cls.predict(decoder(x).detach())\n",
    "plot(c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor(0)\n",
      "graph(%input.1 : Float(64:1),\n",
      "      %net.0.bias : Float(32:1),\n",
      "      %net.1.weight : Float(1:1),\n",
      "      %net.2.bias : Float(4:1),\n",
      "      %39 : Float(64:1, 32:64),\n",
      "      %40 : Float(32:1, 4:32),\n",
      "      %41 : Float(1:4, 4:1),\n",
      "      %42 : Float(1:4, 4:1),\n",
      "      %43 : Float(1:4, 4:1)):\n",
      "  %8 : Float(32:1) = onnx::MatMul(%input.1, %39) # /Users/riri/.pyenv/versions/3.7.3/lib/python3.7/site-packages/torch/nn/functional.py:1676:0\n",
      "  %9 : Float(32:1) = onnx::Add(%8, %net.0.bias)\n",
      "  %10 : Float(32:1) = onnx::PRelu(%9, %net.1.weight) # /Users/riri/.pyenv/versions/3.7.3/lib/python3.7/site-packages/torch/nn/functional.py:1319:0\n",
      "  %12 : Float(4:1) = onnx::MatMul(%10, %40) # /Users/riri/.pyenv/versions/3.7.3/lib/python3.7/site-packages/torch/nn/functional.py:1676:0\n",
      "  %13 : Float(4:1) = onnx::Add(%12, %net.2.bias)\n",
      "  %16 : Float(4:1) = onnx::Sub(%13, %41) # <ipython-input-72-4300c352bb02>:11:0\n",
      "  %17 : FloatTensor = onnx::Mul(%16, %16)\n",
      "  %18 : Tensor = onnx::ReduceSum[keepdims=0](%17)\n",
      "  %19 : Float() = onnx::Sqrt(%18) # /Users/riri/.pyenv/versions/3.7.3/lib/python3.7/site-packages/torch/functional.py:1077:0\n",
      "  %22 : Float(4:1) = onnx::Sub(%13, %42) # <ipython-input-72-4300c352bb02>:11:0\n",
      "  %23 : FloatTensor = onnx::Mul(%22, %22)\n",
      "  %24 : Tensor = onnx::ReduceSum[keepdims=0](%23)\n",
      "  %25 : Float() = onnx::Sqrt(%24) # /Users/riri/.pyenv/versions/3.7.3/lib/python3.7/site-packages/torch/functional.py:1077:0\n",
      "  %28 : Float(4:1) = onnx::Sub(%13, %43) # <ipython-input-72-4300c352bb02>:11:0\n",
      "  %29 : FloatTensor = onnx::Mul(%28, %28)\n",
      "  %30 : Tensor = onnx::ReduceSum[keepdims=0](%29)\n",
      "  %31 : Float() = onnx::Sqrt(%30) # /Users/riri/.pyenv/versions/3.7.3/lib/python3.7/site-packages/torch/functional.py:1077:0\n",
      "  %32 : Tensor = onnx::Unsqueeze[axes=[0]](%19)\n",
      "  %33 : Tensor = onnx::Unsqueeze[axes=[0]](%25)\n",
      "  %34 : Tensor = onnx::Unsqueeze[axes=[0]](%31)\n",
      "  %35 : Float(3:1) = onnx::Concat[axis=0](%32, %33, %34) # <ipython-input-72-4300c352bb02>:12:0\n",
      "  %36 : Tensor = onnx::Constant[value={-1}]()\n",
      "  %37 : Tensor = onnx::Reshape(%35, %36)\n",
      "  %38 : Long() = onnx::ArgMin[axis=0, keepdims=0](%37) # <ipython-input-72-4300c352bb02>:13:0\n",
      "  return (%38)\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/riri/.pyenv/versions/3.7.3/lib/python3.7/site-packages/torch/tensor.py:455: RuntimeWarning: Iterating over a tensor might cause the trace to be incorrect. Passing a tensor of different shape won't change the number of iterations executed (and might lead to errors or silently give incorrect results).\n",
      "  'incorrect results).', category=RuntimeWarning)\n"
     ]
    }
   ],
   "source": [
    "class KMeansLookup(nn.Module):\n",
    "    def __init__(self, centroids, net):\n",
    "        super().__init__()\n",
    "        self.centroids = nn.Parameter(centroids)\n",
    "        self.net = net\n",
    "        \n",
    "    def forward(self, x):\n",
    "        cls_x = self.net(x)\n",
    "        dists = []\n",
    "        for ix, c in enumerate(self.centroids):\n",
    "            dists.append(torch.norm(cls_x - c))\n",
    "        dists = torch.stack(dists)\n",
    "        return dists.argmin()\n",
    "\n",
    "    \n",
    "class FinalModel(nn.Module):\n",
    "    def __init__(self, centroids, net):\n",
    "        super().__init__()\n",
    "        self.km = KMeansLookup(centroids, net)\n",
    "        \n",
    "    def forward(self, x, xp):\n",
    "        r1 = self.km(x)\n",
    "        r2 = self.km(xp)\n",
    "        \n",
    "        ignore_stmt = r1 * 0\n",
    "        inc_stmt = ignore_stmt * 0 + 1\n",
    "        dec_stmt = ignore_stmt * 0 - 1\n",
    "        \n",
    "        return torch.where(\n",
    "            r1 == 1,\n",
    "            torch.where(\n",
    "                r2 == 0, \n",
    "                inc_stmt,\n",
    "                torch.where(\n",
    "                    r2 == 2,\n",
    "                    dec_stmt,\n",
    "                    ignore_stmt\n",
    "                )\n",
    "            ),\n",
    "            ignore_stmt\n",
    "        )\n",
    "#         if r1 == 1:\n",
    "#             if r2 == 0:\n",
    "#                 return r1 * 0 + 1\n",
    "#             elif r2 == 2:\n",
    "#                 return r1 * 0 - 1\n",
    "#         return r2 * 0\n",
    "\n",
    "#a = FinalModel(torch.tensor(cls.cluster_centers_).float(), decoder)\n",
    "a = KMeansLookup(torch.tensor(cls.cluster_centers_).float(), decoder)\n",
    "fmdl = torch.jit.script(a)\n",
    "r = fmdl(x[262])\n",
    "print(r)\n",
    "\n",
    "torch.onnx.export(a, x[0], 'test.onnx', verbose=True, training=False, example_outputs=r)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[17.7500, 18.5000, 18.5000,  ..., 18.0000, 18.0000, 18.5000],\n",
       "        [17.7500, 18.0000, 18.2500,  ..., 17.7500, 17.7500, 18.2500],\n",
       "        [17.7500, 18.0000, 18.2500,  ..., 17.7500, 17.7500, 18.2500],\n",
       "        ...,\n",
       "        [18.0000, 19.0000, 18.7500,  ..., 18.5000, 18.7500, 18.5000],\n",
       "        [18.0000, 19.0000, 18.7500,  ..., 18.5000, 18.7500, 18.5000],\n",
       "        [18.2500, 19.2500, 19.2500,  ..., 17.7500, 18.5000, 19.2500]])"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}