File size: 77,093 Bytes
f2d6c5c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"import sys\n",
"import time\n",
"import subprocess\n",
"import logging\n",
"import warnings\n",
"import gc\n",
"import numpy as np\n",
"import pandas as pd\n",
"import seaborn as sns\n",
"import matplotlib.pyplot as plt\n",
"import matplotlib.patches as mpatches\n",
"from concurrent.futures import ProcessPoolExecutor, as_completed"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"from rdkit import Chem\n",
"from rdkit.Chem import AllChem, DataStructs, Draw\n",
"from rdkit import RDConfig\n",
"from rdkit.Chem import Descriptors, rdMolDescriptors, Lipinski, rdDistGeom, rdPartialCharges\n",
"from rdkit.Chem.AllChem import GetMorganGenerator\n",
"from rdkit.DataStructs.cDataStructs import ConvertToNumpyArray\n",
"from rdkit.Avalon.pyAvalonTools import GetAvalonFP"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"import tensorflow as tf\n",
"from tensorflow import keras\n",
"from tensorflow.keras import layers\n",
"from tensorflow.keras.models import Sequential\n",
"from tensorflow.keras.layers import Dense, Dropout, Activation\n",
"from tensorflow.keras.regularizers import l2\n",
"from tensorflow.keras.optimizers import Adam\n",
"from tensorflow.keras import regularizers"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.model_selection import train_test_split\n",
"from sklearn.linear_model import Ridge\n",
"from sklearn.ensemble import RandomForestRegressor\n",
"from sklearn.neural_network import MLPRegressor\n",
"from sklearn.svm import SVR\n",
"from sklearn.metrics import r2_score, mean_absolute_error, mean_squared_error, root_mean_squared_error"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"import optuna\n",
"from optuna.trial import TrialState"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [],
"source": [
"tf.keras.backend.clear_session()\n",
"gpus = tf.config.experimental.list_physical_devices('GPU')\n",
"if gpus:\n",
" try:\n",
" for gpu in gpus:\n",
" tf.config.experimental.set_memory_growth(gpu, True)\n",
" except RuntimeError as e:\n",
" print(e)"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [],
"source": [
"target_path = \"result/5_ANO_structure\"\n",
"os.makedirs(target_path, exist_ok=True)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [],
"source": [
"data_ws = pd.read_csv('./data/ws496_logS.csv', dtype={'SMILES': 'string'})\n",
"smiles_ws = data_ws['SMILES']\n",
"y_ws = data_ws.iloc[:, 2]\n",
"\n",
"data_delaney = pd.read_csv('./data/delaney-processed.csv', dtype={'smiles': 'string'})\n",
"smiles_de = data_delaney['smiles']\n",
"y_de = data_delaney.iloc[:, 1]\n",
"\n",
"data_lovric2020 = pd.read_csv('./data/Lovric2020_logS0.csv', dtype={'isomeric_smiles': 'string'})\n",
"smiles_lo = data_lovric2020['isomeric_smiles']\n",
"y_lo = data_lovric2020.iloc[:, 1]\n",
"\n",
"data_huuskonen = pd.read_csv('./data/huusk.csv', dtype={'SMILES': 'string'})\n",
"smiles_hu = data_huuskonen['SMILES']\n",
"y_hu = data_huuskonen.iloc[:, -1].astype('float')"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"def mol3d(mol):\n",
" mol = Chem.AddHs(mol)\n",
" optimization_methods = [\n",
" (AllChem.EmbedMolecule, (mol, AllChem.ETKDGv3()), {}),\n",
" (AllChem.UFFOptimizeMolecule, (mol,), {'maxIters': 200}),\n",
" (AllChem.MMFFOptimizeMolecule, (mol,), {'maxIters': 200})\n",
" ]\n",
"\n",
" for method, args, kwargs in optimization_methods:\n",
" try:\n",
" method(*args, **kwargs)\n",
" if mol.GetNumConformers() > 0:\n",
" return mol\n",
" except ValueError as e:\n",
" print(f\"Error: {e} - Trying next optimization method [{method}]\")\n",
"\n",
" print(f\"Invalid mol for 3d {'\\033[94m'}{Chem.MolToSmiles(mol)}{'\\033[0m'} - No conformer generated\")\n",
" return None"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"def convert_smiles_to_mol(smiles, fail_folder=None, index=None, yvalue=None):\n",
" mol = Chem.MolFromSmiles(smiles)\n",
" if mol is None:\n",
" print(f\"[convert_smiles_to_mol] Cannot convert {smiles} to Mols\")\n",
" return None, {\"smiles\": smiles, \"y_value\": yvalue, \"error\": \"Invalid SMILES\"}\n",
"\n",
" try:\n",
" Chem.Kekulize(mol, clearAromaticFlags=True)\n",
" isomeric_smiles = Chem.MolToSmiles(mol, isomericSmiles=True)\n",
" mol = Chem.MolFromSmiles(isomeric_smiles)\n",
" except Exception as e:\n",
" print(f\"[convert_smiles_to_mol] failed {smiles} isomeric_smiles by {e}\")\n",
" if fail_folder and index is not None:\n",
" img_path = os.path.join(fail_folder, f\"mol_{index}.png\")\n",
" img = Draw.MolToImage(mol)\n",
" img.save(img_path)\n",
" return None, {\"smiles\": smiles, \"y_value\": yvalue, \"error\": f\"Isomeric SMILES error: {e}\"}\n",
"\n",
" try:\n",
" Chem.SanitizeMol(mol)\n",
" except Exception as e:\n",
" print(f\"[convert_smiles_to_mol] failed {smiles} SanitizeMol by {e}\")\n",
" if fail_folder and index is not None:\n",
" img_path = os.path.join(fail_folder, f\"mol_{index}.png\")\n",
" img = Draw.MolToImage(mol)\n",
" img.save(img_path)\n",
" return None, {\"smiles\": smiles, \"y_value\": yvalue, \"error\": f\"SanitizeMol error: {e}\"}\n",
"\n",
" return mol, None"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"def process_smiles(smiles, yvalue, fail_folder, index):\n",
" mol, error = convert_smiles_to_mol(smiles, fail_folder, index, yvalue)\n",
" if error:\n",
" return None, None, error\n",
"\n",
" mol_3d = mol3d(mol)\n",
" if mol_3d:\n",
" return smiles, yvalue, None\n",
" else:\n",
" img_path = os.path.join(fail_folder, f\"mol_{index}.png\")\n",
" img = Draw.MolToImage(mol)\n",
" img.save(img_path)\n",
" return None, None, {\"smiles\": smiles, \"y_value\": yvalue}\n",
"\n",
"def process_dataset(smiles_list, y_values, dataset_name, target_path=\"result\", max_workers=None):\n",
" start = time.time()\n",
" valid_smiles, valid_y = [], []\n",
" error_smiles_list = []\n",
" fail_folder = f\"{target_path}/failed/{dataset_name}\"\n",
" os.makedirs(fail_folder, exist_ok=True)\n",
"\n",
" with ProcessPoolExecutor(max_workers=max_workers) as executor:\n",
" futures = [\n",
" executor.submit(process_smiles, smiles, yvalue, fail_folder, i)\n",
" for i, (smiles, yvalue) in enumerate(zip(smiles_list, y_values))\n",
" ]\n",
" for future in as_completed(futures):\n",
" smiles, yvalue, error = future.result()\n",
" if error:\n",
" error_smiles_list.append(error)\n",
" elif smiles is not None and yvalue is not None:\n",
" valid_smiles.append(smiles)\n",
" valid_y.append(yvalue)\n",
"\n",
" if error_smiles_list:\n",
" error_df = pd.DataFrame(error_smiles_list)\n",
" error_df.to_csv(os.path.join(fail_folder, \"failed_smiles.csv\"), index=False)\n",
" print(f\" [{dataset_name:<10}] : {time.time()-start:.4f} sec\")\n",
" return valid_smiles, valid_y"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" [ws496 ] : 0.8649 sec\n",
" [delaney ] : 1.3527 sec\n",
"Error: Bad Conformer Id - Trying next optimization method [<Boost.Python.function object at 0x560fe30ad760>]\n",
"Error: Bad Conformer Id - Trying next optimization method [<Boost.Python.function object at 0x560fe30ae5f0>]\n",
"Invalid mol for 3d \u001b[94m[H]O[C@@]([H])(c1c([H])c([H])nc2c([H])c([H])c(OC([H])([H])[H])c([H])c12)[C@]1([H])[N@]2C([H])([H])C([H])([H])[C@@]([H])(C1([H])[H])[C@@]([H])(C([H])=C([H])[H])C2([H])[H]\u001b[0m - No conformer generated\n",
"Error: Bad Conformer Id - Trying next optimization method [<Boost.Python.function object at 0x560fe30ad760>]\n",
"Error: Bad Conformer Id - Trying next optimization method [<Boost.Python.function object at 0x560fe30ae5f0>]\n",
"Invalid mol for 3d \u001b[94m[H]O[C@]([H])(c1c([H])c([H])nc2c([H])c([H])c(OC([H])([H])[H])c([H])c12)[C@@]1([H])[N@]2C([H])([H])C([H])([H])[C@@]([H])(C1([H])[H])[C@@]([H])(C([H])=C([H])[H])C2([H])[H]\u001b[0m - No conformer generated\n",
" [Lovric2020_logS0] : 8.3057 sec\n",
" [huusk ] : 1.5089 sec\n"
]
}
],
"source": [
"smiles_ws, y_ws = process_dataset(smiles_ws, y_ws, \"ws496\", target_path)\n",
"smiles_de, y_de = process_dataset(smiles_de, y_de, \"delaney\", target_path)\n",
"smiles_lo, y_lo = process_dataset(smiles_lo, y_lo, \"Lovric2020_logS0\", target_path)\n",
"smiles_hu, y_hu = process_dataset(smiles_hu, y_hu, \"huusk\", target_path)"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [],
"source": [
"LEN_OF_FF = 2048\n",
"LEN_OF_MA = 167\n",
"LEN_OF_AV = 512"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [],
"source": [
"def get_fingerprints(mol):\n",
" if mol is None:\n",
" return None, None, None\n",
" \n",
" morgan_generator = GetMorganGenerator(radius=2, fpSize=LEN_OF_FF)\n",
" ecfp = morgan_generator.GetFingerprint(mol)\n",
" ecfp_array = np.zeros((LEN_OF_FF,),dtype=int)\n",
" DataStructs.ConvertToNumpyArray(ecfp, ecfp_array)\n",
" \n",
" maccs = Chem.rdMolDescriptors.GetMACCSKeysFingerprint(mol)\n",
"\n",
" avalon_fp = GetAvalonFP(mol)\n",
" avalon_array = np.zeros((LEN_OF_AV,),dtype=int)\n",
" DataStructs.ConvertToNumpyArray(avalon_fp, avalon_array)\n",
" \n",
" return ecfp_array, maccs, avalon_array\n",
"\n",
"def fp_converter(data, use_parallel=True):\n",
" mols = [Chem.MolFromSmiles(smi) for smi in data]\n",
" \n",
" if use_parallel:\n",
" try: \n",
" with ProcessPoolExecutor() as executor:\n",
" results = list(executor.map(get_fingerprints, mols))\n",
" except Exception as e:\n",
" print(f\"Parallel processing failed due to: {e}. Falling back to sequential processing.\")\n",
" use_parallel = False\n",
" \n",
" if not use_parallel:\n",
" results = [get_fingerprints(mol) for mol in mols]\n",
" \n",
" ECFP, MACCS, AvalonFP = zip(*results)\n",
" \n",
" ECFP_container = np.vstack([arr for arr in ECFP if arr is not None])\n",
" MACCS_container = np.zeros((len(MACCS), LEN_OF_MA), dtype=int)\n",
" AvalonFP_container = np.vstack([arr for arr in AvalonFP if arr is not None])\n",
"\n",
" for i, fp in enumerate(MACCS):\n",
" if fp is not None:\n",
" DataStructs.ConvertToNumpyArray(fp, MACCS_container[i])\n",
" \n",
" return mols, ECFP_container, MACCS_container, AvalonFP_container"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"mol_ws, x_ws, MACCS_ws, AvalonFP_ws = fp_converter(smiles_ws,target_path)\n",
"mol_de, x_de, MACCS_de, AvalonFP_de = fp_converter(smiles_de,target_path)\n",
"mol_lo, x_lo, MACCS_lo, AvalonFP_lo = fp_converter(smiles_lo,target_path)\n",
"mol_hu, x_hu, MACCS_hu, AvalonFP_hu = fp_converter(smiles_hu,target_path)"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [],
"source": [
"def concatenate_to_numpy(*dataframes):\n",
" numpy_arrays = [df.to_numpy() if isinstance(df, pd.DataFrame) else df for df in dataframes]\n",
" if not all(isinstance(arr, np.ndarray) for arr in numpy_arrays):\n",
" raise ValueError(\"All inputs must be either pandas DataFrame or numpy array\")\n",
" return np.concatenate(numpy_arrays, axis=1)"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [],
"source": [
"group_nws = concatenate_to_numpy(x_ws, MACCS_ws, AvalonFP_ws)\n",
"group_nde = concatenate_to_numpy(x_de, MACCS_de, AvalonFP_de)\n",
"group_nlo = concatenate_to_numpy(x_lo, MACCS_lo, AvalonFP_lo)\n",
"group_nhu = concatenate_to_numpy(x_hu, MACCS_hu, AvalonFP_hu)"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [],
"source": [
"BATCHSIZE = 32\n",
"EPOCHS = 1000\n",
"lr = 0.0001\n",
"decay = 1e-4"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {},
"outputs": [],
"source": [
"def search_model(trial, input_dim):\n",
" n_layers = trial.suggest_int(\"n_layers\", 1, 3)\n",
" model = tf.keras.Sequential()\n",
" model.add(tf.keras.layers.Input(shape=(input_dim,)))\n",
" layer_dropout = trial.suggest_int(\"layer_dropout\", 0, 1)\n",
" \n",
" for i in range(n_layers):\n",
" num_hidden = trial.suggest_int(f\"n_units_l_{i}\", 2, 10000)\n",
" num_decay = trial.suggest_categorical(f\"n_decay_l_{i}\", [1e-3, 1e-4, 1e-5])\n",
" model.add(\n",
" tf.keras.layers.Dense(\n",
" num_hidden,\n",
" activation=\"relu\",\n",
" kernel_initializer='glorot_uniform',\n",
" kernel_regularizer=tf.keras.regularizers.l2(num_decay),\n",
" )\n",
" )\n",
" if layer_dropout == 1:\n",
" fdropout1 = trial.suggest_categorical(f\"F_dropout_{i}\", [0.1, 0.2, 0.3])\n",
" model.add(tf.keras.layers.Dropout(rate=fdropout1))\n",
" \n",
" if layer_dropout == 0:\n",
" fdropout2 = trial.suggest_categorical(\"last_dropout\", [0.1, 0.2, 0.3])\n",
" model.add(tf.keras.layers.Dropout(rate=fdropout2))\n",
" \n",
" model.add(tf.keras.layers.Dense(units=1))\n",
" # # Colab\n",
" # learningr = trial.suggest_categorical(\"Learning_rate\",[0.01,0.001,0.0001])\n",
" # model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=learningr),\n",
" # loss=tf.keras.losses.MeanSquaredError(),\n",
" # metrics=[tf.keras.losses.MeanSquaredError(),\n",
" # tf.keras.losses.MeanAbsoluteError(),\n",
" # tf.keras.metrics.RootMeanSquaredError()])\n",
" return model\n",
"\n",
"def save_model(trial, x_data):\n",
" model_path = \"save_model/full_model.keras\"\n",
" if not os.path.exists(model_path):\n",
" try:\n",
" model = search_model(trial, x_data.shape[1])\n",
" os.makedirs(\"save_model\", exist_ok=True)\n",
" model.save(model_path)\n",
" print(f\"Model successfully saved to {model_path}\")\n",
" except Exception as e:\n",
" print(f\"Error saving model: {e}\")\n",
" else:\n",
" print(f\"Model already exists at {model_path}\")\n",
" os.remove(model_path)\n",
" save_model(trial, x_data)"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {},
"outputs": [],
"source": [
"import logging\n",
"import warnings\n",
"\n",
"os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'\n",
"os.environ['CUDA_VISIBLE_DEVICES'] = '0'\n",
"os.environ['TF_GPU_ALLOCATOR'] = 'cuda_malloc_async'\n",
"os.environ['CUDA_DEVICE_ORDER'] = 'PCI_BUS_ID'\n",
"os.environ['TF_XLA_FLAGS'] = '--tf_xla_auto_jit=2 --tf_xla_enable_xla_devices'\n",
"os.environ['XLA_FLAGS'] = '--xla_gpu_cuda_data_dir=/usr/local/cuda --xla_gpu_force_compilation_parallelism=1'\n",
"os.environ['TF_FORCE_GPU_ALLOW_GROWTH'] = 'true'\n",
"os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'\n",
"os.environ['TF_NUMA_NODES'] = '1'\n",
"\n",
"warnings.filterwarnings('ignore')\n",
"\n",
"warnings.simplefilter(action='ignore', category=FutureWarning)\n",
"\n",
"logging.getLogger('tensorflow').setLevel(logging.ERROR)\n",
"\n",
"tf.get_logger().setLevel('ERROR')\n",
"tf.autograph.set_verbosity(0)\n",
"\n",
"def suppress_warnings(condition=True):\n",
" if condition:\n",
" logging.getLogger('tensorflow').setLevel(logging.ERROR)\n",
" os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'\n",
" else:\n",
" logging.getLogger('tensorflow').setLevel(logging.WARNING)\n",
" os.environ['TF_CPP_MIN_LOG_LEVEL'] = '0'\n",
"\n",
"suppress_warnings(condition=True)"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [],
"source": [
"def objective_ws_struct(trial):\n",
" try:\n",
" y_true = np.asarray(y_ws).astype('float')\n",
" np.save('new_fps.npy', group_nws)\n",
" np.save('y_true.npy', y_true)\n",
" \n",
" save_model(trial, group_nws)\n",
"\n",
" lr = trial.suggest_categorical(f\"lr\", [1e-3, 1e-4, 1e-5])\n",
"\n",
" result = subprocess.run(['python3', './extra_code/learning_process.py', \n",
" str(BATCHSIZE), str(EPOCHS), \n",
" str(lr), \n",
" 'new_fps.npy', 'y_true.npy'],\n",
" stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)\n",
"\n",
" if result.stderr:\n",
" filtered_stderr = '\\n'.join([line for line in result.stderr.split('\\n') if \"could not open file to read NUMA node\" not in line and \"Your kernel may have been built without NUMA support\" not in line])\n",
" if filtered_stderr:\n",
" print(f\"Error in subprocess: {filtered_stderr}\", file=sys.stderr)\n",
"\n",
" for line in result.stdout.splitlines():\n",
" if \"R2\" in line:\n",
" if \"(prune)\" in line:\n",
" print(f\"Pruning trial due to poor R2: {line}\")\n",
" r2_result = 0.0\n",
" trial.report(r2_result, step=0)\n",
" raise optuna.exceptions.TrialPruned()\n",
" else:\n",
" r2_result = float(line.split(\":\")[1].strip())\n",
" print(f\"R2 score: {r2_result}\")\n",
" trial.report(r2_result, step=0)\n",
"\n",
" if trial.should_prune():\n",
" raise optuna.exceptions.TrialPruned()\n",
"\n",
" except Exception as e:\n",
" print(f\"Exception occurred: {e}\", file=sys.stderr)\n",
" r2_result = 0.0\n",
"\n",
" gc.collect()\n",
"\n",
" return r2_result"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [],
"source": [
"def objective_de_struct(trial):\n",
" try:\n",
" y_true = np.asarray(y_de).astype('float')\n",
" np.save('new_fps.npy', group_nde)\n",
" np.save('y_true.npy', y_true)\n",
" \n",
" save_model(trial, group_nde)\n",
"\n",
" lr = trial.suggest_categorical(f\"lr\", [1e-3, 1e-4, 1e-5])\n",
"\n",
" result = subprocess.run(['python3', './extra_code/learning_process.py', \n",
" str(BATCHSIZE), str(EPOCHS), \n",
" str(lr), \n",
" 'new_fps.npy', 'y_true.npy'],\n",
" stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)\n",
"\n",
" if result.stderr:\n",
" filtered_stderr = '\\n'.join([line for line in result.stderr.split('\\n') if \"could not open file to read NUMA node\" not in line and \"Your kernel may have been built without NUMA support\" not in line])\n",
" if filtered_stderr:\n",
" print(f\"Error in subprocess: {filtered_stderr}\", file=sys.stderr)\n",
"\n",
" for line in result.stdout.splitlines():\n",
" if \"R2\" in line:\n",
" if \"(prune)\" in line:\n",
" print(f\"Pruning trial due to poor R2: {line}\")\n",
" r2_result = 0.0\n",
" trial.report(r2_result, step=0)\n",
" raise optuna.exceptions.TrialPruned()\n",
" else:\n",
" r2_result = float(line.split(\":\")[1].strip())\n",
" print(f\"R2 score: {r2_result}\")\n",
" trial.report(r2_result, step=0)\n",
"\n",
" if trial.should_prune():\n",
" raise optuna.exceptions.TrialPruned()\n",
"\n",
" except Exception as e:\n",
" print(f\"Exception occurred: {e}\", file=sys.stderr)\n",
" r2_result = 0.0\n",
"\n",
" gc.collect()\n",
"\n",
" return r2_result"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [],
"source": [
"def objective_lo_struct(trial):\n",
" try:\n",
" y_true = np.asarray(y_lo).astype('float')\n",
" np.save('new_fps.npy', group_nlo)\n",
" np.save('y_true.npy', y_true)\n",
" \n",
" save_model(trial, group_nlo)\n",
"\n",
" lr = trial.suggest_categorical(f\"lr\", [1e-3, 1e-4, 1e-5])\n",
"\n",
" result = subprocess.run(['python3', './extra_code/learning_process.py', \n",
" str(BATCHSIZE), str(EPOCHS), \n",
" str(lr), \n",
" 'new_fps.npy', 'y_true.npy'],\n",
" stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)\n",
"\n",
" if result.stderr:\n",
" filtered_stderr = '\\n'.join([line for line in result.stderr.split('\\n') if \"could not open file to read NUMA node\" not in line and \"Your kernel may have been built without NUMA support\" not in line])\n",
" if filtered_stderr:\n",
" print(f\"Error in subprocess: {filtered_stderr}\", file=sys.stderr)\n",
"\n",
" for line in result.stdout.splitlines():\n",
" if \"R2\" in line:\n",
" if \"(prune)\" in line:\n",
" print(f\"Pruning trial due to poor R2: {line}\")\n",
" r2_result = 0.0\n",
" trial.report(r2_result, step=0)\n",
" raise optuna.exceptions.TrialPruned()\n",
" else:\n",
" r2_result = float(line.split(\":\")[1].strip())\n",
" print(f\"R2 score: {r2_result}\")\n",
" trial.report(r2_result, step=0)\n",
"\n",
" if trial.should_prune():\n",
" raise optuna.exceptions.TrialPruned()\n",
"\n",
" except Exception as e:\n",
" print(f\"Exception occurred: {e}\", file=sys.stderr)\n",
" r2_result = 0.0\n",
"\n",
" gc.collect()\n",
"\n",
" return r2_result"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {},
"outputs": [],
"source": [
"def objective_hu_struct(trial):\n",
" try:\n",
" y_true = np.asarray(y_hu).astype('float')\n",
" np.save('new_fps.npy', group_nhu)\n",
" np.save('y_true.npy', y_true)\n",
" \n",
" save_model(trial, group_nhu)\n",
"\n",
" lr = trial.suggest_categorical(f\"lr\", [1e-3, 1e-4, 1e-5])\n",
"\n",
" result = subprocess.run(['python3', './extra_code/learning_process.py', \n",
" str(BATCHSIZE), str(EPOCHS), \n",
" str(lr), \n",
" 'new_fps.npy', 'y_true.npy'],\n",
" stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)\n",
"\n",
" if result.stderr:\n",
" filtered_stderr = '\\n'.join([line for line in result.stderr.split('\\n') if \"could not open file to read NUMA node\" not in line and \"Your kernel may have been built without NUMA support\" not in line])\n",
" if filtered_stderr:\n",
" print(f\"Error in subprocess: {filtered_stderr}\", file=sys.stderr)\n",
"\n",
" for line in result.stdout.splitlines():\n",
" if \"R2\" in line:\n",
" if \"(prune)\" in line:\n",
" print(f\"Pruning trial due to poor R2: {line}\")\n",
" r2_result = 0.0\n",
" trial.report(r2_result, step=0)\n",
" raise optuna.exceptions.TrialPruned()\n",
" else:\n",
" r2_result = float(line.split(\":\")[1].strip())\n",
" print(f\"R2 score: {r2_result}\")\n",
" trial.report(r2_result, step=0)\n",
"\n",
" if trial.should_prune():\n",
" raise optuna.exceptions.TrialPruned()\n",
"\n",
" except Exception as e:\n",
" print(f\"Exception occurred: {e}\", file=sys.stderr)\n",
" r2_result = 0.0\n",
"\n",
" gc.collect()\n",
"\n",
" return r2_result"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {},
"outputs": [],
"source": [
"storage = optuna.storages.RDBStorage(url=\"sqlite:///ano_analysis.db\", engine_kwargs={\"connect_args\": {\"timeout\": 10000}})\n",
"# storage_urls = \"postgresql+psycopg2://postgres:{pwd}}@localhost:{num}}\"\n",
"# storage = optuna.storages.RDBStorage(url=storage_urls)"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [],
"source": [
"try:\n",
" optuna.delete_study(study_name=\"ANO_ws_struct\", storage=storage)\n",
" optuna.delete_study(study_name=\"ANO_de_struct\", storage=storage)\n",
" optuna.delete_study(study_name=\"ANO_lo_struct\", storage=storage)\n",
" optuna.delete_study(study_name=\"ANO_hu_struct\", storage=storage)\n",
"except:\n",
" pass"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [],
"source": [
"TRIALS = 5"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:27:16,193] A new study created in RDB with name: ANO_ws_struct\n",
"I0000 00:00:1729823236.262387 712386 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node\n",
"Your kernel may have been built without NUMA support.\n",
"I0000 00:00:1729823236.262494 712386 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node\n",
"Your kernel may have been built without NUMA support.\n",
"I0000 00:00:1729823236.262547 712386 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node\n",
"Your kernel may have been built without NUMA support.\n",
"I0000 00:00:1729823236.414390 712386 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node\n",
"Your kernel may have been built without NUMA support.\n",
"I0000 00:00:1729823236.414547 712386 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node\n",
"Your kernel may have been built without NUMA support.\n",
"2024-10-25 11:27:16.414564: I tensorflow/core/common_runtime/gpu/gpu_device.cc:2112] Could not identify NUMA node of platform GPU id 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
"2024-10-25 11:27:16.414596: I tensorflow/core/common_runtime/gpu/gpu_process_state.cc:198] Using CUDA malloc Async allocator for GPU: 0\n",
"I0000 00:00:1729823236.414877 712386 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node\n",
"Your kernel may have been built without NUMA support.\n",
"2024-10-25 11:27:16.414914: I tensorflow/core/common_runtime/gpu/gpu_device.cc:2021] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 3586 MB memory: -> device: 0, name: NVIDIA GeForce RTX 3060 Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 8.6\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823240.072256 713235 service.cc:146] XLA service 0x558b4abf8ec0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823240.072318 713235 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823240.199007 713235 service.cc:146] XLA service 0x558b4b330cb0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823240.199045 713235 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823242.903035 713344 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.72685\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:27:28,379] Trial 0 finished with value: 0.72685 and parameters: {'n_layers': 1, 'layer_dropout': 0, 'n_units_l_0': 9922, 'n_decay_l_0': 1e-05, 'last_dropout': 0.1, 'lr': 0.001}. Best is trial 0 with value: 0.72685.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823251.215260 714082 service.cc:146] XLA service 0x55a8ec6b6500 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823251.215311 714082 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823251.333433 714082 service.cc:146] XLA service 0x55a8ec5cd290 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823251.333463 714082 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823263.351498 714193 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.707063\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:28:11,955] Trial 1 finished with value: 0.707063 and parameters: {'n_layers': 2, 'layer_dropout': 0, 'n_units_l_0': 6572, 'n_decay_l_0': 1e-05, 'n_units_l_1': 1332, 'n_decay_l_1': 0.0001, 'last_dropout': 0.3, 'lr': 1e-05}. Best is trial 0 with value: 0.72685.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823294.853158 716419 service.cc:146] XLA service 0x55b5e46a62f0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823294.853225 716419 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823295.002577 716419 service.cc:146] XLA service 0x55b5e46e5fb0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823295.002610 716419 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823297.511032 716525 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.705862\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:28:23,131] Trial 2 finished with value: 0.705862 and parameters: {'n_layers': 1, 'layer_dropout': 1, 'n_units_l_0': 3241, 'n_decay_l_0': 1e-05, 'F_dropout_0': 0.2, 'lr': 0.0001}. Best is trial 0 with value: 0.72685.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-10-25 11:28:23.378722: W external/local_tsl/tsl/framework/cpu_allocator_impl.cc:83] Allocation of 286627920 exceeds 10% of free system memory.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823306.834696 718042 service.cc:146] XLA service 0x55ff6bc60e10 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823306.834744 718042 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823306.978987 718042 service.cc:146] XLA service 0x55ff6bc7c930 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823306.979019 718042 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823333.708484 718154 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.741337\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:30:53,672] Trial 3 finished with value: 0.741337 and parameters: {'n_layers': 3, 'layer_dropout': 0, 'n_units_l_0': 787, 'n_decay_l_0': 0.0001, 'n_units_l_1': 9082, 'n_decay_l_1': 0.001, 'n_units_l_2': 7890, 'n_decay_l_2': 0.001, 'last_dropout': 0.1, 'lr': 0.0001}. Best is trial 3 with value: 0.741337.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823456.995677 725244 service.cc:146] XLA service 0x56214db0e060 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823456.995725 725244 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823457.136743 725244 service.cc:146] XLA service 0x56214da498b0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823457.136787 725244 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823459.392929 725349 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.68373\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:31:13,484] Trial 4 finished with value: 0.68373 and parameters: {'n_layers': 1, 'layer_dropout': 0, 'n_units_l_0': 1253, 'n_decay_l_0': 0.0001, 'last_dropout': 0.2, 'lr': 1e-05}. Best is trial 3 with value: 0.741337.\n"
]
}
],
"source": [
"study_ws_struct = optuna.create_study(study_name='ANO_ws_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.SuccessiveHalvingPruner(reduction_factor=64, min_early_stopping_rate=10),load_if_exists=True) \n",
"# study_ws_fea = optuna.create_study(study_name='ANO_ws_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.HyperbandPruner(min_resource=100,max_resource=1000,reduction_factor=3), load_if_exists=True)\n",
"study_ws_struct.optimize(objective_ws_struct, n_trials=TRIALS)\n",
"pruned_trials_ws_struct = study_ws_struct.get_trials(deepcopy=False, states=[TrialState.PRUNED])\n",
"complete_trials_ws_struct = study_ws_struct.get_trials(deepcopy=False, states=[TrialState.COMPLETE])"
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:31:13,504] A new study created in RDB with name: ANO_de_struct\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823476.380438 735317 service.cc:146] XLA service 0x564b5beee4b0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823476.380497 735317 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823476.549364 735317 service.cc:146] XLA service 0x564b5be2ad00 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823476.549448 735317 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823490.704246 735426 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.803869\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:32:02,303] Trial 0 finished with value: 0.803869 and parameters: {'n_layers': 2, 'layer_dropout': 0, 'n_units_l_0': 1737, 'n_decay_l_0': 1e-05, 'n_units_l_1': 6702, 'n_decay_l_1': 1e-05, 'last_dropout': 0.2, 'lr': 0.001}. Best is trial 0 with value: 0.803869.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823525.878834 736385 service.cc:146] XLA service 0x55bad3cb21e0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823525.878873 736385 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823526.015032 736385 service.cc:146] XLA service 0x55bad3c304f0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823526.015066 736385 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823529.879054 736488 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.826782\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:32:39,308] Trial 1 finished with value: 0.826782 and parameters: {'n_layers': 2, 'layer_dropout': 1, 'n_units_l_0': 9935, 'n_decay_l_0': 1e-05, 'F_dropout_0': 0.2, 'n_units_l_1': 3544, 'n_decay_l_1': 1e-05, 'F_dropout_1': 0.3, 'lr': 0.0001}. Best is trial 1 with value: 0.826782.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823562.604376 737355 service.cc:146] XLA service 0x55eaf2377f20 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823562.604431 737355 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823562.742863 737355 service.cc:146] XLA service 0x55eaf23d2e30 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823562.742895 737355 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823566.592650 737461 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.823751\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:33:33,879] Trial 2 finished with value: 0.823751 and parameters: {'n_layers': 2, 'layer_dropout': 1, 'n_units_l_0': 7233, 'n_decay_l_0': 1e-05, 'F_dropout_0': 0.2, 'n_units_l_1': 4859, 'n_decay_l_1': 0.0001, 'F_dropout_1': 0.3, 'lr': 0.001}. Best is trial 1 with value: 0.826782.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823616.856357 739058 service.cc:146] XLA service 0x55d8efa212a0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823616.856406 739058 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823616.997938 739058 service.cc:146] XLA service 0x55d8ef979320 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823616.997975 739058 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823621.412038 739168 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.796098\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:34:39,639] Trial 3 finished with value: 0.796098 and parameters: {'n_layers': 3, 'layer_dropout': 1, 'n_units_l_0': 809, 'n_decay_l_0': 1e-05, 'F_dropout_0': 0.3, 'n_units_l_1': 3939, 'n_decay_l_1': 0.0001, 'F_dropout_1': 0.2, 'n_units_l_2': 6198, 'n_decay_l_2': 0.0001, 'F_dropout_2': 0.1, 'lr': 0.001}. Best is trial 1 with value: 0.826782.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823682.926391 740686 service.cc:146] XLA service 0x5579d09bb100 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823682.926465 740686 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823683.068876 740686 service.cc:146] XLA service 0x5579d09fa810 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823683.068912 740686 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823696.228323 740798 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.84961\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:37:05,301] Trial 4 finished with value: 0.84961 and parameters: {'n_layers': 2, 'layer_dropout': 0, 'n_units_l_0': 7109, 'n_decay_l_0': 0.001, 'n_units_l_1': 3436, 'n_decay_l_1': 1e-05, 'last_dropout': 0.3, 'lr': 1e-05}. Best is trial 4 with value: 0.84961.\n"
]
}
],
"source": [
"study_de_struct = optuna.create_study(study_name='ANO_de_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.SuccessiveHalvingPruner(reduction_factor=64, min_early_stopping_rate=10),load_if_exists=True) \n",
"# study_de_fea = optuna.create_study(study_name='ANO_de_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.HyperbandPruner(min_resource=100,max_resource=1000,reduction_factor=3), load_if_exists=True)\n",
"study_de_struct.optimize(objective_de_struct, n_trials=TRIALS)\n",
"pruned_trials_de_struct = study_de_struct.get_trials(deepcopy=False, states=[TrialState.PRUNED])\n",
"complete_trials_de_struct = study_de_struct.get_trials(deepcopy=False, states=[TrialState.COMPLETE])"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:37:05,323] A new study created in RDB with name: ANO_lo_struct\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823828.951072 753530 service.cc:146] XLA service 0x560175567120 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823828.951145 753530 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823829.108219 753530 service.cc:146] XLA service 0x56017553de50 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823829.108251 753530 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823833.863752 753634 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.679332\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:39:42,324] Trial 0 finished with value: 0.679332 and parameters: {'n_layers': 2, 'layer_dropout': 1, 'n_units_l_0': 7114, 'n_decay_l_0': 0.001, 'F_dropout_0': 0.1, 'n_units_l_1': 7475, 'n_decay_l_1': 0.0001, 'F_dropout_1': 0.3, 'lr': 1e-05}. Best is trial 0 with value: 0.679332.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729823985.424040 762359 service.cc:146] XLA service 0x55d75759be40 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823985.424082 762359 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729823985.555550 762359 service.cc:146] XLA service 0x55d75744c2d0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729823985.555588 762359 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729823989.791130 762463 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.668488\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:40:50,103] Trial 1 finished with value: 0.668488 and parameters: {'n_layers': 3, 'layer_dropout': 1, 'n_units_l_0': 2152, 'n_decay_l_0': 0.001, 'F_dropout_0': 0.1, 'n_units_l_1': 1830, 'n_decay_l_1': 0.0001, 'F_dropout_1': 0.1, 'n_units_l_2': 4427, 'n_decay_l_2': 0.0001, 'F_dropout_2': 0.3, 'lr': 0.0001}. Best is trial 0 with value: 0.679332.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824052.960321 765604 service.cc:146] XLA service 0x55cc5280bdf0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824052.960390 765604 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824053.116021 765604 service.cc:146] XLA service 0x55cc50253a30 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824053.116054 765604 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824055.695706 765714 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.662751\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:41:01,389] Trial 2 finished with value: 0.662751 and parameters: {'n_layers': 1, 'layer_dropout': 0, 'n_units_l_0': 2892, 'n_decay_l_0': 0.001, 'last_dropout': 0.1, 'lr': 0.001}. Best is trial 0 with value: 0.679332.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824064.281415 766911 service.cc:146] XLA service 0x55b827b832f0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824064.281454 766911 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824064.424930 766911 service.cc:146] XLA service 0x55b827b5a3c0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824064.424972 766911 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824067.087899 767014 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.644237\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:41:12,311] Trial 3 finished with value: 0.644237 and parameters: {'n_layers': 1, 'layer_dropout': 0, 'n_units_l_0': 6028, 'n_decay_l_0': 0.0001, 'last_dropout': 0.2, 'lr': 0.001}. Best is trial 0 with value: 0.679332.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824075.529341 767599 service.cc:146] XLA service 0x563c07a27f10 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824075.529392 767599 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824075.650832 767599 service.cc:146] XLA service 0x563c07a82e20 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824075.650868 767599 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824078.421404 767708 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.619821\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:41:23,815] Trial 4 finished with value: 0.619821 and parameters: {'n_layers': 1, 'layer_dropout': 1, 'n_units_l_0': 8402, 'n_decay_l_0': 1e-05, 'F_dropout_0': 0.2, 'lr': 0.001}. Best is trial 0 with value: 0.679332.\n"
]
}
],
"source": [
"study_lo_struct = optuna.create_study(study_name='ANO_lo_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.SuccessiveHalvingPruner(reduction_factor=64, min_early_stopping_rate=10),load_if_exists=True) \n",
"# study_lo_fea = optuna.create_study(study_name='ANO_lo_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.HyperbandPruner(min_resource=100,max_resource=1000,reduction_factor=3), load_if_exists=True)\n",
"study_lo_struct.optimize(objective_lo_struct, n_trials=TRIALS)\n",
"pruned_trials_lo_struct = study_lo_struct.get_trials(deepcopy=False, states=[TrialState.PRUNED])\n",
"complete_trials_lo_struct = study_lo_struct.get_trials(deepcopy=False, states=[TrialState.COMPLETE])"
]
},
{
"cell_type": "code",
"execution_count": 53,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:41:23,834] A new study created in RDB with name: ANO_hu_struct\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824086.884600 768220 service.cc:146] XLA service 0x5626a2e78390 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824086.884652 768220 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824087.003367 768220 service.cc:146] XLA service 0x5626a2e4fce0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824087.003400 768220 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824105.963199 768323 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.856321\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:42:47,320] Trial 0 finished with value: 0.856321 and parameters: {'n_layers': 2, 'layer_dropout': 0, 'n_units_l_0': 6594, 'n_decay_l_0': 0.0001, 'n_units_l_1': 301, 'n_decay_l_1': 0.001, 'last_dropout': 0.3, 'lr': 1e-05}. Best is trial 0 with value: 0.856321.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824170.843053 772954 service.cc:146] XLA service 0x558ed359fba0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824170.843094 772954 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824170.980423 772954 service.cc:146] XLA service 0x558ed34d34b0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824170.980455 772954 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824174.981970 773059 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.852474\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:46:37,297] Trial 1 finished with value: 0.852474 and parameters: {'n_layers': 2, 'layer_dropout': 1, 'n_units_l_0': 6712, 'n_decay_l_0': 0.0001, 'F_dropout_0': 0.1, 'n_units_l_1': 6556, 'n_decay_l_1': 0.001, 'F_dropout_1': 0.1, 'lr': 1e-05}. Best is trial 0 with value: 0.856321.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824400.201392 788855 service.cc:146] XLA service 0x55a8a31a00b0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824400.201453 788855 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824400.333349 788855 service.cc:146] XLA service 0x55a8a3176de0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824400.333383 788855 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824402.770049 788964 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.839939\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:46:47,755] Trial 2 finished with value: 0.839939 and parameters: {'n_layers': 1, 'layer_dropout': 0, 'n_units_l_0': 4994, 'n_decay_l_0': 0.0001, 'last_dropout': 0.1, 'lr': 0.001}. Best is trial 0 with value: 0.856321.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824410.699218 789515 service.cc:146] XLA service 0x55ac8eda5d40 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824410.699260 789515 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824410.834295 789515 service.cc:146] XLA service 0x55ac8ece7750 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824410.834333 789515 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824435.249758 789626 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.845373\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:48:32,252] Trial 3 finished with value: 0.845373 and parameters: {'n_layers': 3, 'layer_dropout': 0, 'n_units_l_0': 4463, 'n_decay_l_0': 0.0001, 'n_units_l_1': 1966, 'n_decay_l_1': 0.001, 'n_units_l_2': 924, 'n_decay_l_2': 0.0001, 'last_dropout': 0.1, 'lr': 0.0001}. Best is trial 0 with value: 0.856321.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model already exists at save_model/full_model.keras\n",
"Model successfully saved to save_model/full_model.keras\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Error in subprocess: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1729824515.181650 795905 service.cc:146] XLA service 0x55dbccb5d560 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824515.181691 795905 service.cc:154] StreamExecutor device (0): Host, Default Version\n",
"I0000 00:00:1729824515.312754 795905 service.cc:146] XLA service 0x55dbccb71e90 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1729824515.312792 795905 service.cc:154] StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
"I0000 00:00:1729824518.617949 796015 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"R2 score: 0.812906\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[I 2024-10-25 11:49:05,409] Trial 4 finished with value: 0.812906 and parameters: {'n_layers': 2, 'layer_dropout': 1, 'n_units_l_0': 1477, 'n_decay_l_0': 0.0001, 'F_dropout_0': 0.3, 'n_units_l_1': 5762, 'n_decay_l_1': 1e-05, 'F_dropout_1': 0.2, 'lr': 0.001}. Best is trial 0 with value: 0.856321.\n"
]
}
],
"source": [
"study_hu_struct = optuna.create_study(study_name='ANO_hu_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.SuccessiveHalvingPruner(reduction_factor=64, min_early_stopping_rate=10),load_if_exists=True) \n",
"# study_hu_fea = optuna.create_study(study_name='ANO_hu_struct', storage=storage, direction=\"maximize\", pruner=optuna.pruners.HyperbandPruner(min_resource=100,max_resource=1000,reduction_factor=3), load_if_exists=True)\n",
"study_hu_struct.optimize(objective_hu_struct, n_trials=TRIALS)\n",
"pruned_trials_hu_struct = study_hu_struct.get_trials(deepcopy=False, states=[TrialState.PRUNED])\n",
"complete_trials_hu_struct = study_hu_struct.get_trials(deepcopy=False, states=[TrialState.COMPLETE])"
]
},
{
"cell_type": "code",
"execution_count": 58,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Study statistics: [ws_structure] \n",
" Number of finished trials: 5\n",
" Number of pruned trials: 0\n",
" Number of complete trials: 5\n",
"Best trial:\n",
" Value: 0.741337\n",
" Params: \n",
" n_layers: 3\n",
" layer_dropout: 0\n",
" n_units_l_0: 787\n",
" n_decay_l_0: 0.0001\n",
" n_units_l_1: 9082\n",
" n_decay_l_1: 0.001\n",
" n_units_l_2: 7890\n",
" n_decay_l_2: 0.001\n",
" last_dropout: 0.1\n",
" lr: 0.0001\n"
]
}
],
"source": [
"print(\"Study statistics: [ws_structure] \")\n",
"print(\" Number of finished trials: \", len(study_ws_struct.trials))\n",
"print(\" Number of pruned trials: \", len(pruned_trials_ws_struct))\n",
"print(\" Number of complete trials: \", len(complete_trials_ws_struct))\n",
"print(\"Best trial:\")\n",
"trials_tmp = study_ws_struct.best_trial\n",
"print(\" Value: \", trials_tmp.value)\n",
"print(\" Params: \")\n",
"for key, value in trials_tmp.params.items():\n",
" print(\" {}: {}\".format(key, value))"
]
},
{
"cell_type": "code",
"execution_count": 59,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Study statistics: [de_structure] \n",
" Number of finished trials: 5\n",
" Number of pruned trials: 0\n",
" Number of complete trials: 5\n",
"Best trial:\n",
" Value: 0.84961\n",
" Params: \n",
" n_layers: 2\n",
" layer_dropout: 0\n",
" n_units_l_0: 7109\n",
" n_decay_l_0: 0.001\n",
" n_units_l_1: 3436\n",
" n_decay_l_1: 1e-05\n",
" last_dropout: 0.3\n",
" lr: 1e-05\n"
]
}
],
"source": [
"print(\"Study statistics: [de_structure] \")\n",
"print(\" Number of finished trials: \", len(study_de_struct.trials))\n",
"print(\" Number of pruned trials: \", len(pruned_trials_de_struct))\n",
"print(\" Number of complete trials: \", len(complete_trials_de_struct))\n",
"print(\"Best trial:\")\n",
"trials_tmp = study_de_struct.best_trial\n",
"print(\" Value: \", trials_tmp.value)\n",
"print(\" Params: \")\n",
"for key, value in trials_tmp.params.items():\n",
" print(\" {}: {}\".format(key, value))"
]
},
{
"cell_type": "code",
"execution_count": 60,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Study statistics: [lo_structure] \n",
" Number of finished trials: 5\n",
" Number of pruned trials: 0\n",
" Number of complete trials: 5\n",
"Best trial:\n",
" Value: 0.679332\n",
" Params: \n",
" n_layers: 2\n",
" layer_dropout: 1\n",
" n_units_l_0: 7114\n",
" n_decay_l_0: 0.001\n",
" F_dropout_0: 0.1\n",
" n_units_l_1: 7475\n",
" n_decay_l_1: 0.0001\n",
" F_dropout_1: 0.3\n",
" lr: 1e-05\n"
]
}
],
"source": [
"print(\"Study statistics: [lo_structure] \")\n",
"print(\" Number of finished trials: \", len(study_lo_struct.trials))\n",
"print(\" Number of pruned trials: \", len(pruned_trials_lo_struct))\n",
"print(\" Number of complete trials: \", len(complete_trials_lo_struct))\n",
"print(\"Best trial:\")\n",
"trials_tmp = study_lo_struct.best_trial\n",
"print(\" Value: \", trials_tmp.value)\n",
"print(\" Params: \")\n",
"for key, value in trials_tmp.params.items():\n",
" print(\" {}: {}\".format(key, value))"
]
},
{
"cell_type": "code",
"execution_count": 61,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Study statistics: [hu_structure] \n",
" Number of finished trials: 5\n",
" Number of pruned trials: 0\n",
" Number of complete trials: 5\n",
"Best trial:\n",
" Value: 0.856321\n",
" Params: \n",
" n_layers: 2\n",
" layer_dropout: 0\n",
" n_units_l_0: 6594\n",
" n_decay_l_0: 0.0001\n",
" n_units_l_1: 301\n",
" n_decay_l_1: 0.001\n",
" last_dropout: 0.3\n",
" lr: 1e-05\n"
]
}
],
"source": [
"print(\"Study statistics: [hu_structure] \")\n",
"print(\" Number of finished trials: \", len(study_hu_struct.trials))\n",
"print(\" Number of pruned trials: \", len(pruned_trials_hu_struct))\n",
"print(\" Number of complete trials: \", len(complete_trials_hu_struct))\n",
"print(\"Best trial:\")\n",
"trials_tmp = study_hu_struct.best_trial\n",
"print(\" Value: \", trials_tmp.value)\n",
"print(\" Params: \")\n",
"for key, value in trials_tmp.params.items():\n",
" print(\" {}: {}\".format(key, value))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "ai",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.12.2"
},
"orig_nbformat": 4
},
"nbformat": 4,
"nbformat_minor": 2
}
|