Changelog

1.66.0 (2024-09-11)

Features

  • Add max_wait_duration option to custom jobs. (ee65917)

  • Add Ray 2.33 support to SDK Client Builder, remove deprecated protocol_version from ray client context. (708a67b)

  • Add support for version 2.33 for RoV Bigquery read/write, remove dead code from version 2.4 (a4b6c60)

  • Update Ray system tests to be compatible with new RoV 2.33 changes (8c7bf27)

  • Update setup.py for including Ray v2.33, restrict RoV prediction to 2.9.3 for now (71c6f3c)

Bug Fixes

  • Identify nested Vertex Tensorboard profile runs for uploading when standard event data is not present (0a08027)

  • Move region tag to include imports (6d1f7fd)

  • Move region tag to include imports (#4357) (6d1f7fd)

  • Rollback change to tensorboard uploader causing increased latency (879dbcd)

Documentation

  • Update SDK Job Submission with Ray v2.33 updated comments (7fda11f)

1.65.0 (2024-09-04)

⚠ BREAKING CHANGES

  • Tokenization - Deprecated ComputeTokenResult.token_info_list in favor of ComputeTokenResult.tokens_info

Features

  • Add support for system instruction and tools in tokenization. (72fcc06)

  • Add vector search alpha to rag retrieval for hybrid search ranking (6624ebe)

  • Adding Weaviate Vector DB option for RAG corpuses to SDK (9b28202)

  • GenAI - Added system_instruction and tools support to GenerativeModel.count_tokens (50fca69)

  • GenAI - Added Llama3 support in GenAI batch prediction (6166152)

  • PrivateEndpoint.stream_raw_predict (197f333)

  • Support reserved_ip_ranges for VPC network in Ray on Vertex cluster (36a56b9)

  • Tokenization - Deprecated ComputeTokenResult.token_info_list in favor of ComputeTokenResult.tokens_info (efbcb54)

Bug Fixes

  • Tensorboard - Fixed bug in tensorboard uploader where profile logs are not uploaded from nested run directories. (37627de)

  • Tokenizers - Fixed Tokenizer.compute_tokens (c29fa5d)

1.64.0 (2024-08-27)

Features

  • Endpoint - Add support for Prediction dedicated endpoint. predict/rawPredict/streamRawPredict can use dedicated DNS to access the dedicated endpoint. (3d68777)

  • GenAI - Added the model Distillation feature (private preview) (a0d4ff2)

  • Grounding - Allow initialization of grounding.VertexAISearch with full resource name or data store ID, project ID, and location. (f334321)

  • Evaluation - Make Rouge class available in base level init (aed82a1)

  • Feature Store - Read for online store w/private service connect (7af80c6)

  • Ray - Support autoscaling in Ray on Vertex (961da42)

Bug Fixes

  • Fix error in tensorboard uploader thrown when time_series_id is None (d59a052)

  • Evaluation - Fix typo in prompt templates: (c8fa7a8)

Documentation

  • samples: Adding code sample for vector search create streaming index (71464e7)

1.63.0 (2024-08-20)

Features

  • A new field satisfies_pzs is added to message .google.cloud.aiplatform.v1.BatchPredictionJob (#4192) (6919037)

  • Add advanced PDF parsing option for RAG file import (6e1dc06)

  • Add multithreading for custom metric computation. (2c93fc1)

  • Add progress bar for generating inference. (b78714f)

  • Add progress bar to custom metrics. (3974aec)

  • Add Prompt class support for configs and Prompt.generate_content wrapper (7f1e031)

  • GenAI - Added seed parameter to the GenerationConfig class (9f1e073)

  • GenAI - Added the Candidate.avg_logprobs property (de80695)

  • GenAI - Released the Prompt feature to Public Preview (64eeab8)

  • GenAI Evaluation: Add generic model-based PointwiseMetric and PairwiseMetric classes that allow customizing metric prompt templates. Add PointwiseMetricPromptTemplate, PairwiseMetricPromptTemplate classes to help formulate and customize metric prompt templates. Add metric_column_mapping parameter to EvalTask for metric prompt template input variable name mapping. (fd38b49)

  • GenAI Evaluation: Open source model-based metric prompt template examples for Gemini 1.5 Pro autorater. Add MetricPromptTemplateExamples class to help retrieve model-based metric prompt templates. (fd38b49)

  • GenAI Evaluation: Release GenAI Evaluation SDK GA features to vertexai.preview module. (fd38b49)

  • Publish GenAI Evaluation SDK GA features to vertexai.evaluation module. Switch GenAI Evaluation Service client to v1 version. (45e4251)

Bug Fixes

  • Add support of display_name to create_cached_content in python SDK (ecc2d54)

  • Always upload the pickled object and dependencies tarball when creating ReasoningEngine (34ef5a3)

  • Remove grouding attribution (f6ece65)

Documentation

  • Update Prompt docstring for batch prompting (e96b6e6)

1.62.0 (2024-08-13)

Features

  • Add metadata to evaluation result. (375095e)

  • Add Prompt class for multimodal prompt templating (1bdc235)

  • Add support for query method in Vertex AI Extension SDK (0008735)

  • Add support for reservation affinity in custom training jobs. (802609b)

  • Add support for strategy in custom training jobs. (a076191)

  • Adding spot, reservation_affinity to Vertex SDK (3e785bd)

  • Support api keys in initializer and create_client (7404f67)

  • Support creating optimized online store with private service connect (659ba3f)

  • Support disable Cloud logging in Ray on Vertex (accaa97)

  • Support PSC-Interface in Ray on Vertex (accaa97)

Bug Fixes

  • Added credentials, project, and location on PipelineJobSchedule init (281c171)

  • Avoid breakage of langchain from orjson 3.10.7 (c990f73)

  • Deprecate disable_attribution in GoogleSearchRetrieval. (c68d559)

Documentation

  • Update the docstring for compute_tokens method. (849e8d4)

1.61.0 (2024-08-05)

Features

  • Add a warning message for scheduled deprecation of Coherence metric class (7f238fb)

  • Add deprecation messages for all model-based metric classes (71c0fd3)

  • Add support for task type (CODE_RETRIEVAL_QUERY) through get_embeddings. (f2ce1e4)

  • Add system_instruction to LangchainAgent template. (c71c3dd)

  • Adding Slack and Jira data connector for RAG to SDK (d92e7c9)

  • Allow protobuf 5.x (ce9cd5d)

  • LVM - Release ImageGenerationModel to GA (718c199)

  • Support “update” for reasoning engine. (b73ef3e)

  • Update Rapid Evaluation Service QPS. Add a customizable evaluation service QPS parameter. (9ee9289)

Documentation

  • Change init sample to use vertexai (829e0f6)

  • Make small fixes to file import documentation (f7d65c3)

1.60.0 (2024-07-24)

Features

  • Add preflight validations to PipelineJob submit and run methods. (c5a3535)

  • Add support for langchain v0.2+ package versions in default installation (259b638)

  • GenAI - Added tokenization support via GenerativeModel.compute_tokens (cfe0cc6)

  • GenAI - ContextCaching - allow from_cached_content to take the cached_content resource name (8f53902)

  • Make count_tokens generally-available at TextEmbeddingModel. (efb8413)

Bug Fixes

  • Avoid throw error when Part.text is empty in modality content checks (bbd4a49)

  • Correct logit_bias type annotation to accept keys as strings (2676d25)

  • Create FV embedding dimensions sample - dimensions should be an int (2aa221e)

  • Fix the sync option for Model Monitor job creation (22151e2)

  • Include DeploymentResourcePool class in aiplatform top-level sdk module (ecc4f09)

  • Overriding the current TracerProvider when enabling tracing (1476c10)

  • Pass the project ID from vertexai.init to CloudTraceSpanExporter when enable_tracing=True for LangchainAgent (3ec043e)

Documentation

  • GenAI - Update README.md for Vertex Generative AI SDK for Python to add subsections to the right nav. (42af742)

1.59.0 (2024-07-09)

Features

  • Add model and contents fields to ComputeTokensRequest v1 (f6e7b9c)

  • Add model and contents fields to ComputeTokensRequest v1beta1 (f6e7b9c)

  • Add role field to TokensInfo v1 (f6e7b9c)

  • Add role field to TokensInfo v1beta1 (f6e7b9c)

  • GenAI - Tuning - Released the Supervised Fine Tuning feature o GA (ae47639)

Bug Fixes

  • deps: Require proto-plus 1.22.3 (4131e65)

  • Offline store - set application name + remove session param (7395665)

1.58.0 (2024-07-03)

Features

  • Add deploy_metadata to PublisherModel.Deploy v1 (71e41c8)

  • Add deploy_metadata to PublisherModel.Deploy v1beta1 (b5c3cdd)

  • Add display tuning job button for Ipython environments when getting an existing job (872b455)

  • Add private_service_connect_config and service_attachment fields to DedicatedServingEndpoint v1 (71e41c8)

  • Add satisfies_pzs and satisfies_pzi fields to Model v1 (71e41c8)

  • Add satisfies_pzs and satisfies_pzi fields to Model v1beta1 (b5c3cdd)

  • Add satisfies_pzs and satisfies_pzi fields to Tensorboard v1 (71e41c8)

  • Add satisfies_pzs and satisfies_pzi fields to Tensorboard v1beta1 (b5c3cdd)

  • Add UpdateDeploymentResourcePool method to DeploymentResourcePoolService v1 (71e41c8)

  • Add UpdateDeploymentResourcePool method to DeploymentResourcePoolService v1beta1 (b5c3cdd)

  • Add use_effective_order field to BleuSpec v1beta1 (b5c3cdd)

  • GenAI - Evaluation - Add a progress bar for evaluation service requests (bbffb0d)

  • GenAI - Evaluation - Implement rate limiter and refactor parallelization for online evaluation service requests (bbffb0d)

  • GenAI - Evaluation - Return partial evaluation results with error logging (bbffb0d)

  • Migrate DeploymentResourcePool and associated functionality to V1 namespace for GA launch of model co-hosting. (1474d98)

Bug Fixes

  • IndexConfig - use TreeAhConfig as default algorithm_config. (341d287)

  • LVM - Update Video.load_from_file() to support storage.googleapis.com links (b63f960)

Documentation

  • Update comments of AutoscalingSpec v1 (71e41c8)

  • Update comments of AutoscalingSpec v1beta1 (b5c3cdd)

  • Update import paths for Gemini README (46b3042)

1.57.0 (2024-06-26)

Features

  • Add _ModelBasedMetric base class to vertexai.preview.evaluation.metrics and allow metric spec customization (536f1d5)

  • Add _AutomaticMetric type and support customizing metric spec for automatic pointwise metrics. Add Rouge metric to vertexai.preview.evaluation.metrics (f6b6dee)

  • Add 12 model-based pointwise metric classes to vertexai.preview.evaluation.metrics (4742a87)

  • Add display tuning job button for Ipython environments when starting a new job (b1e9a6c)

  • Add encryption_spec to TuningJob (5a6ce78)

  • Add enum value MALFORMED_FUNCTION_CALL to .google.cloud.aiplatform.v1beta1.content.Candidate.FinishReason (5a6ce78)

  • Add fields grounding_chunks and grounding_supports to GroundingMetadata (5a6ce78)

  • Add MALFORMED_FUNCTION_CALL to FinishReason (5a6ce78)

  • Add preflight_validations to PipelineJob (5a6ce78)

  • Add streamRawPredict rpc to prediction service (5a6ce78)

  • Add support for hybrid queries for private endpoint in Matching Engine Index Endpoint. (ce65eab)

  • Add text field in Segment (5a6ce78)

  • Change the icon for the model evaluation link to a lightbulb for ipython environments (7dad7f0)

  • Endpoint.direct_predict (9351e1a)

  • Endpoint.direct_predict_async (9351e1a)

  • Endpoint.direct_raw_predict (9351e1a)

  • Endpoint.direct_raw_predict_async (9351e1a)

  • Endpoint.stream_direct_predict (9351e1a)

  • Endpoint.stream_direct_raw_predict (9351e1a)

  • Endpoint.stream_raw_predict (9351e1a)

  • Expose new text embedding tuning parameters in GA namespace. (249a5fa)

  • GenAI - Batch Prediction - Added support for tuned GenAI models (a90ee8d)

  • Mark default profile run as complete when calling end_upload_tb_log (6397f30)

  • Sample code for Vertex AI Feature Store (2c28b4e)

  • Sample code for Vertex AI Feature Store (2fbf5a5)

  • Tokenization - Added count_tokens support for local tokenization (a8a4c30)

  • Update rich html Ray on Vertex Client Context widget (380c9d9)

Bug Fixes

  • FeatureGroup.create() - remove unused entity id column param (c46f3e9)

  • Fix bug where retry_timeout does not work with evaluation with experiments (a31ac4d)

  • Use default run_name in Tensorboard uploader for direct directory upload. (78a92a1)

Documentation

  • A comment for field distibution in message .google.cloud.aiplatform.v1beta1.model_monitoring_stats.ModelMonitoringStatsDataPoint is changed. (5a6ce78)

  • A comment for field name in message .google.cloud.aiplatform.v1beta1.cached_content.CachedContent is changed (5a6ce78)

  • A comment for field source in message .google.cloud.aiplatform.v1beta1.tool.Retrieval is added. (5a6ce78)

  • Update the documentation for the time_series_dataset and video_dataset classes (2cf4cb9)

1.56.0 (2024-06-18)

Features

  • Add retry_timeout to EvalTask in vertexai.preview.evaluation (4d9ee9d)

  • Add hybrid query example to vector search sample. (510da5e)

  • Add metric classes for 2 pairwise metrics for rapid evaluation SDK. (831c8e4)

  • Add pipeline_job_name to allow PipelineJob.get(pipeline_job_name) (32e3b22)

  • Add sample code show how to create an optimized private online store in Vertex AI Feature Store. (e352175)

  • GenAI - Context Caching - add get() classmethod and refresh() instance method (6be874a)

  • GenAI - Context Caching - also print model_name and expire_time. (d548c11)

  • GenAI - Tuning - Added support for CMEK (eb651bc)

Bug Fixes

  • Do not reset aiplatform.Experiment or aiplatform.ExperimentRun unnecessarily when running tensorboard uploader. (28a091a)

Documentation

  • Update the documentation for the time_series_dataset and video_dataset classes (d5dc7b5)

1.55.0 (2024-06-12)

Features

  • view model evaluations when AustoSxS pipeline completes (17c59c4c2)

  • Add support for user-configurable 1P embedding models and quota for RAG (8b3beb666)

  • Add Webpage Browser Extension to Vertex AI Extension SDK (c222f4bdb)

  • Add enable_tracing to LangchainAgent. (cad035cb3)

  • Implement multithreaded batch inference in Rapid Evaluation SDK for performance improvement (3c6d1732b)

Bug Fixes

  • support VPC and BYOSA case in Ray on Vertex JobSubmissionClient using cluster resource name (662d039c9)

  • Include tensorboard profiler requirements when installing google-cloud-aiplatformtensorboard

  • fix numerical NaN experiment run logging error in EvalTask. (641faec0b)

  • Generate uuid run_name when run_name_prefix is not specified. (4e2d87f9a)

Documentation

  • Update the documentation for the image_dataset class (583cb9313)

  • Include imports in experiment_tracking samples directly in cloud.google.com snippets. (3ffb6cba7)

1.54.1 (2024-06-07)

Bug Fixes

  • _append_tool_name list index out of range (#3914) (323c2f5)

1.54.0 (2024-06-06)

BREAKING CHANGES

  • Remove Vertex SDK data science package (ec4ec8f)

Features

  • Add display experiment run button for Ipython environments (ba65828)

  • Add hybrid search for public find_neighbors() call. (9d35617)

  • Enable Ray Job submission without VPC peering (37875b5)

  • GenAI - Allowed callable functions to return values directly in Automatic Function Calling (768af67)

  • GenAI - Release ToolConfig to GA (bc8b14a)

  • Sample code for Vertex AI Feature Store (6c14e8b)

  • Support VertexTool in langchain template. (28a3c56)

Bug Fixes

  • Allow non-lro delete method (c23c0ad)

  • Deep copy dataset before passing it to evaluation (019b610)

  • Ensure model starts with publishers/ when users provide resource path from models/ (d689331)

  • Fix failed unit tests due to google-cloud-storage upgrade. (945b9e4)

  • Generalize RAG files import from Google Drive (88c6a6a)

  • Set upper bound of setuptools to unbreak public Colab for using vertex_ray namespace (6cc45bb)

1.53.0 (2024-05-30)

Features

  • Add a cloneable protocol for Reasoning Engine. (8960a80)

  • Add labels parameter to the supervised tuning train method (f7c5567)

  • Added reboot command for PersistentResource (7785f8c)

  • Added the new GenerationConfig.response_schema field (#3772) (5436d88)

  • Enable Tensorboard profile plugin in all regions by default. (8a4a41a)

  • GenAI - Added the response_schema parameter to the GenerationConfig class (b5e2c02)

  • LLM - Added the seed parameter to the TextGenerationModel’s predict methods (cb2f4aa)

Bug Fixes

  • Create run_name when run_name_prefix is not specified for Tensorboard uploader. (ac17d87)

  • GenAI - Tuning - Supervised - Fix adapter_size parameter handling to match enum values. (1cc22c3)

  • Model Monitor console uri. (71fbc81)

1.52.0 (2024-05-21)

Features

  • Add FeatureGroup delete (f9011e0)

  • Add support for ToolConfig in the LangChain template (9bda328)

  • Create Vertex Experiment when uploading Tensorboard logs (339f8b6)

  • GenAI - Add BatchPredictionJob for GenAI models (df4a4f2)

  • GenAI - Add cancel, delete, list methods in BatchPredictionJob (7ff8071)

  • GenAI - Added the BatchPredictionJob.submit method (4d091c6)

  • Private Endpoints - Added private service connect support to prediction endpoint. (6bdcfb3)

Bug Fixes

  • Add validation for evaluation dataset fields, update logging info for eval api request count (d6ef500)

  • Fix feature attribution drift visualization for model monitoring SDK (710f33d)

  • Fix the default value of response_column_name in EvalTask.evaluate() (98f9b35)

  • Update get_experiment_df to pass Experiment and allow empty metrics. (de5d0f3)

Documentation

  • Add Vertex Model Monitoring V2 SDK documentation (b47e6ff)

  • Update docstrings for rapid evaluation library. (d6d371d)

1.51.0 (2024-05-10)

Features

  • Add FeatureGroup create function (3938107)

  • Add FeatureGroup init/get (e47d436)

  • Add support for BaseModels in LangChain templates (5eb885e)

  • Added the vision_models.Image._mime_type property to make vision_models.Image compatible with generative_models.Image (6557d88)

  • AutoSxS Pairwise Metric in Rapid Evaluation SDK (b0c5eda)

  • GenAI - Grounding - Released Google Web Search retriever to GA (32b030a)

  • GenAI - Tuning - Supervised - Added support for the adapter_size parameter (88188d2)

  • LLM - Made the tuning location parameters truly optional (bae8429)

  • LLM - Support tuning of new text embedding models by migrating to the new v1.1.3 pipeline. (7fea754)

  • LLM - Text embedding - Added the output_dimensionality and learning_rate_multiplier parameters to text embedding tuning (Preview only) (cc8bc96)

  • LLM - Text Embedding - Added validation for text embedding tuning parameters. (5a300c1)

  • Release Ray on Vertex SDK to GA (f7c5132)

  • Support custom service account for Ray cluster creation and Ray Client connection (e0c6227)

  • Support vector_distance_threshold filtering and file-based retrieval for RAG (cd85d8f)

Bug Fixes

  • A bug in the evaluation library where the job crashes if only custom metrics are specified. (c528b6f)

  • Add DeprecationWarning to vertexai.preview predictive models SDK (3c3727b)

  • Add MAX_TOKENS to the list of successful finish reasons for Rapid Evaluation SDK (195c77e)

  • AttributeError for TorchModelSerializer.deserialize in torch >=2.3.0 (20b1866)

  • GenAI - Fixed handling of multiple tools in AutomaticFunctionCallingResponder (58e6ac9)

  • Remove InternalServerError and Unknown evaluation service error from retriable exceptions (12c147b)

  • Upload the reference model in model registry (510c833)

1.50.0 (2024-05-02)

Features

  • Add Candidate.grounding_metadata property (b22a8b8)

  • Add option to not include time_series_metrics in get_experiment_df call. This will improve execution time for Experiments with large number of runs. (78a95c5)

  • Add tune_model and deploy_tuned_model for TextEmbeddingModel. (42f5d6f)

  • Automatically populate parents for full resource name in Vertex RAG SDK (26657ff)

  • Deploy a tuned text embedding model – it doesn’t matter, if it’s tuned using Node.js, or curl. (8ca9cdf)

  • Make get_embeddings work both for foundational & tuned models. (b8b589c)

  • Python SDK for Vertex Model Monitoring V2. (021d59f)

  • Support public endpoint for Ray Client (57a5f78)

Bug Fixes

  • Add deprecation warnings when using Ray v2.4 (3a36784)

  • Append allowed_plugins in tb-gcp-uploader to default allowed plugins (aab9c3e)

  • LLM - Added missing parameters to the no-op _TunableTextEmbeddingModelMixin.get_tuned_model method (eb05ac4)

  • LVM - Fixed the typo in the VisionModel aspect ratio type annotation (2d19137)

  • Move torch import (e6d34df)

  • Ray - Fixed exception when using Ray 2.4 (2661f52)

1.49.0 (2024-04-27)

Features

  • Add additional parameters for GenerationConfig (0599ca1)

  • Add FeatureNormType to MatchingEngineIndexConfig. (c0e7acc)

  • Add mappings to pipeline templates for text-embedding models. (6279924)

  • Add support for TPU v5 lite pod(v5e) for custom training jobs. Custom training jobs now accept the v5e machine types as listed in https://cloud.google.com/tpu/docs/tpus-in-gke#v5e. (415912e)

  • Fix typo in Vertex Feature Store SDK. (b5404e7)

  • GenAI - Improved the exception messages when candidates, parts or text are not available (e82264d)

  • Support PreflightValidation in Preview PipelineJob submit function. (e88dc0d)

  • Tune_model method for class TextEmbeddingModel. (3eda55d)

  • Vertex RAG for enhanced generative AI (39b5149)

Bug Fixes

  • GCS Bucket subdirectory not being creating while using ReasoningEngine create method (3d22a18)

  • GenAI - Fixed the Part.mime_type property (819a44e)

  • Handle missing import for RunnableConfig when generating schema for LangChain templates (76c5d6d)

  • Parse intermediate steps from LangChain into JSON. (754c89d)

  • Register TensorFlow models from Ray checkpoints for more recent TensorFlow version, addressing the deprecation of SavedModel format in keras 3 (1341e2c)

  • Tensorboard uploader - move remaining tensorboard uploader code to aiplatform (f34094b)

1.48.0 (2024-04-17)

Features

  • Add support for reading requirements from a file. (80db7a0)

  • Adding tpu_topology to Vertex SDK (423c764)

  • Enable continuous upload for profile logs. (f05924d)

  • GenAI - Added the GenerationResponse.prompt_feedback property (efd5a72)

  • GenAI - Added the GenerationResponse.usage_metadata property (0654c35)

  • Support NOT_EQUAL for MatchingEngineIndexEndpoint numeric_restricts. (aa918e3)

  • Support referenced models in SDK. (c9b6b8b)

Bug Fixes

  • Add validation check for extra_packages when creating a reasoning engine. (255dabc)

  • Add validation for langchain tools. (a821d50)

  • Fixed the vertexai.init partial initialization issues (636a654)

  • GenAI - Workaround for streaming when content role is missing in service responses (fa35b91)

Documentation

  • Add Reasoning Engine reference documentation (496fc4b)

  • GenAI - Add Rapid Evaluation SDK reference documentation (40b728b)

1.47.0 (2024-04-06)

Features

  • Add display experiment button for tuning in Ipython environments (9bb687c)

  • Add Persistent Resource ID parameter to Custom Job form_local_script, run, and submit methods. (f5be0b5)

  • Add Persistent Resource Id parameter to Custom Training Job run and submit methods. (f428006)

  • Added GA support for PersistentResource management (98a07dd)

  • Added GA support for running Custom and Hp tuning jobs on Persistent Resources (35ecbac)

  • Added the Experiment.dashboard_url property (c8eec21)

  • GenAI - Added support for SafetySetting.method (probability or severity) (317ab8f)

  • GenAI - Added support for supervised fine-tuning (036d2d0)

  • GenAI - Added support for system instructions (4990eb6)

  • GenAI - Forced function calling feature (806ef9f)

  • Initial template for Langchain on Vertex. (0752a29)

  • LLM - Add RLHF-tuning support for text-bison@002 (1f27c3e)

  • Vertex AI Extension SDK Public Preview (137b5e1)

  • Vertex AI Reasoning Engine SDK Public Preview (6aaa5d0)

  • Vertex Rapid Evaluation SDK and Prompt Template for Vertex Prompt Management Public Preview (8c6ddf5)

Bug Fixes

  • GenAI - Fixed response validation error during streaming (c881998)

  • GenAI - Fixed the GenerativeModel’s handling of tuned models from different region (bf33fb3)

  • GenAI - Fixed the TuningJob dashboard URL (5367fbb)

  • Reinstate persistent resource preview class and tests (765d60d)

Documentation

  • Add run custom job on persistent resource sample. (53fc845)

  • Add run custom job on persistent resource sample. (31100c6)

1.46.0 (2024-03-30)

Features

  • GenAI - Automatic Function Calling feature (eef84c6)

  • Python SDK for Vertex Feature Store. (5015d25)

1.45.0 (2024-03-28)

Features

  • Add an arg to turn off Ray metrics collection during cluster creation (e33d11f)

  • Add Batch Delete and Batch Cancel Pipeline Jobs and unit tests. (cb495e7)

  • Add display model evaluation button for Ipython environments (181dc7a)

  • Add safety filter levels, watermark support and person generation support for Imagen 2 (0c498c5)

  • Add safety filter levels, watermark support and person generation support for Imagen 2 (e2efdbe)

  • Add support for output_dimensionality parameter through get_embeddings. (b1cab3f)

  • Add support for task types (TASK_QUESTION_ANSWERING, TASK_FACT_VERIFICATION) through get_embeddings. (9ec4590)

  • Add v1beta1 service clients for reasoning engine (13ec7e0)

  • GenAI - Support generating JSON Schema from Python function (be4922a)

  • Improve get_experiment_df execution speed (2e56acc)

  • Ray on Vertex enables XGBoost register model with custom version using pre-built container (e45ef96)

  • Support accelerator_type in RLHF tuning (e51c977)

Bug Fixes

  • Fix failing get_experiments_df test. (767712e)

  • GenAI - Capture content blocked case when validating responses (f0086df)

  • LLM - Fixed the InlineContext grounding source (6f59100)

  • Made Endpoint.raw_predict thread-safe by setting authorized_session last (c72c1ef)

1.44.0 (2024-03-14)

Features

  • Add custom tool context manager for telemetry (b30f5a6)

  • Add display experiment button for Ipython environments (0b33bdd)

  • Allow module_name to be specified with the base logger. (c01df66)

  • Enable Ray cluster creation with custom_image for each Resource (f90c9cc)

  • GenAI - Added function_calls shortcut property to Candidate class. (d6490ff)

  • LLM - Added support for the max_context_length tuning parameter (e5daae9)

  • Support shard_size for MatchingEngineIndex create index. (6dbf7d3)

  • Use colab enterprise enviroment variables to infer project_id and region (5baf5f8)

Bug Fixes

  • Batch read tensorboard time series data to allow reading more than 20 time series metrics. (6e3eac2)

  • GenAI - Fixed get_tuned_model deploying the model every time (e1c7870)

  • GenAI - Improved from_dict methods for content types (GenerationResponse, Candidate, Content, Part) (613ce69)

  • Improve import time by moving TensorFlow to lazy import (f294ba8)

  • LVM - Added support for GCS storage.googleapis.com URL import in vision_models.Image (2690e72)

  • Raise AttributeError for accessing non-existent fields of a proto. (b91edf5)

Documentation

  • Fixed the documentation generation (63ad1bf)

  • GenAI - Added the GA classes to the documentation (9eb5a52)

1.43.0 (2024-02-29)

Features

  • Add update_mask to MatchingEngineIndex upsert_datapoints() to support dynamic metadata update. (81f6a25)

  • Add Ray on Vertex BigQuery read/write support for Ray 2.9 (e048e3a)

  • Add read_bigquery and write_bigquery wrappers for Ray on Vertex (4739118)

  • Add transport override to enable the use of REST instead of GRPC (6ab4084)

  • Enable Ray cluster creation and registering TensorFlow checkpoint to Vertex with Ray version 2.9 (ff148cd)

  • LLM - Add support for batch prediction to CodeGenerationModel (code-bison) (fbf2f7c)

  • Support custom timeout for MatchingEngineIndex and MatchingEngineIndexEndpoint APIs. (09d1946)

  • Update ray extras installation in setup.py for Ray 2.9.3 (d947304)

Bug Fixes

  • Allow destination directory to exist before we copy. (0b55762)

  • LVM - Fixed VideoSegmentConfig in Multimodal Embeddings API call (665e78c)

Documentation

  • Update transport docstrings to include that REST is in preview (a442866)

1.42.1 (2024-02-15)

Bug Fixes

  • Handle case when no metadata is returned from model.predict (59e2bca)

  • Remove runtime dependency of setuptools (07c2a25)

Documentation

  • GenAI - Docs - Added quickstart link and minimum SDK version (cf6ab21)

  • GenAI - Updated the README for GA (cfd96d8)

1.42.0 (2024-02-15)

Features

  • Add Candidate.grounding_metadata (310ee49)

  • Add GoogleSearchRetrieval (310ee49)

  • Add metadata in model.predict output for LLMs (4661e58)

  • Add Retrieval (310ee49)

  • Add Tool.google_search_retrieval (310ee49)

  • Add Tool.retrieval (310ee49)

  • Add VertexAiSearch (310ee49)

  • GenAI - Added to_dict() methods to response and content classes (a78748e)

  • GenAI - Added support for Grounding (0c3e294)

  • GenAI - Added the GenerativeModel.start_chat(response_validation: bool = True) parameter (94f7cd9)

  • GenAI - GAPIC - Added support for Grounding (310ee49)

  • GenAI - Release the GenerativeModel to GA (c7e3f07)

  • LLM - Add newly supported model text-bison@002, update pipeline template version and prediction row limit for pipeline-based LLM Evaluation SDK (35e57b3)

  • LVM - Add GCS URI support for Imagen Models (imagetext, imagegeneration) (4109ea8)

  • LVM - Added support for Images from GCS uri for multimodal embeddings (90d95d7)

  • LVM - Added support for Videos from GCS uri for multimodal embeddings (f3bd3bf)

  • Support custom image for Ray cluster creation (d727189)

Bug Fixes

  • Add google-auth as a direct dependency (ecc6454)

  • Add restricts and crowding tag to MatchingEngineIndexEndpoint query response. (83cb52d)

  • Remove usage of distutils (e35ab64)

Documentation

  • Fix incorrect field in log_time_series_metrics_sample (f249353)

  • Update README (14b41b5)

1.41.0 (2024-02-05)

Features

  • Add numeric_filter to MatchingEngineIndexEndpoint match() and find_neighbor() private endpoint queries. (679646a)

  • Added the BatchPredictionJob.submit method - a non-blocking version of BatchPredictionJob.create (b62ddcd)

  • LLM - Add support for RLHF tuning (b2458ec)

  • Remove deletion of model upload staging bucket in system tests (b9b373b)

  • Switch Python generateContent to call Unary API endpoint (9a19545)

Bug Fixes

  • Fix crash when no target_col is specified for Bigframes tensorflow (512b82d)

Documentation

  • Add missing code block in docs/README.rst (16d6b58)

1.40.0 (2024-01-24)

Features

  • Add return_full_datapoint for MatchEngineIndexEndpoint match(). (ad8d9c1)

  • Add batch delete method in preview pipeline job class and unit test. (b0b604e)

  • Add option to not overwrite table in Ray on Vertex BQ Write (a99e992)

  • Add query by id for MatchingEngineIndexEndpoint find_neighbors() public endpoint query. (42c7e08)

  • Add support for low_level_batch_size in match() for MatchingEngineIndexEndpoint private query. (67e593b)

  • Added Public Preview support for PersistentResource management (262a36b)

  • Enable inline context in grounding to TextGenerationModel predict. (a75e81c)

  • Fix dataset export system test (1fbf049)

  • LVM - Added the MultiModalEmbeddingModel.get_embeddings(dimension=...) parameter (1d9bd23)

  • Support empty index for MatchingEngineIndex create index. (a00db07)

  • Support private service connect for MatchingEngineIndexEndpoint match() and read_index_datapoints(). (61cff4b)

Bug Fixes

  • Fix example usage of batch delete pipelinejobs. (3b28d64)

  • Fix experiments failure when backing tensorboard has been deleted. (e7a197e)

  • Fix message to Json parse issue in Dataset class (066f32d)

  • Fix typo in raise exception when tool is unsupported (2301d79)

  • Fixed the PipelineJob bucket creation after a breaking change in Google Cloud Storage client library (a8b01e0), closes #2936

  • Increase memory for Ray on Vertex default machine type (e6bcb17)

  • Use https to connect through Ray dashboard (7af3e67)

1.39.0 (2024-01-05)

Features

  • Add fraction_leaf_nodes_to_search_override. Add support for private endpoint in find_neighbors. (cd31c13)

  • Add notification_channels field to model monitoring alert config. (bb228ce)

  • Add support of newly added fields of ExportData API to SDK (ec3ea30)

  • Allow reuse of deleted experiment run id. (5f6ad8d)

  • GenAI - Added support for “models/

  • Support “reservedIpRanges” parameter in PipelineJob run() and submit() methods. (ab99e00)

  • Support custom target y column name for Bigframes Tensorflow (1634940)

  • Verify client and cluster Ray versions match in create_ray_cluster (17dc9b7)

Bug Fixes

  • Missing request parameter for v1beta explain. (443fa9d)

  • Pin google-cloud-aiplatform[tensorboard] dependency on tensorflow < 2.15.0 due to breaking change introduced in tensorboard 2.15.1 (4e891f7)

Documentation

  • GenAI - Added GenAI to docs (92fd7f0)

  • Update docstring for start_upload_tb_log() (c033c59)

  • Update tensorboard continuous uploader sample (1220746)

1.38.1 (2023-12-13)

Features

  • Adding serving_container_grpc_ports parameter to Model.upload() method (6a00ed7)

  • LLM - Added support for model distillation (28925e9)

  • Support CMEK for scheduled pipeline jobs. (406595d)

Miscellaneous Chores

1.38.0 (2023-12-11)

Features

  • Release GenerativeModel support for Gemini (cd233ef)

  • Add explicit constraints for update_ray_cluster (979a4f3)

  • Check if dataset exists before creation for Ray on Vertex BigQuery Write (544d6fe)

  • LLM - Added support for the logprobs, presence_penalty, frequency_penalty, and logit_bias generation parameters (1449344)

  • Support read_index_datapoints for private network. (c9f7119)

  • Support custom batch size for Bigframes Tensorflow (7dc8771)

  • Update the v1 service definition to add numeric_restricts. (d0c2ffa)

  • Verify client and cluster Ray versions match (10c6ad2)

Bug Fixes

  • read_index_endpoint private endpoint support. (3d8835e)

  • Fix exception message to use vertexai when project is not provided. (0cb1a7b)

  • Update test assumption for streaming endpoint of chat-bison@001 (f9a5b69)

  • When user is not logged in, throw more intuitive message (a8b24ad)

Documentation

  • Add default value to optional field parameter_values (0a4d772)

1.37.0 (2023-12-05)

Features

  • Add additional parameters to Model.upload(). (7b7d7d2)

  • Adding support for concurrent explanations (8e2ad75)

  • Allow configuring container logging settings on models when deploying. (beae48f)

  • Support user provided api endpoint. (92f2b4e)

  • Add grpc_ports to UploadModel ModelContainerSpec, add DirectPredict, DirectRawPredict, StreamingPredict, StreamingRawPredict to PredictionService. (6dfbad7)

Bug Fixes

  • Clarify wording when Ray on Vertex cluster is missing head node ip (4a71c8c)

  • Fix error when allowed_plugins is set to None. (6f2860a)

  • Fixed INTEGER and BOOL parameters casting issue. Fix conversion bug where INTEGER and BOOL parameters are not cast to the correct type. (9a204c4)

  • Make PipelineJobSchedule propagate labels to created PipelineJobs (a34533f)

Documentation

  • Add upload Tensorboard profile log to Experiment sample. (5780513)

  • Update the documentation for the image_dataset class (8562368)

1.36.4 (2023-11-16)

Features

  • Add numeric_restricts to MatchingEngineIndex find_neighbors() for querying (6c1f2cc)

  • Add remove_datapoints() to MatchingEngineIndex. (b86a404)

  • Add upsert_datapoints() to MatchingEngineIndex to support streaming update index. (7ca484d)

  • LLM - include error code into blocked response from TextGenerationModel, ChatModel, CodeChatMode, and CodeGenerationModel. (1f81cf2)

  • Populate Ray Cluster dashboard_address from proto field (dd4b852)

  • add CountTokens API, ComputeTokens API, and ModelContainerSpec features (ba2fb39)

Bug Fixes

  • Add check for empty encryption_spec_key_name for MatchingEngineIndexEndpoint create. (7740132)

  • Fix server error due to no encryption_spec_key_name in MatchingEngineIndex create_tree_ah_index and create_brute_force_index (595b580)

Miscellaneous Chores

1.36.3 (2023-11-14)

Features

  • Add option to not use default tensorboard (a25c669)

  • Add preview HyperparameterTuningJob which can be run on persistent resource (0da8c53)

  • Add Featurestore Bigtable Serving, Feature Registry v1, November bulk GAPIC release (9f46f7c)

Documentation

  • Fix documentation for obsolete link to GCS formatting (95184de)

Miscellaneous Chores

1.36.2 (2023-11-10)

Features

  • Add encryption_spec_key_name to MatchingEngineIndex create_tree_ah_index and (1a9e36f)

  • Add encryption_spec_key_name, enable_private_service_connect,project_allowlist to MatchingEngineIndexEndpoint create. (750e17b)

  • Add index_update_method to MatchingEngineIndex create() (dcb6205)

  • Expose max_retry_cnt parameter for Ray on Vertex BigQuery write (568907c)

  • LLM - Grounding - Added support for the disable_attribution grounding parameter (91e985a)

  • LLM - Support model evaluation when tuning chat models (ChatModel, CodeChatModel) (755c3f9)

  • LVM - Added multi-language support for ImageGenerationModel (791eff5)

Bug Fixes

  • Async call bug in CodeChatModel.send_message_async method (fcf05cb)

Documentation

  • Add Bigframes remote training example to vertexai README (8b993b3)

  • Update the documentation for the tabular_dataset class (6f40f1b)

Miscellaneous Chores

1.36.1 (2023-11-07)

Features

  • Add per_crowding_attribute_neighbor_count, approx_num_neighbors, fraction_leaf_nodes_to_search_override, and return_full_datapoint to MatchingEngineIndexEndpoint find_neighbors (33c551e)

  • Add profiler support to tensorboard uploader sdk (be1df7f)

  • Add support for per_crowding_attribute_num_neighbors approx_num_neighborsto MatchingEngineIndexEndpoint match() (e5c20c3)

  • Add support for per_crowding_attribute_num_neighbors approx_num_neighborsto MatchingEngineIndexEndpoint match() (53d31b5)

  • Add support for per_crowding_attribute_num_neighbors approx_num_neighborsto MatchingEngineIndexEndpoint match() (4e357d5)

  • Enable grounding to ChatModel send_message and send_message_async methods (d4667f2)

  • Enable grounding to TextGenerationModel predict and predict_async methods (b0b4e6b)

  • LLM - Added support for the enable_checkpoint_selection tuning evaluation parameter (eaf4420)

  • LLM - Added tuning support for the \*-bison-32k models (9eba18f)

  • LLM - Released CodeChatModel tuning to GA (621af52)

Bug Fixes

  • Correct class name in system test (b822b57)

Documentation

  • Clean up RoV create_ray_cluster docstring (1473e19)

Miscellaneous Chores

1.36.0 (2023-10-31)

Features

  • Add preview count_tokens method to CodeGenerationModel (96e7f7d)

  • Allow the users to use extra serialization arguments for objects. (ffbd872)

  • Also support unhashable objects to be serialized with extra args (77a741e)

  • LLM - Added count_tokens support to ChatModel (preview) (01989b1)

  • LLM - Added new regions for tuning and tuned model inference (3d43497)

  • LLM - Added support for async streaming (760a025)

  • LLM - Added support for multiple response candidates in code chat models (598d57d)

  • LLM - Added support for multiple response candidates in code generation models (0c371a4)

  • LLM - Enable tuning eval TensorBoard without evaluation data (eaf5d81)

  • LLM - Released CodeGenerationModel tuning to GA (87dfe40)

  • LLM - Support accelerator_type in tuning (98ab2f9)

  • Support experiment autologging when using persistent cluster as executor (c19b6c3)

  • Upgrade BigQuery Datasource to use write() interface (7944348)

Bug Fixes

  • Adding setuptools to dependencies for Python 3.12 and above. (afd540d)

  • Fix Bigframes tensorflow serializer dependencies (b4cdb05)

  • LLM - Fixed the async streaming (41bfcb6)

  • LLM - Make tuning use the global staging bucket if specified (d9ced10)

  • LVM - Fixed negative prompt in ImageGenerationModel (cbe3a0d)

  • Made the Endpoint prediction client initialization lazy (eb6071f)

  • Make sure PipelineRuntimeConfigBuilder is created with the right arguments (ad19838)

  • Make sure the models list is populated before indexing (f1659e8)

  • Raise exception for RoV BQ Write for too many rate limit exceeded (7e09529)

  • Rollback BigQuery Datasource to use do_write() interface (dc1b82a)

1.35.0 (2023-10-10)

Features

  • Add serializer.register_custom_command() (639cf10)

  • Install Bigframes sklearn dependencies automatically (7aaffe5)

  • Install Bigframes tensorflow dependencies automatically (e58689b)

  • Install Bigframes torch dependencies automatically (1d65347)

  • LLM - Added support for multiple chat response candidates (587df74)

  • LLM - Added support for multiple text generation response candidates (c3ae475)

Bug Fixes

  • Duplicate logs in Colab (9b75259)

  • LLM - Fixed tuning and evaluation when explicit credentials are specified (188dffe)

Documentation

  • Add probabilistic inference to TiDE and L2L model code samples. (efe88f9)

1.34.0 (2023-10-02)

Features

  • Add Model Garden support to vertexai.preview.from_pretrained (f978200)

  • Enable vertexai preview persistent cluster executor (0ae969d)

  • LLM - Added the count_tokens method to the preview TextGenerationModel and TextEmbeddingModel classes (6a2f2aa)

  • LLM - Improved representation for blocked responses (222f222)

  • LLM - Released ChatModel tuning to GA (7d667f9)

Bug Fixes

  • Create PipelineJobSchedule in same project and location as associated PipelineJob by default (c22220e)

Documentation

  • Add documentation for the preview namespace (69a67f2)

1.33.1 (2023-09-20)

Bug Fixes

  • Lightning trainer fails to be unwrapped in remote training (8271301)

1.33.0 (2023-09-18)

Features

  • Add Custom Job support to from_pretrained (8b0add1)

  • Added async prediction and explanation support to the Endpoint class (e9eb159)

  • LLM - Added support for async prediction methods (c9c9f10)

  • LLM - CodeChat - Added support for context (f7feeca)

  • Release Ray on Vertex SDK Preview (3be36e6)

Bug Fixes

  • Handle Ray image parsing error (41a3a83)

  • Vizier - Fixed field existence checks for child params in to_proto(). (d516931)

1.32.0 (2023-09-05)

Features

  • LLM - Added stop_sequences parameter to streaming methods and CodeChatModel (d62bb1b)

  • LLM - Improved the handling of temperature and top_p in streaming (6566529)

  • Support bigframes sharded parquet ingestion at remote deserialization (Tensorflow) (a8f85ec)

  • Release Vertex SDK Preview (c60b9ca)

  • Allow setting default service account (d11b8e6)

Bug Fixes

  • Fix feature update since no LRO is created (468e6e7)

  • LLM - CodeGenerationModel now supports safety attributes (c2c8a5e)

  • LLM - Fixed batch prediction on tuned models (2a08535)

  • LLM - Fixed the handling of the TextEmbeddingInput.task_type parameter. (2e3090b)

  • Make statistics Optional for TextEmbedding. (7eaa1d4)

1.31.1 (2023-08-24)

  • fix: LLM - De-hardcoded the max_output_tokens default value for the CodeGenerationModel (f5a20eb)

1.31.0 (2023-08-21)

Features

  • Add disable_retries option to custom jobs. (db518b0)

  • LLM - Added support for stop_sequences in inference (6f7ea84)

  • LLM - Exposed the TextGenerationResponse.raw_prediction_response (f8f2b9c)

  • LLM - Made tuning asynchronous when tuning becomes GA (226ab8b)

  • LLM - release model evaluation for TextGenerationModel to public preview (8df5185)

  • LLM - Released TextGenerationModel tuning to GA (62ff30d)

  • LLM - Support streaming prediction for chat models (ce60cf7)

  • LLM - Support streaming prediction for code chat models (0359f1d)

  • LLM - Support streaming prediction for code generation models (3a8348b)

  • LLM - Support streaming prediction for text generation models (fb527f3)

  • LLM - TextEmbeddingModel - Added support for structural inputs (TextEmbeddingInput), auto_truncate parameter and result statistics (cbf9b6e)

  • LVM - Added support for Image Generation models (b3729c1)

  • LVM - Released ImageCaptioningModel to GA (7575046)

  • LVM - Released ImageQnAModel to GA (fd5cb02)

  • LVM - Released MultiModalEmbeddingModel to GA (e99f366)

  • LVM - Removed the width and height parameters from ImageGenerationModel.generate_images since the service has dropped support for image sizes and aspect ratios (52897e6)

  • Scheduled pipelines client GA. (62b8b23)

Documentation

  • Generate documentation for tune_model and related class (705e1ea)

  • LVM - Added autogenerated documentation for visual models (18e8bb2)

1.30.1 (2023-08-11)

Features

  • LLM - Added tuning support for chat-bison models (3a97c52)

  • LLM - Added tuning support for codechat-bison models (af6e455)

Bug Fixes

  • LLM - Fixed the TextGenerationModel.predict parameters (f3b25ab)

Miscellaneous Chores

1.30.0 (2023-08-10)

Features

  • Add model.evaluate() method to Model class (51df86e)

  • Add support for providing only text to MultiModalEmbeddingModel.get_embeddings() (38ec40a)

Bug Fixes

  • LLM - Fixed filter in list_tuned_model_names (57806fb)

1.29.0 (2023-08-02)

Features

  • Add preview CustomJob which can be run on persistent resource (56906b0)

  • LLM - Support for Batch Prediction for the textembedding models (preview) (a368538)

  • LLM - Support tuning for the code-bison model (preview) (e4b23a2)

  • LVM - Large Vision Models SDK (preview release). Support for image captioning and image QnA (imagetext model) and multi modal embedding (multimodelembedding model) (preview) (9bbf1ea)

Bug Fixes

  • LLM - Fixed get_tuned_model for the future models that are not text-bison (1adf72b)

Documentation

  • Fix auto-generated pydoc for language_models (7d72bd1)

  • LLM - Made it possible to provide message history to CodeChatModel when starting chat. (cf46145)

1.28.1 (2023-07-18)

Features

  • LLM - Released the BatchPrediction to GA for TextGenerationModel (701c3a2)

  • LLM - Support tuning in the “us-central1” location (4aa7745)

Bug Fixes

  • Fix artifact registry link not showing in ui when creating schedules with SDK. (203cb47)

  • Fixed the installation error caused by a PyYAML issue (4b86ce1)

  • Require model name in ModelEvaluation.list() (aed8c76)

Documentation

  • Fixed a docstring for train_steps (1f55b05)

Miscellaneous Chores

1.28.0 (2023-07-08)

Features

  • LLM - Released the Chat models to GA (22aa26d)

1.27.1 (2023-07-06)

Features

  • Add sdk support for xai example-based explanations (f9ca1d5)

Miscellaneous Chores

1.27.0 (2023-06-30)

Features

  • Add submit for CustomTrainingJob and CustomContainerTrainingJob which won’t block until complete. (d6476d0)

  • LLM - Added support for learning_rate in tuning (c6cdd10)

  • LLM - Released the Codey models to GA (89609c9)

Bug Fixes

  • Fix aiplatform.init bug by replacing experiment_name with experiment (c60773a)

  • Fix error when calling update_state() after ExperimentRun.list() (cb255ec)

  • LLM - Exported the ChatMessage class (7bf7634)

  • LLM - Fixed the chat models failing due to safetyAttributes format (459ba86)

  • Vizier - Fixed pyvizier client study creation errors (16299d1)

Documentation

  • Fixed a docstring for _Dataset (b68a941)

  • Fixed a docstring for TimeSeriesDataset (a7dfce2)

  • Populate GA LLM SDK Pydocs (e248285)

  • Update scheduled pipelines client max_run_count docstring with allowed values. (750e161)

1.26.1 (2023-06-21)

Features

  • Add additional scheduled pipelines client getters and unit tests. (9371b4f)

  • Add PipelineJobSchedule update method and unit tests. (69c5f60)

  • Add tunable parameters for Model Garden model training to the “AutoMLImageTrainingJob” in SDK. (50646be)

  • LLM - Added batch prediction (2235305)

  • LLM - Exposed the chat history (bf0e20b)

  • LLM - Exposed the safety attributes (01ba3ca)

Bug Fixes

  • Change scheduled pipelines client dashboard uri to view created schedules. Note: uri will not work until scheduler UI is GA. (d4d8613)

  • Fix bug where scheduled pipeline jobs were not running. (4e7d11a)

  • Remove Schedule read mask because ListSchedules does not support it. (1fda417)

Miscellaneous Chores

Documentation

  • Update scheduled pipelines client wait() docstring. (a7d92e5)

1.26.0 (2023-06-07)

Features

  • Add additional scheduled pipelines client create method unit tests. (0463678)

  • Add pipelineJob create_schedule() method and unit test. (635ae9c)

  • Add scheduled pipelines client create/get methods and unit tests. (4755fc7)

  • Add scheduled pipelines client list/pause/resume methods and unit tests. (ce5dee4)

  • Adding enable_access_logging parameter to Endpoint.deploy() method, minor edit to batch_predict() docstring (794cedd)

  • LLM - Added support for CMEK in tuning (aebf74a)

  • LLM - Released the LLM SDK to GA (76465e2)

  • Support publisher models in BatchPredictionJob.create (13b11c6)

Bug Fixes

  • CustomJob.from_local_script does not pass args to script for custom container images (6ead69d)

  • Fix bug when checking PipelineJob failure status (a154859)

  • Fix the bug that start_upload_tb_log() doesn’t continuously upload (66e6eae)

  • LLM - Fixed parameters set in ChatModel.start_chat being ignored (a0d815d)

  • LLM - Fixed the DataFrame staging on Windows (056b0bd)

  • Resource created by _construct_sdk_resource_from_gapic should use the project from the resource name instead of the default project. (162b2f2)

  • Retry for etag errors on context update. (d3d5f9a)

  • Unbreak additional timeout for MatchingEngineIndexEndpoint deploy_index (af199c0)

Documentation

  • Correct text embedding model ID docstring (8824629)

  • LLM - Fixed the rendering of the example usage code blocks. (eaaee28)

1.25.0 (2023-05-09)

Features

  • Add support for Large Language Models (866c6aa)

  • Add default TensorBoard support. (fa7d3a0)

  • Add support for find_neighbors/read_index_datapoints in matching engine public endpoint (e3a87f0)

  • Added the new root vertexai package (fbd03b1)

Bug Fixes

  • EntityType RPC update returns the updated EntityType - not an LRO. (8f9c714)

  • Fix default AutoML Forecasting transformations list. (77b89c0)

  • Fix type hints for Prediction.predictions. (56518f1)

  • Removed parameter Resume, due to causing confusion and errors. (c82e0b5)

1.24.1 (2023-04-21)

Features

  • Add preview capability to deploy models with shared resources. (29d4e45)

  • Add support for create public index endpoint in matching engine (7e6022b)

  • Add support for return public endpoint dns name in matching engine (1b5ae44)

  • Add the new model types to “AutoMLImageTrainingJob” in SDK. (4d032d5)

  • Adds the Time series Dense Encoder (TiDE) forecasting job. (d8e6744)

  • Remove google internal annotation when export to github. (fd5ff99)

Bug Fixes

  • Support timestamp in Vertex SDK write_feature_values() (4b0722c)

Documentation

  • Add Time series Dense Encoder (TiDE) model code sample. (8e91a58)

  • Fix docstring formatting for exceptions (d75322c)

Miscellaneous Chores

1.24.0 (2023-04-12)

Features

  • Add ExperimentRun.get_logged_custom_jobs method (c116b07)

  • Add get method for Experiment and ExperimentRun (41cd943)

  • Add incremental training to AutoMLImageTrainingJob. (bb92380)

  • Add preview capability to manage DeploymentResourcePools. (5df5da0)

  • Add start_time support for BatchReadFeatureValues wrapper methods. (91d8459)

  • Add TensorBoard log uploader (3fad7bb)

  • Enable deployment of models that do not support deployment (25f3f21)

  • Enable experiment tracking in CustomJob (94a63b8)

  • Update the v1 service definition to add the embedding_id field in MatchRequest. (5a1146e)

Bug Fixes

  • Adding previously created PrivateEndpoint network parameter in Model deploy helper method (3e1b206)

Documentation

  • Adds note to delete endpoint sample (#2060) (9922eb2)

  • Fix create tensorboard sample (2c45123)

  • samples: Add sample for experiment run state update. (111a747)

  • Update docstring for 3 model uploading methods (a71e4a3)

  • Update Vertex Forecasting weight column description. (e0ee183)

1.23.0 (2023-03-15)

Features

  • Implement Model.copy functionality. (94dd82f)

  • Update the v1 service definition to add the fraction_leaf_nodes_to_search_override field which replaces leaf_nodes_to_search_percent_override. (badd386)

Documentation

  • Added missing comma in README (8cb4377)

1.22.1 (2023-02-28)

Features

  • Add support for enable_dashboard_access field for Training jobs in SDK (3500eab)

  • Add the recently added new model type “cloud_1” to the “AutoMLImageTrainingJob” in SDK. (581939b)

Documentation

  • Add temporal fusion transformer (TFT) model code sample. (8ddc062)

  • samples: Add samples for autologging (f8052b8)

Miscellaneous Chores

1.22.0 (2023-02-16)

Features

  • Add a return value (ClassificationMetrics) for the log_classification_metrics() (8ebcdbd)

  • Add metric and parameter autologging to experiments (96e9e12)

  • Add update_version to Model Registry (8621e24)

  • Support a list of GCS URIs in CustomPythonPackageTrainingJob (05bb71f)

  • Support Model Serialization in Vertex Experiments(tensorflow) (f38ddc2)

Bug Fixes

  • Added missing instances_format parameter to batch_prediction_job_samples (82a2afc)

  • Address broken unit tests in certain environments (d06b22d)

  • List method for MLMD schema classes (2401a1d)

  • Unbreak additional timeout for _deploy_call() (076308f)

  • Unbreak additional timeout for MatchingEngine update_embeddings (5d0bc1e)

  • Unbreak timeouts for Dataset create. (328ebac)

  • Use Client.list_blobs instead of Bucket.list_blobs in CPR artifact downloader, to make sure that CPR works with custom service accounts on Vertex Prediction. (bb27619)

Documentation

  • Add a hint to auth Docker to the LocalModel push_image docstring. (e97a6fb)

  • Fix Create and Import Tabular BQ dataset sample (4415c10)

  • Fix LocalModel push_image docstring. (5fdb7fc)

  • Fixed a typo in docstring. (4ee6232)

  • New samples for model serialization (83457ca)

  • Samples for model serialization (7997094)

1.21.0 (2023-01-13)

Features

  • Add default skew threshold to be an optional input at _SkewDetectionConfig and also mark the target_field and data_source of skew config to optional. (7da4164)

  • Add filter to Model Registry list_versions API. (c1cb33f)

  • Add MLMD schema class ExperimentModel (94b2f29)

  • Add Service Account support to BatchPredictionJob (deba06b)

  • Add support for Predict Request Response Logging in Endpoint SDK (372ab8d)

  • Adding Feature Store: Streaming ingestion to GA (6bc4c84)

  • Enable passing experiment_tensorboard to init without experiment (369a0cc)

  • Support Model Serialization in Vertex Experiments(sklearn) (d4deed3)

  • Support Model Serialization in Vertex Experiments(xgboost) (fe75eba)

Bug Fixes

  • Endpoint.undeploy_all() doesn’t undeploy all models (9fb24d7)

  • Fix bug in associating tensorboard to an experiment (6def0b8)

  • Pin shapely version to <2.0.0 (1efd816)

  • Unbreak timeouts for Dataset create, FeatureStore ingest, and MatchingEngine Index create. (3096d1c)

  • Updated proto message formatting logic for batch predict model monitoring (f87fef0)

1.20.0 (2022-12-15)

Features

  • Adds the temporal fusion transformer (TFT) forecasting job (99313e0)

  • Reraise exceptions from API calls (d72bc83)

Documentation

  • samples: Feature Store: Streaming ingestion code sample and test (bc9e2cf)

1.19.1 (2022-12-08)

Features

  • Add explanationSpec to TrainingPipeline-based custom jobs (957703f)

Bug Fixes

  • Add pre-built container(tf2-gpu-2-1) to the container URI list (cdd557e)

  • Fix bug that broke profiler with ‘0-rc2’ tensorflow versions. (8779df5)

  • Fixed argument name in UnmanagedContainerModel (d876b3a)

Documentation

  • Add a sample for create_tensorboard. (52656ca)

  • Fix Experiment resource name format docstring. (f8e5842)

  • Fix get Experiment data frame sample (24e1465)

  • Update docstrings for “data_item_labels” in dataset (b2f8c42)

  • Update README fix product doc link (43a2679)

Miscellaneous Chores

1.19.0 (2022-11-17)

Features

  • Add Feature Store: Streaming Ingestion (write_feature_values()) and introduce Preview namespace to Vertex SDK (bae0315)

  • Add bq_dataset_id parameter to batch_serve_to_df (bb72562)

  • Add annotation_labels to ImportDataConfig in aiplatform v1 dataset.proto (43e2805)

  • Add support for ordery_by in Metadata SDK list methods for Artifact, Execution and Context. (2377606)

  • Support global network parameter. (c7f57ad)

Bug Fixes

  • Correct data file gcs path for import_data_text_sentiment_analysis_sample test (86df4b5)

  • Print error for schema classes (13e2165)

Documentation

  • Update README with new link for AI Platform API (35b83d9)

1.18.3 (2022-11-01)

Documentation

  • Add a sample for get_experiment_run_artifacts (7266352)

1.18.3 (2022-10-31)

Documentation

  • Add a sample for get_experiment_run_artifacts (7266352)

1.18.2 (2022-10-20)

Bug Fixes

  • Added proto message conversion to MDMJob.update fields (#1718) (9e77c61)

  • Log_classification_metrics (#1742) (3588526)

  • PipelineJob should only pass bearer tokens for AR URIs (b43851c)

Documentation

  • Fix create experiment sample (#1716) (cba7fbf)

  • Resurface googleapis.dev and prediction docs (#1724) (24f0c6f)

  • samples: Improve docstring of Vertex AI Python SDK Model Registry samples (#1705) (f97e90f)

1.18.1 (2022-10-10)

Bug Fixes

1.18.0 (2022-10-03)

Features

  • Add deleteFeatureValues in aiplatform v1beta1 featurestore_service.proto (#1670) (9a506ee)

  • Add model_source_info to Model in aiplatform v1beta1 model.proto (#1691) (876fb2a)

  • Add support for HTTPS URI pipeline templates (#1683) (926d0b6)

  • Add support for V1 and V2 classification models for the V1Beta2 API (#1680) (1cda4b4)

  • Support complex metrics in Vertex Experiments (#1698) (ed0492e)

Bug Fixes

  • deps: Require protobuf >= 3.20.2 (#1699) (c5c77ad)

  • Fix endpoint parsing in ModelDeploymentMonitoringJob.update (#1671) (186872d)

  • Project/location parsing for nested resources (#1700) (9e1d796)

  • Show inherited SDK methods in pydoc (#1707) (2b7583b)

Documentation

1.17.1 (2022-09-15)

Features

  • Add enable_simple_view to PipelineJob.list() (#1614) (627fdf9)

  • Add eval metrics types to get_experiment_df (#1648) (944b03f)

  • Adding Python 3.10 support + updating google-vizier version (#1644) (f4766dc)

Miscellaneous Chores

1.17.0 (2022-09-07)

Features

  • Add input artifact when creating a pipeline (#1593) (2cf9fe6)

  • Add model_monitoring_stats_anomalies,model_monitoring_status to BatchPredictionJob in aiplatform v1beta1 batch_prediction_job.proto (#1621) (0a1f4e9)

  • Add read_mask to ListPipelineJobsRequest in aiplatform v1 pipeline_service (#1589) (9e19a40)

  • Add samples for get execution input and output artifacts (#1585) (eb5a4b6)

  • Add support for SDK Method metrics tracking via _USER_AGENT_SDK… (#1591) (28e56ef)

  • Support filters in matching engine vector matching (#1608) (d591d3e)

  • Support model monitoring for batch prediction in Vertex SDK (#1570) (bbec998)

  • Support raw_predict for Endpoint (#1620) (cc7c968)

  • Support ResourceName with Version. (#1609) (737dc2b)

  • Update the samples of hyperparameter tuning in the public doc (#1600) (653b759)

Bug Fixes

  • deps: Allow protobuf < 5.0.0 (#1587) (3d3e0aa)

  • deps: require proto-plus >= 1.22.0 (3d3e0aa)

  • Log_metrics docstring error (#1588) (0385c4c)

  • Study.list() method (#1594) (47eb0ae)

  • Update Model.list_model_evaluations and get_model_evaluation to use the provided version (#1616) (8fb836b)

Documentation

1.16.1 (2022-08-02)

Features

  • Add google.ClassificationMetrics, google.RegressionMetrics, and google.Forecasting Metrics (#1549) (3526b3e)

  • added support for conditional parameters in hyperparameter tuning (#1544) (744cc38)

  • SDK support for model monitoring (#1249) (18c88d1)

  • support case insensitive match on search facets (#1523) (cb4d405)

  • Vertex Vizier support in SDK. (#1434) (b63b3ba)

Bug Fixes

Miscellaneous Chores

1.16.0 (2022-07-27)

Features

  • Add metadata SDK sample for delete method. (#1530) (46aa9b5)

  • Add metadata SDK samples for list artifact and list execution (#1514) (c0d01f1)

  • Add Metadata SDK support and samples for get method (#1516) (d442248)

  • Add samples for Metadata context list, get, and create (#1525) (d913e1d)

  • Change the Metadata SDK _Context class to an external class (#1519) (95b107c)

  • Refactor schema classes to subclass from _Resource (#1536) (93002e8)

  • Support custom containers in CustomJob.from_local_script (#1483) (be0b7e1)

  • Vertex AI Prediction Custom Prediction Routine (34bbd0a)

Bug Fixes

  • Fixed getting the output GCS bucket in PipelineJob.submit (#1542) (69d6c7d)

  • Pass the PipelineJob credentials to create_gcs_bucket_for_pipeline_artifacts_if_it_does_not_exist (#1537) (b53e2b5)

1.15.1 (2022-07-18)

Features

  • add get_associated_experiment method to pipeline_jobs (#1476) (e9f2c3c)

  • Add sample for create artifact and execution using the Metadata SDK. (#1462) (1fc7dd9)

  • Add support for start_execution in MLMD SDK. (#1465) (298958f)

  • Add support for Vertex Tables Q2 regions (#1498) (1b16f90)

  • Added the PipelineJob.from_pipeline_func method (#1415) (6ef05de)

Bug Fixes

  • deps: require google-api-core>=1.32.0,>=2.8.0 (#1512) (6d09dee)

  • Unbreak aiplatform.Experiment.create (#1509) (558c141)

Miscellaneous Chores

1.15.0 (2022-06-29)

Features

  • add default_skew_threshold to TrainingPredictionSkewDetectionConfig in aiplatform v1beta1, v1 model_monitoring.proto (#1411) (7a8e3be)

  • add model_monitoring_config to BatchPredictionJob in aiplatform v1beta1 batch_prediction_job.proto (#1450) (d35df58)

  • add model_version_id to BatchPredictionJob in aiplatform v1 batch_prediction_job.proto (#1453) (9ef057a)

  • add model_version_id to UploadModelResponse in aiplatform v1 model_service.proto (#1442) (1c198f1)

  • Add PrivateEndpoint class and HTTP methods (#1033) (425a32f)

  • add support for accepting an Artifact Registry URL in pipeline_job (#1405) (e138cfd)

  • add support for failure_policy in PipelineJob (#1452) (d0968ea)

  • Improved metadata artifact and execution creation using python / SDK (#1430) (6c4374f)

  • support dataset update (#1416) (e3eb82f)

  • Support for Model Versioning (#1438) (d890685)

  • Vertex AI Experiments GA (#1410) (24d1bb6)

Bug Fixes

  • Fixed docstrings for wildcards and matching engine type (#1220) (d778dee)

  • Removed dirs_exist_ok parameter as it’s not backwards compatible (#1170) (50d4129)

1.14.0 (2022-06-08)

Features

  • add a way to easily clone a PipelineJob (#1239) (efaf6ed)

  • add display_name and metadata to ModelEvaluation in aiplatform model_evaluation.proto (b6bf6dc)

  • add Examples to Explanation related messages in aiplatform v1beta1 explanation.proto (b6bf6dc)

  • Add hierarchy and window configs to Vertex Forecasting training job (#1255) (8560fa8)

  • add holiday regions for vertex forecasting (#1253) (0036ab0)

  • add IAM policy to aiplatform_v1beta1.yaml (b6bf6dc)

  • add latent_space_source to ExplanationMetadata in aiplatform v1 explanation_metadata.proto (b6bf6dc)

  • add latent_space_source to ExplanationMetadata in aiplatform v1beta1 explanation_metadata.proto (b6bf6dc)

  • add preset configuration for example-based explanations in aiplatform v1beta1 explanation.proto (b6bf6dc)

  • add scaling to OnlineServingConfig in aiplatform v1 featurestore.proto (b6bf6dc)

  • add seq2seq forecasting training job (#1196) (643d335)

  • add successful_forecast_point_count to CompletionStats in completion_stats.proto (b6bf6dc)

  • add template_metadata to PipelineJob in aiplatform v1 pipeline_job.proto (b6bf6dc)

  • Add Vertex Forecasting E2E test. (#1248) (e82c179)

  • Added forecasting snippets and fixed bugs with existing snippets (#1210) (4e4bff5)

Bug Fixes

  • change endpoint update method to return resource (#1409) (44e279b)

  • Changed system test to use list_models() correctly (#1397) (a3da19a)

  • Pinned protobuf to prevent issues with pb files. (#1398) (7a54637)

Documentation

1.13.1 (2022-05-26)

Features

Bug Fixes

Documentation

  • update aiplatform SDK arrangement for Sphinx (#1163) (e9510ea)

Miscellaneous Chores

1.13.0 (2022-05-09)

Features

  • add ConvexAutomatedStoppingSpec to StudySpec in aiplatform v1 study.proto (847ad78)

  • add ConvexAutomatedStoppingSpec to StudySpec in aiplatform v1beta1 study.proto (847ad78)

  • add JOB_STATE_UPDATING to JobState in aiplatform v1 job_state.proto (847ad78)

  • add JOB_STATE_UPDATING to JobState in aiplatform v1beta1 job_state.proto (847ad78)

  • add LatestMonitoringPipelineMetadata to ModelDeploymentMonitoringJob in aiplatform v1beta1 model_deployment_monitoring_job.proto (847ad78)

  • add ListModelVersion, DeleteModelVersion, and MergeVersionAliases rpcs to aiplatform v1beta1 model_service.proto (847ad78)

  • add MfsMount in aiplatform v1 machine_resources.proto (847ad78)

  • add MfsMount in aiplatform v1beta1 machine_resources.proto (847ad78)

  • add model_id and parent_model to TrainingPipeline in aiplatform v1beta1 training_pipeline.proto (847ad78)

  • add model_version_id to DeployedModel in aiplatform v1beta1 endpoint.proto (847ad78)

  • add model_version_id to PredictResponse in aiplatform v1beta1 prediction_service.proto (847ad78)

  • add model_version_id to UploadModelRequest and UploadModelResponse in aiplatform v1beta1 model_service.proto (847ad78)

  • add nfs_mounts to WorkPoolSpec in aiplatform v1 custom_job.proto (847ad78)

  • add nfs_mounts to WorkPoolSpec in aiplatform v1beta1 custom_job.proto (847ad78)

  • add Pandas DataFrame support to TabularDataset (#1185) (4fe4558)

  • add PredictRequestResponseLoggingConfig to aiplatform v1beta1 endpoint.proto (847ad78)

  • add reserved_ip_ranges to CustomJobSpec in aiplatform v1 custom_job.proto (#1165) (847ad78)

  • add reserved_ip_ranges to CustomJobSpec in aiplatform v1beta1 custom_job.proto (847ad78)

  • add template_metadata to PipelineJob in aiplatform v1beta1 pipeline_job.proto (#1186) (99aca4a)

  • add version_id to Model in aiplatform v1beta1 model.proto (847ad78)

  • allow creating featurestore without online node (#1180) (3224ae3)

  • Allow users to specify timestamp split for vertex forecasting (#1187) (ee49e00)

  • Make matching engine API public (#1192) (469db6b)

  • rename Similarity to Examples, and similarity to examples in ExplanationParameters in aiplatform v1beta1 explanation.proto (847ad78)

Documentation

  • fix type in docstring for map fields (847ad78)

1.12.1 (2022-04-20)

Features

Bug Fixes

  • change default for create_request_timeout arg to None (#1175) (47791f7)

Documentation

Miscellaneous Chores

1.12.0 (2022-04-07)

Features

  • add categorical_threshold_config to FeaturestoreMonitoringConfig in aiplatform v1 featurestore_monitoring.proto (38f3711)

  • add categorical_threshold_config to FeaturestoreMonitoringConfig in aiplatform v1beta1 featurestore_monitoring.proto (38f3711)

  • add disable_monitoring to Feature in aiplatform v1 feature.proto (38f3711)

  • add disable_monitoring to Feature in aiplatform v1beta1 feature.proto (38f3711)

  • Add done method for pipeline, training, and batch prediction jobs (#1062) (f3338fc)

  • add import_features_analysis to FeaturestoreMonitoringConfig in aiplatform v1 featurestore_monitoring.proto (38f3711)

  • add import_features_analysis to FeaturestoreMonitoringConfig in aiplatform v1beta1 featurestore_monitoring.proto (38f3711)

  • add ImportModelEvaluation in aiplatform v1 model_service.proto (#1105) (ef5930c)

  • add monitoring_config to EntityType in aiplatform v1 entity_type.proto (#1077) (38f3711)

  • add monitoring_stats_anomalies to Feature in aiplatform v1 feature.proto (38f3711)

  • add monitoring_stats_anomalies to Feature in aiplatform v1beta1 feature.proto (38f3711)

  • add numerical_threshold_config to FeaturestoreMonitoringConfig in aiplatform v1 featurestore_monitoring.proto (38f3711)

  • add numerical_threshold_config to FeaturestoreMonitoringConfig in aiplatform v1beta1 featurestore_monitoring.proto (38f3711)

  • add objective to MonitoringStatsSpec in aiplatform v1 featurestore_service.proto (38f3711)

  • add objective to MonitoringStatsSpec in aiplatform v1beta1 featurestore_service.proto (38f3711)

  • add PredictRequestResponseLoggingConfig to Endpoint in aiplatform v1 endpoint.proto (#1072) (be0ccc4)

  • add staleness_days to SnapshotAnalysis in aiplatform v1 featurestore_monitoring.proto (38f3711)

  • add staleness_days to SnapshotAnalysis in aiplatform v1beta1 featurestore_monitoring.proto (38f3711)

  • Add support for Vertex Tables Q1 regions (#1065) (6383d4f)

  • add timeout arg across SDK (#1099) (184f7f3)

  • Add timeout arguments to Endpoint.predict and Endpoint.explain (#1094) (cc59e60)

  • Made display_name parameter optional for most calls (#882) (400b760)

  • sdk: enable loading both JSON and YAML pipelines IR (#1089) (f2e70b1)

  • v1beta1: add service_account to BatchPredictionJob in batch_prediction_job.proto (#1084) (b7a5177)

Bug Fixes

  • add resource manager utils to get project ID from project number (#1068) (f10a1d4)

  • add self.wait() in operations after optional_sync supported resource creation (#1083) (79aeec1)

  • Don’t throw exception when getting representation of unrun GCA objects (#1071) (c9ba060)

  • Fix import error string showing wrong pip install path (#1076) (74ffa19)

  • Fixed getting project ID when running on Vertex AI; Fixes #852 (#943) (876cb33)

  • Give aiplatform logging its own log namespace, let the user configure their own root logger (#1081) (fb78243)

  • Honoring the model’s supported_deployment_resources_types (#865) (db34b85)

  • missing reference to logged_web_access_uris (#1056) (198a1b5)

  • system tests failure from test_upload_and_deploy_xgboost_model (#1149) (c8422a9)

Documentation

  • fix CustomContainerTrainingJob example in docstring (#1101) (d2fb9db)

  • improve bigquery_destination_prefix docstring (#1098) (a46df64)

  • Include time dependency in documentation for weight, time, and target columns. (#1102) (52273c2)

  • samples: read, import, batch_serve, batch_create features (#1046) (80dd40d)

  • Update AutoML Video docstring (#987) (6002d5d)

1.11.0 (2022-03-03)

Features

  • add additional_experiement flag in the tables and forecasting training job (#979) (5fe59a4)

  • add TPU_V2 & TPU_V3 values to AcceleratorType in aiplatform v1/v1beta1 accelerator_type.proto (#1010) (09c2e8a)

  • Added scheduling to CustomTrainingJob, CustomPythonPackageTrainingJob, CustomContainerTrainingJob (#970) (89078e0)

Bug Fixes

  • deps: allow google-cloud-storage < 3.0.0dev (#1008) (1c34154)

  • deps: require google-api-core>=1.31.5, >=2.3.2 (#1050) (dfbd68a)

  • deps: require proto-plus>=1.15.0 (dfbd68a)

  • enforce bq SchemaField field_type and mode using feature value_type (#1019) (095bea2)

  • Fix create_lit_model_from_endpoint not accepting models that don’t return a dictionary. (#1020) (b9a057d)

  • loosen assertions for system test featurestore (#1040) (2ba404f)

  • remove empty scripts kwarg in setup.py (#1014) (ef3fcc8)

  • show logs when TFX pipelines are submitted (#976) (c10923b)

  • update system test_model_upload to use BUILD_SPECIFIC_GCP_PROJECT (#1043) (e7d2719)

Documentation

  • samples: add samples to create/delete featurestore (#980) (5ee6354)

  • samples: added create feature and create entity type samples and tests (#984) (d221e6b)

1.10.0 (2022-02-07)

Features

  • _TrainingScriptPythonPackager to support folders (#812) (3aec6a7)

  • add dedicated_resources to DeployedIndex in aiplatform v1beta1 index_endpoint.proto feat: add Scaling to OnlineServingConfig in aiplatform v1beta1 featurestore.proto chore: sort imports (#991) (7a7f0d4)

  • add dedicated_resources to DeployedIndex message in aiplatform v1 index_endpoint.proto chore: sort imports (#990) (a814923)

  • Add XAI SDK integration to TensorFlow models with LIT integration (#917) (ea2b5cf)

  • Added aiplatform.Model.update method (#952) (44e208a)

  • Enable europe-west6 and northamerica-northeast2 regions (0f6b670)

  • enable feature store batch serve to BigQuery and GCS for csv and tfrecord (#919) (c840728)

  • enable feature store batch serve to Pandas DataFrame; fix: read instances uri for batch serve (#983) (e0fec36)

  • enable feature store online serving (#918) (b8f5f82)

  • enable ingest from pd.DataFrame (#977) (9289f2d)

  • Open LIT with a deployed model (#963) (ea16849)

Bug Fixes

Documentation

  • samples: replace deprecated fields in create_training_pipeline_tabular_forecasting_sample.py (#981) (9ebc972)

1.9.0 (2021-12-29)

Features

  • add create in Featurestore, EntityType, Feature; add create_entity_type in Featurestore; add create_feature, batch_create_features in EntityType; add ingest_from_* for bq and gcs in EntityType; add and update delete with force delete nested resources (#872) (ba11c3d)

  • Add LIT methods for Pandas DataFrame and TensorFlow saved model. (#874) (03cf301)

  • Add support to create TensorboardExperiment (#909) (96ce738)

  • Add support to create TensorboardRun (#912) (8df74a2)

Bug Fixes

  • Fix timestamp proto util to default to timestamp at call time. (#933) (d72a254)

  • Improve handling of undeploying model without redistributing remaining traffic (#898) (8a8a4fa)

  • issues/192254729 (#914) (3ec620c)

  • issues/192254729 (#915) (0f22ff6)

  • use open_in_new_tab in the render method. (#926) (04618e0)

1.8.1 (2021-12-14)

Bug Fixes

  • add clarity to param model_name (#888) (1d81783)

  • add clarity to parameters per user feedback (#886) (37ee0a1)

  • add param for multi-label per user’s feedback (#887) (fda942f)

  • add support for API base path overriding (#908) (45c4086)

  • Important the correct constants and use v1 for tensorboard experiments (#905) (48c2bf1)

  • incorrect uri for IOD yaml (#889) (e108ef8)

  • Minor docstring and snippet fixes (#873) (578e06d)

Documentation

  • Update references to containers and notebook samples. (#890) (67fa1f1)

  • Updated docstrings with exception error classes (#894) (f9aecd2)

1.8.0 (2021-12-03)

Features

  • Add cloud profiler to training_utils (6d5c7c4)

  • add enable_private_service_connect field to Endpoint feat: add id field to DeployedModel feat: add service_attachment field to PrivateEndpoints feat: add endpoint_id to CreateEndpointRequest and method signature to CreateEndpoint feat: add method… (#878) (ca813be)

  • add enable_private_service_connect field to Endpoint feat: add id field to DeployedModel feat: add service_attachment field to PrivateEndpoints feat: add endpoint_id to CreateEndpointRequest and method signature to CreateEndpoint feat: add method… (#879) (47e93b2)

  • add featurestore module including Featurestore, EntityType, and Feature classes; add get, update, delete, list methods in all featurestore classes; add search method in Feature class (#850) (66745a6)

  • Add prediction container URI builder method (#805) (91dd3c0)

  • default to custom job display name if experiment name looks like a custom job ID (#833) (8b9376e)

  • Support uploading local models (#779) (bffbd9d)

  • Tensorboard v1 protos release (#847) (e0fc3d9)

  • updating Tensorboard related code to use v1 (#851) (b613b26)

  • Upgrade Tensorboard from v1beta1 to v1 (#849) (c40ec85)

Bug Fixes

  • Import error for cloud_profiler (#869) (0f124e9)

  • Support multiple instances in custom predict sample (#857) (8cb4839)

Documentation

  • Added comment for evaluation_id to python examples (#860) (004bf5f)

  • Reverted IDs in model_service snippets test (#871) (da747b5)

  • Update name of BQ source parameter in samples (#859) (f11b598)

1.7.1 (2021-11-16)

Features

Bug Fixes

  • add parameters_value in PipelineJob for schema > 2.0.0 (#817) (900a449)

  • exclude support for python 3.10 (#831) (0301a1d)

Miscellaneous Chores

1.7.0 (2021-11-06)

Features

  • Adds support for google.protobuf.Value pipeline parameters in the parameter_values field (#807) (c97199d)

  • Adds support for google.protobuf.Value pipeline parameters in the parameter_values field (#808) (726b620)

  • PipelineJob switch to v1 API from v1beta1 API (#750) (8db7e0c)

Bug Fixes

  • Correct PipelineJob credentials description (#816) (49aaa87)

  • Fixed docstrings for Dataset in AutoMLForecastingTrainingJob (760887b)

Documentation

  • Fix pydocs README to be consistent with repo README (#821) (95dbd60)

  • Update sample with feedback from b/191251050 (#818) (6b2d938)

1.6.2 (2021-11-01)

Features

  • Add PipelineJob.submit to create PipelineJob without monitoring it’s completion. (#798) (7ab05d5)

  • support new protobuf value param types for Pipeline Job client (#797) (2fc05ca)

Bug Fixes

  • Add retries when polling during monitoring runs (#786) (45401c0)

  • use version.py for versioning (#804) (514031f)

  • Widen system test timeout, handle tearing down failed training pipelines (#791) (78879e2)

Miscellaneous Chores

1.6.1 (2021-10-25)

Features

  • Add debugging terminal support for CustomJob, HyperparameterTun… (#699) (2deb505)

  • add support for python 3.10 (#769) (8344804)

  • Add training_utils folder and environment_variables for training (141c008)

  • enable reduction server (#741) (8ef0ded)

  • enabling AutoML Forecasting training response to include BigQuery location of exported evaluated examples (#657) (c1c2326)

  • PipelineJob: allow PipelineSpec as param (#774) (f90a1bd)

  • pre batch creating TensorboardRuns and TensorboardTimeSeries in one_shot mode to speed up uploading (#772) (c9f68c6)

Bug Fixes

  • cast resource labels to dict type (#783) (255edc9)

  • Remove sync parameter from create_endpoint_sample (#695) (0477f5a)

Miscellaneous Chores

1.6.0 (2021-10-12)

Features

  • add featurestore service to aiplatform v1 (#765) (68c88e4)

  • Add one shot profile uploads to tensorboard uploader. (#704) (a83f253)

  • Added column_specs, training_encryption_spec_key_name, model_encryption_spec_key_name to AutoMLForecastingTrainingJob.init and various split methods to AutoMLForecastingTrainingJob.run (#647) (7cb6976)

  • Lazy load Endpoint class (#655) (c795c6f)

1.5.0 (2021-09-30)

Features

  • Add data plane code snippets for feature store service (#713) (e3ea683)

  • add flaky test diagnostic script (#734) (09e48de)

  • add vizier service to aiplatform v1 BUILD.bazel (#731) (1a580ae)

  • code snippets for feature store control plane (#709) (8e06ced)

  • Updating the Tensorboard uploader to use the new batch write API so it runs more efficiently (#710) (9d1b01a)

Bug Fixes

1.4.3 (2021-09-17)

Features

  • PipelineJob: support dict, list, bool typed input parameters fr… (#693) (243b75c)

Bug Fixes

  • Update milli node_hours for image training (#663) (64768c3)

  • XAI Metadata compatibility with Model.upload (#705) (f0570cb)

Miscellaneous Chores

1.4.2 (2021-09-10)

Features

  • add explanation metadata get_metadata_protobuf for reuse (#672) (efb6d18)

1.4.1 (2021-09-07)

Features

  • add prediction service RPC RawPredict to aiplatform_v1beta1 feat: add tensorboard service RPCs to aiplatform_v1beta1: BatchCreateTensorboardRuns, BatchCreateTensorboardTimeSeries, WriteTensorboardExperimentData feat: add model_deployment_monitori… (#670) (b73cd94)

  • add Vizier service to aiplatform v1 (#671) (179150a)

  • add XAI, model monitoring, and index services to aiplatform v1 (#668) (1fbce55)

  • Update tensorboard uploader to use Dispatcher for handling different event types (#651) (d20b520), closes #519

Documentation

1.4.0 (2021-08-30)

Features

  • add filter and timestamp splits (#627) (1a13577)

  • add labels to all resource creation apis (#601) (4e7666a)

  • add PipelineJob.list (a58ea82)

  • add support for export_evaluated_data_items_config in AutoMLTab… (#583) (2a6b0a3)

  • add util functions to get URLs for Tensorboard web app. (#635) (8d88c00)

  • Add wait_for_resource_creation to BatchPredictionJob and unblock async creation when model is pending creation. (#660) (db580ad)

  • Added the VertexAiResourceNoun.to_dict() method (#588) (b478075)

  • expose base_output_dir for custom job (#586) (2f138d1)

  • expose boot disk type and size for CustomTrainingJob, CustomPythonPackageTrainingJob, and CustomContainerTrainingJob (#602) (355ea24)

  • split GAPIC samples by service (#599) (5f15b4f)

Bug Fixes

  • Fixed bug in TabularDataset.column_names (#590) (0fbcd59)

  • pipeline none values (#649) (2f89343)

  • Populate service_account and network in PipelineJob instead of pipeline_spec (#658) (8fde2ce)

  • re-remove extra TB dependencies introduced due to merge conflict (#593) (433b94a)

  • Update BatchPredictionJob.iter_outputs() and BQ docstrings (#631) (28f32fd)

1.3.0 (2021-07-30)

Features

  • add get method for PipelineJob (#561) (fe5e6e4)

  • add Samples section to CONTRIBUTING.rst (#558) (d35c466)

  • add tensorboard resource management (#539) (6f8d3d1)

  • add tf1 metadata builder (#526) (918998c)

  • add wait for creation and more informative exception when properties are not available (#566) (e346117)

  • Adds a new API method FindMostStableBuild (6a99b12)

  • Adds attribution_score_drift_threshold field (6a99b12)

  • Adds attribution_score_skew_thresholds field (6a99b12)

  • Adds BigQuery output table field to batch prediction job output config (6a99b12)

  • Adds CustomJob.enable_web_access field (6a99b12)

  • Adds CustomJob.web_access_uris field (6a99b12)

  • Adds Endpoint.network, Endpoint.private_endpoints fields and PrivateEndpoints message (6a99b12)

  • Adds Execution.State constants: CACHED and CANCELLED (6a99b12)

  • Adds Feature Store features (6a99b12)

  • Adds fields to Study message (6a99b12)

  • Adds IndexEndpoint.private_ip_ranges field (6a99b12)

  • Adds IndexEndpointService.deployed_index_id field (6a99b12)

  • Adds MetadataService.DeleteArtifact and DeleteExecution methods (6a99b12)

  • Adds ModelMonitoringObjectConfig.explanation_config field (6a99b12)

  • Adds ModelMonitoringObjectConfig.ExplanationConfig message field (6a99b12)

  • column specs for tabular transformation (#466) (71d0bd4)

  • enable_caching in PipelineJob to compile time settings (#557) (c9da662)

  • Removes breaking change from v1 version of AI Platform protos (6a99b12)

Bug Fixes

  • change default replica count to 1 for custom training job classes (#579) (c24251f)

  • create pipeline job with user-specified job id (#567) (df68ec3)

  • deps: pin ‘google-{api,cloud}-core’, ‘google-auth’ to allow 2.x versions (#556) (5d79795)

  • enable self signed jwt for grpc (#570) (6a99b12)

Documentation

1.2.0 (2021-07-14)

Features

  • Add additional_experiments field to AutoMlTablesInputs (#540) (96ee726)

  • add always_use_jwt_access (#498) (6df4866)

  • add explain get_metadata function for tf2. (#507) (f6f9a97)

  • Add structure for XAI explain (from XAI SDK) (#502) (cb9ef11)

  • Add two new ModelType constants for Video Action Recognition training jobs (96ee726)

  • Adds AcceleratorType.NVIDIA_TESLA_A100 constant (f3a3d03)

  • Adds additional_experiments field to AutoMlForecastingInputs (8077b3d)

  • Adds additional_experiments field to AutoMlTablesInputs (#544) (8077b3d)

  • Adds AutoscalingMetricSpec message (f3a3d03)

  • Adds BigQuery output table field to batch prediction job output config (f3a3d03)

  • Adds fields to Study message (f3a3d03)

  • Adds JobState.JOB_STATE_EXPIRED constant (f3a3d03)

  • Adds PipelineService methods for Create, Get, List, Delete, Cancel (f3a3d03)

  • Adds two new ModelType constants for Video Action Recognition training jobs (8077b3d)

  • Removes AcceleratorType.TPU_V2 and TPU_V3 constants (#543) (f3a3d03)

Bug Fixes

  • Handle nested fields from BigQuery source when getting default column_names (#522) (3fc1d44)

  • log pipeline completion and raise pipeline failures (#523) (2508fe9)

  • making the uploader depend on tensorflow-proper (#499) (b95e040)

  • Set prediction client when listing Endpoints (#512) (95639ee)

1.1.1 (2021-06-22)

Features

  • add cancel method to pipeline client (#488) (3b19fff)

Bug Fixes

  • check if training_task_metadata is populated before logging backingCustomJob (#494) (2e627f8)

Documentation

Miscellaneous Chores

1.1.0 (2021-06-17)

Features

  • add aiplatform API Vizier service (fdc968f)

  • add featurestore, index, metadata, monitoring, pipeline, and tensorboard services to aiplatform v1beta1 (fdc968f)

  • add invalid_row_count to ImportFeatureValuesResponse and ImportFeatureValuesOperationMetadata (fdc968f)

  • add pipeline client init and run to vertex AI (1f1226f)

  • add tensorboard support for CustomTrainingJob, CustomContainerTrainingJob, CustomPythonPackageTrainingJob (#462) (8cfd611)

  • adds enhanced protos for time series forecasting (fdc968f)

  • adds enhanced protos for time series forecasting (#374) (fdc968f)

  • allow the prediction endpoint to be overridden (#461) (c2cf612)

  • AutoMlImageSegmentationInputs.ModelType adds MOBILE_TF_LOW_LATENCY constant (fdc968f)

  • AutoMlVideoClassificationInputs.ModelType adds MOBILE_JETSON_VERSATILE_1 constant (fdc968f)

  • Expose additional attributes into Vertex SDK to close gap with GAPIC (#477) (572a27c)

  • ImageSegmentationPredictionResult.category_mask field changed to string data type (fdc968f)

  • remove unsupported accelerator types (fdc968f)

  • removes forecasting (time_series_forecasting proto) from public v1beta1 protos (fdc968f)

  • removes unused protos from schema/ folders: schema/io_format.proto, schema/saved_query_metadata.proto (fdc968f)

  • support additional_experiments for AutoML Tables and AutoML Forecasting (#428) (b4211f2)

  • support self-signed JWT flow for service accounts (fdc968f)

Bug Fixes

  • add async client to %name_%version/init.py (fdc968f)

  • add target_column docstring (#473) (c0543cd)

  • configuring timeouts for aiplatform v1 methods (fdc968f)

  • Enable MetadataStore to use credentials when aiplatfrom.init passed experiment and credentials. (#460) (e7bf0d8)

  • exclude docs and tests from package (#481) (b209904)

  • pass credentials to BQ and GCS clients (#469) (481d172)

  • remove display_name from FeatureStore (fdc968f)

  • Remove URI attribute from Endpoint sample (#478) (e3cbdd8)

Documentation

  • changes product name to Vertex AI (fdc968f)

  • correct link to fieldmask (fdc968f)

  • removes tinyurl links (fdc968f)

1.0.1 (2021-05-21)

Bug Fixes

  • use resource name location when passed full resource name (#421) (f40f322)

1.0.0 (2021-05-19)

Features

  • add custom and hp tuning (#388) (aab9e58)

  • add tensorboard support to custom job and hyperparameter tuning job (#404) (fa9bc39)

Bug Fixes

  • tb-gcp-uploader to show flags in “–help” correctly (#409) (9f603dd)

Miscellaneous Chores

0.9.0 (2021-05-17)

Features

  • Add AutoML vision, Custom training job, and generic prediction samples (#300) (cc1a708)

  • Add VPC Peering support to CustomTrainingJob classes (#378) (56273f7)

  • AutoML Forecasting, Metadata Experiment Tracking, Tensorboard uploader (e94c9db)

Bug Fixes

0.8.0 (2021-05-11)

Features

  • Add export model (#353) (12c5be4)

  • add mbsdk video dataset samples (#307) (24d6920)

  • Add service account support to Custom Training and Model deployment (#342) (b4b1b12)

  • add services to aiplatform_v1beta1 (#367) (beb4032)

  • Added create_training_pipeline_custom_job_sample and create_training_pipeline_custom_training_managed_dataset_sample and fixed create_training_pipeline_image_classification_sample (#343) (1c6b998)

  • Added create_training_pipeline_custom_package_job_sample and create_training_pipeline_custom_container_job_sample and reworked create_training_pipeline_custom_job_sample (#351) (7abf8ef)

  • Added default AutoMLTabularTrainingJob column transformations (#357) (4fce8c4)

  • Added deploy_model_with_dedicated_resources_sample, deploy_model_with_automatic_resources_sample, upload_model and get_model samples (#337) (ef4f6f8)

  • Added explain tabular samples (#348) (c95d1ce)

  • aiplatform: Add support for setting User agent header (#364) (d50d26d)

  • expose env var in cust training class run func args (#366) (7ae28b8)

  • MBSDK Tabular samples (#338) (4241738)

  • update featurestore (#377) (bc17163)

Bug Fixes

  • Add all supported uCAIP GA regions (#350) (5e14c59)

  • aiplatform: Fix doc formatting (#359) (857f63d)

  • Bump google-cloud-storage min version to 1.32.0 (#371) (6fda925)

  • default model_display_name to _CustomTrainingJob.display_name when model_serving_container_image_uri is provided (#324) (a5fa7a2)

  • env formatiing (#379) (6bc4c61)

  • remove Optional type hint on deploy (#345) (79b0ab1)

0.7.1 (2021-04-14)

Bug Fixes

  • fix list failing without order_by and local sorting (#320) (06e99db)

0.7.0 (2021-04-14)

Features

  • Add Custom Container Prediction support, move to single API endpoint (#277) (ca7f6d6)

  • Add initial Model Builder SDK samples (#265) (1230dc6)

  • Add list() method to all resource nouns (#294) (3ec9386)

  • add support for multiple client versions, change aiplatform from compat.V1BETA1 to compat.V1 (#290) (89e3212)

  • Make aiplatform.Dataset private (#296) (1f0d5f3)

  • parse project location when passed full resource name to get apis (#297) (674227d)

Bug Fixes

  • add quotes to logged snippet (0ecd0a8)

  • make logging more informative during training (#310) (9a4d991)

  • remove TPU from accelerator test cases (57f4fcf)

0.6.0 (2021-03-22)

Features

Bug Fixes

  • skip create data labeling job sample tests (#254) (116a29b)

0.5.1 (2021-03-01)

Bug Fixes

  • fix create data labeling job samples tests (#244) (3c440de)

  • fix predict sample tests for proto-plus==1.14.2 (#250) (b1c9d88)

  • fix update export model sample, and add sample test (#239) (20b8859)

Documentation

0.5.0 (2021-02-17)

Features

Bug Fixes

0.4.0 (2021-01-08)

Features

  • add create_batch_prediction_job samples (#67) (96a850f)

  • add create_hyperparameter_tuning_job_python_package sample (#76) (5155dee)

  • add create_training_pipeline_custom_training_managed_dataset sample (#75) (b012283)

  • add custom_job samples (#69) (fb165b3)

  • add data_labeling samples (#78) (7daacd5)

  • add get_custom_job and get_hyperparameter_tuning_job samples (#68) (26da7a7)

  • add schema namespace (#140) (1cbd4a5)

  • add video action recognition samples (#77) (4c60ad6)

  • Added tabular forecasting sample (#156) (a23857b)

  • Added tabular forecasting samples (#128) (69fc7fd)

  • adds function/method enhancements, demo samples (#122) (1a302d2)

  • adds text batch prediction samples (#82) (ad09c29)

  • initial generation of enhanced types (#102) (5ddbf16)

  • update create_training_pipeline samples (#142) (624a08d)

  • xai samples (#83) (5cf3859)

Bug Fixes

Documentation

0.3.1 (2020-11-13)

Features

0.3.0 (2020-11-05)

Features

Bug Fixes