File size: 52,721 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 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 | # 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.
"""Placeholder docstring"""
from __future__ import print_function, absolute_import
import os
import re
import sagemaker
from sagemaker import fw_utils, job, utils, s3, session, vpc_utils
from sagemaker.amazon import amazon_estimator
from sagemaker.tensorflow import TensorFlow
from sagemaker.estimator import EstimatorBase
from sagemaker.processing import Processor
def prepare_framework(estimator, s3_operations):
"""Prepare S3 operations and environment variables related to framework.
S3 operations specify where to upload `source_dir`.
Args:
estimator (sagemaker.estimator.Estimator): The framework estimator to
get information from and update.
s3_operations (dict): The dict to specify s3 operations (upload
`source_dir` ).
"""
if estimator.code_location is not None:
bucket, key = s3.parse_s3_url(estimator.code_location)
key = os.path.join(key, estimator._current_job_name, "source", "sourcedir.tar.gz")
elif estimator.uploaded_code is not None:
bucket, key = s3.parse_s3_url(estimator.uploaded_code.s3_prefix)
else:
bucket = estimator.sagemaker_session._default_bucket
key = os.path.join(estimator._current_job_name, "source", "sourcedir.tar.gz")
script = os.path.basename(estimator.entry_point)
if estimator.source_dir and estimator.source_dir.lower().startswith("s3://"):
code_dir = estimator.source_dir
estimator.uploaded_code = fw_utils.UploadedCode(s3_prefix=code_dir, script_name=script)
else:
code_dir = "s3://{}/{}".format(bucket, key)
estimator.uploaded_code = fw_utils.UploadedCode(s3_prefix=code_dir, script_name=script)
s3_operations["S3Upload"] = [
{
"Path": estimator.source_dir or estimator.entry_point,
"Bucket": bucket,
"Key": key,
"Tar": True,
}
]
estimator._hyperparameters[sagemaker.model.DIR_PARAM_NAME] = code_dir
estimator._hyperparameters[sagemaker.model.SCRIPT_PARAM_NAME] = script
estimator._hyperparameters[
sagemaker.model.CONTAINER_LOG_LEVEL_PARAM_NAME
] = estimator.container_log_level
estimator._hyperparameters[sagemaker.model.JOB_NAME_PARAM_NAME] = estimator._current_job_name
estimator._hyperparameters[
sagemaker.model.SAGEMAKER_REGION_PARAM_NAME
] = estimator.sagemaker_session.boto_region_name
def prepare_amazon_algorithm_estimator(estimator, inputs, mini_batch_size=None):
"""Sets up amazon algorithm estimator.
This is done by adding the required `feature_dim` hyperparameter from training data.
Args:
estimator (sagemaker.amazon.amazon_estimator.AmazonAlgorithmEstimatorBase): An estimator
for a built-in Amazon algorithm to get information from and update.
inputs: The training data.
* (sagemaker.amazon.amazon_estimator.RecordSet) - A collection of
Amazon :class:~`Record` objects serialized and stored in S3. For
use with an estimator for an Amazon algorithm.
* (list[sagemaker.amazon.amazon_estimator.RecordSet]) - A list of
:class:~`sagemaker.amazon.amazon_estimator.RecordSet` objects,
where each instance is a different channel of training data.
mini_batch_size:
"""
if isinstance(inputs, list):
for record in inputs:
if isinstance(record, amazon_estimator.RecordSet) and record.channel == "train":
estimator.feature_dim = record.feature_dim
break
elif isinstance(inputs, amazon_estimator.RecordSet):
estimator.feature_dim = inputs.feature_dim
else:
raise TypeError("Training data must be represented in RecordSet or list of RecordSets")
estimator.mini_batch_size = mini_batch_size
def training_base_config(estimator, inputs=None, job_name=None, mini_batch_size=None): # noqa: C901
"""Export Airflow base training config from an estimator
Args:
estimator (sagemaker.estimator.EstimatorBase): The estimator to export
training config from. Can be a BYO estimator, Framework estimator or
Amazon algorithm estimator.
inputs: Information about the training data. Please refer to the ``fit()``
method of
the associated estimator, as this can take any of the following
forms:
* (str) - The S3 location where training data is saved.
* (dict[str, str] or dict[str, sagemaker.inputs.TrainingInput]) - If using multiple
channels for training data, you can specify a dict mapping channel names to
strings or :func:`~sagemaker.inputs.TrainingInput` objects.
* (sagemaker.inputs.TrainingInput) - Channel configuration for S3 data sources that can
provide additional information about the training dataset. See
:func:`sagemaker.inputs.TrainingInput` for full details.
* (sagemaker.amazon.amazon_estimator.RecordSet) - A collection of
Amazon :class:~`Record` objects serialized and stored in S3.
For use with an estimator for an Amazon algorithm.
* (list[sagemaker.amazon.amazon_estimator.RecordSet]) - A list of
:class:~`sagemaker.amazon.amazon_estimator.RecordSet` objects,
where each instance is a different channel of training data.
job_name (str): Specify a training job name if needed.
mini_batch_size (int): Specify this argument only when estimator is a
built-in estimator of an Amazon algorithm. For other estimators,
batch size should be specified in the estimator.
Returns:
dict: Training config that can be directly used by
SageMakerTrainingOperator in Airflow.
"""
if isinstance(estimator, sagemaker.amazon.amazon_estimator.AmazonAlgorithmEstimatorBase):
estimator.prepare_workflow_for_training(
records=inputs, mini_batch_size=mini_batch_size, job_name=job_name
)
else:
estimator.prepare_workflow_for_training(job_name=job_name)
s3_operations = {}
if job_name is not None:
estimator._current_job_name = job_name
else:
base_name = estimator.base_job_name or utils.base_name_from_image(
estimator.training_image_uri(),
default_base_name=EstimatorBase.JOB_CLASS_NAME,
)
estimator._current_job_name = utils.name_from_base(base_name)
if estimator.output_path is None:
default_bucket = estimator.sagemaker_session.default_bucket()
estimator.output_path = "s3://{}/".format(default_bucket)
if isinstance(estimator, sagemaker.estimator.Framework):
prepare_framework(estimator, s3_operations)
elif isinstance(estimator, amazon_estimator.AmazonAlgorithmEstimatorBase):
prepare_amazon_algorithm_estimator(estimator, inputs, mini_batch_size)
job_config = job._Job._load_config(inputs, estimator, expand_role=False, validate_uri=False)
train_config = {
"AlgorithmSpecification": {
"TrainingImage": estimator.training_image_uri(),
"TrainingInputMode": estimator.input_mode,
},
"OutputDataConfig": job_config["output_config"],
"StoppingCondition": job_config["stop_condition"],
"ResourceConfig": job_config["resource_config"],
"RoleArn": job_config["role"],
}
if job_config["input_config"] is not None:
train_config["InputDataConfig"] = job_config["input_config"]
if job_config["vpc_config"] is not None:
train_config["VpcConfig"] = job_config["vpc_config"]
if estimator.use_spot_instances:
# estimator.use_spot_instances may be a Pipeline ParameterBoolean object
# which is parsed during the Pipeline execution runtime
train_config["EnableManagedSpotTraining"] = estimator.use_spot_instances
if estimator.hyperparameters() is not None:
hyperparameters = {str(k): str(v) for (k, v) in estimator.hyperparameters().items()}
if hyperparameters and len(hyperparameters) > 0:
train_config["HyperParameters"] = hyperparameters
if s3_operations:
train_config["S3Operations"] = s3_operations
if (estimator.checkpoint_local_path is not None) & (estimator.checkpoint_s3_uri is not None):
train_config["CheckpointConfig"] = {
"LocalPath": estimator.checkpoint_local_path,
"S3Uri": estimator.checkpoint_s3_uri,
}
return train_config
def training_config(estimator, inputs=None, job_name=None, mini_batch_size=None):
"""Export Airflow training config from an estimator
Args:
estimator (sagemaker.estimator.EstimatorBase): The estimator to export
training config from. Can be a BYO estimator, Framework estimator or
Amazon algorithm estimator.
inputs: Information about the training data. Please refer to the ``fit()``
method of the associated estimator, as this can take any of the following forms:
* (str) - The S3 location where training data is saved.
* (dict[str, str] or dict[str, sagemaker.inputs.TrainingInput]) - If using multiple
channels for training data, you can specify a dict mapping channel names to
strings or :func:`~sagemaker.inputs.TrainingInput` objects.
* (sagemaker.inputs.TrainingInput) - Channel configuration for S3 data sources that can
provide additional information about the training dataset. See
:func:`sagemaker.inputs.TrainingInput` for full details.
* (sagemaker.amazon.amazon_estimator.RecordSet) - A collection of
Amazon :class:~`Record` objects serialized and stored in S3.
For use with an estimator for an Amazon algorithm.
* (list[sagemaker.amazon.amazon_estimator.RecordSet]) - A list of
:class:~`sagemaker.amazon.amazon_estimator.RecordSet` objects,
where each instance is a different channel of training data.
job_name (str): Specify a training job name if needed.
mini_batch_size (int): Specify this argument only when estimator is a
built-in estimator of an Amazon algorithm. For other estimators,
batch size should be specified in the estimator.
Returns:
dict: Training config that can be directly used by
SageMakerTrainingOperator in Airflow.
"""
train_config = training_base_config(estimator, inputs, job_name, mini_batch_size)
train_config["TrainingJobName"] = estimator._current_job_name
if estimator.tags is not None:
train_config["Tags"] = estimator.tags
if estimator.metric_definitions is not None:
train_config["AlgorithmSpecification"]["MetricDefinitions"] = estimator.metric_definitions
return train_config
def tuning_config(tuner, inputs, job_name=None, include_cls_metadata=False, mini_batch_size=None):
"""Export Airflow tuning config from a HyperparameterTuner
Args:
tuner (sagemaker.tuner.HyperparameterTuner): The tuner to export tuning
config from.
inputs: Information about the training data. Please refer to the ``fit()``
method of the associated estimator in the tuner, as this can take any of the
following forms:
* (str) - The S3 location where training data is saved.
* (dict[str, str] or dict[str, sagemaker.inputs.TrainingInput]) - If using multiple
channels for training data, you can specify a dict mapping channel names to
strings or :func:`~sagemaker.inputs.TrainingInput` objects.
* (sagemaker.inputs.TrainingInput) - Channel configuration for S3 data sources that can
provide additional information about the training dataset. See
:func:`sagemaker.inputs.TrainingInput` for full details.
* (sagemaker.amazon.amazon_estimator.RecordSet) - A collection of
Amazon :class:~`Record` objects serialized and stored in S3.
For use with an estimator for an Amazon algorithm.
* (list[sagemaker.amazon.amazon_estimator.RecordSet]) - A list of
:class:~`sagemaker.amazon.amazon_estimator.RecordSet` objects,
where each instance is a different channel of training data.
* (dict[str, one the forms above]): Required by only tuners created via
the factory method ``HyperparameterTuner.create()``. The keys should be the
same estimator names as keys for the ``estimator_dict`` argument of the
``HyperparameterTuner.create()`` method.
job_name (str): Specify a tuning job name if needed.
include_cls_metadata: It can take one of the following two forms.
* (bool) - Whether or not the hyperparameter tuning job should include information
about the estimator class (default: False). This information is passed as a
hyperparameter, so if the algorithm you are using cannot handle unknown
hyperparameters (e.g. an Amazon SageMaker built-in algorithm that does not
have a custom estimator in the Python SDK), then set ``include_cls_metadata``
to ``False``.
* (dict[str, bool]) - This version should be used for tuners created via the factory
method ``HyperparameterTuner.create()``, to specify the flag for individual
estimators provided in the ``estimator_dict`` argument of the method. The keys
would be the same estimator names as in ``estimator_dict``. If one estimator
doesn't need the flag set, then no need to include it in the dictionary. If none
of the estimators need the flag set, then an empty dictionary ``{}`` must be used.
mini_batch_size: It can take one of the following two forms.
* (int) - Specify this argument only when estimator is a built-in estimator of an
Amazon algorithm. For other estimators, batch size should be specified in the
estimator.
* (dict[str, int]) - This version should be used for tuners created via the factory
method ``HyperparameterTuner.create()``, to specify the value for individual
estimators provided in the ``estimator_dict`` argument of the method. The keys
would be the same estimator names as in ``estimator_dict``. If one estimator
doesn't need the value set, then no need to include it in the dictionary. If
none of the estimators need the value set, then an empty dictionary ``{}``
must be used.
Returns:
dict: Tuning config that can be directly used by SageMakerTuningOperator in Airflow.
"""
tuner._prepare_job_name_for_tuning(job_name=job_name)
tune_config = {
"HyperParameterTuningJobName": tuner._current_job_name,
"HyperParameterTuningJobConfig": _extract_tuning_job_config(tuner),
}
if tuner.estimator:
(
tune_config["TrainingJobDefinition"],
s3_operations,
) = _extract_training_config_from_estimator(
tuner, inputs, include_cls_metadata, mini_batch_size
)
else:
(
tune_config["TrainingJobDefinitions"],
s3_operations,
) = _extract_training_config_list_from_estimator_dict(
tuner, inputs, include_cls_metadata, mini_batch_size
)
if s3_operations:
tune_config["S3Operations"] = s3_operations
if tuner.tags:
tune_config["Tags"] = tuner.tags
if tuner.warm_start_config:
tune_config["WarmStartConfig"] = tuner.warm_start_config.to_input_req()
return tune_config
def _extract_tuning_job_config(tuner):
"""Extract tuning job config from a HyperparameterTuner"""
tuning_job_config = {
"Strategy": tuner.strategy,
"ResourceLimits": {
"MaxNumberOfTrainingJobs": tuner.max_jobs,
"MaxParallelTrainingJobs": tuner.max_parallel_jobs,
},
"TrainingJobEarlyStoppingType": tuner.early_stopping_type,
}
if tuner.objective_metric_name:
tuning_job_config["HyperParameterTuningJobObjective"] = {
"Type": tuner.objective_type,
"MetricName": tuner.objective_metric_name,
}
parameter_ranges = tuner.hyperparameter_ranges()
if parameter_ranges:
tuning_job_config["ParameterRanges"] = parameter_ranges
return tuning_job_config
def _extract_training_config_from_estimator(tuner, inputs, include_cls_metadata, mini_batch_size):
"""Extract training job config from a HyperparameterTuner that uses the ``estimator`` field"""
train_config = training_base_config(tuner.estimator, inputs, mini_batch_size)
train_config.pop("HyperParameters", None)
tuner._prepare_static_hyperparameters_for_tuning(include_cls_metadata=include_cls_metadata)
train_config["StaticHyperParameters"] = tuner.static_hyperparameters
if tuner.metric_definitions:
train_config["AlgorithmSpecification"]["MetricDefinitions"] = tuner.metric_definitions
s3_operations = train_config.pop("S3Operations", None)
return train_config, s3_operations
def _extract_training_config_list_from_estimator_dict(
tuner, inputs, include_cls_metadata, mini_batch_size
):
"""Extracts a list of training job configs from a Hyperparameter Tuner.
It uses the ``estimator_dict`` field.
"""
estimator_names = sorted(tuner.estimator_dict.keys())
tuner._validate_dict_argument(name="inputs", value=inputs, allowed_keys=estimator_names)
tuner._validate_dict_argument(
name="include_cls_metadata", value=include_cls_metadata, allowed_keys=estimator_names
)
tuner._validate_dict_argument(
name="mini_batch_size", value=mini_batch_size, allowed_keys=estimator_names
)
train_config_dict = {}
for (estimator_name, estimator) in tuner.estimator_dict.items():
train_config_dict[estimator_name] = training_base_config(
estimator=estimator,
inputs=inputs.get(estimator_name) if inputs else None,
mini_batch_size=mini_batch_size.get(estimator_name) if mini_batch_size else None,
)
tuner._prepare_static_hyperparameters_for_tuning(include_cls_metadata=include_cls_metadata)
train_config_list = []
s3_operations_list = []
for estimator_name in sorted(train_config_dict.keys()):
train_config = train_config_dict[estimator_name]
train_config.pop("HyperParameters", None)
train_config["StaticHyperParameters"] = tuner.static_hyperparameters_dict[estimator_name]
train_config["AlgorithmSpecification"][
"MetricDefinitions"
] = tuner.metric_definitions_dict.get(estimator_name)
train_config["DefinitionName"] = estimator_name
train_config["TuningObjective"] = {
"Type": tuner.objective_type,
"MetricName": tuner.objective_metric_name_dict[estimator_name],
}
train_config["HyperParameterRanges"] = tuner.hyperparameter_ranges_dict()[estimator_name]
s3_operations_list.append(train_config.pop("S3Operations", {}))
train_config_list.append(train_config)
return train_config_list, _merge_s3_operations(s3_operations_list)
def _merge_s3_operations(s3_operations_list):
"""Merge a list of S3 operation dictionaries into one"""
s3_operations_merged = {}
for s3_operations in s3_operations_list:
for (key, operations) in s3_operations.items():
if key not in s3_operations_merged:
s3_operations_merged[key] = []
for operation in operations:
if operation not in s3_operations_merged[key]:
s3_operations_merged[key].append(operation)
return s3_operations_merged
def update_submit_s3_uri(estimator, job_name):
"""Updated the S3 URI of the framework source directory in given estimator.
Args:
estimator (sagemaker.estimator.Framework): The Framework estimator to
update.
job_name (str): The new job name included in the submit S3 URI
Returns:
str: The updated S3 URI of framework source directory
"""
if estimator.uploaded_code is None:
return
pattern = r"(?<=/)[^/]+?(?=/source/sourcedir.tar.gz)"
# update the S3 URI with the latest training job.
# s3://path/old_job/source/sourcedir.tar.gz will become s3://path/new_job/source/sourcedir.tar.gz
submit_uri = estimator.uploaded_code.s3_prefix
submit_uri = re.sub(pattern, job_name, submit_uri)
script_name = estimator.uploaded_code.script_name
estimator.uploaded_code = fw_utils.UploadedCode(submit_uri, script_name)
def update_estimator_from_task(estimator, task_id, task_type):
"""Update training job of the estimator from a task in the DAG
Args:
estimator (sagemaker.estimator.EstimatorBase): The estimator to update
task_id (str): The task id of any
airflow.contrib.operators.SageMakerTrainingOperator or
airflow.contrib.operators.SageMakerTuningOperator that generates
training jobs in the DAG.
task_type (str): Whether the task is from SageMakerTrainingOperator or
SageMakerTuningOperator. Values can be 'training', 'tuning' or None
(which means training job is not from any task).
"""
if task_type is None:
return
if task_type.lower() == "training":
training_job = "{{ ti.xcom_pull(task_ids='%s')['Training']['TrainingJobName'] }}" % task_id
job_name = training_job
elif task_type.lower() == "tuning":
training_job = (
"{{ ti.xcom_pull(task_ids='%s')['Tuning']['BestTrainingJob']['TrainingJobName'] }}"
% task_id
)
# need to strip the double quotes in json to get the string
job_name = (
"{{ ti.xcom_pull(task_ids='%s')['Tuning']['TrainingJobDefinition']"
"['StaticHyperParameters']['sagemaker_job_name'].strip('%s') }}" % (task_id, '"')
)
else:
raise ValueError("task_type must be either 'training', 'tuning' or None.")
estimator._current_job_name = training_job
if isinstance(estimator, sagemaker.estimator.Framework):
update_submit_s3_uri(estimator, job_name)
def prepare_framework_container_def(model, instance_type, s3_operations):
"""This prepares the framework model container information and specifies related S3 operations.
Prepare the framework model container information. Specify related S3
operations for Airflow to perform. (Upload `source_dir` )
Args:
model (sagemaker.model.FrameworkModel): The framework model
instance_type (str): The EC2 instance type to deploy this Model to. For
example, 'ml.p2.xlarge'.
s3_operations (dict): The dict to specify S3 operations (upload
`source_dir` ).
Returns:
dict: The container information of this framework model.
"""
deploy_image = model.image_uri
if not deploy_image:
region_name = model.sagemaker_session.boto_session.region_name
deploy_image = model.serving_image_uri(region_name, instance_type)
base_name = utils.base_name_from_image(deploy_image)
model.name = model.name or utils.name_from_base(base_name)
bucket = model.bucket or model.sagemaker_session._default_bucket
if model.entry_point is not None:
script = os.path.basename(model.entry_point)
key = "{}/source/sourcedir.tar.gz".format(model.name)
if model.source_dir and model.source_dir.lower().startswith("s3://"):
code_dir = model.source_dir
model.uploaded_code = fw_utils.UploadedCode(s3_prefix=code_dir, script_name=script)
else:
code_dir = "s3://{}/{}".format(bucket, key)
model.uploaded_code = fw_utils.UploadedCode(s3_prefix=code_dir, script_name=script)
s3_operations["S3Upload"] = [
{"Path": model.source_dir or script, "Bucket": bucket, "Key": key, "Tar": True}
]
deploy_env = dict(model.env)
deploy_env.update(model._script_mode_env_vars())
try:
if model.model_server_workers:
deploy_env[sagemaker.model.MODEL_SERVER_WORKERS_PARAM_NAME.upper()] = str(
model.model_server_workers
)
except AttributeError:
# This applies to a FrameworkModel which is not SageMaker Deep Learning Framework Model
pass
return sagemaker.container_def(deploy_image, model.model_data, deploy_env)
def model_config(model, instance_type=None, role=None, image_uri=None):
"""Export Airflow model config from a SageMaker model
Args:
model (sagemaker.model.Model): The Model object from which to export the Airflow config
instance_type (str): The EC2 instance type to deploy this Model to. For
example, 'ml.p2.xlarge'
role (str): The ``ExecutionRoleArn`` IAM Role ARN for the model
image_uri (str): An Docker image URI to use for deploying the model
Returns:
dict: Model config that can be directly used by SageMakerModelOperator
in Airflow. It can also be part of the config used by
SageMakerEndpointOperator and SageMakerTransformOperator in Airflow.
"""
s3_operations = {}
model.image_uri = image_uri or model.image_uri
if isinstance(model, sagemaker.model.FrameworkModel):
container_def = prepare_framework_container_def(model, instance_type, s3_operations)
else:
container_def = model.prepare_container_def()
base_name = utils.base_name_from_image(container_def["Image"])
model.name = model.name or utils.name_from_base(base_name)
primary_container = session._expand_container_def(container_def)
config = {
"ModelName": model.name,
"PrimaryContainer": primary_container,
"ExecutionRoleArn": role or model.role,
}
if model.vpc_config:
config["VpcConfig"] = model.vpc_config
if s3_operations:
config["S3Operations"] = s3_operations
return config
def model_config_from_estimator(
estimator,
task_id,
task_type,
instance_type=None,
role=None,
image_uri=None,
name=None,
model_server_workers=None,
vpc_config_override=vpc_utils.VPC_CONFIG_DEFAULT,
):
"""Export Airflow model config from a SageMaker estimator
Args:
estimator (sagemaker.model.EstimatorBase): The SageMaker estimator to
export Airflow config from. It has to be an estimator associated
with a training job.
task_id (str): The task id of any
airflow.contrib.operators.SageMakerTrainingOperator or
airflow.contrib.operators.SageMakerTuningOperator that generates
training jobs in the DAG. The model config is built based on the
training job generated in this operator.
task_type (str): Whether the task is from SageMakerTrainingOperator or
SageMakerTuningOperator. Values can be 'training', 'tuning' or None
(which means training job is not from any task).
instance_type (str): The EC2 instance type to deploy this Model to. For
example, 'ml.p2.xlarge'
role (str): The ``ExecutionRoleArn`` IAM Role ARN for the model
image_uri (str): A Docker image URI to use for deploying the model
name (str): Name of the model
model_server_workers (int): The number of worker processes used by the
inference server. If None, server will use one worker per vCPU. Only
effective when estimator is a SageMaker framework.
vpc_config_override (dict[str, list[str]]): Override for VpcConfig set on
the model. Default: use subnets and security groups from this Estimator.
* 'Subnets' (list[str]): List of subnet ids.
* 'SecurityGroupIds' (list[str]): List of security group ids.
Returns:
dict: Model config that can be directly used by SageMakerModelOperator in Airflow. It can
also be part of the config used by SageMakerEndpointOperator in Airflow.
"""
update_estimator_from_task(estimator, task_id, task_type)
if isinstance(estimator, sagemaker.estimator.Estimator):
model = estimator.create_model(
role=role, image_uri=image_uri, vpc_config_override=vpc_config_override
)
elif isinstance(estimator, sagemaker.amazon.amazon_estimator.AmazonAlgorithmEstimatorBase):
model = estimator.create_model(vpc_config_override=vpc_config_override)
elif isinstance(estimator, TensorFlow):
model = estimator.create_model(
role=role, vpc_config_override=vpc_config_override, entry_point=estimator.entry_point
)
elif isinstance(estimator, sagemaker.estimator.Framework):
model = estimator.create_model(
model_server_workers=model_server_workers,
role=role,
vpc_config_override=vpc_config_override,
entry_point=estimator.entry_point,
)
else:
raise TypeError(
"Estimator must be one of sagemaker.estimator.Estimator, sagemaker.estimator.Framework"
" or sagemaker.amazon.amazon_estimator.AmazonAlgorithmEstimatorBase."
)
model.name = name
return model_config(model, instance_type, role, image_uri)
def transform_config(
transformer,
data,
data_type="S3Prefix",
content_type=None,
compression_type=None,
split_type=None,
job_name=None,
input_filter=None,
output_filter=None,
join_source=None,
):
"""Export Airflow transform config from a SageMaker transformer
Args:
transformer (sagemaker.transformer.Transformer): The SageMaker
transformer to export Airflow config from.
data (str): Input data location in S3.
data_type (str): What the S3 location defines (default: 'S3Prefix').
Valid values:
* 'S3Prefix' - the S3 URI defines a key name prefix. All objects with this prefix will
be used as inputs for the transform job.
* 'ManifestFile' - the S3 URI points to a single manifest file listing each S3 object
to use as an input for the transform job.
content_type (str): MIME type of the input data (default: None).
compression_type (str): Compression type of the input data, if
compressed (default: None). Valid values: 'Gzip', None.
split_type (str): The record delimiter for the input object (default:
'None'). Valid values: 'None', 'Line', 'RecordIO', and 'TFRecord'.
job_name (str): job name (default: None). If not specified, one will be
generated.
input_filter (str): A JSONPath to select a portion of the input to
pass to the algorithm container for inference. If you omit the
field, it gets the value '$', representing the entire input.
For CSV data, each row is taken as a JSON array,
so only index-based JSONPaths can be applied, e.g. $[0], $[1:].
CSV data should follow the `RFC format <https://tools.ietf.org/html/rfc4180>`_.
See `Supported JSONPath Operators
<https://docs.aws.amazon.com/sagemaker/latest/dg/batch-transform-data-processing.html#data-processing-operators>`_
for a table of supported JSONPath operators.
For more information, see the SageMaker API documentation for
`CreateTransformJob
<https://docs.aws.amazon.com/sagemaker/latest/dg/API_CreateTransformJob.html>`_.
Some examples: "$[1:]", "$.features" (default: None).
output_filter (str): A JSONPath to select a portion of the
joined/original output to return as the output.
For more information, see the SageMaker API documentation for
`CreateTransformJob
<https://docs.aws.amazon.com/sagemaker/latest/dg/API_CreateTransformJob.html>`_.
Some examples: "$[1:]", "$.prediction" (default: None).
join_source (str): The source of data to be joined to the transform
output. It can be set to 'Input' meaning the entire input record
will be joined to the inference result. You can use OutputFilter
to select the useful portion before uploading to S3. (default:
None). Valid values: Input, None.
Returns:
dict: Transform config that can be directly used by
SageMakerTransformOperator in Airflow.
"""
if job_name is not None:
transformer._current_job_name = job_name
else:
base_name = transformer.base_transform_job_name
transformer._current_job_name = (
utils.name_from_base(base_name) if base_name is not None else transformer.model_name
)
if transformer.output_path is None:
transformer.output_path = "s3://{}/{}".format(
transformer.sagemaker_session.default_bucket(), transformer._current_job_name
)
job_config = sagemaker.transformer._TransformJob._load_config(
data, data_type, content_type, compression_type, split_type, transformer
)
config = {
"TransformJobName": transformer._current_job_name,
"ModelName": transformer.model_name,
"TransformInput": job_config["input_config"],
"TransformOutput": job_config["output_config"],
"TransformResources": job_config["resource_config"],
}
data_processing = sagemaker.transformer._TransformJob._prepare_data_processing(
input_filter, output_filter, join_source
)
if data_processing is not None:
config["DataProcessing"] = data_processing
if transformer.strategy is not None:
config["BatchStrategy"] = transformer.strategy
if transformer.max_concurrent_transforms is not None:
config["MaxConcurrentTransforms"] = transformer.max_concurrent_transforms
if transformer.max_payload is not None:
config["MaxPayloadInMB"] = transformer.max_payload
if transformer.env is not None:
config["Environment"] = transformer.env
if transformer.tags is not None:
config["Tags"] = transformer.tags
return config
def transform_config_from_estimator(
estimator,
task_id,
task_type,
instance_count,
instance_type,
data,
data_type="S3Prefix",
content_type=None,
compression_type=None,
split_type=None,
job_name=None,
model_name=None,
strategy=None,
assemble_with=None,
output_path=None,
output_kms_key=None,
accept=None,
env=None,
max_concurrent_transforms=None,
max_payload=None,
tags=None,
role=None,
volume_kms_key=None,
model_server_workers=None,
image_uri=None,
vpc_config_override=None,
input_filter=None,
output_filter=None,
join_source=None,
):
"""Export Airflow transform config from a SageMaker estimator
Args:
estimator (sagemaker.model.EstimatorBase): The SageMaker estimator to
export Airflow config from. It has to be an estimator associated
with a training job.
task_id (str): The task id of any
airflow.contrib.operators.SageMakerTrainingOperator or
airflow.contrib.operators.SageMakerTuningOperator that generates
training jobs in the DAG. The transform config is built based on the
training job generated in this operator.
task_type (str): Whether the task is from SageMakerTrainingOperator or
SageMakerTuningOperator. Values can be 'training', 'tuning' or None
(which means training job is not from any task).
instance_count (int): Number of EC2 instances to use.
instance_type (str): Type of EC2 instance to use, for example,
'ml.c4.xlarge'.
data (str): Input data location in S3.
data_type (str): What the S3 location defines (default: 'S3Prefix').
Valid values:
* 'S3Prefix' - the S3 URI defines a key name prefix. All objects with this prefix will
be used as inputs for the transform job.
* 'ManifestFile' - the S3 URI points to a single manifest file listing each S3 object
to use as an input for the transform job.
content_type (str): MIME type of the input data (default: None).
compression_type (str): Compression type of the input data, if
compressed (default: None). Valid values: 'Gzip', None.
split_type (str): The record delimiter for the input object (default:
'None'). Valid values: 'None', 'Line', 'RecordIO', and 'TFRecord'.
job_name (str): transform job name (default: None). If not specified,
one will be generated.
model_name (str): model name (default: None). If not specified, one will
be generated.
strategy (str): The strategy used to decide how to batch records in a
single request (default: None). Valid values: 'MultiRecord' and
'SingleRecord'.
assemble_with (str): How the output is assembled (default: None). Valid
values: 'Line' or 'None'.
output_path (str): S3 location for saving the transform result. If not
specified, results are stored to a default bucket.
output_kms_key (str): Optional. KMS key ID for encrypting the transform
output (default: None).
accept (str): The accept header passed by the client to
the inference endpoint. If it is supported by the endpoint,
it will be the format of the batch transform output.
env (dict): Environment variables to be set for use during the transform
job (default: None).
max_concurrent_transforms (int): The maximum number of HTTP requests to
be made to each individual transform container at one time.
max_payload (int): Maximum size of the payload in a single HTTP request
to the container in MB.
tags (list[dict]): List of tags for labeling a transform job. If none
specified, then the tags used for the training job are used for the
transform job.
role (str): The ``ExecutionRoleArn`` IAM Role ARN for the ``Model``,
which is also used during transform jobs. If not specified, the role
from the Estimator will be used.
volume_kms_key (str): Optional. KMS key ID for encrypting the volume
attached to the ML compute instance (default: None).
model_server_workers (int): Optional. The number of worker processes
used by the inference server. If None, server will use one worker
per vCPU.
image_uri (str): A Docker image URI to use for deploying the model
vpc_config_override (dict[str, list[str]]): Override for VpcConfig set on
the model. Default: use subnets and security groups from this Estimator.
* 'Subnets' (list[str]): List of subnet ids.
* 'SecurityGroupIds' (list[str]): List of security group ids.
input_filter (str): A JSONPath to select a portion of the input to
pass to the algorithm container for inference. If you omit the
field, it gets the value '$', representing the entire input.
For CSV data, each row is taken as a JSON array,
so only index-based JSONPaths can be applied, e.g. $[0], $[1:].
CSV data should follow the `RFC format <https://tools.ietf.org/html/rfc4180>`_.
See `Supported JSONPath Operators
<https://docs.aws.amazon.com/sagemaker/latest/dg/batch-transform-data-processing.html#data-processing-operators>`_
for a table of supported JSONPath operators.
For more information, see the SageMaker API documentation for
`CreateTransformJob
<https://docs.aws.amazon.com/sagemaker/latest/dg/API_CreateTransformJob.html>`_.
Some examples: "$[1:]", "$.features" (default: None).
output_filter (str): A JSONPath to select a portion of the
joined/original output to return as the output.
For more information, see the SageMaker API documentation for
`CreateTransformJob
<https://docs.aws.amazon.com/sagemaker/latest/dg/API_CreateTransformJob.html>`_.
Some examples: "$[1:]", "$.prediction" (default: None).
join_source (str): The source of data to be joined to the transform
output. It can be set to 'Input' meaning the entire input record
will be joined to the inference result. You can use OutputFilter
to select the useful portion before uploading to S3. (default:
None). Valid values: Input, None.
Returns:
dict: Transform config that can be directly used by
SageMakerTransformOperator in Airflow.
"""
model_base_config = model_config_from_estimator(
estimator=estimator,
task_id=task_id,
task_type=task_type,
instance_type=instance_type,
role=role,
image_uri=image_uri,
name=model_name,
model_server_workers=model_server_workers,
vpc_config_override=vpc_config_override,
)
if isinstance(estimator, sagemaker.estimator.Framework):
transformer = estimator.transformer(
instance_count,
instance_type,
strategy,
assemble_with,
output_path,
output_kms_key,
accept,
env,
max_concurrent_transforms,
max_payload,
tags,
role,
model_server_workers,
volume_kms_key,
)
else:
transformer = estimator.transformer(
instance_count,
instance_type,
strategy,
assemble_with,
output_path,
output_kms_key,
accept,
env,
max_concurrent_transforms,
max_payload,
tags,
role,
volume_kms_key,
)
transformer.model_name = model_base_config["ModelName"]
transform_base_config = transform_config(
transformer,
data,
data_type,
content_type,
compression_type,
split_type,
job_name,
input_filter,
output_filter,
join_source,
)
config = {"Model": model_base_config, "Transform": transform_base_config}
return config
def deploy_config(model, initial_instance_count, instance_type, endpoint_name=None, tags=None):
"""Export Airflow deploy config from a SageMaker model
Args:
model (sagemaker.model.Model): The SageMaker model to export the Airflow
config from.
initial_instance_count (int): The initial number of instances to run in
the ``Endpoint`` created from this ``Model``.
instance_type (str): The EC2 instance type to deploy this Model to. For
example, 'ml.p2.xlarge'.
endpoint_name (str): The name of the endpoint to create (default: None).
If not specified, a unique endpoint name will be created.
tags (list[dict]): List of tags for labeling a training job. For more,
see https://docs.aws.amazon.com/sagemaker/latest/dg/API_Tag.html.
Returns:
dict: Deploy config that can be directly used by
SageMakerEndpointOperator in Airflow.
"""
model_base_config = model_config(model, instance_type)
production_variant = sagemaker.production_variant(
model.name, instance_type, initial_instance_count
)
name = model.name
config_options = {"EndpointConfigName": name, "ProductionVariants": [production_variant]}
if tags is not None:
config_options["Tags"] = tags
endpoint_name = endpoint_name or name
endpoint_base_config = {"EndpointName": endpoint_name, "EndpointConfigName": name}
config = {
"Model": model_base_config,
"EndpointConfig": config_options,
"Endpoint": endpoint_base_config,
}
# if there is s3 operations needed for model, move it to root level of config
s3_operations = model_base_config.pop("S3Operations", None)
if s3_operations is not None:
config["S3Operations"] = s3_operations
return config
def deploy_config_from_estimator(
estimator,
task_id,
task_type,
initial_instance_count,
instance_type,
model_name=None,
endpoint_name=None,
tags=None,
**kwargs
):
"""Export Airflow deploy config from a SageMaker estimator
Args:
estimator (sagemaker.model.EstimatorBase): The SageMaker estimator to
export Airflow config from. It has to be an estimator associated
with a training job.
task_id (str): The task id of any
airflow.contrib.operators.SageMakerTrainingOperator or
airflow.contrib.operators.SageMakerTuningOperator that generates
training jobs in the DAG. The endpoint config is built based on the
training job generated in this operator.
task_type (str): Whether the task is from SageMakerTrainingOperator or
SageMakerTuningOperator. Values can be 'training', 'tuning' or None
(which means training job is not from any task).
initial_instance_count (int): Minimum number of EC2 instances to deploy
to an endpoint for prediction.
instance_type (str): Type of EC2 instance to deploy to an endpoint for
prediction, for example, 'ml.c4.xlarge'.
model_name (str): Name to use for creating an Amazon SageMaker model. If
not specified, one will be generated.
endpoint_name (str): Name to use for creating an Amazon SageMaker
endpoint. If not specified, the name of the SageMaker model is used.
tags (list[dict]): List of tags for labeling a training job. For more,
see https://docs.aws.amazon.com/sagemaker/latest/dg/API_Tag.html.
**kwargs: Passed to invocation of ``create_model()``. Implementations
may customize ``create_model()`` to accept ``**kwargs`` to customize
model creation during deploy. For more, see the implementation docs.
Returns:
dict: Deploy config that can be directly used by
SageMakerEndpointOperator in Airflow.
"""
update_estimator_from_task(estimator, task_id, task_type)
model = estimator.create_model(**kwargs)
model.name = model_name
config = deploy_config(model, initial_instance_count, instance_type, endpoint_name, tags)
return config
def processing_config(
processor,
inputs=None,
outputs=None,
job_name=None,
experiment_config=None,
container_arguments=None,
container_entrypoint=None,
kms_key_id=None,
):
"""Export Airflow processing config from a SageMaker processor
Args:
processor (sagemaker.processor.Processor): The SageMaker
processor to export Airflow config from.
inputs (list[:class:`~sagemaker.processing.ProcessingInput`]): Input files for
the processing job. These must be provided as
:class:`~sagemaker.processing.ProcessingInput` objects (default: None).
outputs (list[:class:`~sagemaker.processing.ProcessingOutput`]): Outputs for
the processing job. These can be specified as either path strings or
:class:`~sagemaker.processing.ProcessingOutput` objects (default: None).
job_name (str): Processing job name. If not specified, the processor generates
a default job name, based on the base job name and current timestamp.
experiment_config (dict[str, str]): Experiment management configuration.
Optionally, the dict can contain three keys:
'ExperimentName', 'TrialName', and 'TrialComponentDisplayName'.
The behavior of setting these keys is as follows:
* If `ExperimentName` is supplied but `TrialName` is not a Trial will be
automatically created and the job's Trial Component associated with the Trial.
* If `TrialName` is supplied and the Trial already exists the job's Trial Component
will be associated with the Trial.
* If both `ExperimentName` and `TrialName` are not supplied the trial component
will be unassociated.
* `TrialComponentDisplayName` is used for display in Studio.
container_arguments ([str]): The arguments for a container used to run a processing job.
container_entrypoint ([str]): The entrypoint for a container used to run a processing job.
kms_key_id (str): The AWS Key Management Service (AWS KMS) key that Amazon SageMaker
uses to encrypt the processing job output. KmsKeyId can be an ID of a KMS key,
ARN of a KMS key, alias of a KMS key, or alias of a KMS key.
The KmsKeyId is applied to all outputs.
Returns:
dict: Processing config that can be directly used by
SageMakerProcessingOperator in Airflow.
"""
if job_name is not None:
processor._current_job_name = job_name
else:
base_name = processor.base_job_name
processor._current_job_name = (
utils.name_from_base(base_name)
if base_name is not None
else utils.base_name_from_image(processor.image_uri, Processor.JOB_CLASS_NAME)
)
config = {
"ProcessingJobName": processor._current_job_name,
"ProcessingInputs": input_output_list_converter(inputs),
}
processing_output_config = sagemaker.processing.ProcessingJob.prepare_output_config(
kms_key_id, input_output_list_converter(outputs)
)
config["ProcessingOutputConfig"] = processing_output_config
if experiment_config is not None:
config["ExperimentConfig"] = experiment_config
app_specification = sagemaker.processing.ProcessingJob.prepare_app_specification(
container_arguments, container_entrypoint, processor.image_uri
)
config["AppSpecification"] = app_specification
config["RoleArn"] = processor.role
if processor.env is not None:
config["Environment"] = processor.env
if processor.network_config is not None:
config["NetworkConfig"] = processor.network_config._to_request_dict()
processing_resources = sagemaker.processing.ProcessingJob.prepare_processing_resources(
instance_count=processor.instance_count,
instance_type=processor.instance_type,
volume_kms_key_id=processor.volume_kms_key,
volume_size_in_gb=processor.volume_size_in_gb,
)
config["ProcessingResources"] = processing_resources
if processor.max_runtime_in_seconds is not None:
stopping_condition = sagemaker.processing.ProcessingJob.prepare_stopping_condition(
processor.max_runtime_in_seconds
)
config["StoppingCondition"] = stopping_condition
if processor.tags is not None:
config["Tags"] = processor.tags
return config
def input_output_list_converter(object_list):
"""Converts a list of ProcessingInput or ProcessingOutput objects to a list of dicts
Args:
object_list (list[ProcessingInput or ProcessingOutput]
Returns:
List of dicts
"""
if object_list:
return [obj._to_request_dict() for obj in object_list]
return object_list
|