File size: 81,882 Bytes
9c6594c |
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 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 |
"""Use the Public API to export or update data that you have saved to W&B.
Before using this API, you'll want to log data from your script — check the
[Quickstart](https://docs.wandb.ai/quickstart) for more details.
You might use the Public API to
- update metadata or metrics for an experiment after it has been completed,
- pull down your results as a dataframe for post-hoc analysis in a Jupyter notebook, or
- check your saved model artifacts for those tagged as `ready-to-deploy`.
For more on using the Public API, check out [our guide](https://docs.wandb.com/guides/track/public-api-guide).
"""
import json
import logging
import os
import urllib
from http import HTTPStatus
from typing import (
TYPE_CHECKING,
Any,
Dict,
Iterator,
List,
Literal,
Optional,
Set,
Union,
)
import requests
from pydantic import ValidationError
from typing_extensions import Unpack
from wandb_gql import Client, gql
from wandb_gql.client import RetryError
import wandb
from wandb import env, util
from wandb._iterutils import one
from wandb.apis import public
from wandb.apis.normalize import normalize_exceptions
from wandb.apis.public.const import RETRY_TIMEDELTA
from wandb.apis.public.registries.registries_search import Registries
from wandb.apis.public.registries.registry import Registry
from wandb.apis.public.registries.utils import _fetch_org_entity_from_organization
from wandb.apis.public.utils import (
PathType,
fetch_org_from_settings_or_entity,
gql_compat,
parse_org_from_registry_path,
)
from wandb.proto.wandb_deprecated import Deprecated
from wandb.proto.wandb_internal_pb2 import ServerFeature
from wandb.sdk.artifacts._validators import is_artifact_registry_project
from wandb.sdk.internal.internal_api import Api as InternalApi
from wandb.sdk.internal.thread_local_settings import _thread_local_api_settings
from wandb.sdk.launch.utils import LAUNCH_DEFAULT_PROJECT
from wandb.sdk.lib import retry, runid
from wandb.sdk.lib.deprecate import deprecate
from wandb.sdk.lib.gql_request import GraphQLSession
if TYPE_CHECKING:
from wandb.automations import (
ActionType,
Automation,
EventType,
Integration,
NewAutomation,
SlackIntegration,
WebhookIntegration,
)
from wandb.automations._utils import WriteAutomationsKwargs
logger = logging.getLogger(__name__)
class RetryingClient:
INFO_QUERY = gql(
"""
query ServerInfo{
serverInfo {
cliVersionInfo
latestLocalVersionInfo {
outOfDate
latestVersionString
versionOnThisInstanceString
}
}
}
"""
)
def __init__(self, client: Client):
self._server_info = None
self._client = client
@property
def app_url(self):
return util.app_url(self._client.transport.url.replace("/graphql", "")) + "/"
@retry.retriable(
retry_timedelta=RETRY_TIMEDELTA,
check_retry_fn=util.no_retry_auth,
retryable_exceptions=(RetryError, requests.RequestException),
)
def execute(
self, *args, **kwargs
): # User not encouraged to use this class directly
try:
return self._client.execute(*args, **kwargs)
except requests.exceptions.ReadTimeout:
if "timeout" not in kwargs:
timeout = self._client.transport.default_timeout
wandb.termwarn(
f"A graphql request initiated by the public wandb API timed out (timeout={timeout} sec). "
f"Create a new API with an integer timeout larger than {timeout}, e.g., `api = wandb.Api(timeout={timeout + 10})` "
f"to increase the graphql timeout."
)
raise
@property
def server_info(self):
if self._server_info is None:
self._server_info = self.execute(self.INFO_QUERY).get("serverInfo")
return self._server_info
def version_supported(
self, min_version: str
) -> bool: # User not encouraged to use this class directly
from packaging.version import parse
return parse(min_version) <= parse(
self.server_info["cliVersionInfo"]["max_cli_version"]
)
class Api:
"""Used for querying the wandb server.
Examples:
Most common way to initialize
>>> wandb.Api()
Args:
overrides: (dict) You can set `base_url` if you are using a wandb server
other than https://api.wandb.ai.
You can also set defaults for `entity`, `project`, and `run`.
"""
_HTTP_TIMEOUT = env.get_http_timeout(19)
DEFAULT_ENTITY_QUERY = gql(
"""
query Viewer{
viewer {
id
entity
}
}
"""
)
VIEWER_QUERY = gql(
"""
query Viewer{
viewer {
id
flags
entity
username
email
admin
apiKeys {
edges {
node {
id
name
description
}
}
}
teams {
edges {
node {
name
}
}
}
}
}
"""
)
USERS_QUERY = gql(
"""
query SearchUsers($query: String) {
users(query: $query) {
edges {
node {
id
flags
entity
admin
email
deletedAt
username
apiKeys {
edges {
node {
id
name
description
}
}
}
teams {
edges {
node {
name
}
}
}
}
}
}
}
"""
)
CREATE_PROJECT = gql(
"""
mutation upsertModel(
$description: String
$entityName: String
$id: String
$name: String
$framework: String
$access: String
$views: JSONString
) {
upsertModel(
input: {
description: $description
entityName: $entityName
id: $id
name: $name
framework: $framework
access: $access
views: $views
}
) {
project {
id
name
entityName
description
access
views
}
model {
id
name
entityName
description
access
views
}
inserted
}
}
"""
)
def __init__(
self,
overrides: Optional[Dict[str, Any]] = None,
timeout: Optional[int] = None,
api_key: Optional[str] = None,
) -> None:
self.settings = InternalApi().settings()
_overrides = overrides or {}
self.settings.update(_overrides)
self.settings["base_url"] = self.settings["base_url"].rstrip("/")
if "organization" in _overrides:
self.settings["organization"] = _overrides["organization"]
if "username" in _overrides and "entity" not in _overrides:
wandb.termwarn(
'Passing "username" to Api is deprecated. please use "entity" instead.'
)
self.settings["entity"] = _overrides["username"]
self._api_key = api_key
if self.api_key is None and _thread_local_api_settings.cookies is None:
wandb.login(host=_overrides.get("base_url"))
self._viewer = None
self._projects = {}
self._runs = {}
self._sweeps = {}
self._reports = {}
self._default_entity = None
self._timeout = timeout if timeout is not None else self._HTTP_TIMEOUT
auth = None
if not _thread_local_api_settings.cookies:
auth = ("api", self.api_key)
proxies = self.settings.get("_proxies") or json.loads(
os.environ.get("WANDB__PROXIES", "{}")
)
self._base_client = Client(
transport=GraphQLSession(
headers={
"User-Agent": self.user_agent,
"Use-Admin-Privileges": "true",
**(_thread_local_api_settings.headers or {}),
},
use_json=True,
# this timeout won't apply when the DNS lookup fails. in that case, it will be 60s
# https://bugs.python.org/issue22889
timeout=self._timeout,
auth=auth,
url="{}/graphql".format(self.settings["base_url"]),
cookies=_thread_local_api_settings.cookies,
proxies=proxies,
)
)
self._client = RetryingClient(self._base_client)
def create_project(self, name: str, entity: str) -> None:
"""Create a new project.
Args:
name: (str) The name of the new project.
entity: (str) The entity of the new project.
"""
self.client.execute(self.CREATE_PROJECT, {"entityName": entity, "name": name})
def create_run(
self,
*,
run_id: Optional[str] = None,
project: Optional[str] = None,
entity: Optional[str] = None,
) -> "public.Run":
"""Create a new run.
Args:
run_id: (str, optional) The ID to assign to the run, if given. The run ID is automatically generated by
default, so in general, you do not need to specify this and should only do so at your own risk.
project: (str, optional) If given, the project of the new run.
entity: (str, optional) If given, the entity of the new run.
Returns:
The newly created `Run`.
"""
if entity is None:
entity = self.default_entity
return public.Run.create(self, run_id=run_id, project=project, entity=entity)
def create_run_queue(
self,
name: str,
type: "public.RunQueueResourceType",
entity: Optional[str] = None,
prioritization_mode: Optional["public.RunQueuePrioritizationMode"] = None,
config: Optional[dict] = None,
template_variables: Optional[dict] = None,
) -> "public.RunQueue":
"""Create a new run queue (launch).
Args:
name: (str) Name of the queue to create
type: (str) Type of resource to be used for the queue. One of "local-container", "local-process", "kubernetes", "sagemaker", or "gcp-vertex".
entity: (str) Optional name of the entity to create the queue. If None, will use the configured or default entity.
prioritization_mode: (str) Optional version of prioritization to use. Either "V0" or None
config: (dict) Optional default resource configuration to be used for the queue. Use handlebars (eg. `{{var}}`) to specify template variables.
template_variables: (dict) A dictionary of template variable schemas to be used with the config. Expected format of:
`{
"var-name": {
"schema": {
"type": ("string", "number", or "integer"),
"default": (optional value),
"minimum": (optional minimum),
"maximum": (optional maximum),
"enum": [..."(options)"]
}
}
}`
Returns:
The newly created `RunQueue`
Raises:
ValueError if any of the parameters are invalid
wandb.Error on wandb API errors
"""
# TODO(np): Need to check server capabilities for this feature
# 0. assert params are valid/normalized
if entity is None:
entity = self.settings["entity"] or self.default_entity
if entity is None:
raise ValueError(
"entity must be passed as a parameter, or set in settings"
)
if len(name) == 0:
raise ValueError("name must be non-empty")
if len(name) > 64:
raise ValueError("name must be less than 64 characters")
if type not in [
"local-container",
"local-process",
"kubernetes",
"sagemaker",
"gcp-vertex",
]:
raise ValueError(
"resource_type must be one of 'local-container', 'local-process', 'kubernetes', 'sagemaker', or 'gcp-vertex'"
)
if prioritization_mode:
prioritization_mode = prioritization_mode.upper()
if prioritization_mode not in ["V0"]:
raise ValueError("prioritization_mode must be 'V0' if present")
if config is None:
config = {}
# 1. create required default launch project in the entity
self.create_project(LAUNCH_DEFAULT_PROJECT, entity)
api = InternalApi(
default_settings={
"entity": entity,
"project": self.project(LAUNCH_DEFAULT_PROJECT),
},
retry_timedelta=RETRY_TIMEDELTA,
)
# 2. create default resource config, receive config id
config_json = json.dumps({"resource_args": {type: config}})
create_config_result = api.create_default_resource_config(
entity, type, config_json, template_variables
)
if not create_config_result["success"]:
raise wandb.Error("failed to create default resource config")
config_id = create_config_result["defaultResourceConfigID"]
# 3. create run queue
create_queue_result = api.create_run_queue(
entity,
LAUNCH_DEFAULT_PROJECT,
name,
"PROJECT",
prioritization_mode,
config_id,
)
if not create_queue_result["success"]:
raise wandb.Error("failed to create run queue")
return public.RunQueue(
client=self.client,
name=name,
entity=entity,
prioritization_mode=prioritization_mode,
_access="PROJECT",
_default_resource_config_id=config_id,
_default_resource_config=config,
)
def upsert_run_queue(
self,
name: str,
resource_config: dict,
resource_type: "public.RunQueueResourceType",
entity: Optional[str] = None,
template_variables: Optional[dict] = None,
external_links: Optional[dict] = None,
prioritization_mode: Optional["public.RunQueuePrioritizationMode"] = None,
):
"""Upsert a run queue (launch).
Args:
name: (str) Name of the queue to create
entity: (str) Optional name of the entity to create the queue. If None, will use the configured or default entity.
resource_config: (dict) Optional default resource configuration to be used for the queue. Use handlebars (eg. `{{var}}`) to specify template variables.
resource_type: (str) Type of resource to be used for the queue. One of "local-container", "local-process", "kubernetes", "sagemaker", or "gcp-vertex".
template_variables: (dict) A dictionary of template variable schemas to be used with the config. Expected format of:
`{
"var-name": {
"schema": {
"type": ("string", "number", or "integer"),
"default": (optional value),
"minimum": (optional minimum),
"maximum": (optional maximum),
"enum": [..."(options)"]
}
}
}`
external_links: (dict) Optional dictionary of external links to be used with the queue. Expected format of:
`{
"name": "url"
}`
prioritization_mode: (str) Optional version of prioritization to use. Either "V0" or None
Returns:
The upserted `RunQueue`.
Raises:
ValueError if any of the parameters are invalid
wandb.Error on wandb API errors
"""
if entity is None:
entity = self.settings["entity"] or self.default_entity
if entity is None:
raise ValueError(
"entity must be passed as a parameter, or set in settings"
)
if len(name) == 0:
raise ValueError("name must be non-empty")
if len(name) > 64:
raise ValueError("name must be less than 64 characters")
prioritization_mode = prioritization_mode or "DISABLED"
prioritization_mode = prioritization_mode.upper()
if prioritization_mode not in ["V0", "DISABLED"]:
raise ValueError(
"prioritization_mode must be 'V0' or 'DISABLED' if present"
)
if resource_type not in [
"local-container",
"local-process",
"kubernetes",
"sagemaker",
"gcp-vertex",
]:
raise ValueError(
"resource_type must be one of 'local-container', 'local-process', 'kubernetes', 'sagemaker', or 'gcp-vertex'"
)
self.create_project(LAUNCH_DEFAULT_PROJECT, entity)
api = InternalApi(
default_settings={
"entity": entity,
"project": self.project(LAUNCH_DEFAULT_PROJECT),
},
retry_timedelta=RETRY_TIMEDELTA,
)
# User provides external_links as a dict with name: url format
# but backend stores it as a list of dicts with url and label keys.
external_links = external_links or {}
external_links = {
"links": [
{
"label": key,
"url": value,
}
for key, value in external_links.items()
]
}
upsert_run_queue_result = api.upsert_run_queue(
name,
entity,
resource_type,
{"resource_args": {resource_type: resource_config}},
template_variables=template_variables,
external_links=external_links,
prioritization_mode=prioritization_mode,
)
if not upsert_run_queue_result["success"]:
raise wandb.Error("failed to create run queue")
schema_errors = (
upsert_run_queue_result.get("configSchemaValidationErrors") or []
)
for error in schema_errors:
wandb.termwarn(f"resource config validation: {error}")
return public.RunQueue(
client=self.client,
name=name,
entity=entity,
)
def create_user(self, email, admin=False):
"""Create a new user.
Args:
email: (str) The email address of the user
admin: (bool) Whether this user should be a global instance admin
Returns:
A `User` object
"""
return public.User.create(self, email, admin)
def sync_tensorboard(self, root_dir, run_id=None, project=None, entity=None):
"""Sync a local directory containing tfevent files to wandb."""
from wandb.sync import SyncManager # TODO: circular import madness
run_id = run_id or runid.generate_id()
project = project or self.settings.get("project") or "uncategorized"
entity = entity or self.default_entity
# TODO: pipe through log_path to inform the user how to debug
sm = SyncManager(
project=project,
entity=entity,
run_id=run_id,
mark_synced=False,
app_url=self.client.app_url,
view=False,
verbose=False,
sync_tensorboard=True,
)
sm.add(root_dir)
sm.start()
while not sm.is_done():
_ = sm.poll()
return self.run("/".join([entity, project, run_id]))
@property
def client(self) -> RetryingClient:
return self._client
@property
def user_agent(self) -> str:
return "W&B Public Client {}".format(wandb.__version__)
@property
def api_key(self) -> Optional[str]:
# just use thread local api key if it's set
if _thread_local_api_settings.api_key:
return _thread_local_api_settings.api_key
if self._api_key is not None:
return self._api_key
auth = requests.utils.get_netrc_auth(self.settings["base_url"])
key = None
if auth:
key = auth[-1]
# Environment should take precedence
if os.getenv("WANDB_API_KEY"):
key = os.environ["WANDB_API_KEY"]
self._api_key = key # memoize key
return key
@property
def default_entity(self) -> Optional[str]:
if self._default_entity is None:
res = self._client.execute(self.DEFAULT_ENTITY_QUERY)
self._default_entity = (res.get("viewer") or {}).get("entity")
return self._default_entity
@property
def viewer(self) -> "public.User":
if self._viewer is None:
self._viewer = public.User(
self._client, self._client.execute(self.VIEWER_QUERY).get("viewer")
)
self._default_entity = self._viewer.entity
return self._viewer
def flush(self):
"""Flush the local cache.
The api object keeps a local cache of runs, so if the state of the run may
change while executing your script you must clear the local cache with
`api.flush()` to get the latest values associated with the run.
"""
self._runs = {}
def from_path(self, path):
"""Return a run, sweep, project or report from a path.
Examples:
```
project = api.from_path("my_project")
team_project = api.from_path("my_team/my_project")
run = api.from_path("my_team/my_project/runs/id")
sweep = api.from_path("my_team/my_project/sweeps/id")
report = api.from_path("my_team/my_project/reports/My-Report-Vm11dsdf")
```
Args:
path: (str) The path to the project, run, sweep or report
Returns:
A `Project`, `Run`, `Sweep`, or `BetaReport` instance.
Raises:
wandb.Error if path is invalid or the object doesn't exist
"""
parts = path.strip("/ ").split("/")
if len(parts) == 1:
return self.project(path)
elif len(parts) == 2:
return self.project(parts[1], parts[0])
elif len(parts) == 3:
return self.run(path)
elif len(parts) == 4:
if parts[2].startswith("run"):
return self.run(path)
elif parts[2].startswith("sweep"):
return self.sweep(path)
elif parts[2].startswith("report"):
if "--" not in parts[-1]:
if "-" in parts[-1]:
raise wandb.Error(
"Invalid report path, should be team/project/reports/Name--XXXX"
)
else:
parts[-1] = "--" + parts[-1]
name, id = parts[-1].split("--")
return public.BetaReport(
self.client,
{
"display_name": urllib.parse.unquote(name.replace("-", " ")),
"id": id,
"spec": "{}",
},
parts[0],
parts[1],
)
raise wandb.Error(
"Invalid path, should be TEAM/PROJECT/TYPE/ID where TYPE is runs, sweeps, or reports"
)
def _parse_project_path(self, path):
"""Return project and entity for project specified by path."""
project = self.settings["project"] or "uncategorized"
entity = self.settings["entity"] or self.default_entity
if path is None:
return entity, project
parts = path.split("/", 1)
if len(parts) == 1:
return entity, path
return parts
def _parse_path(self, path):
"""Parse url, filepath, or docker paths.
Allows paths in the following formats:
- url: entity/project/runs/id
- path: entity/project/id
- docker: entity/project:id
Entity is optional and will fall back to the current logged-in user.
"""
project = self.settings["project"] or "uncategorized"
entity = self.settings["entity"] or self.default_entity
parts = (
path.replace("/runs/", "/").replace("/sweeps/", "/").strip("/ ").split("/")
)
if ":" in parts[-1]:
id = parts[-1].split(":")[-1]
parts[-1] = parts[-1].split(":")[0]
elif parts[-1]:
id = parts[-1]
if len(parts) == 1 and project != "uncategorized":
pass
elif len(parts) > 1:
project = parts[1]
if entity and id == project:
project = parts[0]
else:
entity = parts[0]
if len(parts) == 3:
entity = parts[0]
else:
project = parts[0]
return entity, project, id
def _parse_artifact_path(self, path):
"""Return project, entity and artifact name for project specified by path."""
project = self.settings["project"] or "uncategorized"
entity = self.settings["entity"] or self.default_entity
if path is None:
return entity, project
path, colon, alias = path.partition(":")
full_alias = colon + alias
parts = path.split("/")
if len(parts) > 3:
raise ValueError("Invalid artifact path: {}".format(path))
elif len(parts) == 1:
return entity, project, path + full_alias
elif len(parts) == 2:
return entity, parts[0], parts[1] + full_alias
parts[-1] += full_alias
return parts
def projects(
self, entity: Optional[str] = None, per_page: int = 200
) -> "public.Projects":
"""Get projects for a given entity.
Args:
entity: (str) Name of the entity requested. If None, will fall back to the
default entity passed to `Api`. If no default entity, will raise a `ValueError`.
per_page: (int) Sets the page size for query pagination. Usually there is no reason to change this.
Returns:
A `Projects` object which is an iterable collection of `Project` objects.
"""
if entity is None:
entity = self.settings["entity"] or self.default_entity
if entity is None:
raise ValueError(
"entity must be passed as a parameter, or set in settings"
)
if entity not in self._projects:
self._projects[entity] = public.Projects(
self.client, entity, per_page=per_page
)
return self._projects[entity]
def project(self, name: str, entity: Optional[str] = None) -> "public.Project":
"""Return the `Project` with the given name (and entity, if given).
Args:
name: (str) The project name.
entity: (str) Name of the entity requested. If None, will fall back to the
default entity passed to `Api`. If no default entity, will raise a `ValueError`.
Returns:
A `Project` object.
"""
# For registry artifacts, capture potential org user inputted before resolving entity
org = entity if is_artifact_registry_project(name) else ""
if entity is None:
entity = self.settings["entity"] or self.default_entity
# For registry artifacts, resolve org-based entity
if is_artifact_registry_project(name):
settings_entity = self.settings["entity"] or self.default_entity
entity = InternalApi()._resolve_org_entity_name(
entity=settings_entity, organization=org
)
return public.Project(self.client, entity, name, {})
def reports(
self, path: str = "", name: Optional[str] = None, per_page: int = 50
) -> "public.Reports":
"""Get reports for a given project path.
WARNING: This api is in beta and will likely change in a future release
Args:
path: (str) path to project the report resides in, should be in the form: "entity/project"
name: (str, optional) optional name of the report requested.
per_page: (int) Sets the page size for query pagination. Usually there is no reason to change this.
Returns:
A `Reports` object which is an iterable collection of `BetaReport` objects.
"""
entity, project, _ = self._parse_path(path + "/fake_run")
if name:
name = urllib.parse.unquote(name)
key = "/".join([entity, project, str(name)])
else:
key = "/".join([entity, project])
if key not in self._reports:
self._reports[key] = public.Reports(
self.client,
public.Project(self.client, entity, project, {}),
name=name,
per_page=per_page,
)
return self._reports[key]
def create_team(self, team, admin_username=None):
"""Create a new team.
Args:
team: (str) The name of the team
admin_username: (str) optional username of the admin user of the team, defaults to the current user.
Returns:
A `Team` object
"""
return public.Team.create(self, team, admin_username)
def team(self, team: str) -> "public.Team":
"""Return the matching `Team` with the given name.
Args:
team: (str) The name of the team.
Returns:
A `Team` object.
"""
return public.Team(self.client, team)
def user(self, username_or_email: str) -> Optional["public.User"]:
"""Return a user from a username or email address.
Note: This function only works for Local Admins, if you are trying to get your own user object, please use `api.viewer`.
Args:
username_or_email: (str) The username or email address of the user
Returns:
A `User` object or None if a user couldn't be found
"""
res = self._client.execute(self.USERS_QUERY, {"query": username_or_email})
if len(res["users"]["edges"]) == 0:
return None
elif len(res["users"]["edges"]) > 1:
wandb.termwarn(
"Found multiple users, returning the first user matching {}".format(
username_or_email
)
)
return public.User(self._client, res["users"]["edges"][0]["node"])
def users(self, username_or_email: str) -> List["public.User"]:
"""Return all users from a partial username or email address query.
Note: This function only works for Local Admins, if you are trying to get your own user object, please use `api.viewer`.
Args:
username_or_email: (str) The prefix or suffix of the user you want to find
Returns:
An array of `User` objects
"""
res = self._client.execute(self.USERS_QUERY, {"query": username_or_email})
return [
public.User(self._client, edge["node"]) for edge in res["users"]["edges"]
]
def runs(
self,
path: Optional[str] = None,
filters: Optional[Dict[str, Any]] = None,
order: str = "+created_at",
per_page: int = 50,
include_sweeps: bool = True,
):
"""Return a set of runs from a project that match the filters provided.
Fields you can filter by include:
- `createdAt`: The timestamp when the run was created. (in ISO 8601 format, e.g. "2023-01-01T12:00:00Z")
- `displayName`: The human-readable display name of the run. (e.g. "eager-fox-1")
- `duration`: The total runtime of the run in seconds.
- `group`: The group name used to organize related runs together.
- `host`: The hostname where the run was executed.
- `jobType`: The type of job or purpose of the run.
- `name`: The unique identifier of the run. (e.g. "a1b2cdef")
- `state`: The current state of the run.
- `tags`: The tags associated with the run.
- `username`: The username of the user who initiated the run
Additionally, you can filter by items in the run config or summary metrics.
Such as `config.experiment_name`, `summary_metrics.loss`, etc.
For more complex filtering, you can use MongoDB query operators.
For details, see: https://docs.mongodb.com/manual/reference/operator/query
The following operations are supported:
- `$and`
- `$or`
- `$nor`
- `$eq`
- `$ne`
- `$gt`
- `$gte`
- `$lt`
- `$lte`
- `$in`
- `$nin`
- `$exists`
- `$regex`
Examples:
Find runs in my_project where config.experiment_name has been set to "foo"
```
api.runs(
path="my_entity/my_project",
filters={"config.experiment_name": "foo"},
)
```
Find runs in my_project where config.experiment_name has been set to "foo" or "bar"
```
api.runs(
path="my_entity/my_project",
filters={
"$or": [
{"config.experiment_name": "foo"},
{"config.experiment_name": "bar"},
]
},
)
```
Find runs in my_project where config.experiment_name matches a regex (anchors are not supported)
```
api.runs(
path="my_entity/my_project",
filters={"config.experiment_name": {"$regex": "b.*"}},
)
```
Find runs in my_project where the run name matches a regex (anchors are not supported)
```
api.runs(
path="my_entity/my_project",
filters={"display_name": {"$regex": "^foo.*"}},
)
```
Find runs in my_project where config.experiment contains a nested field "category" with value "testing"
```
api.runs(
path="my_entity/my_project",
filters={"config.experiment.category": "testing"},
)
```
Find runs in my_project with a loss value of 0.5 nested in a dictionary under model1 in the summary metrics
```
api.runs(
path="my_entity/my_project",
filters={"summary_metrics.model1.loss": 0.5},
)
```
Find runs in my_project sorted by ascending loss
```
api.runs(path="my_entity/my_project", order="+summary_metrics.loss")
```
Args:
path: (str) path to project, should be in the form: "entity/project"
filters: (dict) queries for specific runs using the MongoDB query language.
You can filter by run properties such as config.key, summary_metrics.key, state, entity, createdAt, etc.
For example: `{"config.experiment_name": "foo"}` would find runs with a config entry
of experiment name set to "foo"
order: (str) Order can be `created_at`, `heartbeat_at`, `config.*.value`, or `summary_metrics.*`.
If you prepend order with a + order is ascending.
If you prepend order with a - order is descending (default).
The default order is run.created_at from oldest to newest.
per_page: (int) Sets the page size for query pagination.
include_sweeps: (bool) Whether to include the sweep runs in the results.
Returns:
A `Runs` object, which is an iterable collection of `Run` objects.
"""
entity, project = self._parse_project_path(path)
filters = filters or {}
key = (path or "") + str(filters) + str(order)
if not self._runs.get(key):
self._runs[key] = public.Runs(
self.client,
entity,
project,
filters=filters,
order=order,
per_page=per_page,
include_sweeps=include_sweeps,
)
return self._runs[key]
@normalize_exceptions
def run(self, path=""):
"""Return a single run by parsing path in the form entity/project/run_id.
Args:
path: (str) path to run in the form `entity/project/run_id`.
If `api.entity` is set, this can be in the form `project/run_id`
and if `api.project` is set this can just be the run_id.
Returns:
A `Run` object.
"""
entity, project, run_id = self._parse_path(path)
if not self._runs.get(path):
self._runs[path] = public.Run(self.client, entity, project, run_id)
return self._runs[path]
def queued_run(
self,
entity,
project,
queue_name,
run_queue_item_id,
project_queue=None,
priority=None,
):
"""Return a single queued run based on the path.
Parses paths of the form entity/project/queue_id/run_queue_item_id.
"""
return public.QueuedRun(
self.client,
entity,
project,
queue_name,
run_queue_item_id,
project_queue=project_queue,
priority=priority,
)
def run_queue(
self,
entity,
name,
):
"""Return the named `RunQueue` for entity.
To create a new `RunQueue`, use `wandb.Api().create_run_queue(...)`.
"""
return public.RunQueue(
self.client,
name,
entity,
)
@normalize_exceptions
def sweep(self, path=""):
"""Return a sweep by parsing path in the form `entity/project/sweep_id`.
Args:
path: (str, optional) path to sweep in the form entity/project/sweep_id. If `api.entity`
is set, this can be in the form project/sweep_id and if `api.project` is set
this can just be the sweep_id.
Returns:
A `Sweep` object.
"""
entity, project, sweep_id = self._parse_path(path)
if not self._sweeps.get(path):
self._sweeps[path] = public.Sweep(self.client, entity, project, sweep_id)
return self._sweeps[path]
@normalize_exceptions
def artifact_types(self, project: Optional[str] = None) -> "public.ArtifactTypes":
"""Return a collection of matching artifact types.
Args:
project: (str, optional) If given, a project name or path to filter on.
Returns:
An iterable `ArtifactTypes` object.
"""
project_path = project
entity, project = self._parse_project_path(project_path)
# If its a Registry project, the entity is considered to be an org instead
if is_artifact_registry_project(project):
settings_entity = self.settings["entity"] or self.default_entity
org = parse_org_from_registry_path(project_path, PathType.PROJECT)
entity = InternalApi()._resolve_org_entity_name(
entity=settings_entity, organization=org
)
return public.ArtifactTypes(self.client, entity, project)
@normalize_exceptions
def artifact_type(
self, type_name: str, project: Optional[str] = None
) -> "public.ArtifactType":
"""Return the matching `ArtifactType`.
Args:
type_name: (str) The name of the artifact type to retrieve.
project: (str, optional) If given, a project name or path to filter on.
Returns:
An `ArtifactType` object.
"""
project_path = project
entity, project = self._parse_project_path(project_path)
# If its an Registry artifact, the entity is an org instead
if is_artifact_registry_project(project):
org = parse_org_from_registry_path(project_path, PathType.PROJECT)
settings_entity = self.settings["entity"] or self.default_entity
entity = InternalApi()._resolve_org_entity_name(
entity=settings_entity, organization=org
)
return public.ArtifactType(self.client, entity, project, type_name)
@normalize_exceptions
def artifact_collections(
self, project_name: str, type_name: str, per_page: int = 50
) -> "public.ArtifactCollections":
"""Return a collection of matching artifact collections.
Args:
project_name: (str) The name of the project to filter on.
type_name: (str) The name of the artifact type to filter on.
per_page: (int) Sets the page size for query pagination. Usually there is no reason to change this.
Returns:
An iterable `ArtifactCollections` object.
"""
entity, project = self._parse_project_path(project_name)
# If iterating through Registry project, the entity is considered to be an org instead
if is_artifact_registry_project(project):
org = parse_org_from_registry_path(project_name, PathType.PROJECT)
settings_entity = self.settings["entity"] or self.default_entity
entity = InternalApi()._resolve_org_entity_name(
entity=settings_entity, organization=org
)
return public.ArtifactCollections(
self.client, entity, project, type_name, per_page=per_page
)
@normalize_exceptions
def artifact_collection(
self, type_name: str, name: str
) -> "public.ArtifactCollection":
"""Return a single artifact collection by type and parsing path in the form `entity/project/name`.
Args:
type_name: (str) The type of artifact collection to fetch.
name: (str) An artifact collection name. May be prefixed with entity/project.
Returns:
An `ArtifactCollection` object.
"""
entity, project, collection_name = self._parse_artifact_path(name)
# If its an Registry artifact, the entity is considered to be an org instead
if is_artifact_registry_project(project):
org = parse_org_from_registry_path(name, PathType.ARTIFACT)
settings_entity = self.settings["entity"] or self.default_entity
entity = InternalApi()._resolve_org_entity_name(
entity=settings_entity, organization=org
)
if entity is None:
raise ValueError(
"Could not determine entity. Please include the entity as part of the collection name path."
)
return public.ArtifactCollection(
self.client, entity, project, collection_name, type_name
)
@normalize_exceptions
def artifact_versions(self, type_name, name, per_page=50):
"""Deprecated, use `artifacts(type_name, name)` instead."""
deprecate(
field_name=Deprecated.api__artifact_versions,
warning_message=(
"Api.artifact_versions(type_name, name) is deprecated, "
"use Api.artifacts(type_name, name) instead."
),
)
return self.artifacts(type_name, name, per_page=per_page)
@normalize_exceptions
def artifacts(
self,
type_name: str,
name: str,
per_page: int = 50,
tags: Optional[List[str]] = None,
) -> "public.Artifacts":
"""Return an `Artifacts` collection from the given parameters.
Args:
type_name: (str) The type of artifacts to fetch.
name: (str) An artifact collection name. May be prefixed with entity/project.
per_page: (int) Sets the page size for query pagination. Usually there is no reason to change this.
tags: (list[str], optional) Only return artifacts with all of these tags.
Returns:
An iterable `Artifacts` object.
"""
entity, project, collection_name = self._parse_artifact_path(name)
# If its an Registry project, the entity is considered to be an org instead
if is_artifact_registry_project(project):
org = parse_org_from_registry_path(name, PathType.ARTIFACT)
settings_entity = self.settings["entity"] or self.default_entity
entity = InternalApi()._resolve_org_entity_name(
entity=settings_entity, organization=org
)
return public.Artifacts(
self.client,
entity,
project,
collection_name,
type_name,
per_page=per_page,
tags=tags,
)
@normalize_exceptions
def _artifact(
self, name: str, type: Optional[str] = None, enable_tracking: bool = False
):
if name is None:
raise ValueError("You must specify name= to fetch an artifact.")
entity, project, artifact_name = self._parse_artifact_path(name)
# If its an Registry artifact, the entity is an org instead
if is_artifact_registry_project(project):
organization = (
name.split("/")[0]
if name.count("/") == 2
else self.settings["organization"]
)
# set entity to match the settings since in above code it was potentially set to an org
settings_entity = self.settings["entity"] or self.default_entity
# Registry artifacts are under the org entity. Because we offer a shorthand and alias for this path,
# we need to fetch the org entity to for the user behind the scenes.
entity = InternalApi()._resolve_org_entity_name(
entity=settings_entity, organization=organization
)
if entity is None:
raise ValueError(
"Could not determine entity. Please include the entity as part of the artifact name path."
)
artifact = wandb.Artifact._from_name(
entity=entity,
project=project,
name=artifact_name,
client=self.client,
enable_tracking=enable_tracking,
)
if type is not None and artifact.type != type:
raise ValueError(
f"type {type} specified but this artifact is of type {artifact.type}"
)
return artifact
@normalize_exceptions
def artifact(self, name: str, type: Optional[str] = None):
"""Return a single artifact by parsing path in the form `project/name` or `entity/project/name`.
Args:
name: (str) An artifact name. May be prefixed with project/ or entity/project/.
If no entity is specified in the name, the Run or API setting's entity is used.
Valid names can be in the following forms:
name:version
name:alias
type: (str, optional) The type of artifact to fetch.
Returns:
An `Artifact` object.
Raises:
ValueError: If the artifact name is not specified.
ValueError: If the artifact type is specified but does not match the type of the fetched artifact.
Note:
This method is intended for external use only. Do not call `api.artifact()` within the wandb repository code.
"""
return self._artifact(name=name, type=type, enable_tracking=True)
@normalize_exceptions
def job(self, name: Optional[str], path: Optional[str] = None) -> "public.Job":
"""Return a `Job` from the given parameters.
Args:
name: (str) The job name.
path: (str, optional) If given, the root path in which to download the job artifact.
Returns:
A `Job` object.
"""
if name is None:
raise ValueError("You must specify name= to fetch a job.")
elif name.count("/") != 2 or ":" not in name:
raise ValueError(
"Invalid job specification. A job must be of the form: <entity>/<project>/<job-name>:<alias-or-version>"
)
return public.Job(self, name, path)
@normalize_exceptions
def list_jobs(self, entity: str, project: str) -> List[Dict[str, Any]]:
"""Return a list of jobs, if any, for the given entity and project.
Args:
entity: (str) The entity for the listed job(s).
project: (str) The project for the listed job(s).
Returns:
A list of matching jobs.
"""
if entity is None:
raise ValueError("Specify an entity when listing jobs")
if project is None:
raise ValueError("Specify a project when listing jobs")
query = gql(
"""
query ArtifactOfType(
$entityName: String!,
$projectName: String!,
$artifactTypeName: String!,
) {
project(name: $projectName, entityName: $entityName) {
artifactType(name: $artifactTypeName) {
artifactCollections {
edges {
node {
artifacts {
edges {
node {
id
state
aliases {
alias
}
artifactSequence {
name
}
}
}
}
}
}
}
}
}
}
"""
)
try:
artifact_query = self._client.execute(
query,
{
"projectName": project,
"entityName": entity,
"artifactTypeName": "job",
},
)
if not artifact_query or not artifact_query["project"]:
wandb.termerror(
f"Project: '{project}' not found in entity: '{entity}' or access denied."
)
return []
if artifact_query["project"]["artifactType"] is None:
return []
artifacts = artifact_query["project"]["artifactType"][
"artifactCollections"
]["edges"]
return [x["node"]["artifacts"] for x in artifacts]
except requests.exceptions.HTTPError:
return False
@normalize_exceptions
def artifact_exists(self, name: str, type: Optional[str] = None) -> bool:
"""Return whether an artifact version exists within a specified project and entity.
Args:
name: (str) An artifact name. May be prefixed with entity/project.
If entity or project is not specified, it will be inferred from the override params if populated.
Otherwise, entity will be pulled from the user settings and project will default to "uncategorized".
Valid names can be in the following forms:
name:version
name:alias
type: (str, optional) The type of artifact
Returns:
True if the artifact version exists, False otherwise.
"""
try:
self._artifact(name, type)
except wandb.errors.CommError:
return False
return True
@normalize_exceptions
def artifact_collection_exists(self, name: str, type: str) -> bool:
"""Return whether an artifact collection exists within a specified project and entity.
Args:
name: (str) An artifact collection name. May be prefixed with entity/project.
If entity or project is not specified, it will be inferred from the override params if populated.
Otherwise, entity will be pulled from the user settings and project will default to "uncategorized".
type: (str) The type of artifact collection
Returns:
True if the artifact collection exists, False otherwise.
"""
try:
self.artifact_collection(type, name)
except wandb.errors.CommError:
return False
return True
def registries(
self,
organization: Optional[str] = None,
filter: Optional[Dict[str, Any]] = None,
) -> Registries:
"""Returns a Registry iterator.
Use the iterator to search and filter registries, collections,
or artifact versions across your organization's registry.
Examples:
Find all registries with the names that contain "model"
```python
import wandb
api = wandb.Api() # specify an org if your entity belongs to multiple orgs
api.registries(filter={"name": {"$regex": "model"}})
```
Find all collections in the registries with the name "my_collection" and the tag "my_tag"
```python
api.registries().collections(filter={"name": "my_collection", "tag": "my_tag"})
```
Find all artifact versions in the registries with a collection name that contains "my_collection" and a version that has the alias "best"
```python
api.registries().collections(
filter={"name": {"$regex": "my_collection"}}
).versions(filter={"alias": "best"})
```
Find all artifact versions in the registries that contain "model" and have the tag "prod" or alias "best"
```python
api.registries(filter={"name": {"$regex": "model"}}).versions(
filter={"$or": [{"tag": "prod"}, {"alias": "best"}]}
)
```
Args:
organization: (str, optional) The organization of the registry to fetch.
If not specified, use the organization specified in the user's settings.
filter: (dict, optional) MongoDB-style filter to apply to each object in the registry iterator.
Fields available to filter for collections are
`name`, `description`, `created_at`, `updated_at`.
Fields available to filter for collections are
`name`, `tag`, `description`, `created_at`, `updated_at`
Fields available to filter for versions are
`tag`, `alias`, `created_at`, `updated_at`, `metadata`
Returns:
A registry iterator.
"""
if not InternalApi()._server_supports(ServerFeature.ARTIFACT_REGISTRY_SEARCH):
raise RuntimeError(
"Registry search API is not enabled on this wandb server version. "
"Please upgrade your server version or contact support at support@wandb.com."
)
organization = organization or fetch_org_from_settings_or_entity(
self.settings, self.default_entity
)
return Registries(self.client, organization, filter)
def registry(self, name: str, organization: Optional[str] = None) -> Registry:
"""Return a registry given a registry name.
Args:
name: The name of the registry. This is without the `wandb-registry-`
prefix.
organization: The organization of the registry.
If no organization is set in the settings, the organization will be
fetched from the entity if the entity only belongs to one
organization.
Returns:
A registry object.
Examples:
Fetch and update a registry
```python
import wandb
api = wandb.Api()
registry = api.registry(name="my-registry", organization="my-org")
registry.description = "This is an updated description"
registry.save()
```
"""
if not InternalApi()._server_supports(ServerFeature.ARTIFACT_REGISTRY_SEARCH):
raise RuntimeError(
"api.registry() is not enabled on this wandb server version. "
"Please upgrade your server version or contact support at support@wandb.com."
)
organization = organization or fetch_org_from_settings_or_entity(
self.settings, self.default_entity
)
org_entity = _fetch_org_entity_from_organization(self.client, organization)
registry = Registry(self.client, organization, org_entity, name)
registry.load()
return registry
def create_registry(
self,
name: str,
visibility: Literal["organization", "restricted"],
organization: Optional[str] = None,
description: Optional[str] = None,
artifact_types: Optional[List[str]] = None,
) -> Registry:
"""Create a new registry.
Args:
name: The name of the registry. Name must be unique within the organization.
visibility: The visibility of the registry.
organization: Anyone in the organization can view this registry. You can
edit their roles later from the settings in the UI.
restricted: Only invited members via the UI can access this registry.
Public sharing is disabled.
organization: The organization of the registry.
If no organization is set in the settings, the organization will be
fetched from the entity if the entity only belongs to one organization.
description: The description of the registry.
artifact_types: The accepted artifact types of the registry. A type is no
more than 128 characters and do not include characters `/` or `:`. If
not specified, all types are accepted.
Allowed types added to the registry cannot be removed later.
Returns:
A registry object.
Examples:
```python
import wandb
api = wandb.Api()
registry = api.create_registry(
name="my-registry",
visibility="restricted",
organization="my-org",
description="This is a test registry",
artifact_types=["model"],
)
```
"""
if not InternalApi()._server_supports(
ServerFeature.INCLUDE_ARTIFACT_TYPES_IN_REGISTRY_CREATION
):
raise RuntimeError(
"create_registry api is not enabled on this wandb server version. "
"Please upgrade your server version or contact support at support@wandb.com."
)
organization = organization or fetch_org_from_settings_or_entity(
self.settings, self.default_entity
)
try:
existing_registry = self.registry(name=name, organization=organization)
except ValueError:
existing_registry = None
if existing_registry:
raise ValueError(
f"Registry {name!r} already exists in organization {organization!r},"
" please use a different name."
)
return Registry.create(
self.client,
organization,
name,
visibility,
description,
artifact_types,
)
def integrations(
self,
entity: Optional[str] = None,
*,
per_page: int = 50,
) -> Iterator["Integration"]:
"""Return an iterator of all integrations for an entity.
Args:
entity: The entity (e.g. team name) for which to
fetch integrations. If not provided, the user's default entity
will be used.
per_page: Number of integrations to fetch per page.
Defaults to 50. Usually there is no reason to change this.
Yields:
Iterator[SlackIntegration | WebhookIntegration]: An iterator of any supported integrations.
"""
from wandb.apis.public.integrations import Integrations
params = {"entityName": entity or self.default_entity}
return Integrations(client=self.client, variables=params, per_page=per_page)
def webhook_integrations(
self, entity: Optional[str] = None, *, per_page: int = 50
) -> Iterator["WebhookIntegration"]:
"""Returns an iterator of webhook integrations for an entity.
Args:
entity: The entity (e.g. team name) for which to
fetch integrations. If not provided, the user's default entity
will be used.
per_page: Number of integrations to fetch per page.
Defaults to 50. Usually there is no reason to change this.
Yields:
Iterator[WebhookIntegration]: An iterator of webhook integrations.
Examples:
Get all registered webhook integrations for the team "my-team":
```python
import wandb
api = wandb.Api()
webhook_integrations = api.webhook_integrations(entity="my-team")
```
Find only webhook integrations that post requests to "https://my-fake-url.com":
```python
webhook_integrations = api.webhook_integrations(entity="my-team")
my_webhooks = [
ig
for ig in webhook_integrations
if ig.url_endpoint.startswith("https://my-fake-url.com")
]
```
"""
from wandb.apis.public.integrations import WebhookIntegrations
params = {"entityName": entity or self.default_entity}
return WebhookIntegrations(
client=self.client, variables=params, per_page=per_page
)
def slack_integrations(
self, *, entity: Optional[str] = None, per_page: int = 50
) -> Iterator["SlackIntegration"]:
"""Returns an iterator of Slack integrations for an entity.
Args:
entity: The entity (e.g. team name) for which to
fetch integrations. If not provided, the user's default entity
will be used.
per_page: Number of integrations to fetch per page.
Defaults to 50. Usually there is no reason to change this.
Yields:
Iterator[SlackIntegration]: An iterator of Slack integrations.
Examples:
Get all registered Slack integrations for the team "my-team":
```python
import wandb
api = wandb.Api()
slack_integrations = api.slack_integrations(entity="my-team")
```
Find only Slack integrations that post to channel names starting with "team-alerts-":
```python
slack_integrations = api.slack_integrations(entity="my-team")
team_alert_integrations = [
ig
for ig in slack_integrations
if ig.channel_name.startswith("team-alerts-")
]
```
"""
from wandb.apis.public.integrations import SlackIntegrations
params = {"entityName": entity or self.default_entity}
return SlackIntegrations(
client=self.client, variables=params, per_page=per_page
)
def _supports_automation(
self,
*,
event: Optional["EventType"] = None,
action: Optional["ActionType"] = None,
) -> bool:
"""Returns whether the server recognizes the automation event and/or action."""
from wandb.automations._utils import (
ALWAYS_SUPPORTED_ACTIONS,
ALWAYS_SUPPORTED_EVENTS,
)
api = InternalApi()
supports_event = (
(event is None)
or (event in ALWAYS_SUPPORTED_EVENTS)
or api._server_supports(f"AUTOMATION_EVENT_{event.value}")
)
supports_action = (
(action is None)
or (action in ALWAYS_SUPPORTED_ACTIONS)
or api._server_supports(f"AUTOMATION_ACTION_{action.value}")
)
return supports_event and supports_action
def _omitted_automation_fragments(self) -> Set[str]:
"""Returns the names of unsupported automation-related fragments.
Older servers won't recognize newer GraphQL types, so a valid request may
unnecessarily error out because it won't recognize fragments defined on those types.
So e.g. if a server does not support `NO_OP` action types, then the following need to be
removed from the body of the GraphQL request:
- Fragment definition:
```
fragment NoOpActionFields on NoOpTriggeredAction {
noOp
}
```
- Fragment spread in selection set:
```
{
...NoOpActionFields
# ... other fields ...
}
```
"""
from wandb.automations import ActionType
from wandb.automations._generated import (
GenericWebhookActionFields,
NoOpActionFields,
NotificationActionFields,
QueueJobActionFields,
)
# Note: we can't currently define this as a constant outside the method
# and still keep it nearby in this module, because it relies on pydantic v2-only imports
fragment_names: dict[ActionType, str] = {
ActionType.NO_OP: NoOpActionFields.__name__,
ActionType.QUEUE_JOB: QueueJobActionFields.__name__,
ActionType.NOTIFICATION: NotificationActionFields.__name__,
ActionType.GENERIC_WEBHOOK: GenericWebhookActionFields.__name__,
}
return set(
name
for action in ActionType
if (not self._supports_automation(action=action))
and (name := fragment_names.get(action))
)
def automation(
self,
name: str,
*,
entity: Optional[str] = None,
) -> "Automation":
"""Returns the only Automation matching the parameters.
Args:
name: The name of the automation to fetch.
entity: The entity to fetch the automation for.
Raises:
ValueError: If zero or multiple Automations match the search criteria.
Examples:
Get an existing automation named "my-automation":
```python
import wandb
api = wandb.Api()
automation = api.automation(name="my-automation")
```
Get an existing automation named "other-automation", from the entity "my-team":
```python
automation = api.automation(name="other-automation", entity="my-team")
```
"""
return one(
self.automations(entity=entity, name=name),
too_short=ValueError("No automations found"),
too_long=ValueError("Multiple automations found"),
)
def automations(
self,
entity: Optional[str] = None,
*,
name: Optional[str] = None,
per_page: int = 50,
) -> Iterator["Automation"]:
"""Returns an iterator over all Automations that match the given parameters.
If no parameters are provided, the returned iterator will contain all
Automations that the user has access to.
Args:
entity: The entity to fetch the automations for.
name: The name of the automation to fetch.
per_page: The number of automations to fetch per page.
Defaults to 50. Usually there is no reason to change this.
Returns:
A list of automations.
Examples:
Fetch all existing automations for the entity "my-team":
```python
import wandb
api = wandb.Api()
automations = api.automations(entity="my-team")
```
"""
from wandb.apis.public.automations import Automations
from wandb.automations._generated import (
GET_AUTOMATIONS_BY_ENTITY_GQL,
GET_AUTOMATIONS_GQL,
)
# For now, we need to use different queries depending on whether entity is given
variables = {"entityName": entity}
if entity is None:
gql_str = GET_AUTOMATIONS_GQL # Automations for viewer
else:
gql_str = GET_AUTOMATIONS_BY_ENTITY_GQL # Automations for entity
# If needed, rewrite the GraphQL field selection set to omit unsupported fields/fragments/types
omit_fragments = self._omitted_automation_fragments()
query = gql_compat(gql_str, omit_fragments=omit_fragments)
iterator = Automations(
client=self.client, variables=variables, per_page=per_page, _query=query
)
# FIXME: this is crude, move this client-side filtering logic into backend
if name is not None:
iterator = filter(lambda x: x.name == name, iterator)
yield from iterator
@normalize_exceptions
def create_automation(
self,
obj: "NewAutomation",
*,
fetch_existing: bool = False,
**kwargs: Unpack["WriteAutomationsKwargs"],
) -> "Automation":
"""Create a new Automation.
Args:
obj:
The automation to create.
fetch_existing:
If True, and a conflicting automation already exists, attempt
to fetch the existing automation instead of raising an error.
**kwargs:
Any additional values to assign to the automation before
creating it. If given, these will override any values that may
already be set on the automation:
- `name`: The name of the automation.
- `description`: The description of the automation.
- `enabled`: Whether the automation is enabled.
- `scope`: The scope of the automation.
- `event`: The event that triggers the automation.
- `action`: The action that is triggered by the automation.
Returns:
The saved Automation.
Examples:
Create a new automation named "my-automation" that sends a Slack notification
when a run within a specific project logs a metric exceeding a custom threshold:
```python
import wandb
from wandb.automations import OnRunMetric, RunEvent, SendNotification
api = wandb.Api()
project = api.project("my-project", entity="my-team")
# Use the first Slack integration for the team
slack_hook = next(api.slack_integrations(entity="my-team"))
event = OnRunMetric(
scope=project,
filter=RunEvent.metric("custom-metric") > 10,
)
action = SendNotification.from_integration(slack_hook)
automation = api.create_automation(
event >> action,
name="my-automation",
description="Send a Slack message whenever 'custom-metric' exceeds 10.",
)
```
"""
from wandb.automations import Automation
from wandb.automations._generated import CREATE_AUTOMATION_GQL, CreateAutomation
from wandb.automations._utils import prepare_to_create
gql_input = prepare_to_create(obj, **kwargs)
if not self._supports_automation(
event=(event := gql_input.triggering_event_type),
action=(action := gql_input.triggered_action_type),
):
raise ValueError(
f"Automation event or action ({event!r} -> {action!r}) "
"is not supported on this wandb server version. "
"Please upgrade your server version, or contact support at "
"support@wandb.com."
)
# If needed, rewrite the GraphQL field selection set to omit unsupported fields/fragments/types
omit_fragments = self._omitted_automation_fragments()
mutation = gql_compat(CREATE_AUTOMATION_GQL, omit_fragments=omit_fragments)
variables = {"params": gql_input.model_dump(exclude_none=True)}
name = gql_input.name
try:
data = self.client.execute(mutation, variable_values=variables)
except requests.HTTPError as e:
status = HTTPStatus(e.response.status_code)
if status is HTTPStatus.CONFLICT: # 409
if fetch_existing:
wandb.termlog(f"Automation {name!r} exists. Fetching it instead.")
return self.automation(name=name)
raise ValueError(
f"Automation {name!r} exists. Unable to create another with the same name."
) from None
raise
try:
result = CreateAutomation.model_validate(data).result
except ValidationError as e:
msg = f"Invalid response while creating automation {name!r}"
raise RuntimeError(msg) from e
if (result is None) or (result.trigger is None):
msg = f"Empty response while creating automation {name!r}"
raise RuntimeError(msg)
return Automation.model_validate(result.trigger)
@normalize_exceptions
def update_automation(
self,
obj: "Automation",
*,
create_missing: bool = False,
**kwargs: Unpack["WriteAutomationsKwargs"],
) -> "Automation":
"""Update an existing automation.
Args:
obj: The automation to update. Must be an existing automation.
create_missing (bool):
If True, and the automation does not exist, create it.
**kwargs:
Any additional values to assign to the automation before
updating it. If given, these will override any values that may
already be set on the automation:
- `name`: The name of the automation.
- `description`: The description of the automation.
- `enabled`: Whether the automation is enabled.
- `scope`: The scope of the automation.
- `event`: The event that triggers the automation.
- `action`: The action that is triggered by the automation.
Returns:
The updated automation.
Examples:
Disable and edit the description of an existing automation ("my-automation"):
```python
import wandb
api = wandb.Api()
automation = api.automation(name="my-automation")
automation.enabled = False
automation.description = "Kept for reference, but no longer used."
updated_automation = api.update_automation(automation)
```
OR:
```python
import wandb
api = wandb.Api()
automation = api.automation(name="my-automation")
updated_automation = api.update_automation(
automation,
enabled=False,
description="Kept for reference, but no longer used.",
)
```
"""
from wandb.automations import ActionType, Automation
from wandb.automations._generated import UPDATE_AUTOMATION_GQL, UpdateAutomation
from wandb.automations._utils import prepare_to_update
# Check if the server even supports updating automations.
#
# NOTE: Unfortunately, there is no current server feature flag for this. As a workaround,
# we check whether the server supports the NO_OP action, which is a reasonably safe proxy
# for whether it supports updating automations.
if not self._supports_automation(action=ActionType.NO_OP):
raise RuntimeError(
"Updating existing automations is not enabled on this wandb server version. "
"Please upgrade your server version, or contact support at support@wandb.com."
)
gql_input = prepare_to_update(obj, **kwargs)
if not self._supports_automation(
event=(event := gql_input.triggering_event_type),
action=(action := gql_input.triggered_action_type),
):
raise ValueError(
f"Automation event or action ({event.value} -> {action.value}) "
"is not supported on this wandb server version. "
"Please upgrade your server version, or contact support at "
"support@wandb.com."
)
# If needed, rewrite the GraphQL field selection set to omit unsupported fields/fragments/types
omit_fragments = self._omitted_automation_fragments()
mutation = gql_compat(UPDATE_AUTOMATION_GQL, omit_fragments=omit_fragments)
variables = {"params": gql_input.model_dump(exclude_none=True)}
name = gql_input.name
try:
data = self.client.execute(mutation, variable_values=variables)
except requests.HTTPError as e:
status = HTTPStatus(e.response.status_code)
if status is HTTPStatus.NOT_FOUND: # 404
if create_missing:
wandb.termlog(f"Automation {name!r} not found. Creating it.")
return self.create_automation(obj)
raise ValueError(
f"Automation {name!r} not found. Unable to edit it."
) from e
# Not a (known) recoverable HTTP error
wandb.termerror(f"Got response status {status!r}: {e.response.text!r}")
raise
try:
result = UpdateAutomation.model_validate(data).result
except ValidationError as e:
msg = f"Invalid response while updating automation {name!r}"
raise RuntimeError(msg) from e
if (result is None) or (result.trigger is None):
msg = f"Empty response while updating automation {name!r}"
raise RuntimeError(msg)
return Automation.model_validate(result.trigger)
@normalize_exceptions
def delete_automation(self, obj: Union["Automation", str]) -> Literal[True]:
"""Delete an automation.
Args:
obj: The automation to delete, or its ID.
Returns:
True if the automation was deleted successfully.
"""
from wandb.automations._generated import DELETE_AUTOMATION_GQL, DeleteAutomation
from wandb.automations._utils import extract_id
id_ = extract_id(obj)
mutation = gql(DELETE_AUTOMATION_GQL)
variables = {"id": id_}
data = self.client.execute(mutation, variable_values=variables)
try:
result = DeleteAutomation.model_validate(data).result
except ValidationError as e:
msg = f"Invalid response while deleting automation {id_!r}"
raise RuntimeError(msg) from e
if result is None:
msg = f"Empty response while deleting automation {id_!r}"
raise RuntimeError(msg)
if not result.success:
raise RuntimeError(f"Failed to delete automation: {id_!r}")
return result.success
|