Spaces:
Running
Running
File size: 82,569 Bytes
226ac39 d92d2aa 226ac39 d92d2aa 226ac39 |
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 |
"""
Complete Tools Registry for Groq Function Calling - All 67 Tools
Defines all available tools in Groq's function calling format.
"""
TOOLS = [
# ============================================
# BASIC TOOLS (16)
# ============================================
# Data Profiling Tools (3)
{
"type": "function",
"function": {
"name": "profile_dataset",
"description": "Get comprehensive statistics about a dataset including shape, data types, memory usage, null counts, and unique values. Use this as the first step to understand any new dataset.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Absolute or relative path to the CSV or Parquet file"
}
},
"required": ["file_path"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_data_quality_issues",
"description": "Detect data quality issues including outliers (using IQR method), duplicate rows, inconsistent formats, and data anomalies. Returns a prioritized list of issues with severity levels.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
}
},
"required": ["file_path"]
}
}
},
{
"type": "function",
"function": {
"name": "analyze_correlations",
"description": "Compute correlation matrix and identify top correlations. If a target column is specified, shows features most correlated with the target. Useful for feature selection and understanding relationships.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"target": {
"type": "string",
"description": "Optional target column name to analyze correlations with"
}
},
"required": ["file_path"]
}
}
},
# Data Cleaning Tools (3)
{
"type": "function",
"function": {
"name": "clean_missing_values",
"description": "Handle missing values using appropriate strategies based on column type. Strategies include median/mean for numeric, mode for categorical, forward_fill for time series, or drop. In 'auto' mode, first drops columns with >threshold missing (default 40%), then imputes remaining columns. Will not impute ID columns.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"strategy": {
"oneOf": [
{
"type": "string",
"enum": ["auto"],
"description": "Use 'auto' to automatically decide strategies for all columns based on data type. First drops columns with >threshold missing, then imputes remaining columns."
},
{
"type": "object",
"description": "Dictionary mapping column names to strategies ('median', 'mean', 'mode', 'forward_fill', 'drop')",
"additionalProperties": {"type": "string"}
}
],
"description": "Either 'auto' (string) to automatically handle all missing values, or a dictionary mapping specific columns to strategies"
},
"output_path": {
"type": "string",
"description": "Path to save cleaned dataset"
},
"threshold": {
"type": "number",
"description": "For 'auto' mode: drop columns with missing percentage above this threshold (default: 0.4 = 40%). Range: 0.0 to 1.0. For example, 0.7 means drop columns with >70% missing values."
}
},
"required": ["file_path", "strategy", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "handle_outliers",
"description": "Detect and handle outliers in numeric columns using IQR method. Methods: 'clip' (cap at boundaries), 'winsorize' (cap at percentiles), or 'remove' (delete rows).",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"method": {
"type": "string",
"enum": ["clip", "winsorize", "remove"],
"description": "Method to handle outliers"
},
"columns": {
"type": "array",
"items": {"type": "string"},
"description": "List of column names to check for outliers. Use 'all' to check all numeric columns."
},
"output_path": {
"type": "string",
"description": "Path to save cleaned dataset"
}
},
"required": ["file_path", "method", "columns", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "fix_data_types",
"description": "Auto-detect and fix incorrect data types. Handles dates, booleans, categoricals, and numeric columns. Fixes common issues like 'null' strings and mixed types.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"type_mapping": {
"type": "object",
"description": "Optional dictionary mapping column names to target types ('int', 'float', 'string', 'date', 'bool', 'category'). Use 'auto' for automatic detection.",
"additionalProperties": {"type": "string"}
},
"output_path": {
"type": "string",
"description": "Path to save dataset with fixed types"
}
},
"required": ["file_path", "output_path"]
}
}
},
# Data Type Conversion Tools (2)
{
"type": "function",
"function": {
"name": "force_numeric_conversion",
"description": "CRITICAL TOOL: Force convert columns to numeric type even if detected as strings/objects. Essential for datasets with numeric columns stored as strings (with commas, spaces, currency symbols). Use this BEFORE encoding when you see 'no numeric features' errors.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"columns": {
"type": "array",
"items": {"type": "string"},
"description": "List of column names to force convert to numeric. Use ['all'] to auto-detect and convert all non-ID columns that look numeric."
},
"output_path": {
"type": "string",
"description": "Path to save dataset with converted types"
},
"errors": {
"type": "string",
"enum": ["coerce", "raise"],
"description": "How to handle conversion errors. 'coerce' makes invalid values null (recommended), 'raise' throws error."
}
},
"required": ["file_path", "columns", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "smart_type_inference",
"description": "Intelligently infer and fix data types for all columns by analyzing patterns. Goes beyond basic type detection to understand semantic meaning. Use when dataset has widespread type issues.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"output_path": {
"type": "string",
"description": "Path to save dataset with inferred types"
},
"aggressive": {
"type": "boolean",
"description": "If true, attempts aggressive conversion on ambiguous columns. Recommended for messy datasets."
}
},
"required": ["file_path", "output_path"]
}
}
},
# Feature Engineering Tools (2)
{
"type": "function",
"function": {
"name": "create_time_features",
"description": "Extract comprehensive time-based features from datetime columns including year, month, day, day_of_week, quarter, is_weekend, and cyclical encodings (sin/cos for month and hour).",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"date_col": {
"type": "string",
"description": "Name of the datetime column to extract features from"
},
"output_path": {
"type": "string",
"description": "Path to save dataset with new features"
}
},
"required": ["file_path", "date_col", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "encode_categorical",
"description": "Encode categorical variables using one-hot encoding, target encoding, or frequency encoding. Handles high-cardinality columns intelligently. Use method='auto' to automatically choose the best encoding.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file"
},
"method": {
"type": "string",
"enum": ["one_hot", "target", "frequency", "auto"],
"description": "Encoding method to use. 'auto' automatically selects the best method."
},
"columns": {
"type": "array",
"items": {"type": "string"},
"description": "List of categorical columns to encode. Use ['all'] to encode all categorical columns. If not specified, defaults to all categorical columns."
},
"target_col": {
"type": "string",
"description": "Required for target encoding: name of the target column"
},
"output_path": {
"type": "string",
"description": "Path to save dataset with encoded features"
}
},
"required": ["file_path", "output_path"]
}
}
},
# Model Training Tools (2)
{
"type": "function",
"function": {
"name": "train_baseline_models",
"description": "Train multiple baseline models (Logistic Regression, Random Forest, XGBoost) and compare their performance. Automatically detects task type (classification/regression) and returns the best model with metrics.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the prepared dataset file"
},
"target_col": {
"type": "string",
"description": "Name of the target column to predict"
},
"task_type": {
"type": "string",
"enum": ["classification", "regression", "auto"],
"description": "Type of ML task. Use 'auto' to detect automatically."
},
"test_size": {
"type": "number",
"description": "Proportion of data to use for testing (default: 0.2)"
},
"random_state": {
"type": "integer",
"description": "Random seed for reproducibility (default: 42)"
}
},
"required": ["file_path", "target_col"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_model_report",
"description": "Generate comprehensive model evaluation report including metrics, confusion matrix (for classification), feature importance, and SHAP values for top features. Saves report as JSON.",
"parameters": {
"type": "object",
"properties": {
"model_path": {
"type": "string",
"description": "Path to saved model file (.pkl or .joblib)"
},
"test_data_path": {
"type": "string",
"description": "Path to test dataset file"
},
"target_col": {
"type": "string",
"description": "Name of the target column"
},
"output_path": {
"type": "string",
"description": "Path to save the report JSON file"
}
},
"required": ["model_path", "test_data_path", "target_col", "output_path"]
}
}
},
# New Data Wrangling Tools (3)
{
"type": "function",
"function": {
"name": "get_smart_summary",
"description": "Generate an LLM-friendly smart summary of a dataset with per-column missing value percentages (sorted by severity), unique value counts, sample data, and numeric statistics. Much more detailed than profile_dataset for decision-making.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the CSV or Parquet file to summarize"
},
"n_samples": {
"type": "integer",
"description": "Number of sample rows to include in the summary (default: 5)"
}
},
"required": ["file_path"]
}
}
},
{
"type": "function",
"function": {
"name": "merge_datasets",
"description": "Merge two datasets using SQL-like join operations (inner, left, right, outer, cross). Supports joining on single or multiple columns with same or different names. Automatically handles duplicate columns with suffixes.",
"parameters": {
"type": "object",
"properties": {
"left_path": {
"type": "string",
"description": "Path to the left (first) dataset file"
},
"right_path": {
"type": "string",
"description": "Path to the right (second) dataset file"
},
"output_path": {
"type": "string",
"description": "Path to save the merged dataset"
},
"how": {
"type": "string",
"enum": ["inner", "left", "right", "outer", "cross"],
"description": "Join type: 'inner' (only matching rows), 'left' (all left + matching right), 'right' (all right + matching left), 'outer' (all rows from both), 'cross' (cartesian product)"
},
"on": {
"type": ["string", "array"],
"items": {"type": "string"},
"description": "Column name(s) to join on (must exist in both datasets). Can be a single column name or list of columns. Use this when join columns have the same name in both datasets."
},
"left_on": {
"type": ["string", "array"],
"items": {"type": "string"},
"description": "Column name(s) in left dataset to join on. Use with right_on when join columns have different names."
},
"right_on": {
"type": ["string", "array"],
"items": {"type": "string"},
"description": "Column name(s) in right dataset to join on. Use with left_on when join columns have different names."
}
},
"required": ["left_path", "right_path", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "concat_datasets",
"description": "Concatenate multiple datasets either vertically (stacking rows, useful for monthly data) or horizontally (adding columns side-by-side). Validates schema compatibility for vertical concat.",
"parameters": {
"type": "object",
"properties": {
"file_paths": {
"type": "array",
"items": {"type": "string"},
"description": "List of paths to dataset files to concatenate (minimum 2 files)"
},
"output_path": {
"type": "string",
"description": "Path to save the concatenated dataset"
},
"axis": {
"type": "string",
"enum": ["vertical", "horizontal"],
"description": "'vertical' to stack rows (union, for monthly data), 'horizontal' to add columns side-by-side (default: 'vertical')"
}
},
"required": ["file_paths", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "reshape_dataset",
"description": "Transform dataset structure using pivot (long→wide format), melt (wide→long format), or transpose (swap rows and columns) operations.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to the dataset file to reshape"
},
"output_path": {
"type": "string",
"description": "Path to save the reshaped dataset"
},
"operation": {
"type": "string",
"enum": ["pivot", "melt", "transpose"],
"description": "Reshape operation: 'pivot' (long→wide, requires index/columns/values), 'melt' (wide→long, requires id_vars/value_vars), 'transpose' (swap rows/columns)"
},
"index": {
"type": "string",
"description": "Column to use as row index (for pivot operation)"
},
"columns": {
"type": "string",
"description": "Column whose values become new column names (for pivot operation)"
},
"values": {
"type": "string",
"description": "Column whose values populate the pivoted table (for pivot operation)"
},
"id_vars": {
"type": "array",
"items": {"type": "string"},
"description": "Columns to keep as identifiers (for melt operation)"
},
"value_vars": {
"type": "array",
"items": {"type": "string"},
"description": "Columns to unpivot (for melt operation). If not specified, uses all columns except id_vars."
}
},
"required": ["file_path", "output_path", "operation"]
}
}
},
# ============================================
# ADVANCED ANALYSIS TOOLS (5)
# ============================================
{
"type": "function",
"function": {
"name": "perform_eda_analysis",
"description": "Comprehensive Exploratory Data Analysis with visualizations, distribution analysis, and automated insights. Generates HTML report with plots.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Optional target column for supervised analysis"},
"output_dir": {"type": "string", "description": "Directory to save EDA report and plots"}
},
"required": ["file_path", "output_dir"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_model_issues",
"description": "Detect overfitting, underfitting, class imbalance, and other model performance issues. Provides diagnostic recommendations.",
"parameters": {
"type": "object",
"properties": {
"model_path": {"type": "string", "description": "Path to trained model"},
"train_data_path": {"type": "string", "description": "Path to training data"},
"test_data_path": {"type": "string", "description": "Path to test data"},
"target_col": {"type": "string", "description": "Target column name"}
},
"required": ["model_path", "train_data_path", "test_data_path", "target_col"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_anomalies",
"description": "Detect anomalies using Isolation Forest, LOF, or statistical methods. Returns anomaly scores and flags.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"method": {"type": "string", "enum": ["isolation_forest", "lof", "statistical"], "description": "Anomaly detection method"},
"contamination": {"type": "number", "description": "Expected proportion of anomalies (default: 0.1)"},
"output_path": {"type": "string", "description": "Path to save dataset with anomaly scores"}
},
"required": ["file_path", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_and_handle_multicollinearity",
"description": "Detect and handle multicollinearity using VIF (Variance Inflation Factor). Removes highly correlated features automatically.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"threshold": {"type": "number", "description": "VIF threshold (default: 10)"},
"method": {"type": "string", "enum": ["drop", "combine"], "description": "How to handle correlated features"},
"output_path": {"type": "string", "description": "Path to save cleaned dataset"}
},
"required": ["file_path", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_statistical_tests",
"description": "Perform statistical hypothesis tests (t-test, chi-square, ANOVA) to analyze relationships between features and target.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"test_type": {"type": "string", "enum": ["auto", "ttest", "chi2", "anova"], "description": "Type of statistical test"}
},
"required": ["file_path", "target_col"]
}
}
},
# ============================================
# ADVANCED FEATURE ENGINEERING (4)
# ============================================
{
"type": "function",
"function": {
"name": "create_interaction_features",
"description": "Create polynomial, PCA, or cross-product interaction features to capture non-linear relationships.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"method": {"type": "string", "enum": ["polynomial", "pca", "cross"], "description": "Interaction method"},
"degree": {"type": "integer", "description": "Polynomial degree (default: 2)"},
"max_features": {"type": "integer", "description": "Maximum new features to create (default: 50)"},
"output_path": {"type": "string", "description": "Path to save enhanced dataset"}
},
"required": ["file_path", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "create_aggregation_features",
"description": "Create aggregation features (mean, sum, count, etc.) grouped by categorical columns. Useful for customer/transaction data.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"group_cols": {"type": "array", "items": {"type": "string"}, "description": "Columns to group by"},
"agg_cols": {"type": "array", "items": {"type": "string"}, "description": "Columns to aggregate"},
"agg_functions": {"type": "array", "items": {"type": "string"}, "description": "Aggregation functions (mean, sum, count, etc.)"},
"output_path": {"type": "string", "description": "Path to save dataset with aggregations"}
},
"required": ["file_path", "group_cols", "agg_cols", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "engineer_text_features",
"description": "Extract features from text columns: TF-IDF, word counts, sentiment, readability scores, and embeddings.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"text_col": {"type": "string", "description": "Text column name"},
"methods": {"type": "array", "items": {"type": "string"}, "description": "Feature extraction methods (tfidf, count, sentiment, readability)"},
"max_features": {"type": "integer", "description": "Max TF-IDF features (default: 100)"},
"output_path": {"type": "string", "description": "Path to save dataset with text features"}
},
"required": ["file_path", "text_col", "methods", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "auto_feature_engineering",
"description": "Use LLM (Gemini/Groq) to automatically generate creative feature engineering ideas and implement them. Works without API key if environment variables are set.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"groq_api_key": {"type": "string", "description": "Groq API key (optional - uses environment variable if not provided)"},
"max_suggestions": {"type": "integer", "description": "Maximum feature suggestions to generate (default: 10)"},
"implement_top_k": {"type": "integer", "description": "Number of top suggestions to implement (default: 5)"},
"output_path": {"type": "string", "description": "Path to save dataset with new features"}
},
"required": ["file_path", "target_col", "output_path"]
}
}
},
# ============================================
# ADVANCED PREPROCESSING (3)
# ============================================
{
"type": "function",
"function": {
"name": "handle_imbalanced_data",
"description": "Handle class imbalance using SMOTE, ADASYN, or class weights. Critical for classification tasks.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"method": {"type": "string", "enum": ["smote", "adasyn", "random_oversample", "random_undersample"], "description": "Balancing method"},
"sampling_strategy": {"type": "string", "description": "Sampling ratio (auto, minority, majority)"},
"output_path": {"type": "string", "description": "Path to save balanced dataset"}
},
"required": ["file_path", "target_col", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_feature_scaling",
"description": "Scale features using StandardScaler, MinMaxScaler, or RobustScaler. Essential for distance-based algorithms.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"method": {"type": "string", "enum": ["standard", "minmax", "robust"], "description": "Scaling method"},
"columns": {"type": "array", "items": {"type": "string"}, "description": "Columns to scale (None = all numeric)"},
"output_path": {"type": "string", "description": "Path to save scaled dataset"}
},
"required": ["file_path", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "split_data_strategically",
"description": "Split data with stratification, time-based splitting, or group-based splitting for better validation.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target column for stratification"},
"method": {"type": "string", "enum": ["stratified", "time_based", "group_based"], "description": "Split method"},
"test_size": {"type": "number", "description": "Test set proportion (default: 0.2)"},
"time_col": {"type": "string", "description": "Time column for time-based split"},
"group_col": {"type": "string", "description": "Group column for group-based split"}
},
"required": ["file_path", "method"]
}
}
},
# ============================================
# ADVANCED TRAINING (3)
# ============================================
{
"type": "function",
"function": {
"name": "hyperparameter_tuning",
"description": "Optimize model hyperparameters using Optuna (Bayesian optimization). Finds best model configuration automatically.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to prepared dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"model_type": {"type": "string", "enum": ["random_forest", "xgboost", "lightgbm"], "description": "Model to tune"},
"n_trials": {"type": "integer", "description": "Number of tuning trials (default: 100)"},
"task_type": {"type": "string", "enum": ["classification", "regression", "auto"], "description": "ML task type"},
"output_path": {"type": "string", "description": "Path to save tuned model"}
},
"required": ["file_path", "target_col", "model_type", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "train_ensemble_models",
"description": "Train ensemble models using stacking, voting, or blending. Combines multiple models for better performance.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to prepared dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"ensemble_method": {"type": "string", "enum": ["stacking", "voting", "blending"], "description": "Ensemble technique"},
"base_models": {"type": "array", "items": {"type": "string"}, "description": "Base model types to ensemble"},
"task_type": {"type": "string", "enum": ["classification", "regression", "auto"], "description": "ML task type"},
"output_path": {"type": "string", "description": "Path to save ensemble model"}
},
"required": ["file_path", "target_col", "ensemble_method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_cross_validation",
"description": "Perform k-fold cross-validation to get robust model performance estimates. Returns mean and std of metrics.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"model_type": {"type": "string", "description": "Model type (random_forest, xgboost, logistic, ridge)"},
"n_splits": {"type": "integer", "description": "Number of CV folds/splits (default: 5)"},
"task_type": {"type": "string", "enum": ["classification", "regression", "auto"], "description": "ML task type"},
"cv_strategy": {"type": "string", "enum": ["kfold", "stratified", "timeseries"], "description": "Cross-validation strategy (default: kfold)"},
"save_oof": {"type": "boolean", "description": "Whether to save out-of-fold predictions (default: false)"}
},
"required": ["file_path", "target_col", "model_type"]
}
}
},
# ============================================
# BUSINESS INTELLIGENCE (4)
# ============================================
{
"type": "function",
"function": {
"name": "perform_cohort_analysis",
"description": "Analyze user cohorts over time (retention, revenue, engagement). Essential for SaaS and e-commerce businesses.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to transaction/event data"},
"user_col": {"type": "string", "description": "User ID column"},
"date_col": {"type": "string", "description": "Date/timestamp column"},
"metric_col": {"type": "string", "description": "Metric to analyze (revenue, events, etc.)"},
"cohort_period": {"type": "string", "enum": ["daily", "weekly", "monthly"], "description": "Cohort grouping period"},
"output_path": {"type": "string", "description": "Path to save cohort analysis results"}
},
"required": ["file_path", "user_col", "date_col", "metric_col", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_rfm_analysis",
"description": "RFM (Recency, Frequency, Monetary) analysis for customer segmentation. Identifies best/worst customers.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to transaction data"},
"customer_col": {"type": "string", "description": "Customer ID column"},
"date_col": {"type": "string", "description": "Transaction date column"},
"amount_col": {"type": "string", "description": "Transaction amount column"},
"output_path": {"type": "string", "description": "Path to save RFM segments"}
},
"required": ["file_path", "customer_col", "date_col", "amount_col", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_causal_relationships",
"description": "Detect potential causal relationships between features using Granger causality and correlation analysis.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target/effect column"},
"feature_cols": {"type": "array", "items": {"type": "string"}, "description": "Potential cause columns"},
"method": {"type": "string", "enum": ["granger", "correlation"], "description": "Causality detection method"}
},
"required": ["file_path", "target_col"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_business_insights",
"description": "Generate automated business insights using descriptive statistics, trends, and anomaly detection. Creates executive summary.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to business data"},
"metric_cols": {"type": "array", "items": {"type": "string"}, "description": "Key business metrics to analyze"},
"date_col": {"type": "string", "description": "Date column for trend analysis"},
"output_path": {"type": "string", "description": "Path to save insights report"}
},
"required": ["file_path", "metric_cols", "output_path"]
}
}
},
# ============================================
# COMPUTER VISION (3)
# ============================================
{
"type": "function",
"function": {
"name": "extract_image_features",
"description": "Extract features from images using pre-trained CNNs (ResNet, VGG). Converts images to feature vectors for ML.",
"parameters": {
"type": "object",
"properties": {
"image_dir": {"type": "string", "description": "Directory containing images"},
"model": {"type": "string", "enum": ["resnet", "vgg", "mobilenet"], "description": "Pre-trained model to use"},
"output_path": {"type": "string", "description": "Path to save feature vectors CSV"}
},
"required": ["image_dir", "model", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_image_clustering",
"description": "Cluster images based on visual similarity using K-means or DBSCAN on extracted features.",
"parameters": {
"type": "object",
"properties": {
"image_dir": {"type": "string", "description": "Directory containing images"},
"n_clusters": {"type": "integer", "description": "Number of clusters (default: auto-detect)"},
"method": {"type": "string", "enum": ["kmeans", "dbscan"], "description": "Clustering method"},
"output_path": {"type": "string", "description": "Path to save clustering results"}
},
"required": ["image_dir", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "analyze_tabular_image_hybrid",
"description": "Combine tabular data with image features for hybrid ML models. Useful for e-commerce/medical data.",
"parameters": {
"type": "object",
"properties": {
"tabular_path": {"type": "string", "description": "Path to tabular data CSV"},
"image_dir": {"type": "string", "description": "Directory with images"},
"image_id_col": {"type": "string", "description": "Column linking tabular data to images"},
"output_path": {"type": "string", "description": "Path to save combined features"}
},
"required": ["tabular_path", "image_dir", "image_id_col", "output_path"]
}
}
},
# ============================================
# NLP/TEXT ANALYTICS (4)
# ============================================
{
"type": "function",
"function": {
"name": "perform_topic_modeling",
"description": "Discover topics in text documents using LDA or NMF. Extract themes from customer reviews, articles, etc.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset with text"},
"text_col": {"type": "string", "description": "Text column name"},
"n_topics": {"type": "integer", "description": "Number of topics to extract (default: 5)"},
"method": {"type": "string", "enum": ["lda", "nmf"], "description": "Topic modeling method"},
"output_path": {"type": "string", "description": "Path to save topics and document-topic matrix"}
},
"required": ["file_path", "text_col", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_named_entity_recognition",
"description": "Extract named entities (person, organization, location) from text using NER models.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset with text"},
"text_col": {"type": "string", "description": "Text column name"},
"output_path": {"type": "string", "description": "Path to save dataset with extracted entities"}
},
"required": ["file_path", "text_col", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "analyze_sentiment_advanced",
"description": "Perform advanced sentiment analysis with aspect-based sentiment (what features customers like/dislike).",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset with text"},
"text_col": {"type": "string", "description": "Text column name"},
"aspects": {"type": "array", "items": {"type": "string"}, "description": "Aspects to analyze sentiment for (e.g., 'price', 'quality')"},
"output_path": {"type": "string", "description": "Path to save sentiment scores"}
},
"required": ["file_path", "text_col", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_text_similarity",
"description": "Calculate text similarity using cosine similarity, Jaccard, or semantic embeddings. Find duplicate/similar documents.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset with text"},
"text_col": {"type": "string", "description": "Text column name"},
"method": {"type": "string", "enum": ["cosine", "jaccard", "semantic"], "description": "Similarity method"},
"threshold": {"type": "number", "description": "Similarity threshold (0-1)"},
"output_path": {"type": "string", "description": "Path to save similarity matrix"}
},
"required": ["file_path", "text_col", "method", "output_path"]
}
}
},
# ============================================
# PRODUCTION/MLOPS (5)
# ============================================
{
"type": "function",
"function": {
"name": "monitor_model_drift",
"description": "Detect data drift and concept drift in production models. Compare training vs production data distributions.",
"parameters": {
"type": "object",
"properties": {
"train_data_path": {"type": "string", "description": "Path to original training data"},
"production_data_path": {"type": "string", "description": "Path to recent production data"},
"features": {"type": "array", "items": {"type": "string"}, "description": "Features to monitor for drift"},
"output_path": {"type": "string", "description": "Path to save drift report"}
},
"required": ["train_data_path", "production_data_path", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "explain_predictions",
"description": "Explain model predictions using SHAP or LIME. Generate feature importance explanations for individual predictions.",
"parameters": {
"type": "object",
"properties": {
"model_path": {"type": "string", "description": "Path to trained model"},
"data_path": {"type": "string", "description": "Path to data to explain"},
"method": {"type": "string", "enum": ["shap", "lime"], "description": "Explanation method"},
"n_samples": {"type": "integer", "description": "Number of samples to explain (default: 10)"},
"output_path": {"type": "string", "description": "Path to save explanations"}
},
"required": ["model_path", "data_path", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_model_card",
"description": "Generate model card documentation with model details, performance metrics, bias analysis, and usage guidelines.",
"parameters": {
"type": "object",
"properties": {
"model_path": {"type": "string", "description": "Path to trained model"},
"train_data_path": {"type": "string", "description": "Path to training data"},
"test_data_path": {"type": "string", "description": "Path to test data"},
"target_col": {"type": "string", "description": "Target column name"},
"output_path": {"type": "string", "description": "Path to save model card JSON"}
},
"required": ["model_path", "train_data_path", "test_data_path", "target_col", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_ab_test_analysis",
"description": "Analyze A/B test results with statistical significance testing. Determine if variant B is better than control A.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to A/B test data"},
"variant_col": {"type": "string", "description": "Column indicating variant (A/B)"},
"metric_col": {"type": "string", "description": "Success metric column"},
"confidence_level": {"type": "number", "description": "Confidence level for significance (default: 0.95)"}
},
"required": ["file_path", "variant_col", "metric_col"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_feature_leakage",
"description": "Detect potential feature leakage by analyzing feature importance and temporal relationships. Prevents data leakage bugs.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"date_col": {"type": "string", "description": "Optional date column for temporal analysis"}
},
"required": ["file_path", "target_col"]
}
}
},
# ============================================
# TIME SERIES (3)
# ============================================
{
"type": "function",
"function": {
"name": "forecast_time_series",
"description": "Forecast future values using ARIMA, Prophet, or LSTM models. Handles seasonal and trend components.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to time series data"},
"date_col": {"type": "string", "description": "Date/timestamp column"},
"value_col": {"type": "string", "description": "Value column to forecast"},
"forecast_periods": {"type": "integer", "description": "Number of periods to forecast"},
"method": {"type": "string", "enum": ["arima", "prophet", "lstm"], "description": "Forecasting method"},
"output_path": {"type": "string", "description": "Path to save forecast results"}
},
"required": ["file_path", "date_col", "value_col", "forecast_periods", "method", "output_path"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_seasonality_trends",
"description": "Detect seasonality patterns and trends in time series data using STL decomposition and statistical tests.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to time series data"},
"date_col": {"type": "string", "description": "Date/timestamp column"},
"value_col": {"type": "string", "description": "Value column to analyze"},
"period": {"type": "integer", "description": "Expected seasonal period (e.g., 12 for monthly)"}
},
"required": ["file_path", "date_col", "value_col"]
}
}
},
{
"type": "function",
"function": {
"name": "create_time_series_features",
"description": "Create comprehensive time series features: lags, rolling stats, exponential moving averages, and Fourier features.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to time series data"},
"date_col": {"type": "string", "description": "Date/timestamp column"},
"value_col": {"type": "string", "description": "Value column"},
"lags": {"type": "array", "items": {"type": "integer"}, "description": "Lag periods to create (e.g., [1, 7, 30])"},
"windows": {"type": "array", "items": {"type": "integer"}, "description": "Rolling window sizes (e.g., [7, 30])"},
"output_path": {"type": "string", "description": "Path to save dataset with time series features"}
},
"required": ["file_path", "date_col", "value_col", "output_path"]
}
}
},
# ============================================
# ADVANCED INSIGHTS TOOLS (6) - NEW
# ============================================
{
"type": "function",
"function": {
"name": "analyze_root_cause",
"description": "Perform root cause analysis to identify why a metric dropped or changed. Analyzes correlations, temporal patterns, and identifies top influencing factors.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Column to analyze (e.g., 'sales')"},
"time_col": {"type": "string", "description": "Optional time column for trend analysis"},
"threshold_drop": {"type": "number", "description": "Percentage drop to flag as significant (default 0.15)"}
},
"required": ["file_path", "target_col"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_trends_and_seasonality",
"description": "Detect trends and seasonal patterns in time series data using statistical methods and autocorrelation.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"value_col": {"type": "string", "description": "Column with values to analyze"},
"time_col": {"type": "string", "description": "Column with timestamps"},
"seasonal_period": {"type": "integer", "description": "Expected seasonal period (auto-detected if None)"}
},
"required": ["file_path", "value_col", "time_col"]
}
}
},
{
"type": "function",
"function": {
"name": "detect_anomalies_advanced",
"description": "Detect anomalies with confidence scores using Isolation Forest or statistical methods.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"columns": {"type": "array", "items": {"type": "string"}, "description": "Columns to analyze (all numeric if None)"},
"contamination": {"type": "number", "description": "Expected proportion of outliers (default 0.1)"},
"method": {"type": "string", "enum": ["isolation_forest", "statistical"], "description": "Detection method"}
},
"required": ["file_path"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_hypothesis_testing",
"description": "Perform statistical hypothesis testing (t-test, ANOVA, chi-square) to compare groups.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"group_col": {"type": "string", "description": "Column defining groups"},
"value_col": {"type": "string", "description": "Column with values to compare"},
"test_type": {"type": "string", "enum": ["t-test", "anova", "chi-square", "auto"], "description": "Test type (auto-detected if 'auto')"}
},
"required": ["file_path", "group_col", "value_col"]
}
}
},
{
"type": "function",
"function": {
"name": "analyze_distribution",
"description": "Analyze distribution of a column including normality tests, skewness, and kurtosis.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"column": {"type": "string", "description": "Column to analyze"},
"tests": {"type": "array", "items": {"type": "string"}, "description": "Tests to perform (normality, skewness)"}
},
"required": ["file_path", "column"]
}
}
},
{
"type": "function",
"function": {
"name": "perform_segment_analysis",
"description": "Perform cluster-based customer/data segmentation using K-means and profile each segment.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"n_segments": {"type": "integer", "description": "Number of segments to create (default 5)"},
"features": {"type": "array", "items": {"type": "string"}, "description": "Features for clustering (all numeric if None)"}
},
"required": ["file_path"]
}
}
},
# ============================================
# AUTOMATED PIPELINE TOOLS (2) - NEW
# ============================================
{
"type": "function",
"function": {
"name": "auto_ml_pipeline",
"description": "Fully automated ML pipeline: auto-detect types, clean missing values, handle outliers, encode categorical, engineer features, and select best features. Zero configuration required!",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to input dataset"},
"target_col": {"type": "string", "description": "Target column name"},
"task_type": {"type": "string", "enum": ["classification", "regression", "auto"], "description": "Task type (auto-detected if 'auto')"},
"output_path": {"type": "string", "description": "Where to save processed data"},
"feature_engineering_level": {"type": "string", "enum": ["basic", "intermediate", "advanced"], "description": "Feature engineering depth"}
},
"required": ["file_path", "target_col"]
}
}
},
{
"type": "function",
"function": {
"name": "auto_feature_selection",
"description": "Automatically select the best features for modeling using mutual information or F-statistics.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Target column"},
"task_type": {"type": "string", "enum": ["classification", "regression", "auto"], "description": "Task type"},
"max_features": {"type": "integer", "description": "Maximum features to keep (default 50)"},
"method": {"type": "string", "enum": ["mutual_info", "f_test", "auto"], "description": "Selection method"},
"output_path": {"type": "string", "description": "Where to save selected features"}
},
"required": ["file_path", "target_col"]
}
}
},
# ============================================
# VISUALIZATION TOOLS (3) - NEW
# ============================================
{
"type": "function",
"function": {
"name": "generate_all_plots",
"description": "Generate ALL plots for a dataset automatically: data quality, EDA, distributions, and correlations. Creates interactive HTML plots.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Optional target column"},
"output_dir": {"type": "string", "description": "Directory to save plots (default ./outputs/plots)"}
},
"required": ["file_path"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_data_quality_plots",
"description": "Generate data quality visualizations: missing values, data types, and outlier detection plots.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"output_dir": {"type": "string", "description": "Directory to save plots"}
},
"required": ["file_path", "output_dir"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_eda_plots",
"description": "Generate exploratory data analysis plots: correlation heatmap, feature relationships, and pairplots.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Optional target column"},
"output_dir": {"type": "string", "description": "Directory to save plots"}
},
"required": ["file_path", "output_dir"]
}
}
},
# ============================================
# INTERACTIVE PLOTLY VISUALIZATIONS (6)
# ============================================
{
"type": "function",
"function": {
"name": "generate_interactive_scatter",
"description": "Create interactive scatter plot with zoom, pan, and hover capabilities. Great for exploring relationships between variables.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"x_col": {"type": "string", "description": "Column for X-axis"},
"y_col": {"type": "string", "description": "Column for Y-axis"},
"color_col": {"type": "string", "description": "Optional column for color coding points"},
"size_col": {"type": "string", "description": "Optional column for bubble size"},
"output_path": {"type": "string", "description": "Path to save HTML file (default: ./outputs/plots/interactive/scatter.html)"}
},
"required": ["file_path", "x_col", "y_col"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_interactive_histogram",
"description": "Create interactive histogram with box plot overlay. Users can explore distribution interactively.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"column": {"type": "string", "description": "Column to plot distribution"},
"bins": {"type": "integer", "description": "Number of bins (default: 30)"},
"color_col": {"type": "string", "description": "Optional column for grouped histograms"},
"output_path": {"type": "string", "description": "Path to save HTML file"}
},
"required": ["file_path", "column"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_interactive_correlation_heatmap",
"description": "Create interactive correlation heatmap with hover values. Better than static matplotlib version.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"output_path": {"type": "string", "description": "Path to save HTML file"}
},
"required": ["file_path"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_interactive_box_plots",
"description": "Create interactive box plots for outlier detection. Supports grouping by categorical variable.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"columns": {"type": "array", "items": {"type": "string"}, "description": "Columns to plot (all numeric if not specified)"},
"group_by": {"type": "string", "description": "Optional categorical column for grouping"},
"output_path": {"type": "string", "description": "Path to save HTML file"}
},
"required": ["file_path"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_interactive_time_series",
"description": "Create interactive time series plot with range slider and zoom. Perfect for temporal data analysis.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"time_col": {"type": "string", "description": "Column with datetime values"},
"value_cols": {"type": "array", "items": {"type": "string"}, "description": "Columns to plot over time"},
"output_path": {"type": "string", "description": "Path to save HTML file"}
},
"required": ["file_path", "time_col", "value_cols"]
}
}
},
{
"type": "function",
"function": {
"name": "generate_plotly_dashboard",
"description": "Generate complete interactive dashboard with multiple visualizations: correlation heatmap, box plots, scatter plots, histograms. One-stop visualization solution.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to dataset"},
"target_col": {"type": "string", "description": "Optional target column for supervised analysis"},
"output_dir": {"type": "string", "description": "Directory to save all plots (default: ./outputs/plots/interactive)"}
},
"required": ["file_path"]
}
}
},
# EDA Report Generation (1) - NEW PHASE 2
{
"type": "function",
"function": {
"name": "generate_ydata_profiling_report",
"description": "Generate comprehensive HTML report using ydata-profiling (formerly pandas-profiling). Provides extensive analysis: overview, variable statistics, interactions, correlations (Pearson, Spearman, Cramér's V), missing values matrix, duplicate analysis, and more. Most detailed and comprehensive profiling tool with automated insights and data quality warnings.",
"parameters": {
"type": "object",
"properties": {
"file_path": {"type": "string", "description": "Path to the dataset CSV/Parquet file"},
"output_path": {"type": "string", "description": "Where to save HTML report (default: ./outputs/reports/ydata_profile.html)"},
"minimal": {"type": "boolean", "description": "If true, generates faster minimal report (useful for large datasets, default: false)"},
"title": {"type": "string", "description": "Report title (default: 'Data Profiling Report')"}
},
"required": ["file_path"]
}
}
},
# ========================================
# CODE INTERPRETER - THE GAME CHANGER 🚀
# ========================================
{
"type": "function",
"function": {
"name": "execute_python_code",
"description": "⭐ CRITICAL TOOL - Execute custom Python code for ANY data science task not covered by existing tools. This is what makes you a TRUE AI AGENT, not just a function-calling bot. Use this when user requests: 1) Custom visualizations (specific Plotly plots, interactive dashboards, unique chart types) 2) Domain-specific calculations 3) Custom data transformations 4) Specific export formats 5) Interactive widgets/filters. Code has access to pandas, polars, numpy, matplotlib, seaborn, plotly. ALWAYS save outputs to files and return file paths.",
"parameters": {
"type": "object",
"properties": {
"code": {
"type": "string",
"description": "Python code to execute. Auto-imported: pandas as pd, polars as pl, numpy as np, matplotlib.pyplot as plt, seaborn as sns, plotly.express as px, plotly.graph_objects as go. Code should save outputs to files in working_directory. Example: fig.write_html('./outputs/code/plot.html')"
},
"working_directory": {
"type": "string",
"description": "Directory to run code in (default: ./outputs/code). Code can read from ./temp/ and write to this directory."
},
"timeout": {
"type": "integer",
"description": "Maximum execution time in seconds (default: 60)"
}
},
"required": ["code"]
}
}
},
{
"type": "function",
"function": {
"name": "execute_code_from_file",
"description": "Execute Python code from an existing .py file. Useful when code is too long to pass as string, or when running pre-written scripts. Same capabilities as execute_python_code.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to .py file to execute"
},
"working_directory": {
"type": "string",
"description": "Directory to run code in (default: ./outputs/code)"
},
"timeout": {
"type": "integer",
"description": "Maximum execution time in seconds (default: 60)"
}
},
"required": ["file_path"]
}
}
},
# ============================================
# CLOUD DATA SOURCES (4) - NEW
# ============================================
{
"type": "function",
"function": {
"name": "load_bigquery_table",
"description": "Load data from Google BigQuery table into a Polars DataFrame. Supports sampling via LIMIT and column selection. Returns CSV path for downstream tools. Use profile_bigquery_table first for large tables.",
"parameters": {
"type": "object",
"properties": {
"project_id": {
"type": "string",
"description": "Google Cloud project ID"
},
"dataset": {
"type": "string",
"description": "BigQuery dataset name"
},
"table": {
"type": "string",
"description": "BigQuery table name"
},
"limit": {
"type": "integer",
"description": "Optional row limit for sampling (e.g., 10000 for large tables)"
},
"columns": {
"type": "array",
"items": {"type": "string"},
"description": "Optional list of column names to load"
},
"where_clause": {
"type": "string",
"description": "Optional SQL WHERE clause for filtering (without WHERE keyword)"
}
},
"required": ["project_id", "dataset", "table"]
}
}
},
{
"type": "function",
"function": {
"name": "write_bigquery_table",
"description": "Write predictions or processed data from CSV/Parquet file to BigQuery table. Supports append, overwrite, or fail modes.",
"parameters": {
"type": "object",
"properties": {
"file_path": {
"type": "string",
"description": "Path to CSV or Parquet file to write"
},
"project_id": {
"type": "string",
"description": "Google Cloud project ID"
},
"dataset": {
"type": "string",
"description": "BigQuery dataset name"
},
"table": {
"type": "string",
"description": "BigQuery table name"
},
"mode": {
"type": "string",
"enum": ["append", "overwrite", "fail"],
"description": "Write mode: append (add rows), overwrite (replace), fail (error if exists)"
}
},
"required": ["file_path", "project_id", "dataset", "table"]
}
}
},
{
"type": "function",
"function": {
"name": "profile_bigquery_table",
"description": "Profile a BigQuery table without loading all data. Returns row count, column types, null counts (sampled), table size, and load recommendations. Use this BEFORE load_bigquery_table for large tables.",
"parameters": {
"type": "object",
"properties": {
"project_id": {
"type": "string",
"description": "Google Cloud project ID"
},
"dataset": {
"type": "string",
"description": "BigQuery dataset name"
},
"table": {
"type": "string",
"description": "BigQuery table name"
}
},
"required": ["project_id", "dataset", "table"]
}
}
},
{
"type": "function",
"function": {
"name": "query_bigquery",
"description": "Execute custom BigQuery SQL query and return results as DataFrame. Useful for complex aggregations, joins, or transformations before analysis.",
"parameters": {
"type": "object",
"properties": {
"project_id": {
"type": "string",
"description": "Google Cloud project ID"
},
"query": {
"type": "string",
"description": "SQL query to execute"
},
"output_path": {
"type": "string",
"description": "Optional path to save results (default: auto-generated)"
},
"limit": {
"type": "integer",
"description": "Optional row limit to append to query"
}
},
"required": ["project_id", "query"]
}
}
}
]
def get_tool_by_name(tool_name: str) -> dict:
"""Get tool definition by name."""
for tool in TOOLS:
if tool["function"]["name"] == tool_name:
return tool
raise ValueError(f"Tool '{tool_name}' not found in registry")
def get_all_tool_names() -> list:
"""Get list of all tool names."""
return [tool["function"]["name"] for tool in TOOLS]
def get_tools_by_category() -> dict:
"""Get tools organized by category."""
return {
"basic": [t["function"]["name"] for t in TOOLS[:16]],
"advanced_analysis": [t["function"]["name"] for t in TOOLS[16:21]],
"advanced_feature_engineering": [t["function"]["name"] for t in TOOLS[21:25]],
"advanced_preprocessing": [t["function"]["name"] for t in TOOLS[25:28]],
"advanced_training": [t["function"]["name"] for t in TOOLS[28:31]],
"business_intelligence": [t["function"]["name"] for t in TOOLS[31:35]],
"computer_vision": [t["function"]["name"] for t in TOOLS[35:38]],
"nlp_text_analytics": [t["function"]["name"] for t in TOOLS[38:42]],
"production_mlops": [t["function"]["name"] for t in TOOLS[42:47]],
"time_series": [t["function"]["name"] for t in TOOLS[47:50]],
"cloud_data_sources": [t["function"]["name"] for t in TOOLS[50:54]]
}
|