autogluon.cloud 0.6.0__tar.gz → 0.7.0__tar.gz

{autogluon_cloud-0.6.0/src/autogluon.cloud.egg-info → autogluon_cloud-0.7.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: autogluon.cloud
-Version: 0.6.0
+Version: 0.7.0
 Summary: Train and deploy ML models in the cloud
 Author: AutoGluon Community
 License-Expression: Apache-2.0
@@ -36,19 +36,15 @@ License-File: NOTICE
 Requires-Dist: autogluon.common<1.7,>=0.7
 Requires-Dist: boto3<2,>=1.37.23
 Requires-Dist: packaging<27,>=23.0
-Requires-Dist: sagemaker<3,>=2.240.0
 Requires-Dist: pyarrow<25,>=19.0.1
 Requires-Dist: PyYAML~=6.0
 Requires-Dist: Pillow<13,>=10.2
 Requires-Dist: huggingface_hub<2,>=0.20
-Requires-Dist: typing_extensions<5,>=4.0
+Requires-Dist: typing_extensions<5,>=4.5
 Requires-Dist: click<9,>=8.0
 Requires-Dist: rich<15,>=13.0
-Provides-Extra: ray
-Requires-Dist: ray[default]<2.59,>=2.10.0; extra == "ray"
 Provides-Extra: all
 Requires-Dist: autogluon<1.7,>=0.7; extra == "all"
-Requires-Dist: autogluon.cloud[ray]; extra == "all"
 Provides-Extra: tests
 Requires-Dist: pytest; extra == "tests"
 Requires-Dist: moto; extra == "tests"
@@ -63,7 +59,7 @@ Dynamic: license-file
 ## Train and Deploy AutoGluon in the Cloud
 
 [![PyPI](https://img.shields.io/pypi/v/autogluon.cloud.svg)](https://pypi.org/project/autogluon.cloud/)
-[![Python Versions](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-blue)](https://pypi.org/project/autogluon.cloud/)
+[![Python Versions](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12%20%7C%203.13-blue)](https://pypi.org/project/autogluon.cloud/)
 [![GitHub license](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](./LICENSE)
 [![Continuous Integration](https://github.com/autogluon/autogluon-cloud/actions/workflows/continuous_integration.yml/badge.svg)](https://github.com/autogluon/autogluon-cloud/actions/workflows/continuous_integration.yml)
 
@@ -71,30 +67,48 @@ Dynamic: license-file
 
 </div>
 
-AutoGluon-Cloud lets you train and deploy state-of-the-art ML models in the cloud in a few lines of code. Run [AutoGluon](https://auto.gluon.ai/stable/index.html) on [Amazon SageMaker](https://aws.amazon.com/sagemaker/) without worrying about infrastructure, dependencies, or a heavy local ML environment. It supports two workflows:
+AutoGluon-Cloud lets you train and deploy state-of-the-art ML models for classification, regression, and time series forecasting on [Amazon SageMaker](https://aws.amazon.com/sagemaker/). All it takes is a few lines of code; AutoGluon-Cloud handles the infrastructure, dependencies, and glue code for you.
 
-- **[Train your own predictor](https://auto.gluon.ai/cloud/stable/tutorials/predictor-tabular.html)** — the same `fit → deploy → predict` workflow as local AutoGluon, with all the heavy lifting offloaded to SageMaker.
-- **[Run pretrained foundation models](https://auto.gluon.ai/cloud/stable/tutorials/foundation-model-timeseries.html)** — deploy state-of-the-art pretrained models like [Chronos-2](https://huggingface.co/amazon/chronos-2) for zero-shot inference, with no training required.
+## 💡 Why AutoGluon-Cloud?
 
-## 💾 Installation & setup
+- **Works like local [AutoGluon](https://auto.gluon.ai/stable/index.html).** Pass in DataFrames, get predictions back — as convenient as working locally, with the compute handled by AWS.
+- **No boilerplate.** No training scripts, inference handlers, or serialization code to write and maintain.
+- **Official AWS containers.** Everything runs in the [AutoGluon Deep Learning Containers](https://aws.github.io/deep-learning-containers/), maintained and security-patched by AWS.
+- **Sensible defaults, fully configurable.** Under the hood it's just SageMaker running in your AWS account, so you stay in full control.
+
+## 💾 Installation
 
 ```bash
 pip install autogluon.cloud
+autogluon-cloud bootstrap  # one-time setup for IAM role and S3 bucket
 ```
 
-Then provision the IAM role and S3 bucket AutoGluon-Cloud needs to run on AWS:
+See the [Setup tutorial](https://auto.gluon.ai/cloud/stable/tutorials/setup.html) for more details.
+
+## 🚀 Foundation models
+
+Zero-shot forecasts with a pretrained model like [Chronos-2](https://huggingface.co/amazon/chronos-2) — no training required. [Full walkthrough](https://auto.gluon.ai/cloud/stable/tutorials/foundation-model-timeseries.html).
 
 ```python
-from autogluon.cloud import bootstrap
+from autogluon.cloud import TimeSeriesFoundationModel
 
-bootstrap()
-```
+# `data` can be a local path, S3 URL, or pandas DataFrame
+data = "https://autogluon.s3.amazonaws.com/datasets/timeseries/m4_hourly_tiny/train.csv"
+
+model = TimeSeriesFoundationModel("chronos-2")
+
+# Batch prediction
+predictions = model.predict(data=data, target="target", prediction_length=24)
 
-See the [Setup tutorial](https://auto.gluon.ai/cloud/stable/tutorials/setup.html) for the full walkthrough, including how to register an existing role and bucket instead.
+# Real-time inference endpoint
+endpoint = model.deploy()
+predictions = endpoint.predict(data=data, target="target", prediction_length=24)
+endpoint.delete_endpoint()
+```
 
-## ⚙️ Train your own model
+## ⚙️ Train your own predictor
 
-Train an AutoGluon predictor on your data and serve it from a SageMaker endpoint — same API as local AutoGluon, all heavy lifting on AWS. Full walkthrough: [tabular](https://auto.gluon.ai/cloud/stable/tutorials/predictor-tabular.html), [time series](https://auto.gluon.ai/cloud/stable/tutorials/predictor-timeseries.html).
+Train an AutoGluon predictor on your data and serve it from SageMaker. Full walkthrough: [time series](https://auto.gluon.ai/cloud/stable/tutorials/predictor-timeseries.html), [tabular](https://auto.gluon.ai/cloud/stable/tutorials/predictor-tabular.html).
 
 ```python
 from autogluon.cloud import TabularCloudPredictor
@@ -111,40 +125,11 @@ cloud_predictor.fit(
     predictor_fit_args={"time_limit": 120},  # passed to TabularPredictor.fit()
 )
 
-# Real-time inference endpoint
-cloud_predictor.deploy()
-result = cloud_predictor.predict_real_time(test_data)
-cloud_predictor.cleanup_deployment()
-
 # Batch prediction
 result = cloud_predictor.predict(test_data)
-```
-
-## 🚀 Run a pretrained foundation model
-
-Skip training entirely — deploy a pretrained model like Chronos-2 to SageMaker and get zero-shot predictions out of the box. Full walkthrough: [time series](https://auto.gluon.ai/cloud/stable/tutorials/foundation-model-timeseries.html).
-
-```python
-from autogluon.cloud import TimeSeriesFoundationModel
-
-# `data` can be a local path, S3 URL, or pandas DataFrame
-data = "https://autogluon.s3.amazonaws.com/datasets/timeseries/m4_hourly_tiny/train.csv"
-
-model = TimeSeriesFoundationModel("chronos-2")
-
-# Batch prediction — no training required
-predictions = model.predict(
-    data=data,
-    target="target",
-    prediction_length=24,
-)
 
 # Real-time inference endpoint
-endpoint = model.deploy()
-predictions = endpoint.predict(
-    data=data,
-    target="target",
-    prediction_length=24,
-)
+endpoint = cloud_predictor.deploy()
+result = endpoint.predict(test_data)
 endpoint.delete_endpoint()
 ```

autogluon_cloud-0.7.0/README.md

@@ -0,0 +1,82 @@
+
+
+<div align="center">
+<img src="https://user-images.githubusercontent.com/16392542/77208906-224aa500-6aba-11ea-96bd-e81806074030.png" width="350">
+
+## Train and Deploy AutoGluon in the Cloud
+
+[![PyPI](https://img.shields.io/pypi/v/autogluon.cloud.svg)](https://pypi.org/project/autogluon.cloud/)
+[![Python Versions](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12%20%7C%203.13-blue)](https://pypi.org/project/autogluon.cloud/)
+[![GitHub license](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](./LICENSE)
+[![Continuous Integration](https://github.com/autogluon/autogluon-cloud/actions/workflows/continuous_integration.yml/badge.svg)](https://github.com/autogluon/autogluon-cloud/actions/workflows/continuous_integration.yml)
+
+[AutoGluon-Cloud Documentation](https://auto.gluon.ai/cloud/stable/index.html) | [AutoGluon Documentation](https://auto.gluon.ai)
+
+</div>
+
+AutoGluon-Cloud lets you train and deploy state-of-the-art ML models for classification, regression, and time series forecasting on [Amazon SageMaker](https://aws.amazon.com/sagemaker/). All it takes is a few lines of code; AutoGluon-Cloud handles the infrastructure, dependencies, and glue code for you.
+
+## 💡 Why AutoGluon-Cloud?
+
+- **Works like local [AutoGluon](https://auto.gluon.ai/stable/index.html).** Pass in DataFrames, get predictions back — as convenient as working locally, with the compute handled by AWS.
+- **No boilerplate.** No training scripts, inference handlers, or serialization code to write and maintain.
+- **Official AWS containers.** Everything runs in the [AutoGluon Deep Learning Containers](https://aws.github.io/deep-learning-containers/), maintained and security-patched by AWS.
+- **Sensible defaults, fully configurable.** Under the hood it's just SageMaker running in your AWS account, so you stay in full control.
+
+## 💾 Installation
+
+```bash
+pip install autogluon.cloud
+autogluon-cloud bootstrap  # one-time setup for IAM role and S3 bucket
+```
+
+See the [Setup tutorial](https://auto.gluon.ai/cloud/stable/tutorials/setup.html) for more details.
+
+## 🚀 Foundation models
+
+Zero-shot forecasts with a pretrained model like [Chronos-2](https://huggingface.co/amazon/chronos-2) — no training required. [Full walkthrough](https://auto.gluon.ai/cloud/stable/tutorials/foundation-model-timeseries.html).
+
+```python
+from autogluon.cloud import TimeSeriesFoundationModel
+
+# `data` can be a local path, S3 URL, or pandas DataFrame
+data = "https://autogluon.s3.amazonaws.com/datasets/timeseries/m4_hourly_tiny/train.csv"
+
+model = TimeSeriesFoundationModel("chronos-2")
+
+# Batch prediction
+predictions = model.predict(data=data, target="target", prediction_length=24)
+
+# Real-time inference endpoint
+endpoint = model.deploy()
+predictions = endpoint.predict(data=data, target="target", prediction_length=24)
+endpoint.delete_endpoint()
+```
+
+## ⚙️ Train your own predictor
+
+Train an AutoGluon predictor on your data and serve it from SageMaker. Full walkthrough: [time series](https://auto.gluon.ai/cloud/stable/tutorials/predictor-timeseries.html), [tabular](https://auto.gluon.ai/cloud/stable/tutorials/predictor-tabular.html).
+
+```python
+from autogluon.cloud import TabularCloudPredictor
+
+# `train_data` and `test_data` can be a local path, S3 URL, or pandas DataFrame
+train_data = "https://autogluon.s3.amazonaws.com/datasets/Inc/train.csv"
+test_data = "https://autogluon.s3.amazonaws.com/datasets/Inc/test.csv"
+
+# Train
+cloud_predictor = TabularCloudPredictor()
+cloud_predictor.fit(
+    train_data=train_data,
+    predictor_init_args={"label": "class"},  # passed to TabularPredictor()
+    predictor_fit_args={"time_limit": 120},  # passed to TabularPredictor.fit()
+)
+
+# Batch prediction
+result = cloud_predictor.predict(test_data)
+
+# Real-time inference endpoint
+endpoint = cloud_predictor.deploy()
+result = endpoint.predict(test_data)
+endpoint.delete_endpoint()
+```

{autogluon_cloud-0.6.0 → autogluon_cloud-0.7.0}/pyproject.toml

@@ -43,23 +43,19 @@ dependencies = [
     # TransformAmiVersion was added to the SageMaker service model in botocore 1.37.23.
     "boto3>=1.37.23,<2",
     "packaging>=23.0,<27",
-    # Supports forwarding inference_ami_version from Model.deploy.
-    "sagemaker>=2.240.0,<3",
     "pyarrow>=19.0.1,<25",  # lower bound to avoid https://github.com/apache/arrow/issues/45283
     "PyYAML~=6.0",
     "Pillow>=10.2,<13",
     "huggingface_hub>=0.20,<2",
-    "typing_extensions>=4.0,<5",
+    "typing_extensions>=4.5,<5",  # 4.5 adds @deprecated (PEP 702)
     # CLI dependencies (autogluon-cloud command)
     "click>=8.0,<9",
     "rich>=13.0,<15",
 ]
 
 [project.optional-dependencies]
-ray = ["ray[default]>=2.10.0,<2.59"]
 all = [  # To allow user to pass ag objects
     "autogluon>=0.7,<1.7",
-    "autogluon.cloud[ray]",
 ]
 tests = [
     "pytest",
@@ -86,7 +82,6 @@ namespaces = true
 
 [tool.setuptools.package-data]
 "autogluon.cloud" = [
-    "default_cluster_configs/*.yaml",
     "utils/autogluon_dlc.json",
     "templates/*.yaml",
 ]
@@ -105,5 +100,7 @@ known-first-party = ["autogluon"]
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]
-addopts = "--strict-markers"
+# Release tests are slow and expensive; run them with `-m release` (see .github/workflows/release_tests.yml).
+addopts = "--strict-markers -m 'not release'"
+markers = ["release: pre-release test that runs every registry model on its default instance"]
 xfail_strict = true

{autogluon_cloud-0.6.0 → autogluon_cloud-0.7.0}/setup.py

@@ -24,7 +24,7 @@ def update_version(version):
     return version
 
 
-version = "0.6.0"
+version = "0.7.0"
 version = update_version(version)
 
 if __name__ == "__main__":

{autogluon_cloud-0.6.0 → autogluon_cloud-0.7.0}/src/autogluon/cloud/__init__.py

@@ -1,19 +1,9 @@
 import logging
-import os
-
-os.environ.setdefault("SAGEMAKER_SUPPRESS_V2_WARNING", "1")
-import sagemaker
-from packaging.version import Version
-
-if Version(sagemaker.__version__) >= Version("3.0"):
-    raise ImportError(
-        f"SageMaker SDK >= 3.0 is currently not supported (found {sagemaker.__version__}). "
-        "Please downgrade: pip install -U 'sagemaker<3'"
-    )
 
 from autogluon.common.utils.log_utils import _add_stream_handler
 
 from .cloud_setup import bootstrap, register, status, teardown
+from .endpoint.multimodal_endpoint import MultiModalEndpoint
 from .endpoint.tabular_endpoint import TabularEndpoint
 from .endpoint.timeseries_endpoint import TimeSeriesEndpoint
 from .model.foundation_model import TabularFoundationModel, TimeSeriesFoundationModel
@@ -24,6 +14,7 @@ logging.getLogger(__name__).setLevel(logging.INFO)
 
 __all__ = [
     "MultiModalCloudPredictor",
+    "MultiModalEndpoint",
     "TabularCloudPredictor",
     "TabularEndpoint",
     "TabularFoundationModel",

{autogluon_cloud-0.6.0 → autogluon_cloud-0.7.0}/src/autogluon/cloud/backend/backend.py

@@ -3,14 +3,12 @@ from __future__ import annotations
 import json
 import os
 from abc import ABC, abstractmethod
-from typing import Any, Dict, List, Optional, Union
+from typing import Any
 
 import pandas as pd
 
-from ..endpoint.endpoint import Endpoint
 
-
-def dumps_ag_args(config: Dict[str, Any]) -> str:
+def dumps_ag_args(config: dict[str, Any]) -> str:
     """Serialize the remote-training config to JSON, raising a user-facing error on failure.
 
     The config carries the user's predictor init/fit arguments. Some objects (search spaces,
@@ -58,8 +56,8 @@ class Backend(ABC):
         self,
         local_output_path: str,
         predictor_type: str,
-        cloud_output_path: Optional[str] = None,
-        resource_prefix: Optional[str] = None,
+        cloud_output_path: str | None = None,
+        resource_prefix: str | None = None,
         **kwargs,
     ) -> None:
         """Initialize the backend."""
@@ -68,12 +66,7 @@ class Backend(ABC):
         self.predictor_type = predictor_type
         self.resource_prefix = resource_prefix or f"ag-cloud-{predictor_type}"
         self.original_features = None
-        self.endpoint: Optional[Endpoint] = None
-
-    @abstractmethod
-    def parse_backend_fit_kwargs(self, kwargs: Dict) -> Dict[str, Any]:
-        """Parse backend specific kwargs and get them ready to be sent to fit call"""
-        raise NotImplementedError
+        self.endpoint_name: str | None = None
 
     @abstractmethod
     def attach_job(self, job_name: str) -> None:
@@ -108,7 +101,7 @@ class Backend(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def get_fit_job_info(self) -> Dict[str, Any]:
+    def get_fit_job_info(self) -> dict[str, Any]:
         """
         Get general info of the training job.
         """
@@ -139,11 +132,6 @@ class Backend(ABC):
         """Fit AG on the backend"""
         raise NotImplementedError
 
-    @abstractmethod
-    def parse_backend_deploy_kwargs(self, kwargs: Dict) -> Dict[str, Any]:
-        """Parse backend specific kwargs and get them ready to be sent to deploy call"""
-        raise NotImplementedError
-
     @abstractmethod
     def deploy(self, **kwargs) -> None:
         """Deploy and endpoint"""
@@ -155,32 +143,27 @@ class Backend(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def attach_endpoint(self, endpoint: Endpoint) -> None:
-        """Attach the backend to an existing endpoint"""
+    def attach_endpoint(self, endpoint: str) -> None:
+        """Attach the backend to an existing endpoint by name"""
         raise NotImplementedError
 
     @abstractmethod
-    def detach_endpoint(self) -> Endpoint:
-        """Detach the current endpoint and return it"""
+    def detach_endpoint(self) -> str:
+        """Detach the current endpoint and return its name"""
         raise NotImplementedError
 
     @abstractmethod
-    def predict_real_time(self, test_data: Union[str, pd.DataFrame], **kwargs) -> Union[pd.DataFrame, pd.Series]:
+    def predict_real_time(self, test_data: str | pd.DataFrame, **kwargs) -> pd.DataFrame | pd.Series:
         """Realtime prediction with the endpoint"""
         raise NotImplementedError
 
     @abstractmethod
-    def predict_proba_real_time(self, test_data: Union[str, pd.DataFrame], **kwargs) -> Union[pd.DataFrame, pd.Series]:
+    def predict_proba_real_time(self, test_data: str | pd.DataFrame, **kwargs) -> pd.DataFrame | pd.Series:
         """Realtime prediction probability with the endpoint"""
         raise NotImplementedError
 
     @abstractmethod
-    def parse_backend_predict_kwargs(self, kwargs: Dict) -> Dict[str, Any]:
-        """Parse backend specific kwargs and get them ready to be sent to predict call"""
-        raise NotImplementedError
-
-    @abstractmethod
-    def get_batch_inference_job_info(self, job_name: Optional[str] = None) -> Dict[str, Any]:
+    def get_batch_inference_job_info(self, job_name: str | None = None) -> dict[str, Any]:
         """
         Get general info of the batch inference job.
         If job_name not specified, return the info of the most recent batch inference job
@@ -188,7 +171,7 @@ class Backend(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def get_batch_inference_job_status(self, job_name: Optional[str] = None) -> str:
+    def get_batch_inference_job_status(self, job_name: str | None = None) -> str:
         """
         Get general status of the batch inference job.
         If job_name not specified, return the info of the most recent batch inference job
@@ -196,17 +179,17 @@ class Backend(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def get_batch_inference_jobs(self) -> List[str]:
+    def get_batch_inference_jobs(self) -> list[str]:
         """Get a list of names of all batch inference jobs"""
         raise NotImplementedError
 
     @abstractmethod
-    def predict(self, test_data: Union[str, pd.DataFrame], **kwargs) -> Union[pd.DataFrame, pd.Series]:
+    def predict(self, test_data: str | pd.DataFrame, **kwargs) -> pd.DataFrame | pd.Series:
         """Batch inference"""
         raise NotImplementedError
 
     @abstractmethod
-    def predict_proba(self, test_data: Union[str, pd.DataFrame], **kwargs) -> Union[pd.DataFrame, pd.Series]:
+    def predict_proba(self, test_data: str | pd.DataFrame, **kwargs) -> pd.DataFrame | pd.Series:
         """Batch inference probability"""
         raise NotImplementedError
 

autogluon_cloud-0.7.0/src/autogluon/cloud/backend/backend_factory.py

@@ -0,0 +1,25 @@
+from .backend import Backend
+from .multimodal_sagemaker_backend import MultiModalSagemakerBackend
+from .sagemaker_backend import SagemakerBackend
+from .tabular_sagemaker_backend import TabularSagemakerBackend
+from .timeseries_sagemaker_backend import TimeSeriesSagemakerBackend
+
+
+class BackendFactory:
+    _BACKENDS = {
+        SagemakerBackend.name: SagemakerBackend,
+        TabularSagemakerBackend.name: TabularSagemakerBackend,
+        MultiModalSagemakerBackend.name: MultiModalSagemakerBackend,
+        TimeSeriesSagemakerBackend.name: TimeSeriesSagemakerBackend,
+    }
+
+    @staticmethod
+    def get_backend_cls(backend: str) -> type[Backend]:
+        if backend in BackendFactory._BACKENDS:
+            return BackendFactory._BACKENDS[backend]
+        raise ValueError(f"{backend} not supported. Supported backends: {sorted(BackendFactory._BACKENDS)}")
+
+    @staticmethod
+    def get_backend(backend: str, **init_args) -> Backend:
+        """Return the corresponding backend"""
+        return BackendFactory.get_backend_cls(backend)(**init_args)

{autogluon_cloud-0.6.0 → autogluon_cloud-0.7.0}/src/autogluon/cloud/backend/constant.py

@@ -2,9 +2,6 @@ SAGEMAKER = "sagemaker"
 TABULAR_SAGEMAKER = "tabular_sagemaker"
 MULTIMODL_SAGEMAKER = "multimodal_sagemaker"
 TIMESERIES_SAGEMAKER = "timeseries_sagemaker"
-RAY = "ray"
-RAY_AWS = "ray_aws"
-TABULAR_RAY_AWS = "tabular_ray_aws"
 
 # Backends the top-level setup API can provision via CloudFormation.
-SUPPORTED_BACKENDS = (SAGEMAKER, RAY_AWS)
+SUPPORTED_BACKENDS = (SAGEMAKER,)