Merge branch 'master' into ISSUE-24869

* master: (35 commits)
  [tune/structure] Refactor `suggest` into `search` package (ray-project#26074)
  Add back ray.state in deprecation wrapper; print stack trace on warning (ray-project#26086)
  Enable isort for base directory (ray-project#26085)
  [AIR] Add __init__.py to ray.air.callbacks (ray-project#26088)
  [Serve] [Docs] Create end-to-end documentation example for Serve REST API and CLI (ray-project#25936)
  [AIR] Remove unnecessary pandas from examples (ray-project#26009)
  [Datasets] [Hotfix] Update `ds.to_pandas()` limit error to reflect current limit API (ray-project#26081)
  [Serve] [Docs] Add Serve REST API Schema to Serve API Docs (ray-project#25786)
  [Core][Doc] remove cython section from advanced doc. ray-project#26062
  [Core] Fix check failure from incorrect death cause (ray-project#26007)
  [hotfix] Fix linkcheck (ray-project#26070)
  [RLlib] Add timeout to filter synchronization. (ray-project#25959)
  [tune/structure] Introduce logger package (ray-project#26049)
  [RLlib] introduce serialization for our custom gym space types. (ray-project#25923)
  Fix unit test test_check_env.py and est_check_multi_agent.py. (ray-project#25993)
  [RLlib] Make QMix use the ReplayBufferAPI (ray-project#25560)
  [CI] deflake test_multi_node_3 by increasing its timeout
  [CI] Use BUILDKITE_JOB_ID for better navigation for flaky tracker (ray-project#26021)
  [AIR/Docs] Improve user guide gallery (ray-project#26016)
  🎨 Update type annotations to include options in `ray.remote()` (ray-project#25999)
  ...

.isort.cfg

@@ -10,11 +10,11 @@ float_to_top=True
 filter_files=True
 
 # Skip docs/* as these are carefully crafted and aren't strictly code.
-# Skip python/ray/*.py to avoid touching python/ray/__init__.py, which
-# is also carefully crafted.
+# Skip python/ray/__init__.py and python/ray/setup-dev.py, which
+# are also carefully crafted.
 # For the rest we will gradually remove them from the blacklist as we
 # reformat the code to follow the style guide.
-skip_glob=python/ray/*.py,doc/*,ci/*,python/ray/_raylet*,python/ray/_private/*,python/ray/air/*,python/ray/cloudpickle/*,python/ray/core/*,dashboard/*,python/ray/data/*,python/ray/experimental/*,python/ray/includes/*,python/ray/internal/*,python/ray/ray_operator/*,python/ray/scripts/*,python/ray/serve/*,python/ray/sgd/*,python/ray/streaming/*,python/ray/tests/*,python/ray/tests/*,python/ray/train/*,python/ray/tune/*,python/ray/util/*,python/ray/workers/*,python/ray/workflow/*,rllib/*,release/*,
+skip_glob=python/ray/__init__.py,python/ray/setup-dev.py,python/ray/dag/*.py,doc/*,ci/*,python/ray/_raylet*,python/ray/_private/*,python/ray/air/*,python/ray/cloudpickle/*,python/ray/core/*,dashboard/*,python/ray/data/*,python/ray/experimental/*,python/ray/includes/*,python/ray/internal/*,python/ray/ray_operator/*,python/ray/scripts/*,python/ray/serve/*,python/ray/sgd/*,python/ray/streaming/*,python/ray/tests/*,python/ray/tests/*,python/ray/train/*,python/ray/tune/*,python/ray/util/*,python/ray/workers/*,python/ray/workflow/*,rllib/*,release/*,
 
 known_local_folder=ray
 known_afterray=psutil,setproctitle

ci/build/get_build_info.py

@@ -48,9 +48,7 @@ def get_build_env():
         return {
             "TRAVIS_COMMIT": os.environ["BUILDKITE_COMMIT"],
             "TRAVIS_JOB_WEB_URL": (
-                os.environ["BUILDKITE_BUILD_URL"]
-                + "#"
-                + os.environ["BUILDKITE_BUILD_ID"]
+                os.environ["BUILDKITE_BUILD_URL"] + "#" + os.environ["BUILDKITE_JOB_ID"]
             ),
             "TRAVIS_OS_NAME": {  # The map is used to stay consistent with Travis
                 "linux": "linux",

ci/repro-ci.py

@@ -188,7 +188,7 @@ def aws_start_instance(self):
         # Check if instance exists:
         running_instances = self.ec2_resource.instances.filter(
             Filters=[
-                {"Name": "tag:repro_name", "Values": [self.aws_instance_name]},
+                {"Name": "tag:Name", "Values": [self.aws_instance_name]},
                 {"Name": "instance-state-name", "Values": ["running"]},
             ]
         )
@@ -218,7 +218,7 @@ def aws_start_instance(self):
             TagSpecifications=[
                 {
                     "ResourceType": "instance",
-                    "Tags": [{"Key": "repro_name", "Value": self.aws_instance_name}],
+                    "Tags": [{"Key": "Name", "Value": self.aws_instance_name}],
                 }
             ],
             MinCount=1,

doc/source/_toc.yml

@@ -173,6 +173,7 @@ parts:
                   - file: serve/deployment-graph/control_flow_based_on_user_inputs
                   - file: serve/deployment-graph/visualize_dag_during_development
                   - file: serve/deployment-graph/http_endpoint_for_dag_graph
+              - file: serve/production
           - file: serve/architecture
           - file: serve/tutorials/index
           - file: serve/faq

doc/source/conf.py

@@ -177,6 +177,7 @@
     r"https://huggingface.co/*",  # seems to be flaky
     r"https://www.meetup.com/*",  # seems to be flaky
     r"https://www.pettingzoo.ml/*",  # seems to be flaky
+    r"http://localhost[:/].*",  # Ignore localhost links
 ]
 
 # -- Options for HTML output ----------------------------------------------

doc/source/ray-air/doc_code/air_key_concepts.py

@@ -75,11 +75,7 @@
 batch_predictor = BatchPredictor.from_checkpoint(result.checkpoint, XGBoostPredictor)
 
 # Bulk batch prediction.
-predicted_labels = (
-    batch_predictor.predict(test_dataset)
-    .map_batches(lambda df: (df > 0.5).astype(int), batch_format="pandas")
-    .to_pandas(limit=float("inf"))
-)
+predicted_probabilities = batch_predictor.predict(test_dataset)
 
 # Pipelined batch prediction: instead of processing the data in bulk, process it
 # incrementally in windows of the given size.

doc/source/ray-air/doc_code/preprocessors.py

@@ -88,19 +88,17 @@
 test_dataset = ray.data.from_items([{"x": x} for x in range(2, 32, 3)])
 
 batch_predictor = BatchPredictor.from_checkpoint(checkpoint, XGBoostPredictor)
-predicted_labels = batch_predictor.predict(test_dataset)
-print(predicted_labels.to_pandas())
-#    predictions
-# 0     0.098437
-# 1     5.604667
-# 2    11.405312
-# 3    15.684700
-# 4    23.990948
-# 5    29.900211
-# 6    34.599442
-# 7    40.696899
-# 8    45.681076
-# 9    50.290031
+predicted_probabilities = batch_predictor.predict(test_dataset)
+predicted_probabilities.show()
+# {'predictions': 0.09843720495700836}
+# {'predictions': 5.604666709899902}
+# {'predictions': 11.405311584472656}
+# {'predictions': 15.684700012207031}
+# {'predictions': 23.990947723388672}
+# {'predictions': 29.900211334228516}
+# {'predictions': 34.59944152832031}
+# {'predictions': 40.6968994140625}
+# {'predictions': 45.68107604980469}
 # __predictor_end__
 
 

doc/source/ray-air/doc_code/tf_starter.py

@@ -105,7 +105,7 @@ def train_func(config: dict):
 
 predictions = batch_predictor.predict(prediction_dataset, dtype=tf.float32)
 
-pandas_predictions = predictions.to_pandas(float("inf"))
+print("PREDICTIONS")
+predictions.show()
 
-print(f"PREDICTIONS\n{pandas_predictions}")
 # __air_tf_batchpred_end__

doc/source/ray-air/doc_code/xgboost_starter.py

@@ -59,17 +59,11 @@
 
 batch_predictor = BatchPredictor.from_checkpoint(result.checkpoint, XGBoostPredictor)
 
-predicted_labels = (
-    batch_predictor.predict(test_dataset)
-    .map_batches(lambda df: (df > 0.5).astype(int), batch_format="pandas")
-    .to_pandas(limit=float("inf"))
-)
-print("PREDICTED LABELS")
-print(f"{predicted_labels}")
+predicted_probabilities = batch_predictor.predict(test_dataset)
+print("PREDICTED PROBABILITIES")
+predicted_probabilities.show()
 
-shap_values = batch_predictor.predict(test_dataset, pred_contribs=True).to_pandas(
-    limit=float("inf")
-)
+shap_values = batch_predictor.predict(test_dataset, pred_contribs=True)
 print("SHAP VALUES")
-print(f"{shap_values}")
+shap_values.show()
 # __air_xgb_batchpred_end__