CGEM_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "922cc733",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import signac\n",
    "import gsd\n",
    "import freud\n",
    "from scipy.stats import uniform\n",
    "from scipy.optimize import curve_fit\n",
    "from scipy.special import erf\n",
    "from joblib import Parallel, delayed\n",
    "\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "096c6b29",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 9 10377\n",
      "1 10386 20754\n",
      "2 20763 31131\n",
      "3 31140 41508\n",
      "4 41517 51885\n",
      "5 51894 62262\n",
      "6 62271 72639\n",
      "7 72648 83016\n",
      "8 83025 93393\n",
      "9 93402 103770\n",
      "10 103779 114147\n",
      "11 114156 124524\n",
      "12 124533 134901\n",
      "13 134910 145278\n",
      "14 145287 155655\n",
      "15 155664 166032\n",
      "16 166041 176409\n",
      "17 176418 186786\n",
      "18 186795 197163\n",
      "19 197172 207540\n",
      "20 207549 217917\n",
      "21 217926 228294\n",
      "22 228303 238671\n",
      "23 238680 249048\n",
      "24 249057 259425\n",
      "25 259434 269802\n",
      "26 269811 280179\n",
      "27 280188 290556\n",
      "28 290565 300933\n",
      "29 300942 311310\n",
      "30 311319 321687\n",
      "31 321696 332064\n",
      "32 332073 342441\n",
      "33 342450 352818\n",
      "34 352827 363195\n",
      "35 363204 373572\n",
      "36 373581 383949\n",
      "37 383958 394326\n",
      "38 394335 404703\n",
      "39 404712 415080\n",
      "40 415089 425457\n",
      "41 425466 435834\n",
      "42 435843 446211\n",
      "43 446220 456588\n",
      "44 456597 466965\n",
      "45 466974 477342\n",
      "46 477351 487719\n",
      "47 487728 498096\n",
      "48 498105 508473\n",
      "49 508482 518850\n",
      "50 518859 529227\n",
      "51 529236 539604\n",
      "52 539613 549981\n",
      "53 549990 560358\n",
      "54 560367 570735\n",
      "55 570744 581112\n",
      "56 581121 591489\n",
      "57 591498 601866\n",
      "58 601875 612243\n",
      "59 612252 622620\n",
      "60 622629 632997\n",
      "61 633006 643374\n",
      "62 643383 653751\n",
      "63 653760 664128\n",
      "64 664137 674505\n",
      "65 674514 684882\n",
      "66 684891 695259\n",
      "67 695268 705636\n",
      "68 705645 716013\n",
      "69 716022 726390\n",
      "70 726399 736767\n",
      "71 736776 747144\n",
      "72 747153 757521\n",
      "73 757530 767898\n",
      "74 767907 778275\n",
      "75 778284 788652\n",
      "76 788661 799029\n",
      "77 799038 809406\n",
      "78 809415 819783\n",
      "79 819792 830160\n",
      "80 830169 840537\n",
      "81 840546 850914\n",
      "82 850923 861291\n",
      "83 861300 871668\n",
      "84 871677 882045\n",
      "85 882054 892422\n",
      "86 892431 902799\n",
      "87 902808 913176\n",
      "88 913185 923553\n",
      "89 923562 933930\n",
      "90 933939 944307\n",
      "91 944316 954684\n",
      "92 954693 965061\n",
      "93 965070 975438\n",
      "94 975447 985815\n",
      "95 985824 996192\n",
      "96 996201 1006569\n",
      "97 1006578 1016946\n",
      "98 1016955 1027323\n",
      "99 1027332 1037700\n",
      "100 1037709 1048077\n",
      "101 1048086 1058454\n",
      "102 1058463 1068831\n",
      "103 1068840 1079208\n",
      "104 1079217 1089585\n",
      "105 1089594 1099962\n",
      "106 1099971 1110339\n",
      "107 1110348 1120716\n",
      "108 1120725 1131093\n",
      "109 1131102 1141470\n",
      "110 1141479 1151847\n",
      "111 1151856 1162224\n",
      "112 1162233 1172601\n",
      "113 1172610 1182978\n",
      "114 1182987 1193355\n",
      "115 1193364 1203732\n",
      "116 1203741 1214109\n",
      "117 1214118 1224486\n",
      "118 1224495 1234863\n",
      "119 1234872 1245240\n",
      "120 1245249 1255617\n",
      "121 1255626 1265994\n",
      "122 1266003 1276371\n",
      "123 1276380 1286748\n",
      "124 1286757 1297125\n",
      "125 1297134 1307502\n",
      "126 1307511 1317879\n",
      "127 1317888 1328256\n",
      "128 1328265 1338633\n",
      "129 1338642 1349010\n",
      "130 1349019 1359387\n",
      "131 1359396 1369764\n",
      "132 1369773 1380141\n",
      "133 1380150 1390518\n",
      "134 1390527 1400895\n",
      "135 1400904 1411272\n",
      "136 1411281 1421649\n",
      "137 1421658 1432026\n",
      "138 1432035 1442403\n",
      "139 1442412 1452780\n",
      "140 1452789 1463157\n",
      "141 1463166 1473534\n",
      "142 1473543 1483911\n",
      "143 1483920 1494288\n",
      "144 1494297 1504665\n",
      "145 1504674 1515042\n",
      "146 1515051 1525419\n",
      "147 1525428 1535796\n",
      "148 1535805 1546173\n",
      "149 1546182 1556550\n",
      "150 1556559 1566927\n",
      "151 1566936 1577304\n",
      "152 1577313 1587681\n",
      "153 1587690 1598058\n",
      "154 1598067 1608435\n",
      "155 1608444 1618812\n",
      "156 1618821 1629189\n",
      "157 1629198 1639566\n",
      "158 1639575 1649943\n",
      "159 1649952 1660320\n",
      "160 1660329 1670697\n",
      "161 1670706 1681074\n",
      "162 1681083 1691451\n",
      "163 1691460 1701828\n",
      "164 1701837 1712205\n",
      "165 1712214 1722582\n",
      "166 1722591 1732959\n",
      "167 1732968 1743336\n",
      "168 1743345 1753713\n",
      "169 1753722 1764090\n",
      "170 1764099 1774467\n",
      "171 1774476 1784844\n",
      "172 1784853 1795221\n",
      "173 1795230 1805598\n",
      "174 1805607 1815975\n",
      "175 1815984 1826352\n",
      "176 1826361 1836729\n",
      "177 1836738 1847106\n",
      "178 1847115 1857483\n",
      "179 1857492 1867860\n",
      "180 1867869 1878237\n",
      "181 1878246 1888614\n",
      "182 1888623 1898991\n",
      "183 1899000 1909368\n",
      "184 1909377 1919745\n",
      "185 1919754 1930122\n",
      "186 1930131 1940499\n",
      "187 1940508 1950876\n",
      "188 1950885 1961253\n",
      "189 1961262 1971630\n",
      "190 1971639 1982007\n",
      "191 1982016 1992384\n",
      "192 1992393 2002761\n",
      "193 2002770 2013138\n",
      "194 2013147 2023515\n",
      "195 2023524 2033892\n",
      "196 2033901 2044269\n",
      "197 2044278 2054646\n",
      "198 2054655 2065023\n",
      "199 2065032 2075400\n",
      "200 2075409 2085777\n",
      "201 2085786 2096154\n",
      "202 2096163 2106531\n",
      "203 2106540 2116908\n",
      "204 2116917 2127285\n",
      "205 2127294 2137662\n",
      "206 2137671 2148039\n",
      "207 2148048 2158416\n",
      "208 2158425 2168793\n",
      "209 2168802 2179170\n",
      "210 2179179 2189547\n",
      "211 2189556 2199924\n",
      "212 2199933 2210301\n",
      "213 2210310 2220678\n",
      "214 2220687 2231055\n",
      "215 2231064 2241432\n",
      "216 2241441 2251809\n",
      "217 2251818 2262186\n",
      "218 2262195 2272563\n",
      "219 2272572 2282940\n",
      "220 2282949 2293317\n",
      "221 2293326 2303694\n",
      "222 2303703 2314071\n",
      "223 2314080 2324448\n",
      "224 2324457 2334825\n",
      "225 2334834 2345202\n",
      "226 2345211 2355579\n",
      "227 2355588 2365956\n",
      "228 2365965 2376333\n",
      "229 2376342 2386710\n",
      "230 2386719 2397087\n",
      "231 2397096 2407464\n",
      "232 2407473 2417841\n",
      "233 2417850 2428218\n",
      "234 2428227 2438595\n",
      "235 2438604 2448972\n",
      "236 2448981 2459349\n",
      "237 2459358 2469726\n",
      "238 2469735 2480103\n",
      "239 2480112 2490480\n",
      "240 2490489 2500857\n",
      "241 2500866 2511234\n",
      "242 2511243 2521611\n",
      "243 2521620 2531988\n",
      "244 2531997 2542365\n",
      "245 2542374 2552742\n",
      "246 2552751 2563119\n",
      "247 2563128 2573496\n",
      "248 2573505 2583873\n",
      "249 2583882 2594250\n",
      "250 2594259 2604627\n",
      "251 2604636 2615004\n",
      "252 2615013 2625381\n",
      "253 2625390 2635758\n",
      "254 2635767 2646135\n",
      "255 2646144 2656512\n",
      "256 2656521 2666889\n",
      "257 2666898 2677266\n",
      "258 2677275 2687643\n",
      "259 2687652 2698020\n",
      "260 2698029 2708397\n",
      "261 2708406 2718774\n",
      "262 2718783 2729151\n",
      "263 2729160 2739528\n",
      "264 2739537 2749905\n",
      "265 2749914 2760282\n",
      "266 2760291 2770659\n",
      "267 2770668 2781036\n",
      "268 2781045 2791413\n",
      "269 2791422 2801790\n",
      "270 2801799 2812167\n",
      "271 2812176 2822544\n",
      "272 2822553 2832921\n",
      "273 2832930 2843298\n",
      "274 2843307 2853675\n",
      "275 2853684 2864052\n",
      "276 2864061 2874429\n",
      "277 2874438 2884806\n",
      "278 2884815 2895183\n",
      "279 2895192 2905560\n",
      "280 2905569 2915937\n",
      "281 2915946 2926314\n",
      "282 2926323 2936691\n",
      "283 2936700 2947068\n",
      "284 2947077 2957445\n",
      "285 2957454 2967822\n",
      "286 2967831 2978199\n",
      "287 2978208 2988576\n",
      "288 2988585 2998953\n",
      "289 2998962 3009330\n",
      "290 3009339 3019707\n",
      "291 3019716 3030084\n",
      "292 3030093 3040461\n",
      "293 3040470 3050838\n",
      "294 3050847 3061215\n",
      "295 3061224 3071592\n",
      "296 3071601 3081969\n",
      "297 3081978 3092346\n",
      "298 3092355 3102723\n",
      "299 3102732 3113100\n",
      "300 3113109 3123477\n",
      "301 3123486 3133854\n",
      "302 3133863 3144231\n",
      "303 3144240 3154608\n",
      "304 3154617 3164985\n",
      "305 3164994 3175362\n",
      "306 3175371 3185739\n",
      "307 3185748 3196116\n",
      "308 3196125 3206493\n",
      "309 3206502 3216870\n",
      "310 3216879 3227247\n",
      "311 3227256 3237624\n",
      "312 3237633 3248001\n",
      "313 3248010 3258378\n",
      "314 3258387 3268755\n",
      "315 3268764 3279132\n",
      "316 3279141 3289509\n",
      "317 3289518 3299886\n",
      "318 3299895 3310263\n",
      "319 3310272 3320640\n",
      "320 3320649 3331017\n",
      "321 3331026 3341394\n",
      "322 3341403 3351771\n",
      "323 3351780 3362148\n",
      "324 3362157 3372525\n",
      "325 3372534 3382902\n",
      "326 3382911 3393279\n",
      "327 3393288 3403656\n",
      "328 3403665 3414033\n",
      "329 3414042 3424410\n",
      "330 3424419 3434787\n",
      "331 3434796 3445164\n",
      "332 3445173 3455541\n",
      "333 3455550 3465918\n",
      "334 3465927 3476295\n",
      "335 3476304 3486672\n",
      "336 3486681 3497049\n",
      "337 3497058 3507426\n",
      "338 3507435 3517803\n",
      "339 3517812 3528180\n",
      "340 3528189 3538557\n",
      "341 3538566 3548934\n",
      "342 3548943 3559311\n",
      "343 3559320 3569688\n",
      "344 3569697 3580065\n",
      "345 3580074 3590442\n",
      "346 3590451 3600819\n",
      "347 3600828 3611196\n",
      "348 3611205 3621573\n",
      "349 3621582 3631950\n",
      "350 3631959 3642327\n",
      "351 3642336 3652704\n",
      "352 3652713 3663081\n",
      "353 3663090 3673458\n",
      "354 3673467 3683835\n",
      "355 3683844 3694212\n",
      "356 3694221 3704589\n",
      "357 3704598 3714966\n",
      "358 3714975 3725343\n",
      "359 3725352 3735720\n",
      "360 3735729 3746097\n",
      "361 3746106 3756474\n",
      "362 3756483 3766851\n",
      "363 3766860 3777228\n",
      "364 3777237 3787605\n",
      "365 3787614 3797982\n",
      "366 3797991 3808359\n",
      "367 3808368 3818736\n",
      "368 3818745 3829113\n",
      "369 3829122 3839490\n",
      "370 3839499 3849867\n",
      "371 3849876 3860244\n",
      "372 3860253 3870621\n",
      "373 3870630 3880998\n",
      "374 3881007 3891375\n",
      "375 3891384 3901752\n",
      "376 3901761 3912129\n",
      "377 3912138 3922506\n",
      "378 3922515 3932883\n",
      "379 3932892 3943260\n",
      "380 3943269 3953637\n",
      "381 3953646 3964014\n",
      "382 3964023 3974391\n",
      "383 3974400 3984768\n",
      "384 3984777 3995145\n",
      "385 3995154 4005522\n",
      "386 4005531 4015899\n",
      "387 4015908 4026276\n",
      "388 4026285 4036653\n",
      "389 4036662 4047030\n",
      "390 4047039 4057407\n",
      "391 4057416 4067784\n",
      "392 4067793 4078161\n",
      "393 4078170 4088538\n",
      "394 4088547 4098915\n",
      "395 4098924 4109292\n",
      "396 4109301 4119669\n",
      "397 4119678 4130046\n",
      "398 4130055 4140423\n",
      "399 4140432 4150800\n",
      "400 4150809 4161177\n",
      "401 4161186 4171554\n",
      "402 4171563 4181931\n",
      "403 4181940 4192308\n",
      "404 4192317 4202685\n",
      "405 4202694 4213062\n",
      "406 4213071 4223439\n",
      "407 4223448 4233816\n",
      "408 4233825 4244193\n",
      "409 4244202 4254570\n",
      "410 4254579 4264947\n",
      "411 4264956 4275324\n",
      "412 4275333 4285701\n",
      "413 4285710 4296078\n",
      "414 4296087 4306455\n",
      "415 4306464 4316832\n",
      "416 4316841 4327209\n",
      "417 4327218 4337586\n",
      "418 4337595 4347963\n",
      "419 4347972 4358340\n",
      "420 4358349 4368717\n",
      "421 4368726 4379094\n",
      "422 4379103 4389471\n",
      "423 4389480 4399848\n",
      "424 4399857 4410225\n",
      "425 4410234 4420602\n",
      "426 4420611 4430979\n",
      "427 4430988 4441356\n",
      "428 4441365 4451733\n",
      "429 4451742 4462110\n",
      "430 4462119 4472487\n",
      "431 4472496 4482864\n",
      "432 4482873 4493241\n",
      "433 4493250 4503618\n",
      "434 4503627 4513995\n",
      "435 4514004 4524372\n",
      "436 4524381 4534749\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "437 4534758 4545126\n",
      "438 4545135 4555503\n",
      "439 4555512 4565880\n",
      "440 4565889 4576257\n",
      "441 4576266 4586634\n",
      "442 4586643 4597011\n",
      "443 4597020 4607388\n",
      "444 4607397 4617765\n",
      "445 4617774 4628142\n",
      "446 4628151 4638519\n",
      "447 4638528 4648896\n",
      "448 4648905 4659273\n",
      "449 4659282 4669650\n",
      "450 4669659 4680027\n",
      "451 4680036 4690404\n",
      "452 4690413 4700781\n",
      "453 4700790 4711158\n",
      "454 4711167 4721535\n",
      "455 4721544 4731912\n",
      "456 4731921 4742289\n",
      "457 4742298 4752666\n",
      "458 4752675 4763043\n",
      "459 4763052 4773420\n",
      "460 4773429 4783797\n",
      "461 4783806 4794174\n",
      "462 4794183 4804551\n",
      "463 4804560 4814928\n",
      "464 4814937 4825305\n",
      "465 4825314 4835682\n",
      "466 4835691 4846059\n",
      "467 4846068 4856436\n",
      "468 4856445 4866813\n",
      "469 4866822 4877190\n",
      "470 4877199 4887567\n",
      "471 4887576 4897944\n",
      "472 4897953 4908321\n",
      "473 4908330 4918698\n",
      "474 4918707 4929075\n",
      "475 4929084 4939452\n",
      "476 4939461 4949829\n",
      "477 4949838 4960206\n",
      "478 4960215 4970583\n",
      "479 4970592 4980960\n",
      "480 4980969 4991337\n",
      "481 4991346 5001714\n",
      "done\n"
     ]
    }
   ],
   "source": [
    "#Extracting information\n",
    "\n",
    "poses = []\n",
    "boxes = []\n",
    "        \n",
    "f = open(\"dump.lammpstrj\")\n",
    "text = f.readlines()\n",
    "\n",
    "num_atoms = int(np.fromstring(text[3][:-1], sep=' ')[0])\n",
    "num_frames = int(len(text) / (9 + num_atoms))\n",
    "\n",
    "\n",
    "for i in range(0, num_frames, 1):\n",
    "        it = 9 + (9+num_atoms)*i\n",
    "        et = it + num_atoms\n",
    "        box0 = np.fromstring(text[it-4][:13], sep=' ')*-20\n",
    "        box1 = np.fromstring(text[it-3][:13], sep=' ')*-20\n",
    "\n",
    "        boxes.append([box0, box1, box1])\n",
    "        #print(box0)\n",
    "        print(i,it, et)\n",
    "        pos = []\n",
    "\n",
    "        for j in range(it, et):\n",
    "            pos.append(np.fromstring(text[j][:-1], sep=' '))\n",
    "\n",
    "        poses.append(np.asarray(pos))                \n",
    "\n",
    "\n",
    "        #break\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "f.close()\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "print('done')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "45117fa2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100\n",
      "200\n",
      "300\n",
      "400\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Computing fields ignoring the fact that they are Gaussian Charges\n",
    "\n",
    "projE = []\n",
    "\n",
    "for i in range(100, len(poses), 100): #Ignoring the first 100 frames, only analyzing every 100 frames\n",
    "\n",
    "\n",
    "    fbox = freud.box.Box.from_box(boxes[i])\n",
    "\n",
    "\n",
    "    pos = poses[i]\n",
    "    Opos = pos[:,2:][pos[:,0] == 2]\n",
    "    Hpos = pos[:,2:][pos[:,0] == 1]\n",
    "    Epos = pos[:,2:][pos[:,0] == 3]\n",
    "    \n",
    "    # Getting O to H vector\n",
    "    vHpos = np.reshape(Hpos, [int(len(Hpos)/2), 2, 3])\n",
    "    vEpos = np.reshape(Epos, [int(len(Epos)/3), 3, 3])\n",
    "    omat = np.copy(vHpos)        \n",
    "    omat[:,0] = fbox.wrap(vHpos[:,0,:] - Opos)\n",
    "    omat[:,1] = fbox.wrap(vHpos[:,1] - Opos)\n",
    "    omat = np.reshape(omat, [len(Hpos), 3])\n",
    "\n",
    "\n",
    "    fields = []\n",
    "\n",
    "    Hdel_list = (np.arange(0, len(Hpos))/2).astype('int') # Need these to ignore intramolecular E-fields\n",
    "    Odel_list = (np.arange(0, len(Opos))).astype('int')\n",
    "    Edel_list = (np.arange(0, len(Epos))/3).astype('int')\n",
    "\n",
    "\n",
    "    for j in range(0,len(Hpos)):\n",
    "\n",
    "        Hpoint = Hpos[j]\n",
    "        tHpos = Hpos[Hdel_list != int(j/2)]\n",
    "        tOpos = Opos[Odel_list != int(j/2)]\n",
    "        tEpos = Epos[Edel_list != int(j/2)]\n",
    "\n",
    "\n",
    "        dpos = fbox.wrap(Hpoint - tHpos)\n",
    "        dis = np.linalg.norm(dpos, axis=1)\n",
    "        dpos = dpos[dis > 1e-2]\n",
    "        dis = dis[dis > 1e-2]\n",
    "        direct = (dpos.T / dis).T\n",
    "        Hfields = (direct.T / dis**2).T\n",
    "\n",
    "        dpos = fbox.wrap(Hpoint - tOpos)\n",
    "        dis = np.linalg.norm(dpos, axis=1)\n",
    "        dpos = dpos[dis > 1e-2]\n",
    "        dis = dis[dis >1e-2]\n",
    "        direct = (dpos.T / dis).T\n",
    "        Ofields = (direct.T / dis**2).T\n",
    "\n",
    "        dpos = fbox.wrap(Hpoint - tEpos)\n",
    "        dis = np.linalg.norm(dpos, axis=1)\n",
    "        dpos = dpos[dis > 1e-2]\n",
    "        dis = dis[dis >1e-2]\n",
    "        direct = (dpos.T / dis).T\n",
    "        Efields = (direct.T / dis**2).T\n",
    "\n",
    "        field = (1*np.sum(Hfields, axis=0)\n",
    "                 + 1*np.sum(Ofields, axis=0)\n",
    "                 + -1*np.sum(Efields, axis=0))\n",
    "\n",
    "\n",
    "        fields.append(field)\n",
    "\n",
    "\n",
    "    projE.append(np.sum(omat*fields, axis=1))\n",
    "    \n",
    "    print(i)\n",
    "\n",
    "\n",
    "            \n",
    "projE = np.array(projE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "id": "203710d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Field Strength (V/A)')"
      ]
     },
     "execution_count": 231,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "heights, bins = np.histogram(-14.4*projs, 20) # -14.4 is for conversion\n",
    "\n",
    "binw = bins[1] - bins[0]\n",
    "bin_centers = bins[:-1] + binw/2\n",
    "\n",
    "\n",
    "plt.plot(bin_centers, heights)\n",
    "\n",
    "plt.ylabel(\"Counts\")\n",
    "plt.xlabel(\"Field Strength (V/A)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "id": "269660f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x12b7037c0>"
      ]
     },
     "execution_count": 229,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Computing CGEM fields\n",
    "\n",
    "from scipy.special import erf\n",
    "\n",
    "\n",
    "Lambda = 2.06064\n",
    "\n",
    "rO = 0.5414 \n",
    "rH = 0.5749\n",
    "rE = 0.784 \n",
    "\n",
    "def CGEM_field(r1, r2, dis):\n",
    "    \n",
    "    \n",
    "    alpha1 = Lambda / 2 / r1**2\n",
    "    alpha2 = Lambda / 2 / r2**2\n",
    "    \n",
    "    \n",
    "    innera = np.sqrt(alpha1*alpha2 / (alpha1 + alpha2))\n",
    "    num1 = 2*innera*dis*np.exp(-innera**2*dis**2) / np.sqrt(np.pi)\n",
    "    num2 = erf(innera*dis)\n",
    "    denom = xx**2\n",
    "    field = -(num1 - num2) / dis**2\n",
    "    \n",
    "    rcut = 10\n",
    "    taper = 20*(dis/rcut)**7 - 70*(dis/rcut)**6 + 84*(dis/rcut)**5 - 35*(dis/rcut)**4 + 1\n",
    "    \n",
    "    field = taper*field\n",
    "    field[dis > rcut] = 0\n",
    "    return field\n",
    "\n",
    "\n",
    "xx = np.linspace(0.01, 10, 1000)\n",
    "\n",
    "yy = CGEM_field(rH, rH, xx)\n",
    "\n",
    "plt.plot(xx, yy*14.4, label=\"H-H\")\n",
    "\n",
    "yy = CGEM_field(rH, rE, xx)\n",
    "\n",
    "plt.plot(xx, yy*14.4, label=\"H-E\")\n",
    "\n",
    "yy = CGEM_field(rH, rO, xx)\n",
    "\n",
    "plt.plot(xx, yy*14.4, label=\"H-O\")\n",
    "\n",
    "\n",
    "plt.plot(xx, 14.4/xx**2, label=\"1/r$^2$\")\n",
    "\n",
    "plt.ylim(-0.1, 20)\n",
    "plt.xlim(0.01, 10)\n",
    "plt.xlabel(\"Distance\")\n",
    "plt.ylabel(\"Field Strength (V/A)\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "id": "5744f675",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100\n",
      "200\n",
      "300\n",
      "400\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Computing fields using Gaussican cores\n",
    "\n",
    "\n",
    "projE = []\n",
    "\n",
    "for i in range(100, len(poses), 100):\n",
    "\n",
    "\n",
    "    fbox = freud.box.Box.from_box(boxes[i])\n",
    "\n",
    "\n",
    "    pos = poses[i]\n",
    "    Opos = pos[:,2:][pos[:,0] == 2]\n",
    "    Hpos = pos[:,2:][pos[:,0] == 1]\n",
    "    Epos = pos[:,2:][pos[:,0] == 3]\n",
    "    \n",
    "    \n",
    "    \n",
    "    vHpos = np.reshape(Hpos, [int(len(Hpos)/2), 2, 3])\n",
    "    vEpos = np.reshape(Epos, [int(len(Epos)/3), 3, 3])\n",
    "    omat = np.copy(vHpos)        \n",
    "    omat[:,0] = fbox.wrap(vHpos[:,0,:] - Opos)\n",
    "    omat[:,1] = fbox.wrap(vHpos[:,1] - Opos)\n",
    "    omat = np.reshape(omat, [len(Hpos), 3])\n",
    "\n",
    "\n",
    "    fields = []\n",
    "\n",
    "\n",
    "\n",
    "    Hdel_list = (np.arange(0, len(Hpos))/2).astype('int')\n",
    "    Odel_list = (np.arange(0, len(Opos))).astype('int')\n",
    "    Edel_list = (np.arange(0, len(Epos))/3).astype('int')\n",
    "\n",
    "\n",
    "    for j in range(0,len(Hpos)):\n",
    "\n",
    "        Hpoint = Hpos[j]\n",
    "        tHpos = Hpos[Hdel_list != int(j/2)]\n",
    "        tOpos = Opos[Odel_list != int(j/2)]\n",
    "        tEpos = Epos[Edel_list != int(j/2)]\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "        dpos = fbox.wrap(Hpoint - tHpos)\n",
    "        dis = np.linalg.norm(dpos, axis=1)\n",
    "        dpos = dpos[dis > 1e-2]\n",
    "        dis = dis[dis > 1e-2]\n",
    "        direct = (dpos.T / dis).T\n",
    "        Hfields = (direct.T * CGEM_field(rH, rH, dis)).T\n",
    "\n",
    "\n",
    "        dpos = fbox.wrap(Hpoint - tOpos)\n",
    "        dis = np.linalg.norm(dpos, axis=1)\n",
    "        dpos = dpos[dis > 1e-2]\n",
    "        dis = dis[dis > 1e-2]\n",
    "        direct = (dpos.T / dis).T\n",
    "        Ofields = (direct.T * CGEM_field(rH, rO, dis)).T\n",
    "\n",
    "        dpos = fbox.wrap(Hpoint - tEpos)\n",
    "        dis = np.linalg.norm(dpos, axis=1)\n",
    "        dpos = dpos[dis > 1e-2]\n",
    "        dis = dis[dis > 1e-2]\n",
    "        direct = (dpos.T / dis).T\n",
    "        Efields = (direct.T * CGEM_field(rH, rE, dis)).T\n",
    "\n",
    "        \n",
    "        field = (1*np.sum(Hfields, axis=0)\n",
    "                 + 1*np.sum(Ofields, axis=0)\n",
    "                 + -1*np.sum(Efields, axis=0))\n",
    "\n",
    "\n",
    "        fields.append(field)\n",
    "        #break\n",
    "        #print(np.sum(Hfields, axis=0), 1*np.sum(Ofields, axis=0), -1*np.sum(Efields, axis=0))\n",
    "\n",
    "\n",
    "    projE.append(np.sum(omat*fields, axis=1))\n",
    "    \n",
    "    print(i)\n",
    "\n",
    "\n",
    "            \n",
    "projE = np.array(projE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "id": "ec0421c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Field Strength (V/A)')"
      ]
     },
     "execution_count": 232,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "un, co = np.unique(np.round(projE, 2), return_counts=True)\n",
    "\n",
    "plt.plot(un*-14.4, co/np.sum(co))\n",
    "\n",
    "plt.ylabel(\"Counts\")\n",
    "plt.xlabel(\"Field Strength (V/A)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7dbd7952",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ee786226",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "72219e8b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "30ca5711",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9c48f1c1",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "33da42f4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ba416719",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3f83671b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7b2d7f71",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "38fb65d5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a0d95d99",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9e1a0c7a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1adbe755",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "558f2e6d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1ccd48b4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "33663e36",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9574ef4e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "55d07b17",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "02db672f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d33903a8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "1895b0c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[array([[-0.11817648, -0.14704418],\n",
      "       [-0.11357788, -0.07023326],\n",
      "       [-0.15294488, -0.03435561],\n",
      "       ...,\n",
      "       [-0.20508859, -0.09426089],\n",
      "       [-0.17198949, -0.02154821],\n",
      "       [-0.08409236, -0.13122699]], dtype=float32), array([[-0.0756577 , -0.130437  ],\n",
      "       [-0.22865911, -0.03595458],\n",
      "       [-0.11957894, -0.03086508],\n",
      "       ...,\n",
      "       [-0.13613248, -0.09592831],\n",
      "       [-0.15084982, -0.08988983],\n",
      "       [-0.12498599, -0.06627281]], dtype=float32), array([[-0.17070773, -0.13071516],\n",
      "       [-0.17382658, -0.2057371 ],\n",
      "       [-0.13768998, -0.15406695],\n",
      "       ...,\n",
      "       [-0.20068549, -0.14893548],\n",
      "       [-0.06652506, -0.10967651],\n",
      "       [-0.05756936, -0.09580622]], dtype=float32), array([[-0.1531995 , -0.08068633],\n",
      "       [-0.13202882, -0.07826146],\n",
      "       [-0.06707557, -0.0445941 ],\n",
      "       ...,\n",
      "       [-0.05032402, -0.14608277],\n",
      "       [-0.15021935, -0.11321475],\n",
      "       [-0.0415628 , -0.11519535]], dtype=float32), array([[-0.0793936 , -0.09746761],\n",
      "       [-0.14726406, -0.1491873 ],\n",
      "       [-0.06869883, -0.15544681],\n",
      "       ...,\n",
      "       [-0.13188106, -0.09135406],\n",
      "       [-0.19299063, -0.1819382 ],\n",
      "       [-0.21679789, -0.05149623]], dtype=float32), array([[-0.09900048, -0.16659372],\n",
      "       [-0.09443817, -0.11088882],\n",
      "       [-0.07177087, -0.16577707],\n",
      "       ...,\n",
      "       [-0.10480226, -0.12443437],\n",
      "       [-0.13401997, -0.1921505 ],\n",
      "       [-0.15202197, -0.09798761]], dtype=float32), array([[-0.17233202, -0.11612502],\n",
      "       [-0.09003374, -0.11561358],\n",
      "       [-0.18527329, -0.07350565],\n",
      "       ...,\n",
      "       [-0.16419588, -0.09151474],\n",
      "       [-0.09224625, -0.07534555],\n",
      "       [-0.14731272, -0.09495284]], dtype=float32), array([[-0.05633143, -0.07460456],\n",
      "       [-0.08908077, -0.06861977],\n",
      "       [-0.17506078, -0.16313908],\n",
      "       ...,\n",
      "       [-0.1348536 , -0.15610597],\n",
      "       [-0.16697893, -0.22892089],\n",
      "       [-0.09185841, -0.11071345]], dtype=float32), array([[-0.14226621, -0.10283579],\n",
      "       [-0.03120159, -0.10694499],\n",
      "       [-0.14394733, -0.12396016],\n",
      "       ...,\n",
      "       [-0.16719016, -0.16156872],\n",
      "       [-0.17439373, -0.10397941],\n",
      "       [-0.23078908, -0.0346659 ]], dtype=float32), array([[-0.03190266, -0.15753941],\n",
      "       [-0.15641819, -0.14339042],\n",
      "       [-0.13992678, -0.09472401],\n",
      "       ...,\n",
      "       [-0.16602641, -0.04227225],\n",
      "       [-0.06639823, -0.17820415],\n",
      "       [-0.13551944, -0.06550633]], dtype=float32), array([[-0.12259448, -0.09398632],\n",
      "       [-0.1734538 , -0.06753714],\n",
      "       [-0.14488935, -0.11651635],\n",
      "       ...,\n",
      "       [-0.1458569 , -0.1151083 ],\n",
      "       [-0.19419959, -0.04415916],\n",
      "       [-0.10646837, -0.0466648 ]], dtype=float32), array([[-0.05150899, -0.10322918],\n",
      "       [-0.01573659, -0.18158977],\n",
      "       [-0.1989763 , -0.03103153],\n",
      "       ...,\n",
      "       [-0.09589686, -0.0580691 ],\n",
      "       [-0.1506679 , -0.15457873],\n",
      "       [-0.11705451, -0.10068844]], dtype=float32), array([[-0.09922481, -0.0924848 ],\n",
      "       [-0.19311821, -0.11035587],\n",
      "       [-0.06189689, -0.13656005],\n",
      "       ...,\n",
      "       [-0.08871934, -0.17190611],\n",
      "       [-0.0329387 , -0.19165657],\n",
      "       [-0.07998927, -0.11668621]], dtype=float32), array([[-0.10308666, -0.02680009],\n",
      "       [-0.15979986, -0.10194985],\n",
      "       [-0.12087107, -0.09595542],\n",
      "       ...,\n",
      "       [-0.10910617, -0.08818957],\n",
      "       [-0.06375409, -0.11969179],\n",
      "       [-0.16605484, -0.18406264]], dtype=float32), array([[-0.03989097, -0.11627042],\n",
      "       [-0.16473092, -0.10209084],\n",
      "       [-0.14900859, -0.13779017],\n",
      "       ...,\n",
      "       [-0.15427049, -0.0759828 ],\n",
      "       [-0.18264773, -0.21595323],\n",
      "       [-0.0544048 , -0.16342193]], dtype=float32), array([[-0.20790818, -0.07078722],\n",
      "       [-0.17730446, -0.12364783],\n",
      "       [-0.07881644, -0.11923087],\n",
      "       ...,\n",
      "       [-0.07303219, -0.20225659],\n",
      "       [-0.12894465, -0.13427848],\n",
      "       [-0.06950684, -0.04576094]], dtype=float32), array([[-0.24117191, -0.11833839],\n",
      "       [-0.11180228, -0.04214431],\n",
      "       [-0.1404255 , -0.14006385],\n",
      "       ...,\n",
      "       [-0.12223572, -0.12811626],\n",
      "       [-0.12348173, -0.10213903],\n",
      "       [-0.1369023 , -0.0906717 ]], dtype=float32), array([[-0.08788019, -0.16036029],\n",
      "       [-0.14843637, -0.17476127],\n",
      "       [-0.17300145, -0.08173778],\n",
      "       ...,\n",
      "       [-0.11712517, -0.12803543],\n",
      "       [-0.1284148 , -0.17728747],\n",
      "       [-0.15630953, -0.04667281]], dtype=float32), array([[-0.08062878, -0.09471991],\n",
      "       [-0.07661928, -0.05342029],\n",
      "       [-0.04907831, -0.1520519 ],\n",
      "       ...,\n",
      "       [-0.1448657 , -0.16049099],\n",
      "       [-0.20484172, -0.11466835],\n",
      "       [-0.16567765, -0.07193804]], dtype=float32), array([[-0.11916392, -0.15680648],\n",
      "       [-0.1658623 , -0.08614019],\n",
      "       [-0.02768235, -0.09731492],\n",
      "       ...,\n",
      "       [-0.15388784, -0.09977043],\n",
      "       [-0.16209418, -0.21259397],\n",
      "       [-0.14417261, -0.20488794]], dtype=float32), array([[-0.19567978, -0.10062794],\n",
      "       [-0.16328499, -0.08443266],\n",
      "       [-0.1947229 , -0.11034902],\n",
      "       ...,\n",
      "       [-0.07342907, -0.19307339],\n",
      "       [-0.15145339, -0.10696738],\n",
      "       [-0.15883163, -0.09859432]], dtype=float32), array([[-0.0489733 , -0.0298955 ],\n",
      "       [-0.12893167, -0.06385522],\n",
      "       [-0.10140426, -0.17115454],\n",
      "       ...,\n",
      "       [-0.09785289, -0.1409583 ],\n",
      "       [-0.13999228, -0.23089498],\n",
      "       [-0.19614801, -0.11037529]], dtype=float32), array([[-0.16153403, -0.14138515],\n",
      "       [-0.18813927, -0.21453358],\n",
      "       [-0.13858442, -0.10956374],\n",
      "       ...,\n",
      "       [-0.11501104, -0.23603982],\n",
      "       [-0.0665558 , -0.15918817],\n",
      "       [-0.06498209, -0.11747326]], dtype=float32), array([[-0.08276529, -0.14990184],\n",
      "       [-0.03400517, -0.14203563],\n",
      "       [-0.09126113, -0.10421597],\n",
      "       ...,\n",
      "       [-0.10540324, -0.0865237 ],\n",
      "       [-0.08862931, -0.15970895],\n",
      "       [-0.14941041, -0.12098919]], dtype=float32), array([[-0.05741496, -0.14976294],\n",
      "       [-0.14066097, -0.13850497],\n",
      "       [-0.06371249, -0.14669532],\n",
      "       ...,\n",
      "       [-0.10783483, -0.22232829],\n",
      "       [-0.12478893, -0.08686519],\n",
      "       [-0.09751441, -0.08792025]], dtype=float32), array([[-0.21670745, -0.12283266],\n",
      "       [-0.1271024 , -0.13292621],\n",
      "       [-0.08389463, -0.16676539],\n",
      "       ...,\n",
      "       [-0.07751255, -0.09318569],\n",
      "       [-0.12996595, -0.10646942],\n",
      "       [-0.17851321, -0.1937277 ]], dtype=float32), array([[-0.1855597 , -0.1491696 ],\n",
      "       [-0.04963815, -0.21570799],\n",
      "       [-0.13234264, -0.13883877],\n",
      "       ...,\n",
      "       [-0.112152  , -0.11293152],\n",
      "       [-0.17637546, -0.10185259],\n",
      "       [-0.15486744, -0.1264057 ]], dtype=float32), array([[-0.1748017 , -0.15805604],\n",
      "       [-0.0235228 , -0.12254386],\n",
      "       [-0.11149178, -0.08104715],\n",
      "       ...,\n",
      "       [-0.15165335, -0.06783067],\n",
      "       [-0.15556914, -0.09036832],\n",
      "       [-0.02956655, -0.00600729]], dtype=float32), array([[-0.11949529, -0.05150716],\n",
      "       [-0.07102235, -0.16311818],\n",
      "       [-0.23657382, -0.0508759 ],\n",
      "       ...,\n",
      "       [-0.04734233, -0.09537803],\n",
      "       [-0.1269604 , -0.16450751],\n",
      "       [-0.05588133, -0.16012678]], dtype=float32), array([[-0.15848051, -0.0878808 ],\n",
      "       [-0.06753282, -0.0849613 ],\n",
      "       [-0.19392236, -0.05634561],\n",
      "       ...,\n",
      "       [-0.14137475, -0.05578512],\n",
      "       [-0.21953979, -0.17250253],\n",
      "       [-0.15240751, -0.08780885]], dtype=float32), array([[-0.14370026, -0.06622484],\n",
      "       [-0.15522411, -0.15657784],\n",
      "       [-0.05863745, -0.18224125],\n",
      "       ...,\n",
      "       [-0.08332646, -0.13818027],\n",
      "       [-0.16559911, -0.1959652 ],\n",
      "       [-0.12753142, -0.2213518 ]], dtype=float32), array([[-0.18974301, -0.03324183],\n",
      "       [-0.06286059, -0.08963893],\n",
      "       [-0.06587213, -0.05588296],\n",
      "       ...,\n",
      "       [-0.04150677, -0.09091444],\n",
      "       [-0.05823202, -0.20064628],\n",
      "       [-0.15408143, -0.10797114]], dtype=float32), array([[-0.15332109, -0.09185649],\n",
      "       [-0.13396329, -0.12282796],\n",
      "       [-0.05910769, -0.16869287],\n",
      "       ...,\n",
      "       [-0.07826383, -0.19351026],\n",
      "       [-0.17481011, -0.15664206],\n",
      "       [-0.18805516, -0.1794512 ]], dtype=float32), array([[-0.14285871, -0.03272702],\n",
      "       [-0.13905387, -0.15638152],\n",
      "       [-0.1380501 , -0.21175297],\n",
      "       ...,\n",
      "       [-0.11067926, -0.11234608],\n",
      "       [-0.01934537, -0.18155143],\n",
      "       [-0.08022152, -0.10653377]], dtype=float32), array([[-0.11292043, -0.11950021],\n",
      "       [-0.20064947, -0.19175853],\n",
      "       [-0.09450316, -0.20527282],\n",
      "       ...,\n",
      "       [-0.14202176, -0.16253449],\n",
      "       [-0.15555353, -0.05058711],\n",
      "       [-0.15104043, -0.11932351]], dtype=float32), array([[-0.19057013, -0.17187999],\n",
      "       [-0.1804141 , -0.15346038],\n",
      "       [-0.14227729, -0.1811685 ],\n",
      "       ...,\n",
      "       [-0.1791135 , -0.1861896 ],\n",
      "       [-0.1168474 , -0.14275922],\n",
      "       [-0.10951332, -0.13284852]], dtype=float32), array([[-0.17817327, -0.17699859],\n",
      "       [-0.15245812, -0.11867376],\n",
      "       [-0.1417627 , -0.04118174],\n",
      "       ...,\n",
      "       [-0.15798786, -0.15927736],\n",
      "       [-0.17364195, -0.14802603],\n",
      "       [-0.18704036, -0.15137583]], dtype=float32), array([[-0.19080773, -0.17732202],\n",
      "       [-0.1465003 , -0.08225854],\n",
      "       [-0.08368611, -0.09968791],\n",
      "       ...,\n",
      "       [-0.19593467, -0.12618257],\n",
      "       [-0.1648172 , -0.10096839],\n",
      "       [-0.08597536, -0.08529589]], dtype=float32), array([[-0.0278255 , -0.13209772],\n",
      "       [-0.07487678, -0.12638661],\n",
      "       [-0.24233705, -0.09660614],\n",
      "       ...,\n",
      "       [-0.16555871, -0.13156943],\n",
      "       [-0.14447342, -0.18588632],\n",
      "       [-0.06132683, -0.12422071]], dtype=float32), array([[-0.12894532, -0.13549568],\n",
      "       [-0.16006537, -0.07564116],\n",
      "       [-0.10103531, -0.09658217],\n",
      "       ...,\n",
      "       [-0.13828403, -0.03772097],\n",
      "       [-0.1632111 , -0.14094983],\n",
      "       [-0.0358477 , -0.19841436]], dtype=float32), array([[-0.1175544 , -0.12093002],\n",
      "       [-0.1373172 , -0.13343185],\n",
      "       [-0.15297653, -0.0860183 ],\n",
      "       ...,\n",
      "       [-0.10432377, -0.11457533],\n",
      "       [-0.07777621, -0.19451702],\n",
      "       [-0.08167211, -0.04936979]], dtype=float32), array([[-0.11817341, -0.03636614],\n",
      "       [-0.02943886, -0.15564537],\n",
      "       [-0.10072929, -0.05651584],\n",
      "       ...,\n",
      "       [-0.05208138, -0.19116855],\n",
      "       [-0.13770975, -0.10180248],\n",
      "       [-0.12697117, -0.10352042]], dtype=float32), array([[-0.04843969, -0.20931114],\n",
      "       [-0.18545437, -0.1312336 ],\n",
      "       [-0.16431458, -0.23595235],\n",
      "       ...,\n",
      "       [-0.13847516, -0.15858176],\n",
      "       [-0.16573256, -0.16998392],\n",
      "       [-0.16963068, -0.08576098]], dtype=float32), array([[-0.09309182, -0.09390479],\n",
      "       [-0.05693859, -0.11729397],\n",
      "       [-0.11499198, -0.07967059],\n",
      "       ...,\n",
      "       [-0.13393477, -0.10266078],\n",
      "       [-0.04924463, -0.0680997 ],\n",
      "       [-0.03794032, -0.05001358]], dtype=float32), array([[-0.13115099, -0.20005086],\n",
      "       [-0.03531853, -0.10783227],\n",
      "       [-0.17447236, -0.12658255],\n",
      "       ...,\n",
      "       [-0.19167249, -0.11026362],\n",
      "       [-0.00781912, -0.09460847],\n",
      "       [-0.11935879, -0.06561916]], dtype=float32), array([[-0.10997014, -0.07711617],\n",
      "       [-0.10373071, -0.10572213],\n",
      "       [-0.18417922, -0.17276993],\n",
      "       ...,\n",
      "       [-0.13214628, -0.10396378],\n",
      "       [-0.08753316, -0.18139115],\n",
      "       [-0.14258146, -0.19139403]], dtype=float32), array([[-0.08197098, -0.15146013],\n",
      "       [-0.07389113, -0.08568378],\n",
      "       [-0.0516216 , -0.20924054],\n",
      "       ...,\n",
      "       [-0.17921665, -0.11317268],\n",
      "       [-0.15696688, -0.20239231],\n",
      "       [-0.10485709, -0.14425805]], dtype=float32), array([[-0.06041029, -0.15437713],\n",
      "       [-0.05452196, -0.11329359],\n",
      "       [-0.13105817, -0.20212714],\n",
      "       ...,\n",
      "       [-0.14564148, -0.12068386],\n",
      "       [-0.15272817, -0.1519359 ],\n",
      "       [-0.06133653, -0.09312334]], dtype=float32), array([[-0.1137688 , -0.13160118],\n",
      "       [-0.2097278 , -0.09246977],\n",
      "       [-0.18490177, -0.10644498],\n",
      "       ...,\n",
      "       [-0.13420843, -0.06455914],\n",
      "       [-0.15171547, -0.07608189],\n",
      "       [-0.13249396, -0.17003684]], dtype=float32), array([[-0.06756291, -0.10709944],\n",
      "       [-0.16032518, -0.11916131],\n",
      "       [-0.16225336, -0.15146211],\n",
      "       ...,\n",
      "       [-0.09728722, -0.04980183],\n",
      "       [-0.12189385, -0.14299248],\n",
      "       [-0.13708362, -0.16507097]], dtype=float32), array([[-0.13823357, -0.07778165],\n",
      "       [-0.09603202, -0.0885376 ],\n",
      "       [-0.10759582, -0.19701234],\n",
      "       ...,\n",
      "       [-0.08501067, -0.10379375],\n",
      "       [-0.23156342, -0.18480101],\n",
      "       [-0.16379368, -0.1793944 ]], dtype=float32), array([[-0.17941865, -0.19084671],\n",
      "       [-0.10304973, -0.20845796],\n",
      "       [-0.1756866 , -0.19619529],\n",
      "       ...,\n",
      "       [-0.10709026, -0.0612867 ],\n",
      "       [-0.17359036, -0.06376286],\n",
      "       [-0.11719277, -0.12880935]], dtype=float32), array([[-0.09813723, -0.06102981],\n",
      "       [-0.13076952, -0.0347729 ],\n",
      "       [-0.14627528, -0.18709414],\n",
      "       ...,\n",
      "       [-0.1591926 , -0.16809674],\n",
      "       [-0.1743315 , -0.09024116],\n",
      "       [-0.12703654, -0.07979438]], dtype=float32), array([[-0.06879442, -0.10010874],\n",
      "       [-0.10648812, -0.11463622],\n",
      "       [-0.04565511, -0.17220454],\n",
      "       ...,\n",
      "       [-0.20215277, -0.11538669],\n",
      "       [-0.05083453, -0.03596466],\n",
      "       [-0.14254378, -0.14217581]], dtype=float32), array([[-0.19998895, -0.08379712],\n",
      "       [-0.11000605, -0.13626358],\n",
      "       [-0.13494757, -0.12587543],\n",
      "       ...,\n",
      "       [-0.14421132, -0.14230566],\n",
      "       [-0.14413011, -0.08822533],\n",
      "       [-0.15999094, -0.14042947]], dtype=float32), array([[-0.10865754, -0.11019506],\n",
      "       [-0.172292  , -0.12728442],\n",
      "       [-0.0693235 , -0.15131089],\n",
      "       ...,\n",
      "       [-0.22235009, -0.17773278],\n",
      "       [-0.17605707, -0.14759716],\n",
      "       [-0.06608705, -0.12471202]], dtype=float32), array([[-0.11589941, -0.15543562],\n",
      "       [-0.15172385, -0.05582351],\n",
      "       [-0.04637762, -0.20459141],\n",
      "       ...,\n",
      "       [-0.22753103, -0.16899388],\n",
      "       [-0.08190566, -0.23326656],\n",
      "       [-0.204809  , -0.10389359]], dtype=float32), array([[-0.17598987, -0.03765145],\n",
      "       [-0.20834346, -0.11621943],\n",
      "       [-0.13641506, -0.0845672 ],\n",
      "       ...,\n",
      "       [-0.15401448, -0.1455168 ],\n",
      "       [-0.12347526, -0.10955377],\n",
      "       [-0.16703519, -0.16552703]], dtype=float32), array([[-0.1338635 , -0.13342507],\n",
      "       [-0.16196494, -0.09801425],\n",
      "       [-0.07117899, -0.0923063 ],\n",
      "       ...,\n",
      "       [-0.15296139, -0.0890063 ],\n",
      "       [-0.16407438, -0.1602951 ],\n",
      "       [-0.09301545, -0.07156   ]], dtype=float32), array([[-0.03748368, -0.10211256],\n",
      "       [-0.12544061, -0.1313232 ],\n",
      "       [-0.10381979, -0.1735504 ],\n",
      "       ...,\n",
      "       [-0.09023498, -0.06349203],\n",
      "       [-0.02649651, -0.09582682],\n",
      "       [-0.22000742, -0.14082378]], dtype=float32), array([[-0.09136149, -0.07767242],\n",
      "       [-0.14992703, -0.0144429 ],\n",
      "       [-0.11754459, -0.05533516],\n",
      "       ...,\n",
      "       [-0.06330275, -0.10555261],\n",
      "       [-0.13222897, -0.11496547],\n",
      "       [-0.17782421, -0.17455347]], dtype=float32), array([[-0.09598844, -0.0868334 ],\n",
      "       [-0.07324035, -0.1842526 ],\n",
      "       [-0.03174331, -0.1266043 ],\n",
      "       ...,\n",
      "       [-0.11821376, -0.12012026],\n",
      "       [-0.1467978 , -0.12715222],\n",
      "       [-0.13425869, -0.09584543]], dtype=float32), array([[-0.16835725, -0.10122532],\n",
      "       [-0.17047757, -0.19854541],\n",
      "       [-0.16925934, -0.0346225 ],\n",
      "       ...,\n",
      "       [-0.11894952, -0.12610427],\n",
      "       [-0.15161207, -0.06749716],\n",
      "       [-0.09450744, -0.1016803 ]], dtype=float32), array([[-0.17997   , -0.1106399 ],\n",
      "       [-0.13443168, -0.16809241],\n",
      "       [-0.21549323, -0.20883591],\n",
      "       ...,\n",
      "       [-0.15885438, -0.15156421],\n",
      "       [-0.13917407, -0.13284415],\n",
      "       [-0.19297802, -0.09605265]], dtype=float32), array([[-0.17626022, -0.20369124],\n",
      "       [-0.17800161, -0.14083299],\n",
      "       [-0.18005478, -0.13636214],\n",
      "       ...,\n",
      "       [-0.15497637, -0.16269311],\n",
      "       [-0.0469254 , -0.0614472 ],\n",
      "       [-0.17037223, -0.19187452]], dtype=float32), array([[-0.11909258, -0.21099138],\n",
      "       [-0.14147717, -0.18272738],\n",
      "       [-0.17937174, -0.03374496],\n",
      "       ...,\n",
      "       [-0.07682525, -0.11307138],\n",
      "       [-0.12271994, -0.12364547],\n",
      "       [-0.10984305, -0.15268044]], dtype=float32), array([[-0.05521431, -0.19480595],\n",
      "       [-0.13371725, -0.06352061],\n",
      "       [-0.16000566, -0.14212903],\n",
      "       ...,\n",
      "       [-0.09245495, -0.09558292],\n",
      "       [-0.09793006, -0.17958279],\n",
      "       [-0.15508638, -0.06658361]], dtype=float32), array([[-0.12833676, -0.1543225 ],\n",
      "       [-0.10195038, -0.11190069],\n",
      "       [-0.14243151, -0.15313834],\n",
      "       ...,\n",
      "       [-0.14932294, -0.1652253 ],\n",
      "       [-0.1404844 , -0.130019  ],\n",
      "       [-0.15293811, -0.1213826 ]], dtype=float32), array([[-0.16972715, -0.1659075 ],\n",
      "       [-0.25579152, -0.11613295],\n",
      "       [-0.10105971, -0.17187543],\n",
      "       ...,\n",
      "       [-0.18332717, -0.14974225],\n",
      "       [-0.10370463, -0.09965788],\n",
      "       [-0.168942  , -0.14843479]], dtype=float32), array([[-0.17091674, -0.07458842],\n",
      "       [-0.10730687, -0.11388034],\n",
      "       [-0.18143034, -0.07688228],\n",
      "       ...,\n",
      "       [-0.22739528, -0.1570503 ],\n",
      "       [-0.04795042, -0.05662959],\n",
      "       [-0.14249079, -0.07157628]], dtype=float32), array([[-0.13317865, -0.11273965],\n",
      "       [-0.13233373, -0.16666418],\n",
      "       [-0.1255956 , -0.0952547 ],\n",
      "       ...,\n",
      "       [-0.13775112, -0.05905051],\n",
      "       [-0.0867906 , -0.17748885],\n",
      "       [-0.19474193, -0.11153268]], dtype=float32), array([[-0.14804168, -0.18587911],\n",
      "       [-0.12961186, -0.13823353],\n",
      "       [-0.16258606, -0.08971591],\n",
      "       ...,\n",
      "       [-0.15088639, -0.12960623],\n",
      "       [-0.16469307, -0.08447277],\n",
      "       [-0.06133368, -0.21470354]], dtype=float32), array([[-0.13201505, -0.16412055],\n",
      "       [-0.08216473, -0.05696087],\n",
      "       [-0.19933417, -0.08891168],\n",
      "       ...,\n",
      "       [-0.12417572, -0.0770592 ],\n",
      "       [-0.1528231 , -0.04639693],\n",
      "       [-0.09342077, -0.15323545]], dtype=float32), array([[-0.11549712, -0.13970262],\n",
      "       [-0.13198811, -0.06884127],\n",
      "       [-0.1290568 , -0.15386963],\n",
      "       ...,\n",
      "       [-0.08466011, -0.06160564],\n",
      "       [-0.16905947, -0.16452405],\n",
      "       [-0.09654797, -0.08396529]], dtype=float32), array([[-0.14652224, -0.0825812 ],\n",
      "       [-0.16967288, -0.16270751],\n",
      "       [-0.11517986, -0.05506199],\n",
      "       ...,\n",
      "       [-0.13638997, -0.13040698],\n",
      "       [-0.14796075, -0.14156021],\n",
      "       [-0.15079106, -0.06834966]], dtype=float32), array([[-0.19752601, -0.16524495],\n",
      "       [-0.10420656, -0.0465981 ],\n",
      "       [-0.06381158, -0.08175069],\n",
      "       ...,\n",
      "       [-0.09104593, -0.10691755],\n",
      "       [-0.07917699, -0.15511681],\n",
      "       [-0.14289045, -0.16617815]], dtype=float32), array([[-0.15472177, -0.17492023],\n",
      "       [-0.15785517, -0.1422662 ],\n",
      "       [-0.19810489, -0.11019158],\n",
      "       ...,\n",
      "       [-0.13502562, -0.13613729],\n",
      "       [-0.14448237, -0.1588096 ],\n",
      "       [-0.08374635, -0.06813328]], dtype=float32), array([[-0.10702991, -0.10931224],\n",
      "       [-0.22801396, -0.15843508],\n",
      "       [-0.08409625, -0.1413038 ],\n",
      "       ...,\n",
      "       [-0.1426898 , -0.07513205],\n",
      "       [-0.08458428, -0.1370708 ],\n",
      "       [-0.18341711, -0.12848665]], dtype=float32), array([[-0.11403464, -0.21930757],\n",
      "       [-0.0916763 , -0.19737622],\n",
      "       [-0.1380417 , -0.1213036 ],\n",
      "       ...,\n",
      "       [-0.1502493 , -0.12612057],\n",
      "       [-0.19811141, -0.17498288],\n",
      "       [-0.12486202, -0.08356053]], dtype=float32), array([[-0.19074854, -0.13041234],\n",
      "       [-0.04642609, -0.14731315],\n",
      "       [-0.20660488, -0.12478527],\n",
      "       ...,\n",
      "       [-0.15264234, -0.09913626],\n",
      "       [-0.09013166, -0.23179443],\n",
      "       [-0.07469551, -0.11540645]], dtype=float32), array([[-0.07674406, -0.18260191],\n",
      "       [-0.03978457, -0.17350647],\n",
      "       [-0.10475735, -0.08185153],\n",
      "       ...,\n",
      "       [-0.13898957, -0.16496013],\n",
      "       [-0.13644503, -0.08804738],\n",
      "       [-0.17945911, -0.12010814]], dtype=float32), array([[-0.044523  , -0.1124459 ],\n",
      "       [-0.1066047 , -0.18235253],\n",
      "       [-0.10289676, -0.0714946 ],\n",
      "       ...,\n",
      "       [-0.15017939, -0.17211856],\n",
      "       [-0.13935034, -0.03745014],\n",
      "       [-0.07516879, -0.11801428]], dtype=float32), array([[-0.1747689 , -0.14627552],\n",
      "       [-0.08568648, -0.08284468],\n",
      "       [-0.17904367, -0.10095244],\n",
      "       ...,\n",
      "       [-0.12388507, -0.1520569 ],\n",
      "       [-0.14037293, -0.02148483],\n",
      "       [-0.09627619, -0.05698512]], dtype=float32), array([[-0.12482192, -0.21806785],\n",
      "       [-0.07326397, -0.03501823],\n",
      "       [-0.08070597, -0.15619609],\n",
      "       ...,\n",
      "       [-0.12070602, -0.03155429],\n",
      "       [-0.09581748, -0.15321462],\n",
      "       [-0.11330551, -0.16430235]], dtype=float32), array([[-0.17893372, -0.15751225],\n",
      "       [-0.11175349, -0.14879191],\n",
      "       [-0.14686328, -0.14685625],\n",
      "       ...,\n",
      "       [-0.11580049, -0.13425206],\n",
      "       [-0.12526383, -0.02627475],\n",
      "       [-0.11436513, -0.14415663]], dtype=float32), array([[-0.19008893, -0.12676723],\n",
      "       [-0.14532545, -0.12918784],\n",
      "       [-0.03584491, -0.17484133],\n",
      "       ...,\n",
      "       [-0.11993406, -0.18816839],\n",
      "       [-0.11417604, -0.16168107],\n",
      "       [-0.15014456, -0.1218549 ]], dtype=float32), array([[-0.0497546 , -0.09618214],\n",
      "       [-0.21904156, -0.09980526],\n",
      "       [-0.12359386, -0.10999627],\n",
      "       ...,\n",
      "       [-0.1295447 , -0.11710048],\n",
      "       [ 0.00185582, -0.23586799],\n",
      "       [-0.1857467 , -0.09033152]], dtype=float32), array([[-0.09025657, -0.08667264],\n",
      "       [-0.02474688, -0.13429625],\n",
      "       [-0.05036566, -0.09937616],\n",
      "       ...,\n",
      "       [-0.1742666 , -0.1617832 ],\n",
      "       [-0.07177159, -0.10489106],\n",
      "       [-0.15663877, -0.18601787]], dtype=float32), array([[-0.15007007, -0.04579732],\n",
      "       [-0.20878425, -0.03086765],\n",
      "       [-0.03639678, -0.07879689],\n",
      "       ...,\n",
      "       [-0.10681011, -0.12020849],\n",
      "       [-0.02369911, -0.06059569],\n",
      "       [-0.12100284, -0.13948108]], dtype=float32), array([[-0.13996193, -0.14736529],\n",
      "       [-0.09496886, -0.15076031],\n",
      "       [-0.10538046, -0.118288  ],\n",
      "       ...,\n",
      "       [-0.08991499, -0.14143595],\n",
      "       [-0.06642923, -0.10441828],\n",
      "       [-0.06828518, -0.1675502 ]], dtype=float32), array([[-0.13806616, -0.04325547],\n",
      "       [-0.168767  , -0.09660642],\n",
      "       [-0.09032822, -0.04727171],\n",
      "       ...,\n",
      "       [-0.20657347, -0.13303468],\n",
      "       [-0.05220417, -0.16468634],\n",
      "       [-0.1464434 , -0.23034202]], dtype=float32), array([[-0.13132477, -0.18184471],\n",
      "       [-0.13208291, -0.10293137],\n",
      "       [-0.10557705, -0.12855886],\n",
      "       ...,\n",
      "       [-0.14241552, -0.16238916],\n",
      "       [-0.14402753, -0.17835292],\n",
      "       [-0.2218487 , -0.15123276]], dtype=float32), array([[-0.1479667 , -0.06154051],\n",
      "       [-0.19395518, -0.04326924],\n",
      "       [-0.19575459, -0.07683694],\n",
      "       ...,\n",
      "       [-0.06850191, -0.11576045],\n",
      "       [-0.12859705, -0.17560929],\n",
      "       [-0.15095368, -0.18910602]], dtype=float32), array([[-0.18200476, -0.13334523],\n",
      "       [-0.16298088, -0.11223492],\n",
      "       [-0.09431463, -0.10133602],\n",
      "       ...,\n",
      "       [-0.21568976, -0.06744772],\n",
      "       [-0.15868333, -0.09952226],\n",
      "       [-0.15677962, -0.18427384]], dtype=float32), array([[-0.12610558, -0.13888876],\n",
      "       [-0.12962644, -0.12959118],\n",
      "       [-0.11736555, -0.07516077],\n",
      "       ...,\n",
      "       [-0.1848169 , -0.22088109],\n",
      "       [-0.11866364, -0.20787084],\n",
      "       [-0.20119558, -0.15753397]], dtype=float32), array([[-0.08084893, -0.18691641],\n",
      "       [-0.09070449, -0.02389285],\n",
      "       [-0.22234158, -0.09150629],\n",
      "       ...,\n",
      "       [-0.13480262, -0.09648053],\n",
      "       [-0.08612172, -0.11796828],\n",
      "       [-0.07796457, -0.14762333]], dtype=float32), array([[-0.11512873, -0.18620878],\n",
      "       [-0.06162035, -0.09949338],\n",
      "       [-0.19461633, -0.104544  ],\n",
      "       ...,\n",
      "       [-0.05653153, -0.0917932 ],\n",
      "       [-0.10976446, -0.18381421],\n",
      "       [-0.12814659, -0.07809196]], dtype=float32), array([[-0.1585981 , -0.07627721],\n",
      "       [-0.14369339, -0.08074534],\n",
      "       [-0.14385194, -0.10968497],\n",
      "       ...,\n",
      "       [-0.09576465, -0.16305242],\n",
      "       [-0.09988698, -0.14203951],\n",
      "       [-0.13105148, -0.04316316]], dtype=float32), array([[-0.15016977, -0.11829355],\n",
      "       [-0.0885694 , -0.11609217],\n",
      "       [-0.0420789 , -0.09618907],\n",
      "       ...,\n",
      "       [-0.14823028, -0.15717205],\n",
      "       [-0.08234622, -0.21639623],\n",
      "       [-0.15531348, -0.12760018]], dtype=float32), array([[-0.19930306, -0.2043485 ],\n",
      "       [-0.19965225, -0.10420739],\n",
      "       [-0.20523134, -0.1581698 ],\n",
      "       ...,\n",
      "       [-0.12749422, -0.1738849 ],\n",
      "       [-0.11781287, -0.1606482 ],\n",
      "       [-0.09316926, -0.06891327]], dtype=float32), array([[-0.14650694, -0.02921626],\n",
      "       [-0.2106895 , -0.13918124],\n",
      "       [-0.16816452, -0.06318112],\n",
      "       ...,\n",
      "       [-0.08882765, -0.16626827],\n",
      "       [-0.03298642, -0.15522163],\n",
      "       [-0.0370278 , -0.09260011]], dtype=float32), array([[-0.17828873, -0.15771714],\n",
      "       [-0.09214156, -0.17764002],\n",
      "       [-0.15465792, -0.10491984],\n",
      "       ...,\n",
      "       [-0.03050534, -0.156508  ],\n",
      "       [-0.07859881, -0.05785437],\n",
      "       [-0.12179411, -0.08990696]], dtype=float32), array([[-0.21284102, -0.13838966],\n",
      "       [-0.16465858, -0.1042154 ],\n",
      "       [-0.07250795, -0.07433683],\n",
      "       ...,\n",
      "       [-0.13448931, -0.11398347],\n",
      "       [-0.11284971, -0.0564485 ],\n",
      "       [-0.10176106, -0.13316415]], dtype=float32), array([[-0.10621536, -0.12025632],\n",
      "       [-0.09873202, -0.14775608],\n",
      "       [-0.15049258, -0.06966597],\n",
      "       ...,\n",
      "       [-0.09855526, -0.07896975],\n",
      "       [-0.09050024, -0.07460484],\n",
      "       [-0.07736115, -0.1414557 ]], dtype=float32), array([[-0.09101056, -0.14268771],\n",
      "       [-0.15216465, -0.11760989],\n",
      "       [-0.08569868, -0.11838463],\n",
      "       ...,\n",
      "       [-0.06442276, -0.07850402],\n",
      "       [-0.04023621, -0.12668103],\n",
      "       [-0.08747856, -0.19629237]], dtype=float32), array([[-0.06337006, -0.15470874],\n",
      "       [-0.17253998, -0.20519081],\n",
      "       [-0.09275047, -0.16939694],\n",
      "       ...,\n",
      "       [-0.09839325, -0.10676086],\n",
      "       [-0.18357027, -0.14988764],\n",
      "       [-0.2015397 , -0.08864633]], dtype=float32), array([[-0.15655759, -0.07303577],\n",
      "       [-0.1606832 , -0.06939139],\n",
      "       [-0.18937601, -0.12066202],\n",
      "       ...,\n",
      "       [-0.07817851, -0.13299483],\n",
      "       [-0.08840489, -0.20791903],\n",
      "       [-0.02888539, -0.08912779]], dtype=float32), array([[-0.07034096, -0.08616195],\n",
      "       [-0.15078036, -0.19846126],\n",
      "       [-0.11235116, -0.06967682],\n",
      "       ...,\n",
      "       [-0.18392438, -0.07450312],\n",
      "       [-0.10863635, -0.09724211],\n",
      "       [-0.12478662, -0.15780383]], dtype=float32), array([[-0.10463329, -0.13850029],\n",
      "       [-0.16861163, -0.12345431],\n",
      "       [-0.00944297, -0.12806535],\n",
      "       ...,\n",
      "       [-0.27063164, -0.12447986],\n",
      "       [-0.16782565, -0.1406549 ],\n",
      "       [-0.16370083, -0.18103981]], dtype=float32), array([[-0.10510568, -0.10002428],\n",
      "       [-0.05038411, -0.17914626],\n",
      "       [-0.05755374, -0.02872365],\n",
      "       ...,\n",
      "       [-0.15654835, -0.00257575],\n",
      "       [-0.15356137, -0.14452651],\n",
      "       [-0.06215907, -0.1571174 ]], dtype=float32), array([[-0.16358313, -0.11349313],\n",
      "       [-0.13926664, -0.06882828],\n",
      "       [-0.08578321, -0.1393313 ],\n",
      "       ...,\n",
      "       [-0.14273669, -0.08313767],\n",
      "       [-0.11012083, -0.0817814 ],\n",
      "       [-0.07549386, -0.13216351]], dtype=float32), array([[-0.08322988, -0.01860059],\n",
      "       [-0.20226371, -0.15316623],\n",
      "       [-0.15915218, -0.11957458],\n",
      "       ...,\n",
      "       [-0.17511629, -0.12147555],\n",
      "       [-0.08211923, -0.1811997 ],\n",
      "       [-0.05785236, -0.07624552]], dtype=float32), array([[-0.09294113, -0.13502836],\n",
      "       [-0.18674164, -0.20747861],\n",
      "       [-0.23174576, -0.17239565],\n",
      "       ...,\n",
      "       [-0.14216489, -0.19810492],\n",
      "       [-0.21194258, -0.16580531],\n",
      "       [-0.11843228, -0.14883964]], dtype=float32), array([[-0.06913724, -0.03805468],\n",
      "       [-0.086382  , -0.06178598],\n",
      "       [-0.14855367, -0.0473504 ],\n",
      "       ...,\n",
      "       [-0.1439897 , -0.15287657],\n",
      "       [-0.14353532, -0.15922019],\n",
      "       [-0.1375872 , -0.12702239]], dtype=float32), array([[-0.12415466, -0.1611248 ],\n",
      "       [-0.09554369, -0.0286546 ],\n",
      "       [-0.15606888, -0.10671141],\n",
      "       ...,\n",
      "       [-0.06701748, -0.09131048],\n",
      "       [-0.05343989, -0.07886886],\n",
      "       [-0.03723019, -0.08673538]], dtype=float32), array([[-0.10792259, -0.09721148],\n",
      "       [-0.04707403, -0.13317841],\n",
      "       [-0.12620722, -0.09895674],\n",
      "       ...,\n",
      "       [-0.04090516, -0.13187326],\n",
      "       [-0.15567815, -0.01836697],\n",
      "       [-0.05646631, -0.12400496]], dtype=float32), array([[-0.06141383, -0.07052136],\n",
      "       [-0.04269809, -0.14121264],\n",
      "       [-0.13701938, -0.09593248],\n",
      "       ...,\n",
      "       [-0.18885353, -0.16133082],\n",
      "       [-0.1473276 , -0.06737536],\n",
      "       [-0.1534757 , -0.11717645]], dtype=float32), array([[-0.22669783, -0.13201712],\n",
      "       [-0.1962434 , -0.2028026 ],\n",
      "       [-0.20471208, -0.11275259],\n",
      "       ...,\n",
      "       [-0.16652882, -0.05283797],\n",
      "       [-0.16496347, -0.06156228],\n",
      "       [-0.04842011, -0.15948527]], dtype=float32), array([[-0.18033394, -0.14497295],\n",
      "       [-0.11409085, -0.12444185],\n",
      "       [-0.1564497 , -0.02424912],\n",
      "       ...,\n",
      "       [-0.15051265, -0.11195561],\n",
      "       [-0.14703397, -0.13880661],\n",
      "       [-0.1638624 , -0.07511269]], dtype=float32), array([[-0.13046876, -0.08698712],\n",
      "       [-0.09415127, -0.09758955],\n",
      "       [-0.1917573 , -0.03812116],\n",
      "       ...,\n",
      "       [-0.13114071, -0.04772912],\n",
      "       [-0.12230234, -0.02584953],\n",
      "       [-0.0223985 , -0.11070378]], dtype=float32), array([[-0.14861421, -0.21506913],\n",
      "       [-0.15446095, -0.11148316],\n",
      "       [-0.14749508, -0.09635814],\n",
      "       ...,\n",
      "       [-0.14488517, -0.05642219],\n",
      "       [-0.17891717, -0.00837738],\n",
      "       [-0.16964373, -0.09589477]], dtype=float32), array([[-0.18632764, -0.09105914],\n",
      "       [-0.08343393, -0.12852387],\n",
      "       [-0.14999777, -0.15999322],\n",
      "       ...,\n",
      "       [-0.24442178, -0.09538306],\n",
      "       [-0.13464864, -0.08556009],\n",
      "       [-0.06296807, -0.1481019 ]], dtype=float32), array([[-0.1907236 , -0.15251735],\n",
      "       [-0.08037115, -0.16178675],\n",
      "       [-0.02250092, -0.17270488],\n",
      "       ...,\n",
      "       [-0.1515604 , -0.07850508],\n",
      "       [-0.09301068, -0.03316961],\n",
      "       [-0.14995651, -0.19184272]], dtype=float32), array([[-0.17746489, -0.11183776],\n",
      "       [-0.09772605, -0.16080733],\n",
      "       [-0.05121041, -0.10503142],\n",
      "       ...,\n",
      "       [-0.15798199, -0.13084257],\n",
      "       [-0.04811297, -0.10949393],\n",
      "       [-0.01118397, -0.21821095]], dtype=float32), array([[-0.09792008, -0.13079223],\n",
      "       [-0.09149974, -0.08967015],\n",
      "       [-0.13470186, -0.13396564],\n",
      "       ...,\n",
      "       [-0.1027092 , -0.1594966 ],\n",
      "       [-0.13656317, -0.19790214],\n",
      "       [-0.06818247, -0.19692118]], dtype=float32), array([[-0.1648369 , -0.11095896],\n",
      "       [-0.14962184, -0.07523717],\n",
      "       [-0.20243156, -0.11674532],\n",
      "       ...,\n",
      "       [-0.12870415, -0.19011535],\n",
      "       [-0.0251692 , -0.0405643 ],\n",
      "       [-0.13069645, -0.09425343]], dtype=float32), array([[-0.10059332, -0.14829542],\n",
      "       [-0.06152335, -0.1832699 ],\n",
      "       [-0.21776909, -0.09849405],\n",
      "       ...,\n",
      "       [-0.13997501, -0.06913049],\n",
      "       [-0.07296169, -0.1122785 ],\n",
      "       [-0.09725742, -0.15151863]], dtype=float32), array([[-0.06881644, -0.05764177],\n",
      "       [-0.18369465, -0.09685294],\n",
      "       [-0.064973  , -0.16441911],\n",
      "       ...,\n",
      "       [-0.09068434, -0.1314368 ],\n",
      "       [-0.067545  , -0.1614722 ],\n",
      "       [-0.03322049, -0.13571972]], dtype=float32), array([[-0.0848949 , -0.1412193 ],\n",
      "       [-0.07424138, -0.02412072],\n",
      "       [-0.04736136, -0.18279144],\n",
      "       ...,\n",
      "       [-0.1501329 , -0.14454544],\n",
      "       [-0.16942988, -0.18956132],\n",
      "       [-0.15425475, -0.1910641 ]], dtype=float32), array([[-0.1463279 , -0.1523383 ],\n",
      "       [-0.0797442 , -0.08287734],\n",
      "       [-0.1268949 , -0.09619731],\n",
      "       ...,\n",
      "       [-0.07814629, -0.03231731],\n",
      "       [-0.06853322, -0.14944859],\n",
      "       [-0.04367679, -0.19801727]], dtype=float32), array([[-0.18262541, -0.12912238],\n",
      "       [-0.09229242, -0.04035888],\n",
      "       [-0.08247368, -0.13641891],\n",
      "       ...,\n",
      "       [-0.17863055, -0.15169194],\n",
      "       [-0.14772466, -0.05883241],\n",
      "       [-0.09458551, -0.18705647]], dtype=float32), array([[-0.05685416, -0.05526835],\n",
      "       [-0.19749598, -0.08491658],\n",
      "       [-0.17908253, -0.08500383],\n",
      "       ...,\n",
      "       [-0.14001477, -0.15265149],\n",
      "       [-0.13967472, -0.17064974],\n",
      "       [-0.10252549, -0.12060271]], dtype=float32), array([[-0.15260786, -0.17423747],\n",
      "       [-0.13480923, -0.14313093],\n",
      "       [-0.19076133, -0.1848474 ],\n",
      "       ...,\n",
      "       [-0.08294709, -0.13380201],\n",
      "       [-0.17112248, -0.10388027],\n",
      "       [-0.15931387, -0.13844106]], dtype=float32), array([[-0.13045229, -0.14685076],\n",
      "       [-0.11074544, -0.10221934],\n",
      "       [-0.05925822, -0.15112858],\n",
      "       ...,\n",
      "       [-0.08169234, -0.0242656 ],\n",
      "       [-0.13472247, -0.16145584],\n",
      "       [-0.0899098 , -0.06691629]], dtype=float32), array([[-0.04301517, -0.08009117],\n",
      "       [-0.17527014, -0.05442261],\n",
      "       [-0.06225785, -0.07935107],\n",
      "       ...,\n",
      "       [-0.16334414, -0.22335242],\n",
      "       [-0.0991762 , -0.11843368],\n",
      "       [-0.09723415, -0.14552355]], dtype=float32), array([[-0.18622358, -0.22420068],\n",
      "       [-0.09879829, -0.16410223],\n",
      "       [-0.0478717 , -0.11878321],\n",
      "       ...,\n",
      "       [-0.06356183, -0.13398643],\n",
      "       [-0.158056  , -0.13184527],\n",
      "       [-0.16653672, -0.06438064]], dtype=float32), array([[-0.14928196, -0.13002089],\n",
      "       [-0.07732505, -0.16729271],\n",
      "       [-0.11363261, -0.14757165],\n",
      "       ...,\n",
      "       [-0.16312484, -0.06026006],\n",
      "       [-0.18409258, -0.09324311],\n",
      "       [-0.17449163, -0.13524842]], dtype=float32), array([[-0.15589722, -0.04978243],\n",
      "       [-0.15364307, -0.22835939],\n",
      "       [-0.14631434, -0.07550741],\n",
      "       ...,\n",
      "       [-0.07240505, -0.14805645],\n",
      "       [-0.12130734, -0.11091615],\n",
      "       [-0.14685   , -0.04719983]], dtype=float32), array([[-0.12750654, -0.09871345],\n",
      "       [-0.13335977, -0.15461005],\n",
      "       [-0.20917505, -0.05815005],\n",
      "       ...,\n",
      "       [-0.07583277, -0.14081247],\n",
      "       [-0.158666  , -0.14048637],\n",
      "       [-0.19413963, -0.11292055]], dtype=float32), array([[-0.19389424, -0.07666577],\n",
      "       [-0.16086821, -0.14783299],\n",
      "       [-0.10962667, -0.14061624],\n",
      "       ...,\n",
      "       [-0.11443435, -0.20096461],\n",
      "       [-0.12458152, -0.13728754],\n",
      "       [-0.10782786, -0.20476004]], dtype=float32), array([[-0.15333691, -0.11182223],\n",
      "       [-0.1306495 , -0.06824978],\n",
      "       [-0.13069749, -0.12186863],\n",
      "       ...,\n",
      "       [-0.15589029, -0.09816211],\n",
      "       [-0.06915247, -0.0808624 ],\n",
      "       [-0.20314285, -0.15847653]], dtype=float32), array([[-0.1209847 , -0.16887867],\n",
      "       [-0.17843528, -0.13213757],\n",
      "       [-0.1373615 , -0.20404467],\n",
      "       ...,\n",
      "       [-0.09601144, -0.14160475],\n",
      "       [-0.16851756, -0.19834411],\n",
      "       [-0.10983457, -0.11629175]], dtype=float32), array([[-0.1472451 , -0.02888496],\n",
      "       [-0.16183867, -0.13759242],\n",
      "       [-0.05820088, -0.09052683],\n",
      "       ...,\n",
      "       [-0.06819744, -0.21346617],\n",
      "       [-0.03823102, -0.21177077],\n",
      "       [-0.18102527, -0.16975583]], dtype=float32), array([[-0.1524759 , -0.23081785],\n",
      "       [-0.15530784, -0.11779378],\n",
      "       [-0.15598845, -0.14370516],\n",
      "       ...,\n",
      "       [-0.09038054, -0.14717399],\n",
      "       [-0.10113941, -0.11076751],\n",
      "       [-0.06014692, -0.1319826 ]], dtype=float32), array([[-0.16967645, -0.17686   ],\n",
      "       [-0.08526786, -0.14882468],\n",
      "       [-0.1053514 , -0.09702079],\n",
      "       ...,\n",
      "       [-0.15842906, -0.10240679],\n",
      "       [-0.10810175, -0.20125964],\n",
      "       [-0.08970094, -0.15741867]], dtype=float32), array([[-0.1078888 , -0.13384634],\n",
      "       [-0.14514855, -0.2582035 ],\n",
      "       [-0.09971835, -0.10492329],\n",
      "       ...,\n",
      "       [-0.09002198, -0.0507311 ],\n",
      "       [-0.13571927, -0.153976  ],\n",
      "       [-0.0700369 , -0.13243125]], dtype=float32), array([[-0.11096965, -0.119427  ],\n",
      "       [-0.15225083, -0.08466126],\n",
      "       [-0.15183775, -0.13091496],\n",
      "       ...,\n",
      "       [-0.15593642, -0.09271974],\n",
      "       [-0.1294466 , -0.05548253],\n",
      "       [-0.14898963, -0.05446853]], dtype=float32), array([[-0.11643179, -0.21268532],\n",
      "       [-0.09890739, -0.14193702],\n",
      "       [-0.11847021, -0.24211815],\n",
      "       ...,\n",
      "       [-0.16103151, -0.20769033],\n",
      "       [-0.17697725, -0.2187895 ],\n",
      "       [-0.06162669, -0.04720311]], dtype=float32), array([[-0.1165112 , -0.04887313],\n",
      "       [-0.08488281, -0.19982658],\n",
      "       [-0.08637466, -0.10641866],\n",
      "       ...,\n",
      "       [-0.19953753, -0.12837675],\n",
      "       [-0.04938107, -0.21544641],\n",
      "       [-0.16988568, -0.17152643]], dtype=float32), array([[-0.12601522, -0.20441702],\n",
      "       [-0.15893197, -0.08324322],\n",
      "       [-0.12085667, -0.11304581],\n",
      "       ...,\n",
      "       [-0.07925469, -0.09726956],\n",
      "       [-0.02524705, -0.08567899],\n",
      "       [-0.14131415, -0.12572406]], dtype=float32), array([[-0.12899481, -0.10850333],\n",
      "       [-0.11615325, -0.13241951],\n",
      "       [-0.12817587, -0.10599783],\n",
      "       ...,\n",
      "       [-0.12757266, -0.16105753],\n",
      "       [-0.1333101 , -0.1646729 ],\n",
      "       [-0.088749  , -0.1174328 ]], dtype=float32), array([[-0.10935669, -0.16216815],\n",
      "       [-0.12348682, -0.08523864],\n",
      "       [-0.1595337 , -0.16192749],\n",
      "       ...,\n",
      "       [-0.02776077, -0.09777635],\n",
      "       [-0.03327383, -0.20974667],\n",
      "       [-0.07454589, -0.11344708]], dtype=float32), array([[-0.131005  , -0.10868739],\n",
      "       [-0.2029183 , -0.07418986],\n",
      "       [-0.175326  , -0.14284381],\n",
      "       ...,\n",
      "       [-0.16022348, -0.13105915],\n",
      "       [-0.12899575, -0.07715087],\n",
      "       [-0.12872006, -0.14064616]], dtype=float32), array([[-0.19769838, -0.08255308],\n",
      "       [-0.08311651, -0.12414366],\n",
      "       [-0.14710242, -0.1525492 ],\n",
      "       ...,\n",
      "       [-0.11445697, -0.15918359],\n",
      "       [-0.08169796, -0.06111081],\n",
      "       [-0.11101795, -0.11841217]], dtype=float32), array([[-0.08829471, -0.03451327],\n",
      "       [-0.12507531, -0.06308614],\n",
      "       [-0.18117598, -0.10455553],\n",
      "       ...,\n",
      "       [-0.13869464, -0.06268255],\n",
      "       [-0.16351554, -0.06000188],\n",
      "       [-0.1713587 , -0.13561565]], dtype=float32), array([[-0.08044469, -0.13287453],\n",
      "       [-0.16937053, -0.10712549],\n",
      "       [-0.10190484, -0.10323751],\n",
      "       ...,\n",
      "       [-0.01802303, -0.12871465],\n",
      "       [-0.08196137, -0.10154223],\n",
      "       [-0.14126718, -0.12116852]], dtype=float32), array([[-0.17685917, -0.08607845],\n",
      "       [-0.13306603, -0.12980478],\n",
      "       [-0.1853836 , -0.04039282],\n",
      "       ...,\n",
      "       [-0.08750174, -0.14952719],\n",
      "       [-0.14129353, -0.16703454],\n",
      "       [-0.19113547, -0.1607371 ]], dtype=float32), array([[-0.09551916, -0.11216971],\n",
      "       [-0.08264321, -0.1792577 ],\n",
      "       [-0.14470585, -0.16628152],\n",
      "       ...,\n",
      "       [-0.18709464, -0.09616934],\n",
      "       [-0.06403591, -0.15303256],\n",
      "       [-0.22760719, -0.13727853]], dtype=float32), array([[-0.14690715, -0.14612494],\n",
      "       [-0.1765751 , -0.05925631],\n",
      "       [-0.21716389, -0.0790281 ],\n",
      "       ...,\n",
      "       [-0.09058627, -0.1596609 ],\n",
      "       [-0.11958309, -0.04273675],\n",
      "       [-0.17915228, -0.17573097]], dtype=float32), array([[-0.10133073, -0.09670094],\n",
      "       [-0.15569243, -0.05672102],\n",
      "       [-0.1655315 , -0.13223958],\n",
      "       ...,\n",
      "       [-0.00897819, -0.18587942],\n",
      "       [-0.06961878, -0.19909981],\n",
      "       [-0.17154607, -0.146218  ]], dtype=float32), array([[-0.03079765, -0.1157061 ],\n",
      "       [-0.12398097, -0.08765363],\n",
      "       [-0.08609091, -0.16906922],\n",
      "       ...,\n",
      "       [-0.10269044, -0.10289335],\n",
      "       [-0.149349  , -0.10264789],\n",
      "       [-0.07004553, -0.18274263]], dtype=float32), array([[-0.1316727 , -0.21136251],\n",
      "       [-0.06820635, -0.11039808],\n",
      "       [-0.05602678, -0.08077163],\n",
      "       ...,\n",
      "       [-0.20685875, -0.15631485],\n",
      "       [-0.13923998, -0.20457447],\n",
      "       [-0.16499336, -0.08859968]], dtype=float32), array([[-0.12117243, -0.1463043 ],\n",
      "       [-0.18711492, -0.13609396],\n",
      "       [-0.0108269 , -0.13340358],\n",
      "       ...,\n",
      "       [-0.1987925 , -0.18296385],\n",
      "       [-0.06152917, -0.07998059],\n",
      "       [-0.10527813, -0.10589058]], dtype=float32), array([[-0.06098827, -0.17477308],\n",
      "       [-0.06352129, -0.12714589],\n",
      "       [-0.19088769, -0.07235607],\n",
      "       ...,\n",
      "       [-0.07142076, -0.13256809],\n",
      "       [-0.08042585, -0.08529431],\n",
      "       [-0.10714334, -0.187973  ]], dtype=float32), array([[-0.10809582, -0.1472662 ],\n",
      "       [-0.18460216, -0.12891658],\n",
      "       [-0.17200781, -0.06980975],\n",
      "       ...,\n",
      "       [-0.20757495, -0.12366237],\n",
      "       [-0.13763641, -0.17199439],\n",
      "       [-0.05321629, -0.16650993]], dtype=float32), array([[-0.17386091, -0.07244583],\n",
      "       [-0.10809094, -0.13442403],\n",
      "       [-0.05941585, -0.08852728],\n",
      "       ...,\n",
      "       [-0.21708867, -0.09572686],\n",
      "       [-0.20729324, -0.1110993 ],\n",
      "       [-0.17679572, -0.0363608 ]], dtype=float32), array([[-0.1525562 , -0.14967985],\n",
      "       [-0.16071108, -0.11922259],\n",
      "       [-0.08067536, -0.05884471],\n",
      "       ...,\n",
      "       [-0.14655717, -0.13066867],\n",
      "       [-0.20716766, -0.15331075],\n",
      "       [-0.02501666, -0.08419176]], dtype=float32), array([[-0.15756111, -0.1228651 ],\n",
      "       [-0.09124933, -0.05028653],\n",
      "       [-0.14808194, -0.12789404],\n",
      "       ...,\n",
      "       [-0.11820013, -0.05218874],\n",
      "       [-0.08560198, -0.14974226],\n",
      "       [-0.10087973, -0.11052264]], dtype=float32), array([[-0.171134  , -0.06728213],\n",
      "       [-0.07915534, -0.1497516 ],\n",
      "       [-0.11641758, -0.0870166 ],\n",
      "       ...,\n",
      "       [-0.09506036, -0.17801705],\n",
      "       [-0.1260047 , -0.12223519],\n",
      "       [-0.19808611, -0.17148745]], dtype=float32), array([[-0.10266368, -0.13187459],\n",
      "       [-0.18534485, -0.1805908 ],\n",
      "       [-0.10549287, -0.16586731],\n",
      "       ...,\n",
      "       [-0.15361263, -0.03411802],\n",
      "       [-0.09987208, -0.06004155],\n",
      "       [-0.12198734, -0.18533196]], dtype=float32), array([[-0.09742684, -0.08576483],\n",
      "       [-0.15564083, -0.18689033],\n",
      "       [-0.17929849, -0.09448498],\n",
      "       ...,\n",
      "       [-0.12264243, -0.15618128],\n",
      "       [-0.20446393, -0.10878462],\n",
      "       [-0.1209803 , -0.13306293]], dtype=float32), array([[-0.16290098, -0.17602131],\n",
      "       [-0.1438188 , -0.20018962],\n",
      "       [-0.18129958, -0.14633857],\n",
      "       ...,\n",
      "       [-0.10836651, -0.17044453],\n",
      "       [-0.15258151, -0.13104148],\n",
      "       [-0.16913797, -0.05983722]], dtype=float32), array([[-0.17146866, -0.05227009],\n",
      "       [-0.13761659, -0.09317806],\n",
      "       [-0.17660636, -0.11907548],\n",
      "       ...,\n",
      "       [-0.12823407, -0.18179527],\n",
      "       [-0.15210634, -0.10561712],\n",
      "       [-0.1260787 , -0.18408765]], dtype=float32), array([[-0.13135235, -0.16182889],\n",
      "       [-0.15088971, -0.07191022],\n",
      "       [-0.14191216, -0.09334902],\n",
      "       ...,\n",
      "       [-0.11750675, -0.1562106 ],\n",
      "       [-0.12633105, -0.11428841],\n",
      "       [-0.10783336, -0.06780442]], dtype=float32), array([[-0.13692892, -0.11010401],\n",
      "       [-0.1958276 , -0.16761205],\n",
      "       [-0.18216103, -0.18315902],\n",
      "       ...,\n",
      "       [-0.1359959 , -0.06840748],\n",
      "       [-0.07941294, -0.18879533],\n",
      "       [-0.11928497, -0.06094375]], dtype=float32), array([[-0.05072796, -0.12233779],\n",
      "       [-0.17553475, -0.1841146 ],\n",
      "       [-0.05793084, -0.0900848 ],\n",
      "       ...,\n",
      "       [-0.03622333, -0.15107872],\n",
      "       [-0.10154506, -0.08263353],\n",
      "       [-0.0563561 , -0.16758403]], dtype=float32), array([[-0.18522725, -0.04865424],\n",
      "       [-0.11627626, -0.18971702],\n",
      "       [-0.15790057, -0.16766366],\n",
      "       ...,\n",
      "       [-0.10088769, -0.12882361],\n",
      "       [-0.1386712 , -0.14035414],\n",
      "       [-0.16028854, -0.13990149]], dtype=float32), array([[-0.12089952, -0.08279874],\n",
      "       [-0.15782215, -0.10117844],\n",
      "       [-0.12156531, -0.1684203 ],\n",
      "       ...,\n",
      "       [-0.13419023, -0.07237287],\n",
      "       [-0.08029632, -0.15169576],\n",
      "       [-0.0917947 , -0.06003851]], dtype=float32), array([[-0.15632458, -0.06884383],\n",
      "       [-0.03844602, -0.1359422 ],\n",
      "       [-0.05086486, -0.14400771],\n",
      "       ...,\n",
      "       [-0.12085515, -0.18196338],\n",
      "       [-0.05621032, -0.18521304],\n",
      "       [-0.19832209, -0.20032482]], dtype=float32), array([[-0.170512  , -0.03291008],\n",
      "       [-0.15708248, -0.08477172],\n",
      "       [-0.13095129, -0.06078037],\n",
      "       ...,\n",
      "       [-0.07497516, -0.16052061],\n",
      "       [-0.11589533, -0.08945495],\n",
      "       [-0.07453763, -0.13532227]], dtype=float32), array([[-0.09803375, -0.2660803 ],\n",
      "       [-0.08058846, -0.1094896 ],\n",
      "       [-0.20214836, -0.09253597],\n",
      "       ...,\n",
      "       [-0.22498675, -0.19337618],\n",
      "       [-0.16893293, -0.20575619],\n",
      "       [-0.14091703, -0.15814257]], dtype=float32), array([[-0.15532649, -0.13405326],\n",
      "       [-0.09757774, -0.07974519],\n",
      "       [-0.07533299, -0.16433683],\n",
      "       ...,\n",
      "       [-0.19629326, -0.20071825],\n",
      "       [-0.1051964 , -0.12848847],\n",
      "       [-0.05581992, -0.13460061]], dtype=float32), array([[-0.17251632, -0.12611711],\n",
      "       [-0.19854628, -0.18822977],\n",
      "       [-0.155515  , -0.04786018],\n",
      "       ...,\n",
      "       [-0.18342492, -0.13088033],\n",
      "       [-0.12441633, -0.06853426],\n",
      "       [-0.12012675, -0.15495524]], dtype=float32), array([[-0.09482178, -0.19603541],\n",
      "       [-0.09166071, -0.12524588],\n",
      "       [-0.16176048, -0.11835409],\n",
      "       ...,\n",
      "       [-0.17497358, -0.14964187],\n",
      "       [-0.16352944, -0.19283086],\n",
      "       [-0.20466232, -0.09874036]], dtype=float32), array([[-0.11674672, -0.16467401],\n",
      "       [-0.12184181, -0.21634373],\n",
      "       [-0.10401207, -0.19289222],\n",
      "       ...,\n",
      "       [-0.06005355, -0.170162  ],\n",
      "       [-0.12136   , -0.11788516],\n",
      "       [-0.16531214, -0.07654323]], dtype=float32), array([[-0.03471667, -0.1238471 ],\n",
      "       [-0.20285437, -0.11344762],\n",
      "       [-0.12152703, -0.20978467],\n",
      "       ...,\n",
      "       [-0.08848405, -0.08079518],\n",
      "       [-0.13416937, -0.14420897],\n",
      "       [-0.14488049, -0.14560279]], dtype=float32), array([[-0.07187076, -0.10481576],\n",
      "       [-0.12083981, -0.10336985],\n",
      "       [-0.07573432, -0.07211976],\n",
      "       ...,\n",
      "       [-0.21868399, -0.17687975],\n",
      "       [-0.10582448, -0.21721146],\n",
      "       [-0.0523061 , -0.08566587]], dtype=float32), array([[-0.17390132, -0.04969346],\n",
      "       [-0.08840306, -0.12908283],\n",
      "       [-0.1046936 , -0.17812043],\n",
      "       ...,\n",
      "       [-0.11218677, -0.11553197],\n",
      "       [-0.10307235, -0.06625732],\n",
      "       [-0.14854439, -0.09106605]], dtype=float32), array([[-0.07741257, -0.12443353],\n",
      "       [-0.07640709, -0.14609027],\n",
      "       [-0.19586176, -0.05720377],\n",
      "       ...,\n",
      "       [-0.1552993 , -0.13423112],\n",
      "       [-0.09713329, -0.22657117],\n",
      "       [-0.03848798, -0.12511387]], dtype=float32), array([[-0.14753672, -0.14692545],\n",
      "       [-0.1331923 , -0.17082971],\n",
      "       [-0.19598947, -0.10326643],\n",
      "       ...,\n",
      "       [-0.1877635 , -0.15331566],\n",
      "       [-0.14216043, -0.08870994],\n",
      "       [-0.13937415, -0.05111982]], dtype=float32), array([[-0.17788762, -0.13970704],\n",
      "       [-0.15477721, -0.12183269],\n",
      "       [-0.091374  , -0.23675592],\n",
      "       ...,\n",
      "       [-0.16816664, -0.07526328],\n",
      "       [-0.18145612, -0.07439436],\n",
      "       [-0.20624895, -0.1261633 ]], dtype=float32), array([[-0.14253083, -0.035023  ],\n",
      "       [-0.09723096, -0.1291865 ],\n",
      "       [-0.10291082, -0.12015654],\n",
      "       ...,\n",
      "       [-0.07626644, -0.1597221 ],\n",
      "       [-0.11058471, -0.06326266],\n",
      "       [-0.10523093, -0.04348093]], dtype=float32), array([[-0.18445052, -0.04918578],\n",
      "       [-0.11824549, -0.11197001],\n",
      "       [-0.13081901, -0.11146297],\n",
      "       ...,\n",
      "       [-0.20183574, -0.11232939],\n",
      "       [-0.21925555, -0.12015378],\n",
      "       [-0.14465089, -0.12856601]], dtype=float32), array([[-0.03707995, -0.09494269],\n",
      "       [-0.10069435, -0.05063383],\n",
      "       [-0.1532831 , -0.11287962],\n",
      "       ...,\n",
      "       [-0.22822239, -0.06437271],\n",
      "       [-0.2324131 , -0.05054288],\n",
      "       [-0.17825127, -0.1964832 ]], dtype=float32), array([[-0.04851685, -0.11780579],\n",
      "       [-0.12552069, -0.08845103],\n",
      "       [-0.08062717, -0.26110148],\n",
      "       ...,\n",
      "       [-0.15293539, -0.1661922 ],\n",
      "       [-0.17515746, -0.08073862],\n",
      "       [-0.14982307, -0.14253378]], dtype=float32), array([[-0.11627793, -0.16549522],\n",
      "       [-0.14685115, -0.06345461],\n",
      "       [-0.17553718, -0.09993685],\n",
      "       ...,\n",
      "       [-0.1162626 , -0.12943943],\n",
      "       [-0.08589599, -0.08978032],\n",
      "       [-0.09686859, -0.26259297]], dtype=float32), array([[-0.23671043, -0.01425646],\n",
      "       [-0.14379588, -0.14697981],\n",
      "       [-0.07070898, -0.10709669],\n",
      "       ...,\n",
      "       [-0.1729481 , -0.17423338],\n",
      "       [-0.22864808, -0.0543252 ],\n",
      "       [-0.12711875, -0.13227563]], dtype=float32), array([[-0.1400801 , -0.06518403],\n",
      "       [-0.13214236, -0.14348678],\n",
      "       [-0.07189264, -0.09744322],\n",
      "       ...,\n",
      "       [-0.18116102, -0.17737435],\n",
      "       [-0.19266331, -0.07656021],\n",
      "       [-0.1123896 , -0.10025506]], dtype=float32), array([[-0.08183695, -0.15041713],\n",
      "       [-0.0511292 , -0.16085248],\n",
      "       [-0.17183182, -0.13946831],\n",
      "       ...,\n",
      "       [-0.04191387, -0.10807993],\n",
      "       [-0.20227817, -0.0154564 ],\n",
      "       [-0.11714048, -0.14259954]], dtype=float32), array([[-0.22739358, -0.10588811],\n",
      "       [-0.13729289, -0.13122645],\n",
      "       [-0.21873434, -0.11628605],\n",
      "       ...,\n",
      "       [-0.12105282, -0.10992647],\n",
      "       [-0.14471972, -0.13696212],\n",
      "       [-0.11749619, -0.1992336 ]], dtype=float32), array([[-0.11695176, -0.10380145],\n",
      "       [-0.14687616, -0.09084306],\n",
      "       [-0.08488999, -0.18111753],\n",
      "       ...,\n",
      "       [-0.09397879, -0.10588406],\n",
      "       [-0.1327397 , -0.11962442],\n",
      "       [-0.15661274, -0.1550442 ]], dtype=float32), array([[-0.19842146, -0.1451667 ],\n",
      "       [-0.17527902, -0.06161905],\n",
      "       [-0.1656112 , -0.1617283 ],\n",
      "       ...,\n",
      "       [-0.0988306 , -0.1704418 ],\n",
      "       [-0.15426928, -0.08304853],\n",
      "       [-0.1412202 , -0.10869478]], dtype=float32), array([[-0.17866045, -0.11444526],\n",
      "       [-0.0698211 , -0.12157635],\n",
      "       [-0.11353534, -0.1507717 ],\n",
      "       ...,\n",
      "       [-0.01928116, -0.15453456],\n",
      "       [-0.21188271, -0.08135577],\n",
      "       [-0.13770933, -0.17517212]], dtype=float32), array([[-0.08893085, -0.12412277],\n",
      "       [-0.18408391, -0.12312021],\n",
      "       [-0.15637612, -0.09074796],\n",
      "       ...,\n",
      "       [-0.11098973, -0.19254313],\n",
      "       [-0.09567975, -0.20216335],\n",
      "       [-0.12537004, -0.08981661]], dtype=float32), array([[-0.22121637, -0.12376599],\n",
      "       [-0.1564613 , -0.09772362],\n",
      "       [-0.0769034 , -0.13810807],\n",
      "       ...,\n",
      "       [-0.18728498, -0.12165422],\n",
      "       [-0.05468102, -0.10288042],\n",
      "       [-0.06304737, -0.14948437]], dtype=float32), array([[-0.1250941 , -0.09370433],\n",
      "       [-0.0614358 , -0.11310454],\n",
      "       [-0.0906661 , -0.14437495],\n",
      "       ...,\n",
      "       [-0.09426285, -0.18712476],\n",
      "       [-0.21309397, -0.16484171],\n",
      "       [-0.02494379, -0.06801534]], dtype=float32), array([[-0.18029106, -0.10756697],\n",
      "       [-0.04710976, -0.15554407],\n",
      "       [-0.15218325, -0.10665306],\n",
      "       ...,\n",
      "       [-0.08825457, -0.17344631],\n",
      "       [-0.17508312, -0.17645176],\n",
      "       [-0.10349241, -0.04835346]], dtype=float32), array([[-0.19953181, -0.13650814],\n",
      "       [-0.15288746, -0.06266413],\n",
      "       [-0.22265945, -0.13362044],\n",
      "       ...,\n",
      "       [-0.15573268, -0.1374354 ],\n",
      "       [-0.13905436, -0.16254959],\n",
      "       [-0.11545847, -0.1618774 ]], dtype=float32), array([[-0.16371886, -0.13534108],\n",
      "       [-0.19286864, -0.06265116],\n",
      "       [-0.07924348, -0.1374324 ],\n",
      "       ...,\n",
      "       [-0.13439202, -0.10658887],\n",
      "       [-0.19816056, -0.14820915],\n",
      "       [-0.12789539, -0.11859451]], dtype=float32), array([[-0.11562144, -0.11180218],\n",
      "       [-0.07733906, -0.09304847],\n",
      "       [-0.13119066, -0.08048934],\n",
      "       ...,\n",
      "       [-0.20287476, -0.18151213],\n",
      "       [-0.15737021, -0.18643747],\n",
      "       [-0.16229117, -0.10513199]], dtype=float32), array([[-0.16205221, -0.17632404],\n",
      "       [-0.13319167, -0.1502729 ],\n",
      "       [-0.11878501, -0.0851553 ],\n",
      "       ...,\n",
      "       [-0.03670886, -0.06570602],\n",
      "       [-0.15775397, -0.10764007],\n",
      "       [-0.15705165, -0.12247375]], dtype=float32), array([[-0.1567635 , -0.14483312],\n",
      "       [-0.20787373, -0.11322505],\n",
      "       [-0.08849116, -0.23458996],\n",
      "       ...,\n",
      "       [-0.06803692, -0.12449788],\n",
      "       [-0.18581843, -0.14535306],\n",
      "       [-0.04702599, -0.1007445 ]], dtype=float32), array([[-0.11574752, -0.1972947 ],\n",
      "       [-0.12596527, -0.20063761],\n",
      "       [-0.06373217, -0.18183362],\n",
      "       ...,\n",
      "       [-0.1369388 , -0.06332646],\n",
      "       [-0.11439995, -0.12874913],\n",
      "       [-0.15030162, -0.06138959]], dtype=float32), array([[-0.11392289, -0.09397469],\n",
      "       [-0.17035833, -0.17199752],\n",
      "       [-0.09136169, -0.10284289],\n",
      "       ...,\n",
      "       [-0.06664003, -0.10507213],\n",
      "       [-0.04786253, -0.18263799],\n",
      "       [-0.12778185, -0.12190958]], dtype=float32), array([[-0.06038449, -0.15084948],\n",
      "       [-0.17833418, -0.11752126],\n",
      "       [-0.04243968, -0.13063614],\n",
      "       ...,\n",
      "       [-0.1745193 , -0.08566262],\n",
      "       [-0.05766495, -0.16835096],\n",
      "       [-0.18381594, -0.17222889]], dtype=float32), array([[-0.13932557, -0.05042678],\n",
      "       [-0.24433053, -0.07803299],\n",
      "       [-0.04023122, -0.09672986],\n",
      "       ...,\n",
      "       [-0.08994501, -0.11035281],\n",
      "       [-0.15774181, -0.12062894],\n",
      "       [-0.03972339, -0.12009801]], dtype=float32), array([[-0.08485422, -0.07188088],\n",
      "       [-0.07170273, -0.05994263],\n",
      "       [-0.13133898, -0.05193817],\n",
      "       ...,\n",
      "       [-0.20107052, -0.13901168],\n",
      "       [-0.12935588, -0.21889782],\n",
      "       [-0.17788824, -0.14495836]], dtype=float32), array([[-0.18781196, -0.18523835],\n",
      "       [-0.07345939, -0.11784054],\n",
      "       [-0.15454882, -0.10102918],\n",
      "       ...,\n",
      "       [-0.10899741, -0.15237899],\n",
      "       [-0.06390195, -0.138387  ],\n",
      "       [-0.06848392, -0.20052464]], dtype=float32), array([[-0.11313739, -0.21913636],\n",
      "       [-0.09796956, -0.18099354],\n",
      "       [-0.08390207, -0.13546589],\n",
      "       ...,\n",
      "       [-0.12943676, -0.04991848],\n",
      "       [-0.18673263, -0.06318815],\n",
      "       [-0.18013296, -0.20119041]], dtype=float32), array([[-0.17753574, -0.18251254],\n",
      "       [-0.12992318, -0.1675317 ],\n",
      "       [-0.14439933, -0.13700257],\n",
      "       ...,\n",
      "       [-0.11369226, -0.08020747],\n",
      "       [-0.11637998, -0.13798729],\n",
      "       [-0.14735138, -0.10180146]], dtype=float32), array([[-0.06285824, -0.16152439],\n",
      "       [-0.15488337, -0.14800757],\n",
      "       [-0.11141976, -0.11802388],\n",
      "       ...,\n",
      "       [-0.1391948 , -0.12007805],\n",
      "       [-0.13859048, -0.11508022],\n",
      "       [-0.09512717, -0.12802473]], dtype=float32), array([[-0.15441534, -0.14071737],\n",
      "       [-0.13565902, -0.16975734],\n",
      "       [-0.05660586, -0.05049402],\n",
      "       ...,\n",
      "       [-0.15017144, -0.2009713 ],\n",
      "       [-0.01546471, -0.13754036],\n",
      "       [-0.11366512, -0.09740047]], dtype=float32), array([[-0.12270783, -0.13608515],\n",
      "       [-0.17584269, -0.08193059],\n",
      "       [-0.13460766, -0.09597985],\n",
      "       ...,\n",
      "       [-0.1364169 , -0.03896739],\n",
      "       [-0.1484977 , -0.10290703],\n",
      "       [-0.09121718, -0.16029678]], dtype=float32), array([[-0.04852864, -0.06372949],\n",
      "       [-0.12048043, -0.09169984],\n",
      "       [-0.05071722, -0.18119529],\n",
      "       ...,\n",
      "       [-0.15586707, -0.16939512],\n",
      "       [-0.15343331, -0.15018374],\n",
      "       [-0.10704334, -0.1280427 ]], dtype=float32), array([[-0.09308285, -0.11046261],\n",
      "       [-0.19658543, -0.17360711],\n",
      "       [-0.21007095, -0.09330291],\n",
      "       ...,\n",
      "       [-0.18884498, -0.08649861],\n",
      "       [-0.06781594, -0.12865186],\n",
      "       [-0.05830261, -0.10162674]], dtype=float32), array([[-0.01940064, -0.17428982],\n",
      "       [-0.11732451, -0.13785386],\n",
      "       [-0.15428522, -0.10667065],\n",
      "       ...,\n",
      "       [-0.1907319 , -0.13096793],\n",
      "       [-0.09857265, -0.19327736],\n",
      "       [-0.07514641, -0.14638129]], dtype=float32), array([[-0.17823231, -0.15872753],\n",
      "       [-0.16638947, -0.19252191],\n",
      "       [-0.12672581, -0.12702376],\n",
      "       ...,\n",
      "       [-0.15882516, -0.13680512],\n",
      "       [-0.21775457, -0.07341072],\n",
      "       [-0.22343941, -0.13446052]], dtype=float32), array([[-0.08912203, -0.12451753],\n",
      "       [-0.05797474, -0.13680175],\n",
      "       [-0.09286024, -0.06830727],\n",
      "       ...,\n",
      "       [-0.07705353, -0.07066928],\n",
      "       [-0.21203732, -0.17897718],\n",
      "       [-0.18755162, -0.1640239 ]], dtype=float32), array([[-0.19561626, -0.19643517],\n",
      "       [-0.16363333, -0.1588491 ],\n",
      "       [-0.11335237, -0.15451553],\n",
      "       ...,\n",
      "       [-0.02119353, -0.07427615],\n",
      "       [-0.14797379, -0.11245339],\n",
      "       [-0.09201125, -0.13177682]], dtype=float32), array([[-0.15369691, -0.11591158],\n",
      "       [-0.06732772, -0.1312637 ],\n",
      "       [-0.10914438, -0.16258265],\n",
      "       ...,\n",
      "       [-0.06579625, -0.15263428],\n",
      "       [-0.12841146, -0.16377291],\n",
      "       [-0.05706999, -0.09427493]], dtype=float32), array([[-0.13285148, -0.18729308],\n",
      "       [-0.14360633, -0.12000059],\n",
      "       [-0.12695044, -0.15011755],\n",
      "       ...,\n",
      "       [-0.11929064, -0.11369541],\n",
      "       [-0.09766865, -0.19509414],\n",
      "       [-0.05869835, -0.16209282]], dtype=float32), array([[-0.18364172, -0.06174807],\n",
      "       [-0.13745123, -0.16255762],\n",
      "       [-0.07170018, -0.1688781 ],\n",
      "       ...,\n",
      "       [-0.19438292, -0.18250692],\n",
      "       [-0.11351047, -0.06531762],\n",
      "       [-0.15127191, -0.05482296]], dtype=float32), array([[-0.18119691, -0.08879463],\n",
      "       [-0.13844424, -0.14582083],\n",
      "       [-0.19711134, -0.04685061],\n",
      "       ...,\n",
      "       [-0.24805655, -0.04296969],\n",
      "       [-0.12972084, -0.06334154],\n",
      "       [-0.11514574, -0.10134302]], dtype=float32), array([[-0.14385596, -0.18234192],\n",
      "       [-0.15170744, -0.08857523],\n",
      "       [-0.1808902 , -0.1016986 ],\n",
      "       ...,\n",
      "       [-0.15603088, -0.12045798],\n",
      "       [-0.11929762, -0.17927793],\n",
      "       [-0.15388331, -0.11382324]], dtype=float32), array([[-0.13237406, -0.06188735],\n",
      "       [-0.15998149, -0.13484988],\n",
      "       [-0.06346584, -0.08990227],\n",
      "       ...,\n",
      "       [-0.12200771, -0.10719059],\n",
      "       [-0.12975913, -0.1407378 ],\n",
      "       [-0.20555472, -0.07393875]], dtype=float32), array([[-0.09592152, -0.2076897 ],\n",
      "       [-0.06701149, -0.05903566],\n",
      "       [-0.16015789, -0.20636445],\n",
      "       ...,\n",
      "       [-0.07194016, -0.06893075],\n",
      "       [-0.10313277, -0.17344944],\n",
      "       [-0.04546003, -0.08510149]], dtype=float32), array([[-0.14217049, -0.17138967],\n",
      "       [-0.1963726 , -0.05619824],\n",
      "       [-0.09850898, -0.15102695],\n",
      "       ...,\n",
      "       [-0.16977301, -0.16389012],\n",
      "       [-0.08986978, -0.08588076],\n",
      "       [-0.15013172, -0.20078018]], dtype=float32), array([[-0.14617956, -0.17485042],\n",
      "       [-0.0885157 , -0.14474118],\n",
      "       [-0.11591743, -0.11918763],\n",
      "       ...,\n",
      "       [-0.12862161, -0.17734465],\n",
      "       [-0.15004888, -0.11560066],\n",
      "       [-0.08009083, -0.05479952]], dtype=float32), array([[-0.12804459, -0.14215903],\n",
      "       [-0.09371801, -0.11955363],\n",
      "       [-0.17715004, -0.11422667],\n",
      "       ...,\n",
      "       [-0.1113048 , -0.18787079],\n",
      "       [-0.18028215, -0.14371917],\n",
      "       [-0.156421  , -0.17643686]], dtype=float32), array([[-0.14131276, -0.19746901],\n",
      "       [-0.21096094, -0.2178378 ],\n",
      "       [-0.09803271, -0.02956411],\n",
      "       ...,\n",
      "       [-0.17197827, -0.11012115],\n",
      "       [-0.13159241, -0.13914078],\n",
      "       [-0.1736979 , -0.18851632]], dtype=float32), array([[-0.05003079, -0.17308438],\n",
      "       [-0.23491979, -0.18793674],\n",
      "       [-0.1686611 , -0.08439256],\n",
      "       ...,\n",
      "       [-0.1838298 , -0.13182701],\n",
      "       [-0.10913795, -0.11763504],\n",
      "       [-0.20546623, -0.11556701]], dtype=float32), array([[-0.17190403, -0.21519701],\n",
      "       [-0.11007866, -0.13147986],\n",
      "       [-0.20339361, -0.06930718],\n",
      "       ...,\n",
      "       [-0.05878769, -0.09714979],\n",
      "       [-0.09646805, -0.10091089],\n",
      "       [-0.14872691, -0.10070084]], dtype=float32), array([[-0.14631553, -0.0992654 ],\n",
      "       [-0.18480748, -0.12420776],\n",
      "       [-0.06577934, -0.0726196 ],\n",
      "       ...,\n",
      "       [-0.10192493, -0.13185558],\n",
      "       [-0.01262841, -0.10753299],\n",
      "       [-0.15130019, -0.12176459]], dtype=float32), array([[-0.14104313, -0.17601505],\n",
      "       [-0.1236851 , -0.03333365],\n",
      "       [-0.18127047, -0.03824033],\n",
      "       ...,\n",
      "       [-0.08137209, -0.11671407],\n",
      "       [-0.16508238, -0.07558078],\n",
      "       [-0.13512154, -0.07109016]], dtype=float32), array([[-0.12291612, -0.17866537],\n",
      "       [-0.12522551, -0.08979254],\n",
      "       [-0.18866664, -0.22715166],\n",
      "       ...,\n",
      "       [-0.12630042, -0.13865106],\n",
      "       [-0.15159595, -0.1443872 ],\n",
      "       [-0.15179874, -0.09373608]], dtype=float32), array([[-0.13724092, -0.15833288],\n",
      "       [-0.08007976, -0.08353141],\n",
      "       [-0.20078036, -0.05875821],\n",
      "       ...,\n",
      "       [-0.17330083, -0.0537283 ],\n",
      "       [-0.05412776, -0.17494342],\n",
      "       [-0.08264194, -0.09455696]], dtype=float32), array([[-0.16340098, -0.07499877],\n",
      "       [-0.02433134, -0.0603982 ],\n",
      "       [-0.14814351, -0.02490081],\n",
      "       ...,\n",
      "       [-0.15060091, -0.15615623],\n",
      "       [-0.07029933, -0.10381696],\n",
      "       [-0.12908551, -0.17795746]], dtype=float32), array([[-0.1255244 , -0.09558886],\n",
      "       [-0.21917962, -0.0757898 ],\n",
      "       [-0.07800627, -0.15088022],\n",
      "       ...,\n",
      "       [-0.06995051, -0.18402599],\n",
      "       [-0.13632609, -0.12123965],\n",
      "       [-0.11572609, -0.13861771]], dtype=float32), array([[-0.14111602, -0.21498257],\n",
      "       [-0.14553896, -0.15007013],\n",
      "       [-0.16358976, -0.03545001],\n",
      "       ...,\n",
      "       [-0.1121766 , -0.08517601],\n",
      "       [-0.11082826, -0.12714243],\n",
      "       [-0.08538425, -0.16830541]], dtype=float32), array([[-0.08613628, -0.21526767],\n",
      "       [-0.08508602, -0.08256374],\n",
      "       [-0.05076203, -0.17870161],\n",
      "       ...,\n",
      "       [-0.13407792, -0.12019335],\n",
      "       [-0.09348026, -0.05473959],\n",
      "       [-0.17186901, -0.09969019]], dtype=float32), array([[-0.07787026, -0.06587486],\n",
      "       [-0.05893416, -0.2164498 ],\n",
      "       [-0.10329997, -0.08357586],\n",
      "       ...,\n",
      "       [-0.06752399, -0.12403549],\n",
      "       [-0.13667661, -0.14409132],\n",
      "       [-0.13838227, -0.14875719]], dtype=float32), array([[-0.05929606, -0.13726917],\n",
      "       [-0.1426246 , -0.12285343],\n",
      "       [-0.18094128, -0.14474158],\n",
      "       ...,\n",
      "       [-0.04984731, -0.14290494],\n",
      "       [-0.15206905, -0.10761671],\n",
      "       [-0.13221245, -0.09936881]], dtype=float32), array([[-0.15366296, -0.1343673 ],\n",
      "       [-0.09573697, -0.16049562],\n",
      "       [-0.13410586, -0.21299812],\n",
      "       ...,\n",
      "       [-0.20023137, -0.09606233],\n",
      "       [-0.17969735, -0.10641741],\n",
      "       [-0.13371956, -0.1141579 ]], dtype=float32), array([[-0.11746358, -0.10008228],\n",
      "       [-0.06420743, -0.13194385],\n",
      "       [-0.11966159, -0.11977117],\n",
      "       ...,\n",
      "       [-0.20312116, -0.10053974],\n",
      "       [-0.18715943, -0.14903682],\n",
      "       [-0.08386277, -0.12402858]], dtype=float32), array([[-0.06647216, -0.06313089],\n",
      "       [-0.10397911, -0.09082449],\n",
      "       [-0.04747823, -0.08921322],\n",
      "       ...,\n",
      "       [-0.12624826, -0.1427075 ],\n",
      "       [-0.16382684, -0.10406116],\n",
      "       [-0.1441442 , -0.09668346]], dtype=float32)]\n",
      "2\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#np.save('projs1', projs[:-1])\n",
    "#np.save('allHs1', tallHposes)\n",
    "print(projs)\n",
    "#print(projs[0][0][0])\n",
    "#print(projs[0][0][1])\n",
    "#print(np.max(np.concatenate(projs)))\n",
    "un, co = np.unique(np.round(np.concatenate(projs[:]), 2), return_counts=True)\n",
    "plt.plot(un*14.4, co)\n",
    "\n",
    "print(len(projs[0][0]))\n",
    "#print(proj2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "id": "4d972b68",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-0.11817648 -0.14704418]\n",
      " [-0.11357788 -0.07023326]\n",
      " [-0.15294488 -0.03435561]\n",
      " ...\n",
      " [-0.20508859 -0.09426089]\n",
      " [-0.17198949 -0.02154821]\n",
      " [-0.08409236 -0.13122699]]\n"
     ]
    }
   ],
   "source": [
    "print(projs[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "id": "02f39df1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "275\n",
      "[[-22.941494  -25.138401  -13.825902 ]\n",
      " [ 19.879177   -2.8533611  11.644999 ]\n",
      " [-33.98606   -21.880503  -20.584702 ]\n",
      " ...\n",
      " [ 23.57166    10.611601    1.8800907]\n",
      " [ 31.251827  -24.075996    1.0437946]\n",
      " [ 20.884037   -7.8736534  27.1727   ]]\n",
      "[[[-23.917385   -24.929502   -13.889199  ]\n",
      "  [-22.670193   -25.174896   -12.8641    ]]\n",
      "\n",
      " [[ 19.692524    -1.985651    11.184305  ]\n",
      "  [ 20.441257    -2.68647     12.455099  ]]\n",
      "\n",
      " [[-33.22287    -22.398201   -20.971401  ]\n",
      "  [-34.75776    -22.495403   -20.422503  ]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 23.095524    11.421999     1.5387516 ]\n",
      "  [ 23.646923     9.935755     1.1469231 ]]\n",
      "\n",
      " [[ 30.778004   -23.211702     0.87501144]\n",
      "  [ 32.225582   -23.966805     0.8440628 ]]\n",
      "\n",
      " [[ 21.130276    -7.726881   -27.603003  ]\n",
      "  [ 21.058987    -7.0381336   26.6518    ]]]\n"
     ]
    }
   ],
   "source": [
    "print(len(allHpos[1]))\n",
    "print(tallOposes[0])\n",
    "print(tallHposes[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "dd07b910",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[array([[ 3.6086829e-04, -7.7091601e-05],\n",
      "       [ 1.6514765e-04,  1.7764288e-04]], dtype=float32)]\n",
      "[array([[ 3.6086829e-04, -7.7091601e-05],\n",
      "       [ 1.6514765e-04,  1.7764288e-04]], dtype=float32)]\n"
     ]
    }
   ],
   "source": [
    "print(projs[0])\n",
    "print(projs[-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "b6350dbe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "[[[ -9.862946   22.636406    4.129076 ]\n",
      "  [ -8.707554   22.422295    5.262905 ]]\n",
      "\n",
      " [[ -9.261147   16.452301   15.367294 ]\n",
      "  [ -8.552753   17.918201   15.241299 ]]\n",
      "\n",
      " [[-21.549252   18.674503   -5.1904373]\n",
      "  [-23.168644   18.7044     -5.3980484]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 16.581757    2.6156693 -23.5966   ]\n",
      "  [ 16.88295     1.7442818 -22.248802 ]]\n",
      "\n",
      " [[ 26.670258  -27.102596    6.6259346]\n",
      "  [ 26.32486   -26.320896    8.017452 ]]\n",
      "\n",
      " [[ 23.445595  -10.117102  -21.713898 ]\n",
      "  [ 22.616417   -9.438902  -20.481398 ]]]\n",
      "2\n"
     ]
    }
   ],
   "source": [
    "#print(np.concatenate(np.concatenate(projs[0], axis=0)))\n",
    "print(len(projs))\n",
    "\n",
    "#f = np.concatenate(np.concatenate(projs[0], axis=0))\n",
    "print(tallHposes[0])\n",
    "print(len(tallHposes[0][0]))\n",
    "#print(len(projs[0][0]))\n",
    "#print(len(tallHposes[0][0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "ec6c7e18",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32.80566723996666\n",
      "3470040\n",
      "[[-23.917385 -24.929502 -13.889199]\n",
      " [-22.670193 -25.174896 -12.8641  ]\n",
      " [ 19.692524  -1.985651  11.184305]\n",
      " ...\n",
      " [ 27.99128   27.609303 -14.420998]\n",
      " [ -8.350338  -8.127218 -17.9956  ]\n",
      " [ -9.549137  -9.236071 -17.9953  ]]\n",
      "32.80566723996666\n",
      "33717\n",
      "32.80566723996666\n",
      "5216\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "i = 1\n",
    "print(fits[i][0])\n",
    "xx = [fits[i][0], fits[i][0]+2]\n",
    "proj = np.concatenate(projs)\n",
    "Hposes = np.concatenate(np.concatenate(tallHposes[:]), axis=0)\n",
    "print(len(Hposes))\n",
    "print(Hposes)\n",
    "for x in xx:\n",
    "    print(fits[i][0])\n",
    "    #Hposes = np.resize(tallHposes[i], []\n",
    "    wh = np.where(np.abs(Hposes[:,0] > x))\n",
    "    \n",
    "    print(len(wh[0]))\n",
    "    proposed = np.concatenate(proj)[wh]\n",
    "    un, co = np.unique(np.round(proposed*14.4, 2), return_counts=True)\n",
    "    plt.plot(un, co/np.sum(co))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "6dbfc30a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[array([[-0.11817648, -0.14704418],\n",
      "       [-0.11357788, -0.07023326],\n",
      "       [-0.15294488, -0.03435561],\n",
      "       ...,\n",
      "       [-0.20508859, -0.09426089],\n",
      "       [-0.17198949, -0.02154821],\n",
      "       [-0.08409236, -0.13122699]], dtype=float32), array([[-0.0756577 , -0.130437  ],\n",
      "       [-0.22865911, -0.03595458],\n",
      "       [-0.11957894, -0.03086508],\n",
      "       ...,\n",
      "       [-0.13613248, -0.09592831],\n",
      "       [-0.15084982, -0.08988983],\n",
      "       [-0.12498599, -0.06627281]], dtype=float32), array([[-0.17070773, -0.13071516],\n",
      "       [-0.17382658, -0.2057371 ],\n",
      "       [-0.13768998, -0.15406695],\n",
      "       ...,\n",
      "       [-0.20068549, -0.14893548],\n",
      "       [-0.06652506, -0.10967651],\n",
      "       [-0.05756936, -0.09580622]], dtype=float32), array([[-0.1531995 , -0.08068633],\n",
      "       [-0.13202882, -0.07826146],\n",
      "       [-0.06707557, -0.0445941 ],\n",
      "       ...,\n",
      "       [-0.05032402, -0.14608277],\n",
      "       [-0.15021935, -0.11321475],\n",
      "       [-0.0415628 , -0.11519535]], dtype=float32), array([[-0.0793936 , -0.09746761],\n",
      "       [-0.14726406, -0.1491873 ],\n",
      "       [-0.06869883, -0.15544681],\n",
      "       ...,\n",
      "       [-0.13188106, -0.09135406],\n",
      "       [-0.19299063, -0.1819382 ],\n",
      "       [-0.21679789, -0.05149623]], dtype=float32), array([[-0.09900048, -0.16659372],\n",
      "       [-0.09443817, -0.11088882],\n",
      "       [-0.07177087, -0.16577707],\n",
      "       ...,\n",
      "       [-0.10480226, -0.12443437],\n",
      "       [-0.13401997, -0.1921505 ],\n",
      "       [-0.15202197, -0.09798761]], dtype=float32), array([[-0.17233202, -0.11612502],\n",
      "       [-0.09003374, -0.11561358],\n",
      "       [-0.18527329, -0.07350565],\n",
      "       ...,\n",
      "       [-0.16419588, -0.09151474],\n",
      "       [-0.09224625, -0.07534555],\n",
      "       [-0.14731272, -0.09495284]], dtype=float32), array([[-0.05633143, -0.07460456],\n",
      "       [-0.08908077, -0.06861977],\n",
      "       [-0.17506078, -0.16313908],\n",
      "       ...,\n",
      "       [-0.1348536 , -0.15610597],\n",
      "       [-0.16697893, -0.22892089],\n",
      "       [-0.09185841, -0.11071345]], dtype=float32), array([[-0.14226621, -0.10283579],\n",
      "       [-0.03120159, -0.10694499],\n",
      "       [-0.14394733, -0.12396016],\n",
      "       ...,\n",
      "       [-0.16719016, -0.16156872],\n",
      "       [-0.17439373, -0.10397941],\n",
      "       [-0.23078908, -0.0346659 ]], dtype=float32), array([[-0.03190266, -0.15753941],\n",
      "       [-0.15641819, -0.14339042],\n",
      "       [-0.13992678, -0.09472401],\n",
      "       ...,\n",
      "       [-0.16602641, -0.04227225],\n",
      "       [-0.06639823, -0.17820415],\n",
      "       [-0.13551944, -0.06550633]], dtype=float32), array([[-0.12259448, -0.09398632],\n",
      "       [-0.1734538 , -0.06753714],\n",
      "       [-0.14488935, -0.11651635],\n",
      "       ...,\n",
      "       [-0.1458569 , -0.1151083 ],\n",
      "       [-0.19419959, -0.04415916],\n",
      "       [-0.10646837, -0.0466648 ]], dtype=float32), array([[-0.05150899, -0.10322918],\n",
      "       [-0.01573659, -0.18158977],\n",
      "       [-0.1989763 , -0.03103153],\n",
      "       ...,\n",
      "       [-0.09589686, -0.0580691 ],\n",
      "       [-0.1506679 , -0.15457873],\n",
      "       [-0.11705451, -0.10068844]], dtype=float32), array([[-0.09922481, -0.0924848 ],\n",
      "       [-0.19311821, -0.11035587],\n",
      "       [-0.06189689, -0.13656005],\n",
      "       ...,\n",
      "       [-0.08871934, -0.17190611],\n",
      "       [-0.0329387 , -0.19165657],\n",
      "       [-0.07998927, -0.11668621]], dtype=float32), array([[-0.10308666, -0.02680009],\n",
      "       [-0.15979986, -0.10194985],\n",
      "       [-0.12087107, -0.09595542],\n",
      "       ...,\n",
      "       [-0.10910617, -0.08818957],\n",
      "       [-0.06375409, -0.11969179],\n",
      "       [-0.16605484, -0.18406264]], dtype=float32), array([[-0.03989097, -0.11627042],\n",
      "       [-0.16473092, -0.10209084],\n",
      "       [-0.14900859, -0.13779017],\n",
      "       ...,\n",
      "       [-0.15427049, -0.0759828 ],\n",
      "       [-0.18264773, -0.21595323],\n",
      "       [-0.0544048 , -0.16342193]], dtype=float32), array([[-0.20790818, -0.07078722],\n",
      "       [-0.17730446, -0.12364783],\n",
      "       [-0.07881644, -0.11923087],\n",
      "       ...,\n",
      "       [-0.07303219, -0.20225659],\n",
      "       [-0.12894465, -0.13427848],\n",
      "       [-0.06950684, -0.04576094]], dtype=float32), array([[-0.24117191, -0.11833839],\n",
      "       [-0.11180228, -0.04214431],\n",
      "       [-0.1404255 , -0.14006385],\n",
      "       ...,\n",
      "       [-0.12223572, -0.12811626],\n",
      "       [-0.12348173, -0.10213903],\n",
      "       [-0.1369023 , -0.0906717 ]], dtype=float32), array([[-0.08788019, -0.16036029],\n",
      "       [-0.14843637, -0.17476127],\n",
      "       [-0.17300145, -0.08173778],\n",
      "       ...,\n",
      "       [-0.11712517, -0.12803543],\n",
      "       [-0.1284148 , -0.17728747],\n",
      "       [-0.15630953, -0.04667281]], dtype=float32), array([[-0.08062878, -0.09471991],\n",
      "       [-0.07661928, -0.05342029],\n",
      "       [-0.04907831, -0.1520519 ],\n",
      "       ...,\n",
      "       [-0.1448657 , -0.16049099],\n",
      "       [-0.20484172, -0.11466835],\n",
      "       [-0.16567765, -0.07193804]], dtype=float32), array([[-0.11916392, -0.15680648],\n",
      "       [-0.1658623 , -0.08614019],\n",
      "       [-0.02768235, -0.09731492],\n",
      "       ...,\n",
      "       [-0.15388784, -0.09977043],\n",
      "       [-0.16209418, -0.21259397],\n",
      "       [-0.14417261, -0.20488794]], dtype=float32), array([[-0.19567978, -0.10062794],\n",
      "       [-0.16328499, -0.08443266],\n",
      "       [-0.1947229 , -0.11034902],\n",
      "       ...,\n",
      "       [-0.07342907, -0.19307339],\n",
      "       [-0.15145339, -0.10696738],\n",
      "       [-0.15883163, -0.09859432]], dtype=float32), array([[-0.0489733 , -0.0298955 ],\n",
      "       [-0.12893167, -0.06385522],\n",
      "       [-0.10140426, -0.17115454],\n",
      "       ...,\n",
      "       [-0.09785289, -0.1409583 ],\n",
      "       [-0.13999228, -0.23089498],\n",
      "       [-0.19614801, -0.11037529]], dtype=float32), array([[-0.16153403, -0.14138515],\n",
      "       [-0.18813927, -0.21453358],\n",
      "       [-0.13858442, -0.10956374],\n",
      "       ...,\n",
      "       [-0.11501104, -0.23603982],\n",
      "       [-0.0665558 , -0.15918817],\n",
      "       [-0.06498209, -0.11747326]], dtype=float32), array([[-0.08276529, -0.14990184],\n",
      "       [-0.03400517, -0.14203563],\n",
      "       [-0.09126113, -0.10421597],\n",
      "       ...,\n",
      "       [-0.10540324, -0.0865237 ],\n",
      "       [-0.08862931, -0.15970895],\n",
      "       [-0.14941041, -0.12098919]], dtype=float32), array([[-0.05741496, -0.14976294],\n",
      "       [-0.14066097, -0.13850497],\n",
      "       [-0.06371249, -0.14669532],\n",
      "       ...,\n",
      "       [-0.10783483, -0.22232829],\n",
      "       [-0.12478893, -0.08686519],\n",
      "       [-0.09751441, -0.08792025]], dtype=float32), array([[-0.21670745, -0.12283266],\n",
      "       [-0.1271024 , -0.13292621],\n",
      "       [-0.08389463, -0.16676539],\n",
      "       ...,\n",
      "       [-0.07751255, -0.09318569],\n",
      "       [-0.12996595, -0.10646942],\n",
      "       [-0.17851321, -0.1937277 ]], dtype=float32), array([[-0.1855597 , -0.1491696 ],\n",
      "       [-0.04963815, -0.21570799],\n",
      "       [-0.13234264, -0.13883877],\n",
      "       ...,\n",
      "       [-0.112152  , -0.11293152],\n",
      "       [-0.17637546, -0.10185259],\n",
      "       [-0.15486744, -0.1264057 ]], dtype=float32), array([[-0.1748017 , -0.15805604],\n",
      "       [-0.0235228 , -0.12254386],\n",
      "       [-0.11149178, -0.08104715],\n",
      "       ...,\n",
      "       [-0.15165335, -0.06783067],\n",
      "       [-0.15556914, -0.09036832],\n",
      "       [-0.02956655, -0.00600729]], dtype=float32), array([[-0.11949529, -0.05150716],\n",
      "       [-0.07102235, -0.16311818],\n",
      "       [-0.23657382, -0.0508759 ],\n",
      "       ...,\n",
      "       [-0.04734233, -0.09537803],\n",
      "       [-0.1269604 , -0.16450751],\n",
      "       [-0.05588133, -0.16012678]], dtype=float32), array([[-0.15848051, -0.0878808 ],\n",
      "       [-0.06753282, -0.0849613 ],\n",
      "       [-0.19392236, -0.05634561],\n",
      "       ...,\n",
      "       [-0.14137475, -0.05578512],\n",
      "       [-0.21953979, -0.17250253],\n",
      "       [-0.15240751, -0.08780885]], dtype=float32), array([[-0.14370026, -0.06622484],\n",
      "       [-0.15522411, -0.15657784],\n",
      "       [-0.05863745, -0.18224125],\n",
      "       ...,\n",
      "       [-0.08332646, -0.13818027],\n",
      "       [-0.16559911, -0.1959652 ],\n",
      "       [-0.12753142, -0.2213518 ]], dtype=float32), array([[-0.18974301, -0.03324183],\n",
      "       [-0.06286059, -0.08963893],\n",
      "       [-0.06587213, -0.05588296],\n",
      "       ...,\n",
      "       [-0.04150677, -0.09091444],\n",
      "       [-0.05823202, -0.20064628],\n",
      "       [-0.15408143, -0.10797114]], dtype=float32), array([[-0.15332109, -0.09185649],\n",
      "       [-0.13396329, -0.12282796],\n",
      "       [-0.05910769, -0.16869287],\n",
      "       ...,\n",
      "       [-0.07826383, -0.19351026],\n",
      "       [-0.17481011, -0.15664206],\n",
      "       [-0.18805516, -0.1794512 ]], dtype=float32), array([[-0.14285871, -0.03272702],\n",
      "       [-0.13905387, -0.15638152],\n",
      "       [-0.1380501 , -0.21175297],\n",
      "       ...,\n",
      "       [-0.11067926, -0.11234608],\n",
      "       [-0.01934537, -0.18155143],\n",
      "       [-0.08022152, -0.10653377]], dtype=float32), array([[-0.11292043, -0.11950021],\n",
      "       [-0.20064947, -0.19175853],\n",
      "       [-0.09450316, -0.20527282],\n",
      "       ...,\n",
      "       [-0.14202176, -0.16253449],\n",
      "       [-0.15555353, -0.05058711],\n",
      "       [-0.15104043, -0.11932351]], dtype=float32), array([[-0.19057013, -0.17187999],\n",
      "       [-0.1804141 , -0.15346038],\n",
      "       [-0.14227729, -0.1811685 ],\n",
      "       ...,\n",
      "       [-0.1791135 , -0.1861896 ],\n",
      "       [-0.1168474 , -0.14275922],\n",
      "       [-0.10951332, -0.13284852]], dtype=float32), array([[-0.17817327, -0.17699859],\n",
      "       [-0.15245812, -0.11867376],\n",
      "       [-0.1417627 , -0.04118174],\n",
      "       ...,\n",
      "       [-0.15798786, -0.15927736],\n",
      "       [-0.17364195, -0.14802603],\n",
      "       [-0.18704036, -0.15137583]], dtype=float32), array([[-0.19080773, -0.17732202],\n",
      "       [-0.1465003 , -0.08225854],\n",
      "       [-0.08368611, -0.09968791],\n",
      "       ...,\n",
      "       [-0.19593467, -0.12618257],\n",
      "       [-0.1648172 , -0.10096839],\n",
      "       [-0.08597536, -0.08529589]], dtype=float32), array([[-0.0278255 , -0.13209772],\n",
      "       [-0.07487678, -0.12638661],\n",
      "       [-0.24233705, -0.09660614],\n",
      "       ...,\n",
      "       [-0.16555871, -0.13156943],\n",
      "       [-0.14447342, -0.18588632],\n",
      "       [-0.06132683, -0.12422071]], dtype=float32), array([[-0.12894532, -0.13549568],\n",
      "       [-0.16006537, -0.07564116],\n",
      "       [-0.10103531, -0.09658217],\n",
      "       ...,\n",
      "       [-0.13828403, -0.03772097],\n",
      "       [-0.1632111 , -0.14094983],\n",
      "       [-0.0358477 , -0.19841436]], dtype=float32), array([[-0.1175544 , -0.12093002],\n",
      "       [-0.1373172 , -0.13343185],\n",
      "       [-0.15297653, -0.0860183 ],\n",
      "       ...,\n",
      "       [-0.10432377, -0.11457533],\n",
      "       [-0.07777621, -0.19451702],\n",
      "       [-0.08167211, -0.04936979]], dtype=float32), array([[-0.11817341, -0.03636614],\n",
      "       [-0.02943886, -0.15564537],\n",
      "       [-0.10072929, -0.05651584],\n",
      "       ...,\n",
      "       [-0.05208138, -0.19116855],\n",
      "       [-0.13770975, -0.10180248],\n",
      "       [-0.12697117, -0.10352042]], dtype=float32), array([[-0.04843969, -0.20931114],\n",
      "       [-0.18545437, -0.1312336 ],\n",
      "       [-0.16431458, -0.23595235],\n",
      "       ...,\n",
      "       [-0.13847516, -0.15858176],\n",
      "       [-0.16573256, -0.16998392],\n",
      "       [-0.16963068, -0.08576098]], dtype=float32), array([[-0.09309182, -0.09390479],\n",
      "       [-0.05693859, -0.11729397],\n",
      "       [-0.11499198, -0.07967059],\n",
      "       ...,\n",
      "       [-0.13393477, -0.10266078],\n",
      "       [-0.04924463, -0.0680997 ],\n",
      "       [-0.03794032, -0.05001358]], dtype=float32), array([[-0.13115099, -0.20005086],\n",
      "       [-0.03531853, -0.10783227],\n",
      "       [-0.17447236, -0.12658255],\n",
      "       ...,\n",
      "       [-0.19167249, -0.11026362],\n",
      "       [-0.00781912, -0.09460847],\n",
      "       [-0.11935879, -0.06561916]], dtype=float32), array([[-0.10997014, -0.07711617],\n",
      "       [-0.10373071, -0.10572213],\n",
      "       [-0.18417922, -0.17276993],\n",
      "       ...,\n",
      "       [-0.13214628, -0.10396378],\n",
      "       [-0.08753316, -0.18139115],\n",
      "       [-0.14258146, -0.19139403]], dtype=float32), array([[-0.08197098, -0.15146013],\n",
      "       [-0.07389113, -0.08568378],\n",
      "       [-0.0516216 , -0.20924054],\n",
      "       ...,\n",
      "       [-0.17921665, -0.11317268],\n",
      "       [-0.15696688, -0.20239231],\n",
      "       [-0.10485709, -0.14425805]], dtype=float32), array([[-0.06041029, -0.15437713],\n",
      "       [-0.05452196, -0.11329359],\n",
      "       [-0.13105817, -0.20212714],\n",
      "       ...,\n",
      "       [-0.14564148, -0.12068386],\n",
      "       [-0.15272817, -0.1519359 ],\n",
      "       [-0.06133653, -0.09312334]], dtype=float32), array([[-0.1137688 , -0.13160118],\n",
      "       [-0.2097278 , -0.09246977],\n",
      "       [-0.18490177, -0.10644498],\n",
      "       ...,\n",
      "       [-0.13420843, -0.06455914],\n",
      "       [-0.15171547, -0.07608189],\n",
      "       [-0.13249396, -0.17003684]], dtype=float32), array([[-0.06756291, -0.10709944],\n",
      "       [-0.16032518, -0.11916131],\n",
      "       [-0.16225336, -0.15146211],\n",
      "       ...,\n",
      "       [-0.09728722, -0.04980183],\n",
      "       [-0.12189385, -0.14299248],\n",
      "       [-0.13708362, -0.16507097]], dtype=float32), array([[-0.13823357, -0.07778165],\n",
      "       [-0.09603202, -0.0885376 ],\n",
      "       [-0.10759582, -0.19701234],\n",
      "       ...,\n",
      "       [-0.08501067, -0.10379375],\n",
      "       [-0.23156342, -0.18480101],\n",
      "       [-0.16379368, -0.1793944 ]], dtype=float32), array([[-0.17941865, -0.19084671],\n",
      "       [-0.10304973, -0.20845796],\n",
      "       [-0.1756866 , -0.19619529],\n",
      "       ...,\n",
      "       [-0.10709026, -0.0612867 ],\n",
      "       [-0.17359036, -0.06376286],\n",
      "       [-0.11719277, -0.12880935]], dtype=float32), array([[-0.09813723, -0.06102981],\n",
      "       [-0.13076952, -0.0347729 ],\n",
      "       [-0.14627528, -0.18709414],\n",
      "       ...,\n",
      "       [-0.1591926 , -0.16809674],\n",
      "       [-0.1743315 , -0.09024116],\n",
      "       [-0.12703654, -0.07979438]], dtype=float32), array([[-0.06879442, -0.10010874],\n",
      "       [-0.10648812, -0.11463622],\n",
      "       [-0.04565511, -0.17220454],\n",
      "       ...,\n",
      "       [-0.20215277, -0.11538669],\n",
      "       [-0.05083453, -0.03596466],\n",
      "       [-0.14254378, -0.14217581]], dtype=float32), array([[-0.19998895, -0.08379712],\n",
      "       [-0.11000605, -0.13626358],\n",
      "       [-0.13494757, -0.12587543],\n",
      "       ...,\n",
      "       [-0.14421132, -0.14230566],\n",
      "       [-0.14413011, -0.08822533],\n",
      "       [-0.15999094, -0.14042947]], dtype=float32), array([[-0.10865754, -0.11019506],\n",
      "       [-0.172292  , -0.12728442],\n",
      "       [-0.0693235 , -0.15131089],\n",
      "       ...,\n",
      "       [-0.22235009, -0.17773278],\n",
      "       [-0.17605707, -0.14759716],\n",
      "       [-0.06608705, -0.12471202]], dtype=float32), array([[-0.11589941, -0.15543562],\n",
      "       [-0.15172385, -0.05582351],\n",
      "       [-0.04637762, -0.20459141],\n",
      "       ...,\n",
      "       [-0.22753103, -0.16899388],\n",
      "       [-0.08190566, -0.23326656],\n",
      "       [-0.204809  , -0.10389359]], dtype=float32), array([[-0.17598987, -0.03765145],\n",
      "       [-0.20834346, -0.11621943],\n",
      "       [-0.13641506, -0.0845672 ],\n",
      "       ...,\n",
      "       [-0.15401448, -0.1455168 ],\n",
      "       [-0.12347526, -0.10955377],\n",
      "       [-0.16703519, -0.16552703]], dtype=float32), array([[-0.1338635 , -0.13342507],\n",
      "       [-0.16196494, -0.09801425],\n",
      "       [-0.07117899, -0.0923063 ],\n",
      "       ...,\n",
      "       [-0.15296139, -0.0890063 ],\n",
      "       [-0.16407438, -0.1602951 ],\n",
      "       [-0.09301545, -0.07156   ]], dtype=float32), array([[-0.03748368, -0.10211256],\n",
      "       [-0.12544061, -0.1313232 ],\n",
      "       [-0.10381979, -0.1735504 ],\n",
      "       ...,\n",
      "       [-0.09023498, -0.06349203],\n",
      "       [-0.02649651, -0.09582682],\n",
      "       [-0.22000742, -0.14082378]], dtype=float32), array([[-0.09136149, -0.07767242],\n",
      "       [-0.14992703, -0.0144429 ],\n",
      "       [-0.11754459, -0.05533516],\n",
      "       ...,\n",
      "       [-0.06330275, -0.10555261],\n",
      "       [-0.13222897, -0.11496547],\n",
      "       [-0.17782421, -0.17455347]], dtype=float32), array([[-0.09598844, -0.0868334 ],\n",
      "       [-0.07324035, -0.1842526 ],\n",
      "       [-0.03174331, -0.1266043 ],\n",
      "       ...,\n",
      "       [-0.11821376, -0.12012026],\n",
      "       [-0.1467978 , -0.12715222],\n",
      "       [-0.13425869, -0.09584543]], dtype=float32), array([[-0.16835725, -0.10122532],\n",
      "       [-0.17047757, -0.19854541],\n",
      "       [-0.16925934, -0.0346225 ],\n",
      "       ...,\n",
      "       [-0.11894952, -0.12610427],\n",
      "       [-0.15161207, -0.06749716],\n",
      "       [-0.09450744, -0.1016803 ]], dtype=float32), array([[-0.17997   , -0.1106399 ],\n",
      "       [-0.13443168, -0.16809241],\n",
      "       [-0.21549323, -0.20883591],\n",
      "       ...,\n",
      "       [-0.15885438, -0.15156421],\n",
      "       [-0.13917407, -0.13284415],\n",
      "       [-0.19297802, -0.09605265]], dtype=float32), array([[-0.17626022, -0.20369124],\n",
      "       [-0.17800161, -0.14083299],\n",
      "       [-0.18005478, -0.13636214],\n",
      "       ...,\n",
      "       [-0.15497637, -0.16269311],\n",
      "       [-0.0469254 , -0.0614472 ],\n",
      "       [-0.17037223, -0.19187452]], dtype=float32), array([[-0.11909258, -0.21099138],\n",
      "       [-0.14147717, -0.18272738],\n",
      "       [-0.17937174, -0.03374496],\n",
      "       ...,\n",
      "       [-0.07682525, -0.11307138],\n",
      "       [-0.12271994, -0.12364547],\n",
      "       [-0.10984305, -0.15268044]], dtype=float32), array([[-0.05521431, -0.19480595],\n",
      "       [-0.13371725, -0.06352061],\n",
      "       [-0.16000566, -0.14212903],\n",
      "       ...,\n",
      "       [-0.09245495, -0.09558292],\n",
      "       [-0.09793006, -0.17958279],\n",
      "       [-0.15508638, -0.06658361]], dtype=float32), array([[-0.12833676, -0.1543225 ],\n",
      "       [-0.10195038, -0.11190069],\n",
      "       [-0.14243151, -0.15313834],\n",
      "       ...,\n",
      "       [-0.14932294, -0.1652253 ],\n",
      "       [-0.1404844 , -0.130019  ],\n",
      "       [-0.15293811, -0.1213826 ]], dtype=float32), array([[-0.16972715, -0.1659075 ],\n",
      "       [-0.25579152, -0.11613295],\n",
      "       [-0.10105971, -0.17187543],\n",
      "       ...,\n",
      "       [-0.18332717, -0.14974225],\n",
      "       [-0.10370463, -0.09965788],\n",
      "       [-0.168942  , -0.14843479]], dtype=float32), array([[-0.17091674, -0.07458842],\n",
      "       [-0.10730687, -0.11388034],\n",
      "       [-0.18143034, -0.07688228],\n",
      "       ...,\n",
      "       [-0.22739528, -0.1570503 ],\n",
      "       [-0.04795042, -0.05662959],\n",
      "       [-0.14249079, -0.07157628]], dtype=float32), array([[-0.13317865, -0.11273965],\n",
      "       [-0.13233373, -0.16666418],\n",
      "       [-0.1255956 , -0.0952547 ],\n",
      "       ...,\n",
      "       [-0.13775112, -0.05905051],\n",
      "       [-0.0867906 , -0.17748885],\n",
      "       [-0.19474193, -0.11153268]], dtype=float32), array([[-0.14804168, -0.18587911],\n",
      "       [-0.12961186, -0.13823353],\n",
      "       [-0.16258606, -0.08971591],\n",
      "       ...,\n",
      "       [-0.15088639, -0.12960623],\n",
      "       [-0.16469307, -0.08447277],\n",
      "       [-0.06133368, -0.21470354]], dtype=float32), array([[-0.13201505, -0.16412055],\n",
      "       [-0.08216473, -0.05696087],\n",
      "       [-0.19933417, -0.08891168],\n",
      "       ...,\n",
      "       [-0.12417572, -0.0770592 ],\n",
      "       [-0.1528231 , -0.04639693],\n",
      "       [-0.09342077, -0.15323545]], dtype=float32), array([[-0.11549712, -0.13970262],\n",
      "       [-0.13198811, -0.06884127],\n",
      "       [-0.1290568 , -0.15386963],\n",
      "       ...,\n",
      "       [-0.08466011, -0.06160564],\n",
      "       [-0.16905947, -0.16452405],\n",
      "       [-0.09654797, -0.08396529]], dtype=float32), array([[-0.14652224, -0.0825812 ],\n",
      "       [-0.16967288, -0.16270751],\n",
      "       [-0.11517986, -0.05506199],\n",
      "       ...,\n",
      "       [-0.13638997, -0.13040698],\n",
      "       [-0.14796075, -0.14156021],\n",
      "       [-0.15079106, -0.06834966]], dtype=float32), array([[-0.19752601, -0.16524495],\n",
      "       [-0.10420656, -0.0465981 ],\n",
      "       [-0.06381158, -0.08175069],\n",
      "       ...,\n",
      "       [-0.09104593, -0.10691755],\n",
      "       [-0.07917699, -0.15511681],\n",
      "       [-0.14289045, -0.16617815]], dtype=float32), array([[-0.15472177, -0.17492023],\n",
      "       [-0.15785517, -0.1422662 ],\n",
      "       [-0.19810489, -0.11019158],\n",
      "       ...,\n",
      "       [-0.13502562, -0.13613729],\n",
      "       [-0.14448237, -0.1588096 ],\n",
      "       [-0.08374635, -0.06813328]], dtype=float32), array([[-0.10702991, -0.10931224],\n",
      "       [-0.22801396, -0.15843508],\n",
      "       [-0.08409625, -0.1413038 ],\n",
      "       ...,\n",
      "       [-0.1426898 , -0.07513205],\n",
      "       [-0.08458428, -0.1370708 ],\n",
      "       [-0.18341711, -0.12848665]], dtype=float32), array([[-0.11403464, -0.21930757],\n",
      "       [-0.0916763 , -0.19737622],\n",
      "       [-0.1380417 , -0.1213036 ],\n",
      "       ...,\n",
      "       [-0.1502493 , -0.12612057],\n",
      "       [-0.19811141, -0.17498288],\n",
      "       [-0.12486202, -0.08356053]], dtype=float32), array([[-0.19074854, -0.13041234],\n",
      "       [-0.04642609, -0.14731315],\n",
      "       [-0.20660488, -0.12478527],\n",
      "       ...,\n",
      "       [-0.15264234, -0.09913626],\n",
      "       [-0.09013166, -0.23179443],\n",
      "       [-0.07469551, -0.11540645]], dtype=float32), array([[-0.07674406, -0.18260191],\n",
      "       [-0.03978457, -0.17350647],\n",
      "       [-0.10475735, -0.08185153],\n",
      "       ...,\n",
      "       [-0.13898957, -0.16496013],\n",
      "       [-0.13644503, -0.08804738],\n",
      "       [-0.17945911, -0.12010814]], dtype=float32), array([[-0.044523  , -0.1124459 ],\n",
      "       [-0.1066047 , -0.18235253],\n",
      "       [-0.10289676, -0.0714946 ],\n",
      "       ...,\n",
      "       [-0.15017939, -0.17211856],\n",
      "       [-0.13935034, -0.03745014],\n",
      "       [-0.07516879, -0.11801428]], dtype=float32), array([[-0.1747689 , -0.14627552],\n",
      "       [-0.08568648, -0.08284468],\n",
      "       [-0.17904367, -0.10095244],\n",
      "       ...,\n",
      "       [-0.12388507, -0.1520569 ],\n",
      "       [-0.14037293, -0.02148483],\n",
      "       [-0.09627619, -0.05698512]], dtype=float32), array([[-0.12482192, -0.21806785],\n",
      "       [-0.07326397, -0.03501823],\n",
      "       [-0.08070597, -0.15619609],\n",
      "       ...,\n",
      "       [-0.12070602, -0.03155429],\n",
      "       [-0.09581748, -0.15321462],\n",
      "       [-0.11330551, -0.16430235]], dtype=float32), array([[-0.17893372, -0.15751225],\n",
      "       [-0.11175349, -0.14879191],\n",
      "       [-0.14686328, -0.14685625],\n",
      "       ...,\n",
      "       [-0.11580049, -0.13425206],\n",
      "       [-0.12526383, -0.02627475],\n",
      "       [-0.11436513, -0.14415663]], dtype=float32), array([[-0.19008893, -0.12676723],\n",
      "       [-0.14532545, -0.12918784],\n",
      "       [-0.03584491, -0.17484133],\n",
      "       ...,\n",
      "       [-0.11993406, -0.18816839],\n",
      "       [-0.11417604, -0.16168107],\n",
      "       [-0.15014456, -0.1218549 ]], dtype=float32), array([[-0.0497546 , -0.09618214],\n",
      "       [-0.21904156, -0.09980526],\n",
      "       [-0.12359386, -0.10999627],\n",
      "       ...,\n",
      "       [-0.1295447 , -0.11710048],\n",
      "       [ 0.00185582, -0.23586799],\n",
      "       [-0.1857467 , -0.09033152]], dtype=float32), array([[-0.09025657, -0.08667264],\n",
      "       [-0.02474688, -0.13429625],\n",
      "       [-0.05036566, -0.09937616],\n",
      "       ...,\n",
      "       [-0.1742666 , -0.1617832 ],\n",
      "       [-0.07177159, -0.10489106],\n",
      "       [-0.15663877, -0.18601787]], dtype=float32), array([[-0.15007007, -0.04579732],\n",
      "       [-0.20878425, -0.03086765],\n",
      "       [-0.03639678, -0.07879689],\n",
      "       ...,\n",
      "       [-0.10681011, -0.12020849],\n",
      "       [-0.02369911, -0.06059569],\n",
      "       [-0.12100284, -0.13948108]], dtype=float32), array([[-0.13996193, -0.14736529],\n",
      "       [-0.09496886, -0.15076031],\n",
      "       [-0.10538046, -0.118288  ],\n",
      "       ...,\n",
      "       [-0.08991499, -0.14143595],\n",
      "       [-0.06642923, -0.10441828],\n",
      "       [-0.06828518, -0.1675502 ]], dtype=float32), array([[-0.13806616, -0.04325547],\n",
      "       [-0.168767  , -0.09660642],\n",
      "       [-0.09032822, -0.04727171],\n",
      "       ...,\n",
      "       [-0.20657347, -0.13303468],\n",
      "       [-0.05220417, -0.16468634],\n",
      "       [-0.1464434 , -0.23034202]], dtype=float32), array([[-0.13132477, -0.18184471],\n",
      "       [-0.13208291, -0.10293137],\n",
      "       [-0.10557705, -0.12855886],\n",
      "       ...,\n",
      "       [-0.14241552, -0.16238916],\n",
      "       [-0.14402753, -0.17835292],\n",
      "       [-0.2218487 , -0.15123276]], dtype=float32), array([[-0.1479667 , -0.06154051],\n",
      "       [-0.19395518, -0.04326924],\n",
      "       [-0.19575459, -0.07683694],\n",
      "       ...,\n",
      "       [-0.06850191, -0.11576045],\n",
      "       [-0.12859705, -0.17560929],\n",
      "       [-0.15095368, -0.18910602]], dtype=float32), array([[-0.18200476, -0.13334523],\n",
      "       [-0.16298088, -0.11223492],\n",
      "       [-0.09431463, -0.10133602],\n",
      "       ...,\n",
      "       [-0.21568976, -0.06744772],\n",
      "       [-0.15868333, -0.09952226],\n",
      "       [-0.15677962, -0.18427384]], dtype=float32), array([[-0.12610558, -0.13888876],\n",
      "       [-0.12962644, -0.12959118],\n",
      "       [-0.11736555, -0.07516077],\n",
      "       ...,\n",
      "       [-0.1848169 , -0.22088109],\n",
      "       [-0.11866364, -0.20787084],\n",
      "       [-0.20119558, -0.15753397]], dtype=float32), array([[-0.08084893, -0.18691641],\n",
      "       [-0.09070449, -0.02389285],\n",
      "       [-0.22234158, -0.09150629],\n",
      "       ...,\n",
      "       [-0.13480262, -0.09648053],\n",
      "       [-0.08612172, -0.11796828],\n",
      "       [-0.07796457, -0.14762333]], dtype=float32), array([[-0.11512873, -0.18620878],\n",
      "       [-0.06162035, -0.09949338],\n",
      "       [-0.19461633, -0.104544  ],\n",
      "       ...,\n",
      "       [-0.05653153, -0.0917932 ],\n",
      "       [-0.10976446, -0.18381421],\n",
      "       [-0.12814659, -0.07809196]], dtype=float32), array([[-0.1585981 , -0.07627721],\n",
      "       [-0.14369339, -0.08074534],\n",
      "       [-0.14385194, -0.10968497],\n",
      "       ...,\n",
      "       [-0.09576465, -0.16305242],\n",
      "       [-0.09988698, -0.14203951],\n",
      "       [-0.13105148, -0.04316316]], dtype=float32), array([[-0.15016977, -0.11829355],\n",
      "       [-0.0885694 , -0.11609217],\n",
      "       [-0.0420789 , -0.09618907],\n",
      "       ...,\n",
      "       [-0.14823028, -0.15717205],\n",
      "       [-0.08234622, -0.21639623],\n",
      "       [-0.15531348, -0.12760018]], dtype=float32), array([[-0.19930306, -0.2043485 ],\n",
      "       [-0.19965225, -0.10420739],\n",
      "       [-0.20523134, -0.1581698 ],\n",
      "       ...,\n",
      "       [-0.12749422, -0.1738849 ],\n",
      "       [-0.11781287, -0.1606482 ],\n",
      "       [-0.09316926, -0.06891327]], dtype=float32), array([[-0.14650694, -0.02921626],\n",
      "       [-0.2106895 , -0.13918124],\n",
      "       [-0.16816452, -0.06318112],\n",
      "       ...,\n",
      "       [-0.08882765, -0.16626827],\n",
      "       [-0.03298642, -0.15522163],\n",
      "       [-0.0370278 , -0.09260011]], dtype=float32), array([[-0.17828873, -0.15771714],\n",
      "       [-0.09214156, -0.17764002],\n",
      "       [-0.15465792, -0.10491984],\n",
      "       ...,\n",
      "       [-0.03050534, -0.156508  ],\n",
      "       [-0.07859881, -0.05785437],\n",
      "       [-0.12179411, -0.08990696]], dtype=float32), array([[-0.21284102, -0.13838966],\n",
      "       [-0.16465858, -0.1042154 ],\n",
      "       [-0.07250795, -0.07433683],\n",
      "       ...,\n",
      "       [-0.13448931, -0.11398347],\n",
      "       [-0.11284971, -0.0564485 ],\n",
      "       [-0.10176106, -0.13316415]], dtype=float32), array([[-0.10621536, -0.12025632],\n",
      "       [-0.09873202, -0.14775608],\n",
      "       [-0.15049258, -0.06966597],\n",
      "       ...,\n",
      "       [-0.09855526, -0.07896975],\n",
      "       [-0.09050024, -0.07460484],\n",
      "       [-0.07736115, -0.1414557 ]], dtype=float32), array([[-0.09101056, -0.14268771],\n",
      "       [-0.15216465, -0.11760989],\n",
      "       [-0.08569868, -0.11838463],\n",
      "       ...,\n",
      "       [-0.06442276, -0.07850402],\n",
      "       [-0.04023621, -0.12668103],\n",
      "       [-0.08747856, -0.19629237]], dtype=float32), array([[-0.06337006, -0.15470874],\n",
      "       [-0.17253998, -0.20519081],\n",
      "       [-0.09275047, -0.16939694],\n",
      "       ...,\n",
      "       [-0.09839325, -0.10676086],\n",
      "       [-0.18357027, -0.14988764],\n",
      "       [-0.2015397 , -0.08864633]], dtype=float32), array([[-0.15655759, -0.07303577],\n",
      "       [-0.1606832 , -0.06939139],\n",
      "       [-0.18937601, -0.12066202],\n",
      "       ...,\n",
      "       [-0.07817851, -0.13299483],\n",
      "       [-0.08840489, -0.20791903],\n",
      "       [-0.02888539, -0.08912779]], dtype=float32), array([[-0.07034096, -0.08616195],\n",
      "       [-0.15078036, -0.19846126],\n",
      "       [-0.11235116, -0.06967682],\n",
      "       ...,\n",
      "       [-0.18392438, -0.07450312],\n",
      "       [-0.10863635, -0.09724211],\n",
      "       [-0.12478662, -0.15780383]], dtype=float32), array([[-0.10463329, -0.13850029],\n",
      "       [-0.16861163, -0.12345431],\n",
      "       [-0.00944297, -0.12806535],\n",
      "       ...,\n",
      "       [-0.27063164, -0.12447986],\n",
      "       [-0.16782565, -0.1406549 ],\n",
      "       [-0.16370083, -0.18103981]], dtype=float32), array([[-0.10510568, -0.10002428],\n",
      "       [-0.05038411, -0.17914626],\n",
      "       [-0.05755374, -0.02872365],\n",
      "       ...,\n",
      "       [-0.15654835, -0.00257575],\n",
      "       [-0.15356137, -0.14452651],\n",
      "       [-0.06215907, -0.1571174 ]], dtype=float32), array([[-0.16358313, -0.11349313],\n",
      "       [-0.13926664, -0.06882828],\n",
      "       [-0.08578321, -0.1393313 ],\n",
      "       ...,\n",
      "       [-0.14273669, -0.08313767],\n",
      "       [-0.11012083, -0.0817814 ],\n",
      "       [-0.07549386, -0.13216351]], dtype=float32), array([[-0.08322988, -0.01860059],\n",
      "       [-0.20226371, -0.15316623],\n",
      "       [-0.15915218, -0.11957458],\n",
      "       ...,\n",
      "       [-0.17511629, -0.12147555],\n",
      "       [-0.08211923, -0.1811997 ],\n",
      "       [-0.05785236, -0.07624552]], dtype=float32), array([[-0.09294113, -0.13502836],\n",
      "       [-0.18674164, -0.20747861],\n",
      "       [-0.23174576, -0.17239565],\n",
      "       ...,\n",
      "       [-0.14216489, -0.19810492],\n",
      "       [-0.21194258, -0.16580531],\n",
      "       [-0.11843228, -0.14883964]], dtype=float32), array([[-0.06913724, -0.03805468],\n",
      "       [-0.086382  , -0.06178598],\n",
      "       [-0.14855367, -0.0473504 ],\n",
      "       ...,\n",
      "       [-0.1439897 , -0.15287657],\n",
      "       [-0.14353532, -0.15922019],\n",
      "       [-0.1375872 , -0.12702239]], dtype=float32), array([[-0.12415466, -0.1611248 ],\n",
      "       [-0.09554369, -0.0286546 ],\n",
      "       [-0.15606888, -0.10671141],\n",
      "       ...,\n",
      "       [-0.06701748, -0.09131048],\n",
      "       [-0.05343989, -0.07886886],\n",
      "       [-0.03723019, -0.08673538]], dtype=float32), array([[-0.10792259, -0.09721148],\n",
      "       [-0.04707403, -0.13317841],\n",
      "       [-0.12620722, -0.09895674],\n",
      "       ...,\n",
      "       [-0.04090516, -0.13187326],\n",
      "       [-0.15567815, -0.01836697],\n",
      "       [-0.05646631, -0.12400496]], dtype=float32), array([[-0.06141383, -0.07052136],\n",
      "       [-0.04269809, -0.14121264],\n",
      "       [-0.13701938, -0.09593248],\n",
      "       ...,\n",
      "       [-0.18885353, -0.16133082],\n",
      "       [-0.1473276 , -0.06737536],\n",
      "       [-0.1534757 , -0.11717645]], dtype=float32), array([[-0.22669783, -0.13201712],\n",
      "       [-0.1962434 , -0.2028026 ],\n",
      "       [-0.20471208, -0.11275259],\n",
      "       ...,\n",
      "       [-0.16652882, -0.05283797],\n",
      "       [-0.16496347, -0.06156228],\n",
      "       [-0.04842011, -0.15948527]], dtype=float32), array([[-0.18033394, -0.14497295],\n",
      "       [-0.11409085, -0.12444185],\n",
      "       [-0.1564497 , -0.02424912],\n",
      "       ...,\n",
      "       [-0.15051265, -0.11195561],\n",
      "       [-0.14703397, -0.13880661],\n",
      "       [-0.1638624 , -0.07511269]], dtype=float32), array([[-0.13046876, -0.08698712],\n",
      "       [-0.09415127, -0.09758955],\n",
      "       [-0.1917573 , -0.03812116],\n",
      "       ...,\n",
      "       [-0.13114071, -0.04772912],\n",
      "       [-0.12230234, -0.02584953],\n",
      "       [-0.0223985 , -0.11070378]], dtype=float32), array([[-0.14861421, -0.21506913],\n",
      "       [-0.15446095, -0.11148316],\n",
      "       [-0.14749508, -0.09635814],\n",
      "       ...,\n",
      "       [-0.14488517, -0.05642219],\n",
      "       [-0.17891717, -0.00837738],\n",
      "       [-0.16964373, -0.09589477]], dtype=float32), array([[-0.18632764, -0.09105914],\n",
      "       [-0.08343393, -0.12852387],\n",
      "       [-0.14999777, -0.15999322],\n",
      "       ...,\n",
      "       [-0.24442178, -0.09538306],\n",
      "       [-0.13464864, -0.08556009],\n",
      "       [-0.06296807, -0.1481019 ]], dtype=float32), array([[-0.1907236 , -0.15251735],\n",
      "       [-0.08037115, -0.16178675],\n",
      "       [-0.02250092, -0.17270488],\n",
      "       ...,\n",
      "       [-0.1515604 , -0.07850508],\n",
      "       [-0.09301068, -0.03316961],\n",
      "       [-0.14995651, -0.19184272]], dtype=float32), array([[-0.17746489, -0.11183776],\n",
      "       [-0.09772605, -0.16080733],\n",
      "       [-0.05121041, -0.10503142],\n",
      "       ...,\n",
      "       [-0.15798199, -0.13084257],\n",
      "       [-0.04811297, -0.10949393],\n",
      "       [-0.01118397, -0.21821095]], dtype=float32), array([[-0.09792008, -0.13079223],\n",
      "       [-0.09149974, -0.08967015],\n",
      "       [-0.13470186, -0.13396564],\n",
      "       ...,\n",
      "       [-0.1027092 , -0.1594966 ],\n",
      "       [-0.13656317, -0.19790214],\n",
      "       [-0.06818247, -0.19692118]], dtype=float32), array([[-0.1648369 , -0.11095896],\n",
      "       [-0.14962184, -0.07523717],\n",
      "       [-0.20243156, -0.11674532],\n",
      "       ...,\n",
      "       [-0.12870415, -0.19011535],\n",
      "       [-0.0251692 , -0.0405643 ],\n",
      "       [-0.13069645, -0.09425343]], dtype=float32), array([[-0.10059332, -0.14829542],\n",
      "       [-0.06152335, -0.1832699 ],\n",
      "       [-0.21776909, -0.09849405],\n",
      "       ...,\n",
      "       [-0.13997501, -0.06913049],\n",
      "       [-0.07296169, -0.1122785 ],\n",
      "       [-0.09725742, -0.15151863]], dtype=float32), array([[-0.06881644, -0.05764177],\n",
      "       [-0.18369465, -0.09685294],\n",
      "       [-0.064973  , -0.16441911],\n",
      "       ...,\n",
      "       [-0.09068434, -0.1314368 ],\n",
      "       [-0.067545  , -0.1614722 ],\n",
      "       [-0.03322049, -0.13571972]], dtype=float32), array([[-0.0848949 , -0.1412193 ],\n",
      "       [-0.07424138, -0.02412072],\n",
      "       [-0.04736136, -0.18279144],\n",
      "       ...,\n",
      "       [-0.1501329 , -0.14454544],\n",
      "       [-0.16942988, -0.18956132],\n",
      "       [-0.15425475, -0.1910641 ]], dtype=float32), array([[-0.1463279 , -0.1523383 ],\n",
      "       [-0.0797442 , -0.08287734],\n",
      "       [-0.1268949 , -0.09619731],\n",
      "       ...,\n",
      "       [-0.07814629, -0.03231731],\n",
      "       [-0.06853322, -0.14944859],\n",
      "       [-0.04367679, -0.19801727]], dtype=float32), array([[-0.18262541, -0.12912238],\n",
      "       [-0.09229242, -0.04035888],\n",
      "       [-0.08247368, -0.13641891],\n",
      "       ...,\n",
      "       [-0.17863055, -0.15169194],\n",
      "       [-0.14772466, -0.05883241],\n",
      "       [-0.09458551, -0.18705647]], dtype=float32), array([[-0.05685416, -0.05526835],\n",
      "       [-0.19749598, -0.08491658],\n",
      "       [-0.17908253, -0.08500383],\n",
      "       ...,\n",
      "       [-0.14001477, -0.15265149],\n",
      "       [-0.13967472, -0.17064974],\n",
      "       [-0.10252549, -0.12060271]], dtype=float32), array([[-0.15260786, -0.17423747],\n",
      "       [-0.13480923, -0.14313093],\n",
      "       [-0.19076133, -0.1848474 ],\n",
      "       ...,\n",
      "       [-0.08294709, -0.13380201],\n",
      "       [-0.17112248, -0.10388027],\n",
      "       [-0.15931387, -0.13844106]], dtype=float32), array([[-0.13045229, -0.14685076],\n",
      "       [-0.11074544, -0.10221934],\n",
      "       [-0.05925822, -0.15112858],\n",
      "       ...,\n",
      "       [-0.08169234, -0.0242656 ],\n",
      "       [-0.13472247, -0.16145584],\n",
      "       [-0.0899098 , -0.06691629]], dtype=float32), array([[-0.04301517, -0.08009117],\n",
      "       [-0.17527014, -0.05442261],\n",
      "       [-0.06225785, -0.07935107],\n",
      "       ...,\n",
      "       [-0.16334414, -0.22335242],\n",
      "       [-0.0991762 , -0.11843368],\n",
      "       [-0.09723415, -0.14552355]], dtype=float32), array([[-0.18622358, -0.22420068],\n",
      "       [-0.09879829, -0.16410223],\n",
      "       [-0.0478717 , -0.11878321],\n",
      "       ...,\n",
      "       [-0.06356183, -0.13398643],\n",
      "       [-0.158056  , -0.13184527],\n",
      "       [-0.16653672, -0.06438064]], dtype=float32), array([[-0.14928196, -0.13002089],\n",
      "       [-0.07732505, -0.16729271],\n",
      "       [-0.11363261, -0.14757165],\n",
      "       ...,\n",
      "       [-0.16312484, -0.06026006],\n",
      "       [-0.18409258, -0.09324311],\n",
      "       [-0.17449163, -0.13524842]], dtype=float32), array([[-0.15589722, -0.04978243],\n",
      "       [-0.15364307, -0.22835939],\n",
      "       [-0.14631434, -0.07550741],\n",
      "       ...,\n",
      "       [-0.07240505, -0.14805645],\n",
      "       [-0.12130734, -0.11091615],\n",
      "       [-0.14685   , -0.04719983]], dtype=float32), array([[-0.12750654, -0.09871345],\n",
      "       [-0.13335977, -0.15461005],\n",
      "       [-0.20917505, -0.05815005],\n",
      "       ...,\n",
      "       [-0.07583277, -0.14081247],\n",
      "       [-0.158666  , -0.14048637],\n",
      "       [-0.19413963, -0.11292055]], dtype=float32), array([[-0.19389424, -0.07666577],\n",
      "       [-0.16086821, -0.14783299],\n",
      "       [-0.10962667, -0.14061624],\n",
      "       ...,\n",
      "       [-0.11443435, -0.20096461],\n",
      "       [-0.12458152, -0.13728754],\n",
      "       [-0.10782786, -0.20476004]], dtype=float32), array([[-0.15333691, -0.11182223],\n",
      "       [-0.1306495 , -0.06824978],\n",
      "       [-0.13069749, -0.12186863],\n",
      "       ...,\n",
      "       [-0.15589029, -0.09816211],\n",
      "       [-0.06915247, -0.0808624 ],\n",
      "       [-0.20314285, -0.15847653]], dtype=float32), array([[-0.1209847 , -0.16887867],\n",
      "       [-0.17843528, -0.13213757],\n",
      "       [-0.1373615 , -0.20404467],\n",
      "       ...,\n",
      "       [-0.09601144, -0.14160475],\n",
      "       [-0.16851756, -0.19834411],\n",
      "       [-0.10983457, -0.11629175]], dtype=float32), array([[-0.1472451 , -0.02888496],\n",
      "       [-0.16183867, -0.13759242],\n",
      "       [-0.05820088, -0.09052683],\n",
      "       ...,\n",
      "       [-0.06819744, -0.21346617],\n",
      "       [-0.03823102, -0.21177077],\n",
      "       [-0.18102527, -0.16975583]], dtype=float32), array([[-0.1524759 , -0.23081785],\n",
      "       [-0.15530784, -0.11779378],\n",
      "       [-0.15598845, -0.14370516],\n",
      "       ...,\n",
      "       [-0.09038054, -0.14717399],\n",
      "       [-0.10113941, -0.11076751],\n",
      "       [-0.06014692, -0.1319826 ]], dtype=float32), array([[-0.16967645, -0.17686   ],\n",
      "       [-0.08526786, -0.14882468],\n",
      "       [-0.1053514 , -0.09702079],\n",
      "       ...,\n",
      "       [-0.15842906, -0.10240679],\n",
      "       [-0.10810175, -0.20125964],\n",
      "       [-0.08970094, -0.15741867]], dtype=float32), array([[-0.1078888 , -0.13384634],\n",
      "       [-0.14514855, -0.2582035 ],\n",
      "       [-0.09971835, -0.10492329],\n",
      "       ...,\n",
      "       [-0.09002198, -0.0507311 ],\n",
      "       [-0.13571927, -0.153976  ],\n",
      "       [-0.0700369 , -0.13243125]], dtype=float32), array([[-0.11096965, -0.119427  ],\n",
      "       [-0.15225083, -0.08466126],\n",
      "       [-0.15183775, -0.13091496],\n",
      "       ...,\n",
      "       [-0.15593642, -0.09271974],\n",
      "       [-0.1294466 , -0.05548253],\n",
      "       [-0.14898963, -0.05446853]], dtype=float32), array([[-0.11643179, -0.21268532],\n",
      "       [-0.09890739, -0.14193702],\n",
      "       [-0.11847021, -0.24211815],\n",
      "       ...,\n",
      "       [-0.16103151, -0.20769033],\n",
      "       [-0.17697725, -0.2187895 ],\n",
      "       [-0.06162669, -0.04720311]], dtype=float32), array([[-0.1165112 , -0.04887313],\n",
      "       [-0.08488281, -0.19982658],\n",
      "       [-0.08637466, -0.10641866],\n",
      "       ...,\n",
      "       [-0.19953753, -0.12837675],\n",
      "       [-0.04938107, -0.21544641],\n",
      "       [-0.16988568, -0.17152643]], dtype=float32), array([[-0.12601522, -0.20441702],\n",
      "       [-0.15893197, -0.08324322],\n",
      "       [-0.12085667, -0.11304581],\n",
      "       ...,\n",
      "       [-0.07925469, -0.09726956],\n",
      "       [-0.02524705, -0.08567899],\n",
      "       [-0.14131415, -0.12572406]], dtype=float32), array([[-0.12899481, -0.10850333],\n",
      "       [-0.11615325, -0.13241951],\n",
      "       [-0.12817587, -0.10599783],\n",
      "       ...,\n",
      "       [-0.12757266, -0.16105753],\n",
      "       [-0.1333101 , -0.1646729 ],\n",
      "       [-0.088749  , -0.1174328 ]], dtype=float32), array([[-0.10935669, -0.16216815],\n",
      "       [-0.12348682, -0.08523864],\n",
      "       [-0.1595337 , -0.16192749],\n",
      "       ...,\n",
      "       [-0.02776077, -0.09777635],\n",
      "       [-0.03327383, -0.20974667],\n",
      "       [-0.07454589, -0.11344708]], dtype=float32), array([[-0.131005  , -0.10868739],\n",
      "       [-0.2029183 , -0.07418986],\n",
      "       [-0.175326  , -0.14284381],\n",
      "       ...,\n",
      "       [-0.16022348, -0.13105915],\n",
      "       [-0.12899575, -0.07715087],\n",
      "       [-0.12872006, -0.14064616]], dtype=float32), array([[-0.19769838, -0.08255308],\n",
      "       [-0.08311651, -0.12414366],\n",
      "       [-0.14710242, -0.1525492 ],\n",
      "       ...,\n",
      "       [-0.11445697, -0.15918359],\n",
      "       [-0.08169796, -0.06111081],\n",
      "       [-0.11101795, -0.11841217]], dtype=float32), array([[-0.08829471, -0.03451327],\n",
      "       [-0.12507531, -0.06308614],\n",
      "       [-0.18117598, -0.10455553],\n",
      "       ...,\n",
      "       [-0.13869464, -0.06268255],\n",
      "       [-0.16351554, -0.06000188],\n",
      "       [-0.1713587 , -0.13561565]], dtype=float32), array([[-0.08044469, -0.13287453],\n",
      "       [-0.16937053, -0.10712549],\n",
      "       [-0.10190484, -0.10323751],\n",
      "       ...,\n",
      "       [-0.01802303, -0.12871465],\n",
      "       [-0.08196137, -0.10154223],\n",
      "       [-0.14126718, -0.12116852]], dtype=float32), array([[-0.17685917, -0.08607845],\n",
      "       [-0.13306603, -0.12980478],\n",
      "       [-0.1853836 , -0.04039282],\n",
      "       ...,\n",
      "       [-0.08750174, -0.14952719],\n",
      "       [-0.14129353, -0.16703454],\n",
      "       [-0.19113547, -0.1607371 ]], dtype=float32), array([[-0.09551916, -0.11216971],\n",
      "       [-0.08264321, -0.1792577 ],\n",
      "       [-0.14470585, -0.16628152],\n",
      "       ...,\n",
      "       [-0.18709464, -0.09616934],\n",
      "       [-0.06403591, -0.15303256],\n",
      "       [-0.22760719, -0.13727853]], dtype=float32), array([[-0.14690715, -0.14612494],\n",
      "       [-0.1765751 , -0.05925631],\n",
      "       [-0.21716389, -0.0790281 ],\n",
      "       ...,\n",
      "       [-0.09058627, -0.1596609 ],\n",
      "       [-0.11958309, -0.04273675],\n",
      "       [-0.17915228, -0.17573097]], dtype=float32), array([[-0.10133073, -0.09670094],\n",
      "       [-0.15569243, -0.05672102],\n",
      "       [-0.1655315 , -0.13223958],\n",
      "       ...,\n",
      "       [-0.00897819, -0.18587942],\n",
      "       [-0.06961878, -0.19909981],\n",
      "       [-0.17154607, -0.146218  ]], dtype=float32), array([[-0.03079765, -0.1157061 ],\n",
      "       [-0.12398097, -0.08765363],\n",
      "       [-0.08609091, -0.16906922],\n",
      "       ...,\n",
      "       [-0.10269044, -0.10289335],\n",
      "       [-0.149349  , -0.10264789],\n",
      "       [-0.07004553, -0.18274263]], dtype=float32), array([[-0.1316727 , -0.21136251],\n",
      "       [-0.06820635, -0.11039808],\n",
      "       [-0.05602678, -0.08077163],\n",
      "       ...,\n",
      "       [-0.20685875, -0.15631485],\n",
      "       [-0.13923998, -0.20457447],\n",
      "       [-0.16499336, -0.08859968]], dtype=float32), array([[-0.12117243, -0.1463043 ],\n",
      "       [-0.18711492, -0.13609396],\n",
      "       [-0.0108269 , -0.13340358],\n",
      "       ...,\n",
      "       [-0.1987925 , -0.18296385],\n",
      "       [-0.06152917, -0.07998059],\n",
      "       [-0.10527813, -0.10589058]], dtype=float32), array([[-0.06098827, -0.17477308],\n",
      "       [-0.06352129, -0.12714589],\n",
      "       [-0.19088769, -0.07235607],\n",
      "       ...,\n",
      "       [-0.07142076, -0.13256809],\n",
      "       [-0.08042585, -0.08529431],\n",
      "       [-0.10714334, -0.187973  ]], dtype=float32), array([[-0.10809582, -0.1472662 ],\n",
      "       [-0.18460216, -0.12891658],\n",
      "       [-0.17200781, -0.06980975],\n",
      "       ...,\n",
      "       [-0.20757495, -0.12366237],\n",
      "       [-0.13763641, -0.17199439],\n",
      "       [-0.05321629, -0.16650993]], dtype=float32), array([[-0.17386091, -0.07244583],\n",
      "       [-0.10809094, -0.13442403],\n",
      "       [-0.05941585, -0.08852728],\n",
      "       ...,\n",
      "       [-0.21708867, -0.09572686],\n",
      "       [-0.20729324, -0.1110993 ],\n",
      "       [-0.17679572, -0.0363608 ]], dtype=float32), array([[-0.1525562 , -0.14967985],\n",
      "       [-0.16071108, -0.11922259],\n",
      "       [-0.08067536, -0.05884471],\n",
      "       ...,\n",
      "       [-0.14655717, -0.13066867],\n",
      "       [-0.20716766, -0.15331075],\n",
      "       [-0.02501666, -0.08419176]], dtype=float32), array([[-0.15756111, -0.1228651 ],\n",
      "       [-0.09124933, -0.05028653],\n",
      "       [-0.14808194, -0.12789404],\n",
      "       ...,\n",
      "       [-0.11820013, -0.05218874],\n",
      "       [-0.08560198, -0.14974226],\n",
      "       [-0.10087973, -0.11052264]], dtype=float32), array([[-0.171134  , -0.06728213],\n",
      "       [-0.07915534, -0.1497516 ],\n",
      "       [-0.11641758, -0.0870166 ],\n",
      "       ...,\n",
      "       [-0.09506036, -0.17801705],\n",
      "       [-0.1260047 , -0.12223519],\n",
      "       [-0.19808611, -0.17148745]], dtype=float32), array([[-0.10266368, -0.13187459],\n",
      "       [-0.18534485, -0.1805908 ],\n",
      "       [-0.10549287, -0.16586731],\n",
      "       ...,\n",
      "       [-0.15361263, -0.03411802],\n",
      "       [-0.09987208, -0.06004155],\n",
      "       [-0.12198734, -0.18533196]], dtype=float32), array([[-0.09742684, -0.08576483],\n",
      "       [-0.15564083, -0.18689033],\n",
      "       [-0.17929849, -0.09448498],\n",
      "       ...,\n",
      "       [-0.12264243, -0.15618128],\n",
      "       [-0.20446393, -0.10878462],\n",
      "       [-0.1209803 , -0.13306293]], dtype=float32), array([[-0.16290098, -0.17602131],\n",
      "       [-0.1438188 , -0.20018962],\n",
      "       [-0.18129958, -0.14633857],\n",
      "       ...,\n",
      "       [-0.10836651, -0.17044453],\n",
      "       [-0.15258151, -0.13104148],\n",
      "       [-0.16913797, -0.05983722]], dtype=float32), array([[-0.17146866, -0.05227009],\n",
      "       [-0.13761659, -0.09317806],\n",
      "       [-0.17660636, -0.11907548],\n",
      "       ...,\n",
      "       [-0.12823407, -0.18179527],\n",
      "       [-0.15210634, -0.10561712],\n",
      "       [-0.1260787 , -0.18408765]], dtype=float32), array([[-0.13135235, -0.16182889],\n",
      "       [-0.15088971, -0.07191022],\n",
      "       [-0.14191216, -0.09334902],\n",
      "       ...,\n",
      "       [-0.11750675, -0.1562106 ],\n",
      "       [-0.12633105, -0.11428841],\n",
      "       [-0.10783336, -0.06780442]], dtype=float32), array([[-0.13692892, -0.11010401],\n",
      "       [-0.1958276 , -0.16761205],\n",
      "       [-0.18216103, -0.18315902],\n",
      "       ...,\n",
      "       [-0.1359959 , -0.06840748],\n",
      "       [-0.07941294, -0.18879533],\n",
      "       [-0.11928497, -0.06094375]], dtype=float32), array([[-0.05072796, -0.12233779],\n",
      "       [-0.17553475, -0.1841146 ],\n",
      "       [-0.05793084, -0.0900848 ],\n",
      "       ...,\n",
      "       [-0.03622333, -0.15107872],\n",
      "       [-0.10154506, -0.08263353],\n",
      "       [-0.0563561 , -0.16758403]], dtype=float32), array([[-0.18522725, -0.04865424],\n",
      "       [-0.11627626, -0.18971702],\n",
      "       [-0.15790057, -0.16766366],\n",
      "       ...,\n",
      "       [-0.10088769, -0.12882361],\n",
      "       [-0.1386712 , -0.14035414],\n",
      "       [-0.16028854, -0.13990149]], dtype=float32), array([[-0.12089952, -0.08279874],\n",
      "       [-0.15782215, -0.10117844],\n",
      "       [-0.12156531, -0.1684203 ],\n",
      "       ...,\n",
      "       [-0.13419023, -0.07237287],\n",
      "       [-0.08029632, -0.15169576],\n",
      "       [-0.0917947 , -0.06003851]], dtype=float32), array([[-0.15632458, -0.06884383],\n",
      "       [-0.03844602, -0.1359422 ],\n",
      "       [-0.05086486, -0.14400771],\n",
      "       ...,\n",
      "       [-0.12085515, -0.18196338],\n",
      "       [-0.05621032, -0.18521304],\n",
      "       [-0.19832209, -0.20032482]], dtype=float32), array([[-0.170512  , -0.03291008],\n",
      "       [-0.15708248, -0.08477172],\n",
      "       [-0.13095129, -0.06078037],\n",
      "       ...,\n",
      "       [-0.07497516, -0.16052061],\n",
      "       [-0.11589533, -0.08945495],\n",
      "       [-0.07453763, -0.13532227]], dtype=float32), array([[-0.09803375, -0.2660803 ],\n",
      "       [-0.08058846, -0.1094896 ],\n",
      "       [-0.20214836, -0.09253597],\n",
      "       ...,\n",
      "       [-0.22498675, -0.19337618],\n",
      "       [-0.16893293, -0.20575619],\n",
      "       [-0.14091703, -0.15814257]], dtype=float32), array([[-0.15532649, -0.13405326],\n",
      "       [-0.09757774, -0.07974519],\n",
      "       [-0.07533299, -0.16433683],\n",
      "       ...,\n",
      "       [-0.19629326, -0.20071825],\n",
      "       [-0.1051964 , -0.12848847],\n",
      "       [-0.05581992, -0.13460061]], dtype=float32), array([[-0.17251632, -0.12611711],\n",
      "       [-0.19854628, -0.18822977],\n",
      "       [-0.155515  , -0.04786018],\n",
      "       ...,\n",
      "       [-0.18342492, -0.13088033],\n",
      "       [-0.12441633, -0.06853426],\n",
      "       [-0.12012675, -0.15495524]], dtype=float32), array([[-0.09482178, -0.19603541],\n",
      "       [-0.09166071, -0.12524588],\n",
      "       [-0.16176048, -0.11835409],\n",
      "       ...,\n",
      "       [-0.17497358, -0.14964187],\n",
      "       [-0.16352944, -0.19283086],\n",
      "       [-0.20466232, -0.09874036]], dtype=float32), array([[-0.11674672, -0.16467401],\n",
      "       [-0.12184181, -0.21634373],\n",
      "       [-0.10401207, -0.19289222],\n",
      "       ...,\n",
      "       [-0.06005355, -0.170162  ],\n",
      "       [-0.12136   , -0.11788516],\n",
      "       [-0.16531214, -0.07654323]], dtype=float32), array([[-0.03471667, -0.1238471 ],\n",
      "       [-0.20285437, -0.11344762],\n",
      "       [-0.12152703, -0.20978467],\n",
      "       ...,\n",
      "       [-0.08848405, -0.08079518],\n",
      "       [-0.13416937, -0.14420897],\n",
      "       [-0.14488049, -0.14560279]], dtype=float32), array([[-0.07187076, -0.10481576],\n",
      "       [-0.12083981, -0.10336985],\n",
      "       [-0.07573432, -0.07211976],\n",
      "       ...,\n",
      "       [-0.21868399, -0.17687975],\n",
      "       [-0.10582448, -0.21721146],\n",
      "       [-0.0523061 , -0.08566587]], dtype=float32), array([[-0.17390132, -0.04969346],\n",
      "       [-0.08840306, -0.12908283],\n",
      "       [-0.1046936 , -0.17812043],\n",
      "       ...,\n",
      "       [-0.11218677, -0.11553197],\n",
      "       [-0.10307235, -0.06625732],\n",
      "       [-0.14854439, -0.09106605]], dtype=float32), array([[-0.07741257, -0.12443353],\n",
      "       [-0.07640709, -0.14609027],\n",
      "       [-0.19586176, -0.05720377],\n",
      "       ...,\n",
      "       [-0.1552993 , -0.13423112],\n",
      "       [-0.09713329, -0.22657117],\n",
      "       [-0.03848798, -0.12511387]], dtype=float32), array([[-0.14753672, -0.14692545],\n",
      "       [-0.1331923 , -0.17082971],\n",
      "       [-0.19598947, -0.10326643],\n",
      "       ...,\n",
      "       [-0.1877635 , -0.15331566],\n",
      "       [-0.14216043, -0.08870994],\n",
      "       [-0.13937415, -0.05111982]], dtype=float32), array([[-0.17788762, -0.13970704],\n",
      "       [-0.15477721, -0.12183269],\n",
      "       [-0.091374  , -0.23675592],\n",
      "       ...,\n",
      "       [-0.16816664, -0.07526328],\n",
      "       [-0.18145612, -0.07439436],\n",
      "       [-0.20624895, -0.1261633 ]], dtype=float32), array([[-0.14253083, -0.035023  ],\n",
      "       [-0.09723096, -0.1291865 ],\n",
      "       [-0.10291082, -0.12015654],\n",
      "       ...,\n",
      "       [-0.07626644, -0.1597221 ],\n",
      "       [-0.11058471, -0.06326266],\n",
      "       [-0.10523093, -0.04348093]], dtype=float32), array([[-0.18445052, -0.04918578],\n",
      "       [-0.11824549, -0.11197001],\n",
      "       [-0.13081901, -0.11146297],\n",
      "       ...,\n",
      "       [-0.20183574, -0.11232939],\n",
      "       [-0.21925555, -0.12015378],\n",
      "       [-0.14465089, -0.12856601]], dtype=float32), array([[-0.03707995, -0.09494269],\n",
      "       [-0.10069435, -0.05063383],\n",
      "       [-0.1532831 , -0.11287962],\n",
      "       ...,\n",
      "       [-0.22822239, -0.06437271],\n",
      "       [-0.2324131 , -0.05054288],\n",
      "       [-0.17825127, -0.1964832 ]], dtype=float32), array([[-0.04851685, -0.11780579],\n",
      "       [-0.12552069, -0.08845103],\n",
      "       [-0.08062717, -0.26110148],\n",
      "       ...,\n",
      "       [-0.15293539, -0.1661922 ],\n",
      "       [-0.17515746, -0.08073862],\n",
      "       [-0.14982307, -0.14253378]], dtype=float32), array([[-0.11627793, -0.16549522],\n",
      "       [-0.14685115, -0.06345461],\n",
      "       [-0.17553718, -0.09993685],\n",
      "       ...,\n",
      "       [-0.1162626 , -0.12943943],\n",
      "       [-0.08589599, -0.08978032],\n",
      "       [-0.09686859, -0.26259297]], dtype=float32), array([[-0.23671043, -0.01425646],\n",
      "       [-0.14379588, -0.14697981],\n",
      "       [-0.07070898, -0.10709669],\n",
      "       ...,\n",
      "       [-0.1729481 , -0.17423338],\n",
      "       [-0.22864808, -0.0543252 ],\n",
      "       [-0.12711875, -0.13227563]], dtype=float32), array([[-0.1400801 , -0.06518403],\n",
      "       [-0.13214236, -0.14348678],\n",
      "       [-0.07189264, -0.09744322],\n",
      "       ...,\n",
      "       [-0.18116102, -0.17737435],\n",
      "       [-0.19266331, -0.07656021],\n",
      "       [-0.1123896 , -0.10025506]], dtype=float32), array([[-0.08183695, -0.15041713],\n",
      "       [-0.0511292 , -0.16085248],\n",
      "       [-0.17183182, -0.13946831],\n",
      "       ...,\n",
      "       [-0.04191387, -0.10807993],\n",
      "       [-0.20227817, -0.0154564 ],\n",
      "       [-0.11714048, -0.14259954]], dtype=float32), array([[-0.22739358, -0.10588811],\n",
      "       [-0.13729289, -0.13122645],\n",
      "       [-0.21873434, -0.11628605],\n",
      "       ...,\n",
      "       [-0.12105282, -0.10992647],\n",
      "       [-0.14471972, -0.13696212],\n",
      "       [-0.11749619, -0.1992336 ]], dtype=float32), array([[-0.11695176, -0.10380145],\n",
      "       [-0.14687616, -0.09084306],\n",
      "       [-0.08488999, -0.18111753],\n",
      "       ...,\n",
      "       [-0.09397879, -0.10588406],\n",
      "       [-0.1327397 , -0.11962442],\n",
      "       [-0.15661274, -0.1550442 ]], dtype=float32), array([[-0.19842146, -0.1451667 ],\n",
      "       [-0.17527902, -0.06161905],\n",
      "       [-0.1656112 , -0.1617283 ],\n",
      "       ...,\n",
      "       [-0.0988306 , -0.1704418 ],\n",
      "       [-0.15426928, -0.08304853],\n",
      "       [-0.1412202 , -0.10869478]], dtype=float32), array([[-0.17866045, -0.11444526],\n",
      "       [-0.0698211 , -0.12157635],\n",
      "       [-0.11353534, -0.1507717 ],\n",
      "       ...,\n",
      "       [-0.01928116, -0.15453456],\n",
      "       [-0.21188271, -0.08135577],\n",
      "       [-0.13770933, -0.17517212]], dtype=float32), array([[-0.08893085, -0.12412277],\n",
      "       [-0.18408391, -0.12312021],\n",
      "       [-0.15637612, -0.09074796],\n",
      "       ...,\n",
      "       [-0.11098973, -0.19254313],\n",
      "       [-0.09567975, -0.20216335],\n",
      "       [-0.12537004, -0.08981661]], dtype=float32), array([[-0.22121637, -0.12376599],\n",
      "       [-0.1564613 , -0.09772362],\n",
      "       [-0.0769034 , -0.13810807],\n",
      "       ...,\n",
      "       [-0.18728498, -0.12165422],\n",
      "       [-0.05468102, -0.10288042],\n",
      "       [-0.06304737, -0.14948437]], dtype=float32), array([[-0.1250941 , -0.09370433],\n",
      "       [-0.0614358 , -0.11310454],\n",
      "       [-0.0906661 , -0.14437495],\n",
      "       ...,\n",
      "       [-0.09426285, -0.18712476],\n",
      "       [-0.21309397, -0.16484171],\n",
      "       [-0.02494379, -0.06801534]], dtype=float32), array([[-0.18029106, -0.10756697],\n",
      "       [-0.04710976, -0.15554407],\n",
      "       [-0.15218325, -0.10665306],\n",
      "       ...,\n",
      "       [-0.08825457, -0.17344631],\n",
      "       [-0.17508312, -0.17645176],\n",
      "       [-0.10349241, -0.04835346]], dtype=float32), array([[-0.19953181, -0.13650814],\n",
      "       [-0.15288746, -0.06266413],\n",
      "       [-0.22265945, -0.13362044],\n",
      "       ...,\n",
      "       [-0.15573268, -0.1374354 ],\n",
      "       [-0.13905436, -0.16254959],\n",
      "       [-0.11545847, -0.1618774 ]], dtype=float32), array([[-0.16371886, -0.13534108],\n",
      "       [-0.19286864, -0.06265116],\n",
      "       [-0.07924348, -0.1374324 ],\n",
      "       ...,\n",
      "       [-0.13439202, -0.10658887],\n",
      "       [-0.19816056, -0.14820915],\n",
      "       [-0.12789539, -0.11859451]], dtype=float32), array([[-0.11562144, -0.11180218],\n",
      "       [-0.07733906, -0.09304847],\n",
      "       [-0.13119066, -0.08048934],\n",
      "       ...,\n",
      "       [-0.20287476, -0.18151213],\n",
      "       [-0.15737021, -0.18643747],\n",
      "       [-0.16229117, -0.10513199]], dtype=float32), array([[-0.16205221, -0.17632404],\n",
      "       [-0.13319167, -0.1502729 ],\n",
      "       [-0.11878501, -0.0851553 ],\n",
      "       ...,\n",
      "       [-0.03670886, -0.06570602],\n",
      "       [-0.15775397, -0.10764007],\n",
      "       [-0.15705165, -0.12247375]], dtype=float32), array([[-0.1567635 , -0.14483312],\n",
      "       [-0.20787373, -0.11322505],\n",
      "       [-0.08849116, -0.23458996],\n",
      "       ...,\n",
      "       [-0.06803692, -0.12449788],\n",
      "       [-0.18581843, -0.14535306],\n",
      "       [-0.04702599, -0.1007445 ]], dtype=float32), array([[-0.11574752, -0.1972947 ],\n",
      "       [-0.12596527, -0.20063761],\n",
      "       [-0.06373217, -0.18183362],\n",
      "       ...,\n",
      "       [-0.1369388 , -0.06332646],\n",
      "       [-0.11439995, -0.12874913],\n",
      "       [-0.15030162, -0.06138959]], dtype=float32), array([[-0.11392289, -0.09397469],\n",
      "       [-0.17035833, -0.17199752],\n",
      "       [-0.09136169, -0.10284289],\n",
      "       ...,\n",
      "       [-0.06664003, -0.10507213],\n",
      "       [-0.04786253, -0.18263799],\n",
      "       [-0.12778185, -0.12190958]], dtype=float32), array([[-0.06038449, -0.15084948],\n",
      "       [-0.17833418, -0.11752126],\n",
      "       [-0.04243968, -0.13063614],\n",
      "       ...,\n",
      "       [-0.1745193 , -0.08566262],\n",
      "       [-0.05766495, -0.16835096],\n",
      "       [-0.18381594, -0.17222889]], dtype=float32), array([[-0.13932557, -0.05042678],\n",
      "       [-0.24433053, -0.07803299],\n",
      "       [-0.04023122, -0.09672986],\n",
      "       ...,\n",
      "       [-0.08994501, -0.11035281],\n",
      "       [-0.15774181, -0.12062894],\n",
      "       [-0.03972339, -0.12009801]], dtype=float32), array([[-0.08485422, -0.07188088],\n",
      "       [-0.07170273, -0.05994263],\n",
      "       [-0.13133898, -0.05193817],\n",
      "       ...,\n",
      "       [-0.20107052, -0.13901168],\n",
      "       [-0.12935588, -0.21889782],\n",
      "       [-0.17788824, -0.14495836]], dtype=float32), array([[-0.18781196, -0.18523835],\n",
      "       [-0.07345939, -0.11784054],\n",
      "       [-0.15454882, -0.10102918],\n",
      "       ...,\n",
      "       [-0.10899741, -0.15237899],\n",
      "       [-0.06390195, -0.138387  ],\n",
      "       [-0.06848392, -0.20052464]], dtype=float32), array([[-0.11313739, -0.21913636],\n",
      "       [-0.09796956, -0.18099354],\n",
      "       [-0.08390207, -0.13546589],\n",
      "       ...,\n",
      "       [-0.12943676, -0.04991848],\n",
      "       [-0.18673263, -0.06318815],\n",
      "       [-0.18013296, -0.20119041]], dtype=float32), array([[-0.17753574, -0.18251254],\n",
      "       [-0.12992318, -0.1675317 ],\n",
      "       [-0.14439933, -0.13700257],\n",
      "       ...,\n",
      "       [-0.11369226, -0.08020747],\n",
      "       [-0.11637998, -0.13798729],\n",
      "       [-0.14735138, -0.10180146]], dtype=float32), array([[-0.06285824, -0.16152439],\n",
      "       [-0.15488337, -0.14800757],\n",
      "       [-0.11141976, -0.11802388],\n",
      "       ...,\n",
      "       [-0.1391948 , -0.12007805],\n",
      "       [-0.13859048, -0.11508022],\n",
      "       [-0.09512717, -0.12802473]], dtype=float32), array([[-0.15441534, -0.14071737],\n",
      "       [-0.13565902, -0.16975734],\n",
      "       [-0.05660586, -0.05049402],\n",
      "       ...,\n",
      "       [-0.15017144, -0.2009713 ],\n",
      "       [-0.01546471, -0.13754036],\n",
      "       [-0.11366512, -0.09740047]], dtype=float32), array([[-0.12270783, -0.13608515],\n",
      "       [-0.17584269, -0.08193059],\n",
      "       [-0.13460766, -0.09597985],\n",
      "       ...,\n",
      "       [-0.1364169 , -0.03896739],\n",
      "       [-0.1484977 , -0.10290703],\n",
      "       [-0.09121718, -0.16029678]], dtype=float32), array([[-0.04852864, -0.06372949],\n",
      "       [-0.12048043, -0.09169984],\n",
      "       [-0.05071722, -0.18119529],\n",
      "       ...,\n",
      "       [-0.15586707, -0.16939512],\n",
      "       [-0.15343331, -0.15018374],\n",
      "       [-0.10704334, -0.1280427 ]], dtype=float32), array([[-0.09308285, -0.11046261],\n",
      "       [-0.19658543, -0.17360711],\n",
      "       [-0.21007095, -0.09330291],\n",
      "       ...,\n",
      "       [-0.18884498, -0.08649861],\n",
      "       [-0.06781594, -0.12865186],\n",
      "       [-0.05830261, -0.10162674]], dtype=float32), array([[-0.01940064, -0.17428982],\n",
      "       [-0.11732451, -0.13785386],\n",
      "       [-0.15428522, -0.10667065],\n",
      "       ...,\n",
      "       [-0.1907319 , -0.13096793],\n",
      "       [-0.09857265, -0.19327736],\n",
      "       [-0.07514641, -0.14638129]], dtype=float32), array([[-0.17823231, -0.15872753],\n",
      "       [-0.16638947, -0.19252191],\n",
      "       [-0.12672581, -0.12702376],\n",
      "       ...,\n",
      "       [-0.15882516, -0.13680512],\n",
      "       [-0.21775457, -0.07341072],\n",
      "       [-0.22343941, -0.13446052]], dtype=float32), array([[-0.08912203, -0.12451753],\n",
      "       [-0.05797474, -0.13680175],\n",
      "       [-0.09286024, -0.06830727],\n",
      "       ...,\n",
      "       [-0.07705353, -0.07066928],\n",
      "       [-0.21203732, -0.17897718],\n",
      "       [-0.18755162, -0.1640239 ]], dtype=float32), array([[-0.19561626, -0.19643517],\n",
      "       [-0.16363333, -0.1588491 ],\n",
      "       [-0.11335237, -0.15451553],\n",
      "       ...,\n",
      "       [-0.02119353, -0.07427615],\n",
      "       [-0.14797379, -0.11245339],\n",
      "       [-0.09201125, -0.13177682]], dtype=float32), array([[-0.15369691, -0.11591158],\n",
      "       [-0.06732772, -0.1312637 ],\n",
      "       [-0.10914438, -0.16258265],\n",
      "       ...,\n",
      "       [-0.06579625, -0.15263428],\n",
      "       [-0.12841146, -0.16377291],\n",
      "       [-0.05706999, -0.09427493]], dtype=float32), array([[-0.13285148, -0.18729308],\n",
      "       [-0.14360633, -0.12000059],\n",
      "       [-0.12695044, -0.15011755],\n",
      "       ...,\n",
      "       [-0.11929064, -0.11369541],\n",
      "       [-0.09766865, -0.19509414],\n",
      "       [-0.05869835, -0.16209282]], dtype=float32), array([[-0.18364172, -0.06174807],\n",
      "       [-0.13745123, -0.16255762],\n",
      "       [-0.07170018, -0.1688781 ],\n",
      "       ...,\n",
      "       [-0.19438292, -0.18250692],\n",
      "       [-0.11351047, -0.06531762],\n",
      "       [-0.15127191, -0.05482296]], dtype=float32), array([[-0.18119691, -0.08879463],\n",
      "       [-0.13844424, -0.14582083],\n",
      "       [-0.19711134, -0.04685061],\n",
      "       ...,\n",
      "       [-0.24805655, -0.04296969],\n",
      "       [-0.12972084, -0.06334154],\n",
      "       [-0.11514574, -0.10134302]], dtype=float32), array([[-0.14385596, -0.18234192],\n",
      "       [-0.15170744, -0.08857523],\n",
      "       [-0.1808902 , -0.1016986 ],\n",
      "       ...,\n",
      "       [-0.15603088, -0.12045798],\n",
      "       [-0.11929762, -0.17927793],\n",
      "       [-0.15388331, -0.11382324]], dtype=float32), array([[-0.13237406, -0.06188735],\n",
      "       [-0.15998149, -0.13484988],\n",
      "       [-0.06346584, -0.08990227],\n",
      "       ...,\n",
      "       [-0.12200771, -0.10719059],\n",
      "       [-0.12975913, -0.1407378 ],\n",
      "       [-0.20555472, -0.07393875]], dtype=float32), array([[-0.09592152, -0.2076897 ],\n",
      "       [-0.06701149, -0.05903566],\n",
      "       [-0.16015789, -0.20636445],\n",
      "       ...,\n",
      "       [-0.07194016, -0.06893075],\n",
      "       [-0.10313277, -0.17344944],\n",
      "       [-0.04546003, -0.08510149]], dtype=float32), array([[-0.14217049, -0.17138967],\n",
      "       [-0.1963726 , -0.05619824],\n",
      "       [-0.09850898, -0.15102695],\n",
      "       ...,\n",
      "       [-0.16977301, -0.16389012],\n",
      "       [-0.08986978, -0.08588076],\n",
      "       [-0.15013172, -0.20078018]], dtype=float32), array([[-0.14617956, -0.17485042],\n",
      "       [-0.0885157 , -0.14474118],\n",
      "       [-0.11591743, -0.11918763],\n",
      "       ...,\n",
      "       [-0.12862161, -0.17734465],\n",
      "       [-0.15004888, -0.11560066],\n",
      "       [-0.08009083, -0.05479952]], dtype=float32), array([[-0.12804459, -0.14215903],\n",
      "       [-0.09371801, -0.11955363],\n",
      "       [-0.17715004, -0.11422667],\n",
      "       ...,\n",
      "       [-0.1113048 , -0.18787079],\n",
      "       [-0.18028215, -0.14371917],\n",
      "       [-0.156421  , -0.17643686]], dtype=float32), array([[-0.14131276, -0.19746901],\n",
      "       [-0.21096094, -0.2178378 ],\n",
      "       [-0.09803271, -0.02956411],\n",
      "       ...,\n",
      "       [-0.17197827, -0.11012115],\n",
      "       [-0.13159241, -0.13914078],\n",
      "       [-0.1736979 , -0.18851632]], dtype=float32), array([[-0.05003079, -0.17308438],\n",
      "       [-0.23491979, -0.18793674],\n",
      "       [-0.1686611 , -0.08439256],\n",
      "       ...,\n",
      "       [-0.1838298 , -0.13182701],\n",
      "       [-0.10913795, -0.11763504],\n",
      "       [-0.20546623, -0.11556701]], dtype=float32), array([[-0.17190403, -0.21519701],\n",
      "       [-0.11007866, -0.13147986],\n",
      "       [-0.20339361, -0.06930718],\n",
      "       ...,\n",
      "       [-0.05878769, -0.09714979],\n",
      "       [-0.09646805, -0.10091089],\n",
      "       [-0.14872691, -0.10070084]], dtype=float32), array([[-0.14631553, -0.0992654 ],\n",
      "       [-0.18480748, -0.12420776],\n",
      "       [-0.06577934, -0.0726196 ],\n",
      "       ...,\n",
      "       [-0.10192493, -0.13185558],\n",
      "       [-0.01262841, -0.10753299],\n",
      "       [-0.15130019, -0.12176459]], dtype=float32), array([[-0.14104313, -0.17601505],\n",
      "       [-0.1236851 , -0.03333365],\n",
      "       [-0.18127047, -0.03824033],\n",
      "       ...,\n",
      "       [-0.08137209, -0.11671407],\n",
      "       [-0.16508238, -0.07558078],\n",
      "       [-0.13512154, -0.07109016]], dtype=float32), array([[-0.12291612, -0.17866537],\n",
      "       [-0.12522551, -0.08979254],\n",
      "       [-0.18866664, -0.22715166],\n",
      "       ...,\n",
      "       [-0.12630042, -0.13865106],\n",
      "       [-0.15159595, -0.1443872 ],\n",
      "       [-0.15179874, -0.09373608]], dtype=float32), array([[-0.13724092, -0.15833288],\n",
      "       [-0.08007976, -0.08353141],\n",
      "       [-0.20078036, -0.05875821],\n",
      "       ...,\n",
      "       [-0.17330083, -0.0537283 ],\n",
      "       [-0.05412776, -0.17494342],\n",
      "       [-0.08264194, -0.09455696]], dtype=float32), array([[-0.16340098, -0.07499877],\n",
      "       [-0.02433134, -0.0603982 ],\n",
      "       [-0.14814351, -0.02490081],\n",
      "       ...,\n",
      "       [-0.15060091, -0.15615623],\n",
      "       [-0.07029933, -0.10381696],\n",
      "       [-0.12908551, -0.17795746]], dtype=float32), array([[-0.1255244 , -0.09558886],\n",
      "       [-0.21917962, -0.0757898 ],\n",
      "       [-0.07800627, -0.15088022],\n",
      "       ...,\n",
      "       [-0.06995051, -0.18402599],\n",
      "       [-0.13632609, -0.12123965],\n",
      "       [-0.11572609, -0.13861771]], dtype=float32), array([[-0.14111602, -0.21498257],\n",
      "       [-0.14553896, -0.15007013],\n",
      "       [-0.16358976, -0.03545001],\n",
      "       ...,\n",
      "       [-0.1121766 , -0.08517601],\n",
      "       [-0.11082826, -0.12714243],\n",
      "       [-0.08538425, -0.16830541]], dtype=float32), array([[-0.08613628, -0.21526767],\n",
      "       [-0.08508602, -0.08256374],\n",
      "       [-0.05076203, -0.17870161],\n",
      "       ...,\n",
      "       [-0.13407792, -0.12019335],\n",
      "       [-0.09348026, -0.05473959],\n",
      "       [-0.17186901, -0.09969019]], dtype=float32), array([[-0.07787026, -0.06587486],\n",
      "       [-0.05893416, -0.2164498 ],\n",
      "       [-0.10329997, -0.08357586],\n",
      "       ...,\n",
      "       [-0.06752399, -0.12403549],\n",
      "       [-0.13667661, -0.14409132],\n",
      "       [-0.13838227, -0.14875719]], dtype=float32), array([[-0.05929606, -0.13726917],\n",
      "       [-0.1426246 , -0.12285343],\n",
      "       [-0.18094128, -0.14474158],\n",
      "       ...,\n",
      "       [-0.04984731, -0.14290494],\n",
      "       [-0.15206905, -0.10761671],\n",
      "       [-0.13221245, -0.09936881]], dtype=float32), array([[-0.15366296, -0.1343673 ],\n",
      "       [-0.09573697, -0.16049562],\n",
      "       [-0.13410586, -0.21299812],\n",
      "       ...,\n",
      "       [-0.20023137, -0.09606233],\n",
      "       [-0.17969735, -0.10641741],\n",
      "       [-0.13371956, -0.1141579 ]], dtype=float32), array([[-0.11746358, -0.10008228],\n",
      "       [-0.06420743, -0.13194385],\n",
      "       [-0.11966159, -0.11977117],\n",
      "       ...,\n",
      "       [-0.20312116, -0.10053974],\n",
      "       [-0.18715943, -0.14903682],\n",
      "       [-0.08386277, -0.12402858]], dtype=float32), array([[-0.06647216, -0.06313089],\n",
      "       [-0.10397911, -0.09082449],\n",
      "       [-0.04747823, -0.08921322],\n",
      "       ...,\n",
      "       [-0.12624826, -0.1427075 ],\n",
      "       [-0.16382684, -0.10406116],\n",
      "       [-0.1441442 , -0.09668346]], dtype=float32)]\n"
     ]
    }
   ],
   "source": [
    "print(projs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "4311a71c",
   "metadata": {},
   "outputs": [],
   "source": [
    "projs = np.asarray(projs)\n",
    "tallHposes = np.asarray(tallHposes)\n",
    "\n",
    "\n",
    "np.save(\"biprojs.npy\", projs)\n",
    "np.save(\"tallHposes.npy\", tallHposes)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "id": "65a5561f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6804\n",
      "6804\n",
      "20\n",
      "1\n",
      "done\n"
     ]
    }
   ],
   "source": [
    "from scipy.spatial.distance import cdist\n",
    "CHallHposes = []\n",
    "\n",
    "CHprojs = []\n",
    "for k in range(1,2):\n",
    "    tHposes = []\n",
    "    boxes = allbox[k]\n",
    "    fbox = freud.box.Box.from_box(boxes[k])    \n",
    "    Oposes = allOpos[k]\n",
    "    Hposes = allHpos[k]\n",
    "    poses = allpos[k]\n",
    "    proj = []\n",
    "    Hposes = np.reshape(Hposes, [len(Hposes), int(len(Hposes[0])/2), 2, 3])\n",
    "\n",
    "    dismatrix = np.copy(Hposes)\n",
    "    omatrix = np.copy(Hposes)\n",
    "    print(len(Oposes[0]))\n",
    "    print(len(Hposes[0]))\n",
    "    for i in range(2):\n",
    "\n",
    "        for j in range(len(dismatrix)):\n",
    "            fbox = freud.box.Box.from_box(boxes[j])\n",
    "\n",
    "            dismatrix[j,:,i,:] = fbox.wrap(Hposes[j,:,i,:] - Oposes[j])\n",
    "            omatrix[j,:,i,:] = (dismatrix[j,:,i,:].T / \n",
    "                                np.linalg.norm(dismatrix[j,:,i,:], axis=1)).T\n",
    "\n",
    "    \n",
    "    for i in range(20, len(Hposes), 1000):\n",
    "        sOpos = Oposes[i]\n",
    "        sHpos = np.concatenate(Hposes[i])\n",
    "\n",
    "        pos = poses[i]\n",
    "        \n",
    "        whO = np.where(pos[:,0] == 4)\n",
    "        whH = np.where(pos[:,0] == 5)\n",
    "\n",
    "        whC1 = np.where((pos[:,0] == 1)*(pos[:,1] == -0.222))\n",
    "        whC2 = np.where((pos[:,0] == 1)*(pos[:,1] == -0.148))\n",
    "\n",
    "        whH1 = np.where(pos[:,0] == 2)\n",
    "        whH2 = np.where(pos[:,0] == 3)\n",
    "        \n",
    "        C1pos = pos[whC1][:,2:5]\n",
    "        C2pos = pos[whC2][:,2:5]\n",
    "\n",
    "        H1pos = pos[whH1][:,2:5]\n",
    "        H2pos = pos[whH2][:,2:5]\n",
    "        Opos = pos[whO][:,2:5]\n",
    "        Hpos = pos[whH][:,2:5]\n",
    "\n",
    "        \n",
    "        fbox = freud.box.Box.from_box(boxes[i])\n",
    "        fields = []\n",
    "        omat = np.concatenate(omatrix[i])\n",
    "        \n",
    "        Hdel_list = (np.arange(0, len(Hpos))/2).astype('int')\n",
    "        Odel_list = (np.arange(0, len(Opos))).astype('int')\n",
    "#         aCfield = []\n",
    "#         aOfield = []\n",
    "#         aH1field = []\n",
    "#         aH2field = []\n",
    "#         aHfield = []\n",
    "\n",
    "\n",
    "        for j in range(len(Hpos)):\n",
    "            \n",
    "            #print(np.where(Hdel_list == int(j/2)))\n",
    "            pos = Hpos[j]\n",
    "            tHpos = Hpos[Hdel_list != int(j/2)]\n",
    "            tOpos = Opos[Odel_list != int(j/2)]\n",
    "            #Opos = Oposes[j]\n",
    "            dpos = fbox.wrap(tHpos - pos)\n",
    "            dis = np.linalg.norm(dpos, axis=1)\n",
    "            dpos = dpos[dis > 1e-2]\n",
    "            dis = dis[dis > 1e-2]\n",
    "            direct = (dpos.T / dis).T\n",
    "            Hfields = (direct.T / dis**2).T\n",
    "            \n",
    "            dpos = fbox.wrap(tOpos - pos)\n",
    "            dis = np.linalg.norm(dpos, axis=1)\n",
    "            dpos = dpos[dis > 1e-2]\n",
    "            dis = dis[dis >1e-2]\n",
    "            direct = (dpos.T / dis).T\n",
    "            Ofields = (direct.T / dis**2).T\n",
    "            \n",
    "            dpos = fbox.wrap(C1pos - pos)\n",
    "            dis = np.linalg.norm(dpos, axis=1)\n",
    "            dpos = dpos[dis > 1e-2]\n",
    "            dis = dis[dis >1e-2]\n",
    "            direct = (dpos.T / dis).T\n",
    "            C1fields = (direct.T / dis**2).T\n",
    "                        \n",
    "            dpos = fbox.wrap(C2pos - pos)\n",
    "            dis = np.linalg.norm(dpos, axis=1)\n",
    "            dpos = dpos[dis > 1e-2]\n",
    "            dis = dis[dis >1e-2]\n",
    "            direct = (dpos.T / dis).T\n",
    "            C2fields = (direct.T / dis**2).T\n",
    "            \n",
    "            dpos = fbox.wrap(H1pos - pos)\n",
    "            dis = np.linalg.norm(dpos, axis=1)\n",
    "            dpos = dpos[dis > 1e-2]\n",
    "            dis = dis[dis >1e-2]\n",
    "            direct = (dpos.T / dis).T\n",
    "            H1fields = (direct.T / dis**2).T\n",
    "            \n",
    "            dpos = fbox.wrap(H2pos - pos)\n",
    "            dis = np.linalg.norm(dpos, axis=1)\n",
    "            dpos = dpos[dis > 1e-2]\n",
    "            dis = dis[dis >1e-2]\n",
    "            direct = (dpos.T / dis).T\n",
    "            H2fields = (direct.T / dis**2).T\n",
    "            \n",
    "            \n",
    "            field = (0.4238*np.sum(Hfields, axis=0) \n",
    "                     -0.8476*np.sum(Ofields, axis=0)\n",
    "                     -0.222*np.sum(C1fields, axis=0)\n",
    "                     -0.148*np.sum(C2fields, axis=0)                   \n",
    "                     +0.074*np.sum(H1fields, axis=0)\n",
    "                     +0.074*np.sum(H2fields, axis=0))\n",
    "            \n",
    "#             aCfields = np.sum(np.concatenate([C1fields, C2fields]), axis=0)\n",
    "#             aOfields = np.sum(Ofields, axis=0)\n",
    "#             aH1fields = np.sum(H1fields, axis=0)\n",
    "#             aH2fields = np.sum(H2fields, axis=0)\n",
    "#             aHfields = np.sum(Hfields, axis=0)\n",
    "\n",
    "            \n",
    "            fields.append(field)\n",
    "            \n",
    "            \n",
    "#             aCfield.append(aCfields)\n",
    "#             aOfield.append(aOfields)\n",
    "#             aHfield.append(aHfields)\n",
    "#             aH1field.append(aH1fields)\n",
    "#             aH2field.append(aH2fields)\n",
    "\n",
    "        \n",
    "        tHposes.append(sHpos)\n",
    "        \n",
    "            #print(j)\n",
    "            \n",
    "        proj.append(np.sum(omat*fields, axis=1))\n",
    "            \n",
    "        #projs.append(proj)    \n",
    "            \n",
    "            \n",
    "        print(i)\n",
    "            \n",
    "        #print(Hpos)\n",
    "        #distance = cdist(Hpos, Hpos)\n",
    "    print(k)\n",
    "    CHprojs.append(proj)\n",
    "    CHallHposes.append(np.asarray(tHposes))  \n",
    "\n",
    "    #break #es\n",
    "            \n",
    "            \n",
    "#np.save(\"cCHprojs\", np.asarray(CHprojs))\n",
    "#np.save(\"cCHtallHs\", np.asarray(CHallHposes)) \n",
    "    \n",
    "print('done')         "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "7d96622f",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'CHprojs' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn [99], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m un, co \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39munique(np\u001b[38;5;241m.\u001b[39mround(CHprojs[:], \u001b[38;5;241m2\u001b[39m), return_counts\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[1;32m      2\u001b[0m plt\u001b[38;5;241m.\u001b[39mplot(un\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m14.4\u001b[39m, co)\n\u001b[1;32m      3\u001b[0m \u001b[38;5;66;03m#plt.xlim(-1, 1)\u001b[39;00m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'CHprojs' is not defined"
     ]
    }
   ],
   "source": [
    "un, co = np.unique(np.round(CHprojs[:], 2), return_counts=True)\n",
    "plt.plot(un*14.4, co)\n",
    "#plt.xlim(-1, 1)\n",
    "print(projs)\n",
    "print(len(Hposes))\n",
    "print(len(projs[0][0]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "341710bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10368\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "feb9dd52",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}