File size: 24,348 Bytes
4021124 | 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 | # Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License"). You
# may not use this file except in compliance with the License. A copy of
# the License is located at
#
# http://aws.amazon.com/apache2.0/
#
# or in the "license" file accompanying this file. This file is
# distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
# ANY KIND, either express or implied. See the License for the specific
# language governing permissions and limitations under the License.
"""This module contains code to create and manage SageMaker ``Artifact``."""
from __future__ import absolute_import
import logging
import math
from datetime import datetime
from typing import Iterator, Union, Any, Optional, List
from sagemaker.apiutils import _base_types, _utils
from sagemaker.lineage import _api_types
from sagemaker.lineage._api_types import ArtifactSource, ArtifactSummary
from sagemaker.lineage.query import (
LineageQuery,
LineageFilter,
LineageSourceEnum,
LineageEntityEnum,
LineageQueryDirectionEnum,
)
from sagemaker.lineage._utils import get_module, _disassociate, get_resource_name_from_arn
from sagemaker.lineage.association import Association
LOGGER = logging.getLogger("sagemaker")
class Artifact(_base_types.Record):
"""An Amazon SageMaker artifact, which is part of a SageMaker lineage.
Examples:
.. code-block:: python
from sagemaker.lineage import artifact
my_artifact = artifact.Artifact.create(
artifact_name='MyArtifact',
artifact_type='S3File',
source_uri='s3://...')
my_artifact.properties["added"] = "property"
my_artifact.save()
for artfct in artifact.Artifact.list():
print(artfct)
my_artifact.delete()
Attributes:
artifact_arn (str): The ARN of the artifact.
artifact_name (str): The name of the artifact.
artifact_type (str): The type of the artifact.
source (obj): The source of the artifact with a URI and types.
properties (dict): Dictionary of properties.
tags (List[dict[str, str]]): A list of tags to associate with the artifact.
creation_time (datetime): When the artifact was created.
created_by (obj): Contextual info on which account created the artifact.
last_modified_time (datetime): When the artifact was last modified.
last_modified_by (obj): Contextual info on which account created the artifact.
"""
artifact_arn: str = None
artifact_name: str = None
artifact_type: str = None
source: ArtifactSource = None
properties: dict = None
tags: list = None
creation_time: datetime = None
created_by: str = None
last_modified_time: datetime = None
last_modified_by: str = None
_boto_create_method: str = "create_artifact"
_boto_load_method: str = "describe_artifact"
_boto_update_method: str = "update_artifact"
_boto_delete_method: str = "delete_artifact"
_boto_update_members = [
"artifact_arn",
"artifact_name",
"properties",
"properties_to_remove",
]
_boto_delete_members = ["artifact_arn"]
_custom_boto_types = {"source": (_api_types.ArtifactSource, False)}
def save(self) -> "Artifact":
"""Save the state of this Artifact to SageMaker.
Note that this method must be run from a SageMaker context such as Studio or a training job
due to restrictions on the CreateArtifact API.
Returns:
Artifact: A SageMaker `Artifact` object.
"""
return self._invoke_api(self._boto_update_method, self._boto_update_members)
def delete(self, disassociate: bool = False):
"""Delete the artifact object.
Args:
disassociate (bool): When set to true, disassociate incoming and outgoing association.
"""
if disassociate:
_disassociate(source_arn=self.artifact_arn, sagemaker_session=self.sagemaker_session)
_disassociate(
destination_arn=self.artifact_arn,
sagemaker_session=self.sagemaker_session,
)
self._invoke_api(self._boto_delete_method, self._boto_delete_members)
@classmethod
def load(cls, artifact_arn: str, sagemaker_session=None) -> "Artifact":
"""Load an existing artifact and return an ``Artifact`` object representing it.
Args:
artifact_arn (str): ARN of the artifact
sagemaker_session (sagemaker.session.Session): Session object which
manages interactions with Amazon SageMaker APIs and any other
AWS services needed. If not specified, one is created using the
default AWS configuration chain.
Returns:
Artifact: A SageMaker ``Artifact`` object
"""
artifact = cls._construct(
cls._boto_load_method,
artifact_arn=artifact_arn,
sagemaker_session=sagemaker_session,
)
return artifact
def downstream_trials(self, sagemaker_session=None) -> list:
"""Use the lineage API to retrieve all downstream trials that use this artifact.
Args:
sagemaker_session (obj): Sagemaker Session to use. If not provided a default session
will be created.
Returns:
[Trial]: A list of SageMaker `Trial` objects.
"""
# don't specify destination type because for Trial Components it could be one of
# SageMaker[TrainingJob|ProcessingJob|TransformJob|ExperimentTrialComponent]
outgoing_associations: Iterator = Association.list(
source_arn=self.artifact_arn, sagemaker_session=sagemaker_session
)
trial_component_arns: list = list(map(lambda x: x.destination_arn, outgoing_associations))
return self._get_trial_from_trial_component(trial_component_arns)
def downstream_trials_v2(self) -> list:
"""Use a lineage query to retrieve all downstream trials that use this artifact.
Returns:
[Trial]: A list of SageMaker `Trial` objects.
"""
return self._trials(direction=LineageQueryDirectionEnum.DESCENDANTS)
def upstream_trials(self) -> List:
"""Use the lineage query to retrieve all upstream trials that use this artifact.
Returns:
[Trial]: A list of SageMaker `Trial` objects.
"""
return self._trials(direction=LineageQueryDirectionEnum.ASCENDANTS)
def _trials(
self, direction: LineageQueryDirectionEnum = LineageQueryDirectionEnum.BOTH
) -> List:
"""Use the lineage query to retrieve all trials that use this artifact.
Args:
direction (LineageQueryDirectionEnum, optional): The query direction.
Returns:
[Trial]: A list of SageMaker `Trial` objects.
"""
query_filter = LineageFilter(entities=[LineageEntityEnum.TRIAL_COMPONENT])
query_result = LineageQuery(self.sagemaker_session).query(
start_arns=[self.artifact_arn],
query_filter=query_filter,
direction=direction,
include_edges=False,
)
trial_component_arns: list = list(map(lambda x: x.arn, query_result.vertices))
return self._get_trial_from_trial_component(trial_component_arns)
def _get_trial_from_trial_component(self, trial_component_arns: list) -> List:
"""Retrieve all upstream trial runs which that use the trial component arns.
Args:
trial_component_arns (list): list of trial component arns
Returns:
[Trial]: A list of SageMaker `Trial` objects.
"""
if not trial_component_arns:
# no outgoing associations for this artifact
return []
get_module("smexperiments")
from smexperiments import trial_component, search_expression
max_search_by_arn: int = 60
num_search_batches = math.ceil(len(trial_component_arns) % max_search_by_arn)
trial_components: list = []
sagemaker_session = self.sagemaker_session or _utils.default_session()
sagemaker_client = sagemaker_session.sagemaker_client
for i in range(num_search_batches):
start: int = i * max_search_by_arn
end: int = start + max_search_by_arn
arn_batch: list = trial_component_arns[start:end]
se: Any = self._get_search_expression(arn_batch, search_expression)
search_result: Any = trial_component.TrialComponent.search(
search_expression=se, sagemaker_boto_client=sagemaker_client
)
trial_components: list = trial_components + list(search_result)
trials: set = set()
for tc in list(trial_components):
for parent in tc.parents:
trials.add(parent["TrialName"])
return list(trials)
def _get_search_expression(self, arns: list, search_expression: object) -> object:
"""Convert a set of arns to a search expression.
Args:
arns (list): Trial Component arns to search for.
search_expression (obj): smexperiments.search_expression
Returns:
search_expression (obj): Arns converted to a Trial Component search expression.
"""
max_arn_per_filter: int = 3
num_filters: Union[float, int] = math.ceil(len(arns) / max_arn_per_filter)
filters: list = []
for i in range(num_filters):
start: int = i * max_arn_per_filter
end: int = i + max_arn_per_filter
batch_arns: list = arns[start:end]
search_filter = search_expression.Filter(
name="TrialComponentArn",
operator=search_expression.Operator.EQUALS,
value=",".join(batch_arns),
)
filters.append(search_filter)
search_expression = search_expression.SearchExpression(
filters=filters,
boolean_operator=search_expression.BooleanOperator.OR,
)
return search_expression
def set_tag(self, tag=None):
"""Add a tag to the object.
Args:
tag (obj): Key value pair to set tag.
Returns:
list({str:str}): a list of key value pairs
"""
return self._set_tags(resource_arn=self.artifact_arn, tags=[tag])
def set_tags(self, tags=None):
"""Add tags to the object.
Args:
tags ([{key:value}]): list of key value pairs.
Returns:
list({str:str}): a list of key value pairs
"""
return self._set_tags(resource_arn=self.artifact_arn, tags=tags)
@classmethod
def create(
cls,
artifact_name: Optional[str] = None,
source_uri: Optional[str] = None,
source_types: Optional[list] = None,
artifact_type: Optional[str] = None,
properties: Optional[dict] = None,
tags: Optional[dict] = None,
sagemaker_session=None,
) -> "Artifact":
"""Create an artifact and return an ``Artifact`` object representing it.
Args:
artifact_name (str, optional): Name of the artifact
source_uri (str, optional): Source URI of the artifact
source_types (list, optional): Source types
artifact_type (str, optional): Type of the artifact
properties (dict, optional): key/value properties
tags (dict, optional): AWS tags for the artifact
sagemaker_session (sagemaker.session.Session): Session object which
manages interactions with Amazon SageMaker APIs and any other
AWS services needed. If not specified, one is created using the
default AWS configuration chain.
Returns:
Artifact: A SageMaker ``Artifact`` object.
"""
return super(Artifact, cls)._construct(
cls._boto_create_method,
artifact_name=artifact_name,
source=_api_types.ArtifactSource(source_uri=source_uri, source_types=source_types),
artifact_type=artifact_type,
properties=properties,
tags=tags,
sagemaker_session=sagemaker_session,
)
@classmethod
def list(
cls,
source_uri: Optional[str] = None,
artifact_type: Optional[str] = None,
created_before: Optional[datetime] = None,
created_after: Optional[datetime] = None,
sort_by: Optional[str] = None,
sort_order: Optional[str] = None,
max_results: Optional[int] = None,
next_token: Optional[str] = None,
sagemaker_session=None,
) -> Iterator[ArtifactSummary]:
"""Return a list of artifact summaries.
Args:
source_uri (str, optional): A source URI.
artifact_type (str, optional): An artifact type.
created_before (datetime.datetime, optional): Return artifacts created before this
instant.
created_after (datetime.datetime, optional): Return artifacts created after this
instant.
sort_by (str, optional): Which property to sort results by.
One of 'SourceArn', 'CreatedBefore','CreatedAfter'
sort_order (str, optional): One of 'Ascending', or 'Descending'.
max_results (int, optional): maximum number of artifacts to retrieve
next_token (str, optional): token for next page of results
sagemaker_session (sagemaker.session.Session): Session object which
manages interactions with Amazon SageMaker APIs and any other
AWS services needed. If not specified, one is created using the
default AWS configuration chain.
Returns:
collections.Iterator[ArtifactSummary]: An iterator
over ``ArtifactSummary`` objects.
"""
return super(Artifact, cls)._list(
"list_artifacts",
_api_types.ArtifactSummary.from_boto,
"ArtifactSummaries",
source_uri=source_uri,
artifact_type=artifact_type,
created_before=created_before,
created_after=created_after,
sort_by=sort_by,
sort_order=sort_order,
max_results=max_results,
next_token=next_token,
sagemaker_session=sagemaker_session,
)
def s3_uri_artifacts(self, s3_uri: str) -> dict:
"""Retrieve a list of artifacts that use provided s3 uri.
Args:
s3_uri (str): A S3 URI.
Returns:
A list of ``Artifacts``
"""
return self.sagemaker_session.sagemaker_client.list_artifacts(SourceUri=s3_uri)
class ModelArtifact(Artifact):
"""A SageMaker lineage artifact representing a model.
Common model specific lineage traversals to discover how the model is connected
to other entities.
"""
from sagemaker.lineage.context import Context
def endpoints(self) -> list:
"""Get association summaries for endpoints deployed with this model.
Returns:
[AssociationSummary]: A list of associations representing the endpoints using the model.
"""
endpoint_development_actions: Iterator = Association.list(
source_arn=self.artifact_arn,
destination_type="Action",
sagemaker_session=self.sagemaker_session,
)
endpoint_context_list: list = [
endpoint_context_associations
for endpoint_development_action in endpoint_development_actions
for endpoint_context_associations in Association.list(
source_arn=endpoint_development_action.destination_arn,
destination_type="Context",
sagemaker_session=self.sagemaker_session,
)
]
return endpoint_context_list
def endpoint_contexts(
self, direction: LineageQueryDirectionEnum = LineageQueryDirectionEnum.DESCENDANTS
) -> List[Context]:
"""Get contexts representing endpoints from the models's lineage.
Args:
direction (LineageQueryDirectionEnum, optional): The query direction.
Returns:
list of Contexts: Contexts representing an endpoint.
"""
query_filter = LineageFilter(
entities=[LineageEntityEnum.CONTEXT], sources=[LineageSourceEnum.ENDPOINT]
)
query_result = LineageQuery(self.sagemaker_session).query(
start_arns=[self.artifact_arn],
query_filter=query_filter,
direction=direction,
include_edges=False,
)
endpoint_contexts = []
for vertex in query_result.vertices:
endpoint_contexts.append(vertex.to_lineage_object())
return endpoint_contexts
def dataset_artifacts(
self, direction: LineageQueryDirectionEnum = LineageQueryDirectionEnum.ASCENDANTS
) -> List[Artifact]:
"""Get artifacts representing datasets from the model's lineage.
Args:
direction (LineageQueryDirectionEnum, optional): The query direction.
Returns:
list of Artifacts: Artifacts representing a dataset.
"""
query_filter = LineageFilter(
entities=[LineageEntityEnum.ARTIFACT], sources=[LineageSourceEnum.DATASET]
)
query_result = LineageQuery(self.sagemaker_session).query(
start_arns=[self.artifact_arn],
query_filter=query_filter,
direction=direction,
include_edges=False,
)
dataset_artifacts = []
for vertex in query_result.vertices:
dataset_artifacts.append(vertex.to_lineage_object())
return dataset_artifacts
def training_job_arns(
self, direction: LineageQueryDirectionEnum = LineageQueryDirectionEnum.ASCENDANTS
) -> List[str]:
"""Get ARNs for all training jobs that appear in the model's lineage.
Returns:
list of str: Training job ARNs.
"""
query_filter = LineageFilter(
entities=[LineageEntityEnum.TRIAL_COMPONENT], sources=[LineageSourceEnum.TRAINING_JOB]
)
query_result = LineageQuery(self.sagemaker_session).query(
start_arns=[self.artifact_arn],
query_filter=query_filter,
direction=direction,
include_edges=False,
)
training_job_arns = []
for vertex in query_result.vertices:
trial_component_name = get_resource_name_from_arn(vertex.arn)
trial_component = self.sagemaker_session.sagemaker_client.describe_trial_component(
TrialComponentName=trial_component_name
)
training_job_arns.append(trial_component["Source"]["SourceArn"])
return training_job_arns
def pipeline_execution_arn(
self, direction: LineageQueryDirectionEnum = LineageQueryDirectionEnum.ASCENDANTS
) -> str:
"""Get the ARN for the pipeline execution associated with this model (if any).
Returns:
str: A pipeline execution ARN.
"""
training_job_arns = self.training_job_arns(direction=direction)
for training_job_arn in training_job_arns:
tags = self.sagemaker_session.sagemaker_client.list_tags(ResourceArn=training_job_arn)[
"Tags"
]
for tag in tags:
if tag["Key"] == "sagemaker:pipeline-execution-arn":
return tag["Value"]
return None
class DatasetArtifact(Artifact):
"""A SageMaker Lineage artifact representing a dataset.
Encapsulates common dataset specific lineage traversals to discover how the dataset is
connect to related entities.
"""
from sagemaker.lineage.context import Context
def trained_models(self) -> List[Association]:
"""Given a dataset artifact, get associated trained models.
Returns:
list(Association): List of Contexts representing model artifacts.
"""
trial_components: Iterator = Association.list(
source_arn=self.artifact_arn, sagemaker_session=self.sagemaker_session
)
result: list = []
for trial_component in trial_components:
if "experiment-trial-component" in trial_component.destination_arn:
models = Association.list(
source_arn=trial_component.destination_arn,
destination_type="Context",
sagemaker_session=self.sagemaker_session,
)
result.extend(models)
return result
def endpoint_contexts(
self, direction: LineageQueryDirectionEnum = LineageQueryDirectionEnum.DESCENDANTS
) -> List[Context]:
"""Get contexts representing endpoints from the dataset's lineage.
Args:
direction (LineageQueryDirectionEnum, optional): The query direction.
Returns:
list of Contexts: Contexts representing an endpoint.
"""
query_filter = LineageFilter(
entities=[LineageEntityEnum.CONTEXT], sources=[LineageSourceEnum.ENDPOINT]
)
query_result = LineageQuery(self.sagemaker_session).query(
start_arns=[self.artifact_arn],
query_filter=query_filter,
direction=direction,
include_edges=False,
)
endpoint_contexts = []
for vertex in query_result.vertices:
endpoint_contexts.append(vertex.to_lineage_object())
return endpoint_contexts
def upstream_datasets(self) -> List[Artifact]:
"""Use the lineage query to retrieve upstream artifacts that use this dataset artifact.
Returns:
list of Artifacts: Artifacts representing an dataset.
"""
return self._datasets(direction=LineageQueryDirectionEnum.ASCENDANTS)
def downstream_datasets(self) -> List[Artifact]:
"""Use the lineage query to retrieve downstream artifacts that use this dataset.
Returns:
list of Artifacts: Artifacts representing an dataset.
"""
return self._datasets(direction=LineageQueryDirectionEnum.DESCENDANTS)
def _datasets(
self, direction: LineageQueryDirectionEnum = LineageQueryDirectionEnum.BOTH
) -> List[Artifact]:
"""Use the lineage query to retrieve all artifacts that use this dataset.
Args:
direction (LineageQueryDirectionEnum, optional): The query direction.
Returns:
list of Artifacts: Artifacts representing an dataset.
"""
query_filter = LineageFilter(
entities=[LineageEntityEnum.ARTIFACT], sources=[LineageSourceEnum.DATASET]
)
query_result = LineageQuery(self.sagemaker_session).query(
start_arns=[self.artifact_arn],
query_filter=query_filter,
direction=direction,
include_edges=False,
)
return [vertex.to_lineage_object() for vertex in query_result.vertices]
class ImageArtifact(Artifact):
"""A SageMaker lineage artifact representing an image.
Common model specific lineage traversals to discover how the image is connected
to other entities.
"""
def datasets(self, direction: LineageQueryDirectionEnum) -> List[Artifact]:
"""Use the lineage query to retrieve datasets that use this image artifact.
Args:
direction (LineageQueryDirectionEnum): The query direction.
Returns:
list of Artifacts: Artifacts representing a dataset.
"""
query_filter = LineageFilter(
entities=[LineageEntityEnum.ARTIFACT], sources=[LineageSourceEnum.DATASET]
)
query_result = LineageQuery(self.sagemaker_session).query(
start_arns=[self.artifact_arn],
query_filter=query_filter,
direction=direction,
include_edges=False,
)
return [vertex.to_lineage_object() for vertex in query_result.vertices]
|