Release Notes

This page documents production updates to AI Platform. You can periodically check this page for announcements about new or updated features, bug fixes, known issues, and deprecated functionality.

To get the latest product updates delivered to you, add the URL of this page to your feed reader.

September 9, 2019

Runtime version 1.14 now supports training with TPUs using TensorFlow 1.14.

September 6, 2019

When you deploy a model version to AI Platform Prediction, you can now configure AI Platform Prediction to log a sample of online prediction requests received by the model together with the responses it sends to these requests. AI Platform Prediction saves these request-response pairs to BigQuery. This feature is in beta.

Learn how to how to enable request-response logging and read about the configuration options for this type of logging.

August 28, 2019

Training with custom containers is now generally available.

NVIDIA Tesla P4 and NVIDIA Tesla T4 GPUs are now generally available for training. Read about using GPUs for training and learn about GPU pricing.

The documentation for AI Platform Notebooks has moved to a new location.

August 26, 2019

AI Platform Training now supports using Cloud TPU devices with TPU v3 configurations to accelerate your training jobs. TPU v3 accelerators for AI Platform Training are available in beta.

Learn more about how to configure your training job to use TPU v3 accelerators and read about TPU v3 pricing.

August 22, 2019

Continuous evaluation for AI Platform Prediction is now available in beta. When you create a continuous evaluation job, AI Platform Data Labeling Service assigns human reviewers to provide ground truth labels for a portion of your model version's online predictions; alternatively, you can provide your own ground truth labels. Then Data Labeling Service compares these labels to your model version's predictions to calculate daily evaluation metrics.

Learn more about continuous evaluation.

August 16, 2019

AI Platform runtime versions 1.13 and 1.14 now include numpy 1.16.4 instead of numpy 1.16.0. View the runtime version list for the full list of packages included in runtime versions 1.13 and 1.14.

August 1, 2019

The AI Platform Training and Prediction documentation has been reorganized. Previously, documentation for using AI Platform with specific machine learning frameworks was separated into sections. You can now navigate to all Training and Prediction documentation from the AI Platform documentation home page.

July 19, 2019

AI Platform runtime version 1.14 is now available for training and prediction. This version supports TensorFlow 1.14.0 and includes other packages as listed in the runtime version list.

Training with TPUs is not supported in runtime version 1.14 at this time.

AI Platform runtime version 1.12 now supports TensorFlow 1.12.3. View the runtime version list for the full list of packages included in runtime version 1.12.

July 17, 2019

The prediction input format for the following built-in algorithms has changed:

Instead of a raw string, make sure to format each instance as a JSON with a "csv_row" key and "key" key. This "key" is useful for doing batch predictions using AI Platform. For online predictions, you can pass in a dummy value to the "key" key in your input JSON request. For example:

{"csv_row": "1, 2, 3, 4, 0, abc", "key" : "dummy-key"}

See the Census Income tutorial for an example.

June 19, 2019

The asia-southeast1 (Singapore) region is now available for training and batch prediction. You can use P4 or T4 GPUs for training in the region. Read about pricing for training in asia-southeast1, including pricing for accelerators.

June 18, 2019

Runtime version 1.13 now supports training with TPUs using TensorFlow 1.13.

Support for training with TPUs in runtime version 1.11 ended on June 6, 2019.

June 12, 2019

You can now view monitoring data for training jobs directly within the AI Platform Job Details page in the GCP Console. The following charts are available:

  • CPU, GPU, and memory utilization, broken down by master, worker, and parameter servers.
  • Network usage: the rate per second of bytes sent and received.

Learn more about how to monitor resource utilization for your training jobs.

There are new options for filtering jobs within the AI Platform Jobs page in the GCP Console. You can filter jobs by Type and by whether or not the job used HyperTune.

Learn more about how to filter your training jobs.

You can now view and sort hyperparameter tuning trials within the AI Platform Job Details page in the GCP Console. If your training job uses hyperparameter tuning, your Job Details page includes a HyperTune trials table, where you can view metrics such as RMSE, learning rate, and training steps. You can also access logs for each trial. This table makes it easier to compare individual trials.

Learn more about how to view your hyperparameter tuning trials.

June 5, 2019

You can now specify a service account for your model version to use when you deploy a custom prediction routine to AI Platform Prediction. This feature is in beta.

June 3, 2019

You can now create AI Platform Notebooks instances with R and core R packages installed. Learn how to install R dependencies, and read guides for using R with BigQuery in AI Platform Notebooks and using R and Python in the same notebook.

May 3, 2019

T4 GPUs are now in beta for AI Platform Training. For more information, see the guides to using GPUs, their regional availability, and their pricing.

AI Platform runtime version 1.12 now supports TensorFlow 1.12.2. View the runtime version list for the full list of packages included in runtime version 1.12.

April 25, 2019

AI Platform Prediction now supports custom prediction routines in beta. Custom prediction routines let you provide AI Platform with custom code to use when it serves online predictions from your deployed model. This can be useful for preprocessing prediction input, postprocessing your model's predictions, and more.

Work through a tutorial on deploying a custom prediction routine with Keras or one on deploying a custom prediction routine with scikit-learn.

AI Platform Prediction now supports custom transformers for scikit-learn pipelines in beta. This lets you provide AI Platform with custom code to use during online prediction. Your deployed scikit-learn pipeline uses this code when it serves predictions.

Work through a tutorial on training and deploying a custom scikit-learn pipeline.

AI Platform Prediction now supports logging of your prediction nodes' stderr and stdout streams to Stackdriver logging during online prediction. Stream logging is in beta. You can enable this type of logging in addition to—or in place of—the access logging that was already available. It can be useful for understanding how your deployment handles prediction requests.

April 1, 2019

AI Platform now offers reduced pricing for training, online prediction and batch prediction.

Learn more about AI Platform pricing.

March 28, 2019

AI Platform now offers training with built-in algorithms. You can submit your data for automatic preprocessing, and train a model on the TensorFlow linear learner, TensorFlow wide and deep, and XGBoost algorithms without writing any code.

Learn more about training with built-in algorithms.

March 25, 2019

AI Platform runtime version 1.13 now supports TensorFlow 1.13.1. View the runtime version list for the full list of packages included in runtime version 1.13.

March 8, 2019

Support for training with TPUs in AI Platform runtime version 1.9 ended on March 8, 2019. See the currently supported versions in the runtime version list.

March 6, 2019

AI Platform runtime version 1.13 is now available for training and prediction. This version supports TensorFlow 1.13 and includes other packages as listed in the runtime version list.

Training with TPUs is not supported in runtime version 1.13 at this time.

March 1, 2019

AI Platform Notebooks is now available in beta. AI Platform Notebooks enables you to create and manage virtual machine (VM) instances that are pre-packaged with JupyterLab and a suite of deep learning software.

Visit the AI Platform Notebooks overview and the guide to creating a new notebook instance to learn more.

February 13, 2019

Cloud TPU is now generally available for training TensorFlow models. Tensor Processing Units (TPUs) are Google's custom-developed accelerators for machine-learning workloads.

See how to use TPUs to train your models on AI Platform, and read more about their pricing.

February 7, 2019

Training with custom containers is now available in Beta. This feature allows you to run your training application on AI Platform using a custom Docker image. You can build your custom container with the ML frameworks of your choice. Get started with training a PyTorch model by using custom containers.

You can now configure training jobs with certain Compute Engine machine types. This provides additional flexibility for allocating computing resources to your training jobs. This feature is available in Beta.

When you configure your job with Compute Engine machine types, you may attach a custom set of GPUs.

Read more about Compute Engine machine types, GPU attachments, and their pricing.

P4 GPUs are now in Beta for training. For more information, see the guides to using GPUs, their regional availability, and their pricing.

February 1, 2019

Quad core CPUs are now available in Beta for online prediction. The names of the machine types are changed, and pricing is updated.

  • Set machineType on projects.models.versions.create to specify the machine type to use for serving. Use mls1-c4-m2 for quad core CPUs. The default is the single core CPU, mls1-c1-m2.
  • The following machine names used in Alpha are deprecated: mls1-highmem-1 and mls1-highcpu-4.
  • For more information, see the guide to online prediction.
  • See the updated pricing for serving machine types.

January 25, 2019

Online prediction is now available in the us-east4 region. See the guide to region availability.

January 10, 2019

V100 GPUs are now generally available for training. For more information, see the guides to using GPUs and pricing.

December 19, 2018

The AI Platform runtime versions 1.11 and 1.12 are now available for training and prediction. These versions support TensorFlow 1.11 and 1.12 respectively, and other packages as listed in the runtime version list.

TPU training support has been added for AI Platform runtime versions 1.11 and 1.12. Version 1.10 is not supported. See the currently supported versions in the runtime version list.

Each AI Platform runtime version now includes joblib. The earliest runtime version that includes joblib is version 1.4.

October 26, 2018

TPU training support for Cloud ML runtime version 1.8 ended on Oct 26, 2018. See the currently supported versions in the runtime version list.

October 11, 2018

The AI Platform runtime version 1.11 is rolled back due to errors caused by a CuDNN version mismatch during GPU training. The current workaround is to use runtime version 1.10. For more details, see the runtime version list.

October 5, 2018

The AI Platform runtime version 1.11 is now available for training and prediction. This version supports TensorFlow 1.11 and other packages as listed in the runtime version list.

August 31, 2018

The AI Platform runtime version 1.10 is now available for training and prediction. This version supports TensorFlow 1.10 and other packages as listed in the runtime version list.

August 27, 2018

V100 GPUs are now in Beta for training. Using V100 GPUs now incurs charges. For more information, see the guides to using GPUs and pricing.

P100 GPUs are now generally available for training. For more information, see the guides to using GPUs and pricing.

Two new regions: us-west1, europe-west4 are now available for training. See regions page for info.

August 24, 2018

TPU training support for Cloud ML runtime version 1.7 ended on Aug 24, 2018. See the currently supported versions in the runtime version list.

August 9th, 2018

We're delighted to announce significant price reductions for online prediction with AI Platform.

The following table shows the previous pricing and the new pricing:

Region Previous price per node per hour New price per node per hour
US $0.30 USD $0.056 USD
Europe $0.348 USD $0.061 USD
Asia Pacific $0.348 USD $0.071 USD

See the pricing guide for details.

August 8th, 2018

We're delighted to announce promotional pricing for Cloud TPU with AI Platform, resulting in significant price reductions.

The following table shows the previous pricing and the new pricing:

Region: US Previous price per TPU per hour New price per TPU per hour
Scale tier: BASIC_TPU (Beta) $9.7674 USD $6.8474 USD
Custom machine type: cloud_tpu (Beta) $9.4900 USD $6.5700 USD

Note that the table shows pricing in the US region only. There is no change in Cloud TPU availability on AI Platform. See the pricing guide for details.

August 6, 2018

The AI Platform runtime version 1.9 is now available for training and prediction. This version supports TensorFlow 1.9 and other packages as listed in the runtime version list.

July 23, 2018

AI Platform now supports scikit-learn and XGBoost for training. This feature is generally available. See the guide to training with scikit-learn and XGBoost on AI Platform.

Online prediction support for scikit-learn and XGBoost is now generally available.

July 12, 2018

You can add labels to your AI Platform resources—jobs, models, and model versions—then use those labels to organize the resources into categories. Labels are also available on operations—in this case the labels are derived from the resource to which the operation applies. Read more about adding and using labels.

June 26, 2018

The following additional regions are now fully available:

  • us-east1
  • asia-northeast1

See more details about region availability.

June 13, 2018

TPU training support for Cloud ML runtime version 1.6 ended on June 13, 2018. See the currently supported versions in the Runtime Version List.

May 29, 2018

You can now use Cloud TPU (Beta) with TensorFlow 1.8 and AI Platform runtime version 1.8.

Background information: Cloud TPU became available in AI Platform on May 14th in runtime versions 1.6 and 1.7. Last week saw the release of runtime version 1.8, but at that time Cloud TPU was not yet available with TensorFlow 1.8. Now it is. See how to use TPUs to train your models on AI Platform.

May 16, 2018

The AI Platform runtime version 1.8 is now available for training and prediction. This version supports TensorFlow 1.8 and other packages as listed in the runtime version list.

May 15, 2018

You can now update the minimum number of nodes for autoscaling on an existing model version, as well as specifying the attribute when creating a new version.

May 14, 2018

AI Platform now offers Cloud TPU (Beta) for training TensorFlow models. Tensor Processing Units (TPUs) are Google’s custom-developed ASICs, used to accelerate machine-learning workloads. See how to use TPUs to train your models on AI Platform.

April 26, 2018

The AI Platform runtime version 1.7 is now available for training and prediction. This version supports TensorFlow 1.7 and other packages as listed in the runtime version list.

April 16, 2018

Hyperparameter algorithms: When tuning the hyperparameters in your training job, you can now specify a search algorithm in the HyperparameterSpec. Available values are:

  • GRID_SEARCH: A simple grid search within the feasible space. This option is particularly useful if you want to specify a number of trials that is more than the number of points in the feasible space. In such cases, if you do not specify a grid search, the AI Platform default algorithm may generate duplicate suggestions. To use grid search, all parameters must be of type INTEGER, CATEGORICAL, or DISCRETE.
  • RANDOM_SEARCH: A simple random search within the feasible space.

If you do not specify an algorithm, your job uses the default AI Platform algorithm, which drives the parameter search to arrive at the optimal solution with a more effective search over the parameter space. For more about hyperparameter tuning, see the hyperparameter tuning overview.

April 5, 2018

AI Platform now supports scikit-learn and XGBoost for online prediction. This feature is in Beta.

  • Set framework on projects.models.versions.create to specify your machine learning framework when creating a model version. Valid values are TENSORFLOW, SCIKIT_LEARN, and XGBOOST. The default is TENSORFLOW. If you specify SCIKIT_LEARN or XGBOOST, you must also set runtimeVersion to 1.4 or greater on the model version.
  • See the guide to scikit-learn and XGBoost on AI Platform.

Python 3.5 is available for online prediction.

March 20, 2018

The AI Platform runtime version 1.6 is now available for training and prediction. This version supports TensorFlow 1.6 and other packages as listed in the runtime version list.

March 13, 2018

The AI Platform runtime version for TensorFlow 1.5 is now available for training and prediction. For more information, see the Runtime Version List.

February 8, 2018

Added new features for hyperparameter tuning: automated early stopping of trials, resuming a previous hyperparameter tuning job, and additional efficiency optimizations when you run similar jobs. For more information, See the hyperparameter tuning overview.

December 14, 2017

The AI Platform runtime version for TensorFlow 1.4 is now available for training and prediction. For more information, see the Runtime Version List.

Python 3 is now available for training as part of the AI Platform runtime version for TensorFlow 1.4. For more information, see the Runtime Version List.

Online prediction is now generally available for single core serving. See the guide to online prediction and the blog post.

Pricing has been reduced and simplified for both training and prediction. See the pricing details, the blog post, and the comparison of old and current prices in the pricing FAQ.

P100 GPUs are now in Beta. Using P100 GPUs now incurs charges. For more information, see Using GPUs and Pricing.

October 26, 2017

Audit logging for AI Platform is now in Beta. For more information, see Viewing Audit Logs.

September 25, 2017

Predefined IAM roles for AI Platform are available for general use. For more information, see Access Control.

June 27, 2017

The AI Platform runtime version for TensorFlow 1.2 is now available for training and prediction. For more information, see the Runtime Version List.

The older runtime versions with TensorFlow 0.11 and 0.12 are no longer supported on AI Platform. For more information, see the Runtime Version List and the support timelines for older runtime versions.

May 9, 2017

Announced general availability of GPU-enabled machines. For more information, see Using GPUs for Training Models in the Cloud.

April 27, 2017

GPUs are now available in the us-central1 region. For the full list of regions that support GPUs, see Using GPUs for Training Models in the Cloud.

v1 (March 8th, 2017)

Announced general availability of AI Platform. Version 1 of AI Platform is available for general use for training models, deploying models, and generating batch predictions. The hyperparameter tuning feature is also available for general use, but online prediction and GPU-enabled machines remain in beta.

Online prediction is now in the Beta launch stage. Its use is now subject to the AI Platform pricing policy, and follows the same pricing formula as batch prediction. While it remains in Beta, online prediction is not intended for use in critical applications.

The environments that AI Platform uses to train models and get predictions have been defined as AI Platform runtime versions. You can specify a supported runtime version to use when training, defining a model resource, or requesting batch predictions. The primary difference in runtime versions at this time is the version of TensorFlow supported by each, but more differences may arise over time. You can find the details in the runtime version list.

You can now run batch prediction jobs against TensorFlow SavedModels that are stored in Google Cloud Storage, not hosted as a model version in AI Platform. Instead of supplying a model or version ID when you create your job, you can use the URI of your SavedModel.

The Google Cloud Machine Learning SDK, formerly released as Alpha, is deprecated, and will no longer be supported effective May 7, 2017. Most of the functionality exposed by the SDK has moved to the new TensorFlow package, tf.Transform. You can use whatever technology or tool you like to preprocess your input data. However, we recommend tf.Transform as well as services that are available on Google Cloud Platform, including Google Cloud Dataflow, Google Cloud Dataproc, and Google BigQuery.

v1beta1 (September 29th, 2016)

Online prediction is an Alpha feature. Though AI Platform overall is in its Beta phase, online prediction is still undergoing significant changes to improve performance. You will not be charged for online prediction while it remains in Alpha.

Preprocessing and the rest of the AI Platform SDK are Alpha features. The SDK is undergoing active development to better integrate AI Platform with Apache Beam.

이 페이지가 도움이 되었나요? 평가를 부탁드립니다.

다음에 대한 의견 보내기...

도움이 필요하시나요? 지원 페이지를 방문하세요.