fix: Add deprecation warnings when using Ray v2.4

PiperOrigin-RevId: 629468075

google/cloud/aiplatform/preview/vertex_ray/cluster_init.py

@@ -19,6 +19,7 @@
 import logging
 import time
 from typing import Dict, List, Optional
+import warnings
 
 from google.cloud.aiplatform import initializer
 from google.cloud.aiplatform import utils
@@ -127,7 +128,10 @@ def create_ray_cluster(
         logging.info(
             "[Ray on Vertex]: No VPC network configured. It is required for client connection."
         )
-
+    if ray_version == "2.4":
+        warnings.warn(
+            _gapic_utils._V2_4_WARNING_MESSAGE, DeprecationWarning, stacklevel=2
+        )
     local_ray_verion = _validation_utils.get_local_ray_version()
     if ray_version != local_ray_verion:
         if custom_images is None and head_node_type.custom_image is None:

google/cloud/aiplatform/preview/vertex_ray/data.py

@@ -18,6 +18,7 @@
 import ray.data
 from ray.data.dataset import Dataset
 from typing import Any, Dict, Optional
+import warnings
 
 from google.cloud.aiplatform.preview.vertex_ray.bigquery_datasource import (
     BigQueryDatasource,
@@ -30,6 +31,10 @@
 except ImportError:
     _BigQueryDatasink = None
 
+from google.cloud.aiplatform.preview.vertex_ray.util._validation_utils import (
+    _V2_4_WARNING_MESSAGE,
+)
+
 
 def read_bigquery(
     project_id: Optional[str] = None,
@@ -56,6 +61,7 @@ def write_bigquery(
     ray_remote_args: Dict[str, Any] = None,
 ) -> Any:
     if ray.__version__ == "2.4.0":
+        warnings.warn(_V2_4_WARNING_MESSAGE, DeprecationWarning, stacklevel=2)
         return ds.write_datasource(
             BigQueryDatasource(),
             project_id=project_id,

google/cloud/aiplatform/preview/vertex_ray/predict/sklearn/register.py

@@ -24,6 +24,7 @@
 import ray.cloudpickle as cpickle
 import tempfile
 from typing import Optional, TYPE_CHECKING
+import warnings
 
 from google.cloud import aiplatform
 from google.cloud.aiplatform import initializer
@@ -33,6 +34,9 @@
 from google.cloud.aiplatform.preview.vertex_ray.predict.util import (
     predict_utils,
 )
+from google.cloud.aiplatform.preview.vertex_ray.util._validation_utils import (
+    _V2_4_WARNING_MESSAGE,
+)
 
 
 try:
@@ -123,6 +127,7 @@ def _get_estimator_from(
 
     ray_version = ray.__version__
     if ray_version == "2.4.0":
+        warnings.warn(_V2_4_WARNING_MESSAGE, DeprecationWarning, stacklevel=2)
         if not isinstance(checkpoint, ray_sklearn.SklearnCheckpoint):
             raise ValueError(
                 "[Ray on Vertex AI]: arg checkpoint should be a"

google/cloud/aiplatform/preview/vertex_ray/predict/tensorflow/register.py

@@ -20,6 +20,7 @@
 import logging
 import ray
 from typing import Callable, Optional, Union, TYPE_CHECKING
+import warnings
 
 from google.cloud import aiplatform
 from google.cloud.aiplatform import initializer
@@ -28,6 +29,9 @@
 from google.cloud.aiplatform.preview.vertex_ray.predict.util import (
     predict_utils,
 )
+from google.cloud.aiplatform.preview.vertex_ray.util._validation_utils import (
+    _V2_4_WARNING_MESSAGE,
+)
 
 
 try:
@@ -141,6 +145,7 @@ def _get_tensorflow_model_from(
     """
     ray_version = ray.__version__
     if ray_version == "2.4.0":
+        warnings.warn(_V2_4_WARNING_MESSAGE, DeprecationWarning, stacklevel=2)
         if not isinstance(checkpoint, ray_tensorflow.TensorflowCheckpoint):
             raise ValueError(
                 "[Ray on Vertex AI]: arg checkpoint should be a"

google/cloud/aiplatform/preview/vertex_ray/predict/torch/register.py

@@ -20,8 +20,12 @@
 import ray
 from ray.air._internal.torch_utils import load_torch_model
 import tempfile
+from google.cloud.aiplatform.preview.vertex_ray.util._validation_utils import (
+    _V2_4_WARNING_MESSAGE,
+)
 from google.cloud.aiplatform.utils import gcs_utils
 from typing import Optional
+import warnings
 
 
 try:
@@ -61,6 +65,7 @@ def get_pytorch_model_from(
     """
     ray_version = ray.__version__
     if ray_version == "2.4.0":
+        warnings.warn(_V2_4_WARNING_MESSAGE, DeprecationWarning, stacklevel=2)
         if not isinstance(checkpoint, ray_torch.TorchCheckpoint):
             raise ValueError(
                 "[Ray on Vertex AI]: arg checkpoint should be a"

google/cloud/aiplatform/preview/vertex_ray/predict/xgboost/register.py

@@ -23,6 +23,7 @@
 import ray
 import tempfile
 from typing import Optional, TYPE_CHECKING
+import warnings
 
 from google.cloud import aiplatform
 from google.cloud.aiplatform import initializer
@@ -32,6 +33,9 @@
 from google.cloud.aiplatform.preview.vertex_ray.predict.util import (
     predict_utils,
 )
+from google.cloud.aiplatform.preview.vertex_ray.util._validation_utils import (
+    _V2_4_WARNING_MESSAGE,
+)
 
 
 try:
@@ -133,6 +137,7 @@ def _get_xgboost_model_from(
     """
     ray_version = ray.__version__
     if ray_version == "2.4.0":
+        warnings.warn(_V2_4_WARNING_MESSAGE, DeprecationWarning, stacklevel=2)
         if not isinstance(checkpoint, ray_xgboost.XGBoostCheckpoint):
             raise ValueError(
                 "[Ray on Vertex AI]: arg checkpoint should be a"

google/cloud/aiplatform/preview/vertex_ray/util/_validation_utils.py

@@ -27,6 +27,10 @@
 
 SUPPORTED_RAY_VERSIONS = immutabledict({"2.4": "2.4.0", "2.9": "2.9.3"})
 SUPPORTED_PY_VERSION = ["3.10"]
+_V2_4_WARNING_MESSAGE = (
+    "After May 30, 2024, using Ray version = 2.4 will result in an error. "
+    "Please use Ray version = 2.9.3 (default) instead."
+)
 
 # Artifact Repository available regions.
 _AVAILABLE_REGIONS = ["us", "europe", "asia"]