sagemaker-python-sdk
A library for training and deploying machine learning models on Amazon SageMaker
APACHE-2.0 License
Bot releases are hidden (Show)
Published by laurenyu almost 6 years ago
- feature: Estimator: add input mode to training channels
- feature: Estimator: add model_uri and model_channel_name parameters
- enhancement: Local Mode: support output_path. Can be either file:// or s3://
- enhancement: Added image uris for SageMaker built-in algorithms for SIN/LHR/BOM/SFO/YUL
- feature: Estimators: add support for MXNet 1.3.0, which introduces a new training script format
- feature: Documentation: add explanation for the new training script format used with MXNet
- feature: Estimators: add
distributionsfor customizing distributed training with the new training script format
Published by mvsusp almost 6 years ago
- feature: add support for TensorFlow 1.11.0
Published by icywang86rui about 6 years ago
- feature: Local Mode: Add support for Batch Inference
- feature: Add timestamp to secondary status in training job output
- bug-fix: Local Mode: Set correct default values for additional_volumes and additional_env_vars
- enhancement: Local Mode: support nvidia-docker2 natively
- warning: Frameworks: add warning for upcoming breaking change that makes framework_version required
Published by ChoiByungWook about 6 years ago
- enhancement: Enable setting VPC config when creating/deploying models
- enhancement: Local Mode: accept short lived credentials with a warning message
- bug-fix: Local Mode: pass in job name as parameter for training environment variable
Published by jesterhazy about 6 years ago
- enhancement: Local Mode: add training environment variables for AWS region and job name
- doc-fix: Instruction on how to use preview version of PyTorch - 1.0.0.dev.
- doc-fix: add role to MXNet estimator example in readme
- bug-fix: default TensorFlow json serializer accepts dict of numpy arrays
Published by ChoiByungWook about 6 years ago
- bug-fix: setting health check timeout limit on local mode to 30s
- bug-fix: make Hyperparameters in local mode optional.
- enhancement: add support for volume KMS key to Transformer
- feature: add support for GovCloud
Published by winstonaws about 6 years ago
- feature: add train_volume_kms_key parameter to Estimator classes
- doc-fix: add deprecation warning for current MXNet training script format
- doc-fix: add docs on deploying TensorFlow model directly from existing model
- doc-fix: fix code example for using Gzip compression for TensorFlow training data
Published by yangaws about 6 years ago
- feature: add support for TensorFlow 1.10.0
Published by jesterhazy about 6 years ago
- doc-fix: fix rst warnings in README.rst
Published by nadiaya about 6 years ago
- bug-fix: Local Mode: Create output/data directory expected by SageMaker Container.
- bug-fix: Estimator accepts the vpc configs made capable by 1.9.1
Published by yangaws about 6 years ago
- feature: add support for TensorFlow 1.9
Published by ChoiByungWook about 6 years ago
- bug-fix: Estimators: Fix serialization of single records
- bug-fix: deprecate enable_cloudwatch_metrics from Framework Estimators.
- enhancement: Enable VPC config in training job creation
Published by laurenyu about 6 years ago
- feature: Estimators: add support for MXNet 1.2.1
Published by yangaws about 6 years ago
- bug-fix: removing PCA from tuner
- feature: Estimators: add support for Amazon k-nearest neighbors(KNN) algorithm
Published by icywang86rui about 6 years ago
- bug-fix: Prediction output for the TF_JSON_SERIALIZER
- enhancement: Add better training job status report
Published by andremoeller about 6 years ago
- bug-fix: get_execution_role no longer fails if user can't call get_role
- bug-fix: Session: use existing model instead of failing during create_model()
- enhancement: Estimator: allow for different role from the Estimator's when creating a Model or Transformer
Published by laurenyu over 6 years ago
- feature: Transformer: add support for batch transform jobs
- feature: Documentation: add instructions for using Pipe Mode with TensorFlow
Published by yangaws over 6 years ago
- feature: Added multiclass classification support for linear learner algorithm.
Published by icywang86rui over 6 years ago
- feature: Add Chainer 4.1.0 support
- bug-fix: Use chainer_full_version fixture in Chainer integration tests
Published by yangaws over 6 years ago
- feature: Added Docker Registry for all 1p algorithms in amazon_estimator.py
- feature: Added get_image_uri method for 1p algorithms in amazon_estimator.py
- Support SageMaker algorithms in FRA and SYD regions