Merge remote-tracking branch 'origin/main' into feature/feature-store-int-fix

Author: harsh97

.github/workflows/run-unittests-py38-cov-report.yml

@@ -57,17 +57,9 @@ jobs:
       - uses: ./.github/workflows/set-dummy-conf
         name: "Test config setup"
 
-      - name: "Test env setup"
+      - uses: ./.github/workflows/test-env-setup
+        name: "Test env setup"
         timeout-minutes: 20
-        shell: bash
-        run: |
-          set -x # print commands that are executed
-
-          sudo apt-get install libkrb5-dev graphviz
-          $CONDA/bin/conda init
-          source /home/runner/.bashrc
-
-          pip install -r dev-requirements.txt
 
       - name: "Run unitary tests folder with maximum ADS dependencies"
         timeout-minutes: 60

.github/workflows/run-unittests-py39-py310.yml

@@ -58,17 +58,9 @@ jobs:
       - uses: ./.github/workflows/set-dummy-conf
         name: "Test config setup"
 
-      - name: "Test env setup"
+      - uses: ./.github/workflows/test-env-setup
+        name: "Test env setup"
         timeout-minutes: 20
-        shell: bash
-        run: |
-          set -x # print commands that are executed
-
-          sudo apt-get install libkrb5-dev graphviz
-          $CONDA/bin/conda init
-          source /home/runner/.bashrc
-
-          pip install -r dev-requirements.txt
 
       - name: "Run unitary tests folder with maximum ADS dependencies"
         timeout-minutes: 60

.github/workflows/test-env-setup/action.yml

@@ -0,0 +1,18 @@
+# This composite action is to avoid duplicating code across test actions.
+# Ref: https://docs.github.com/en/actions/creating-actions/creating-a-composite-action
+
+name: "Test env setup"
+description: "Creates conda test env with optional dependencies to run all unit tests"
+runs:
+  using: composite
+  steps:
+    - shell: bash
+      run: |
+        set -x # print commands that are executed
+
+        sudo apt update
+        sudo apt-get install libkrb5-dev graphviz
+        $CONDA/bin/conda init
+        source /home/runner/.bashrc
+
+        pip install -r dev-requirements.txt

THIRD_PARTY_LICENSES.txt

@@ -247,6 +247,12 @@ psutil
 * Source code: https://github.com/giampaolo/psutil
 * Project home: https://github.com/giampaolo/psutil
 
+py-cpuinfo
+* Copyright (c) 2014-2022 Matthew Brennan Jones <matthew.brennan.jones@gmail.com>
+* License: The MIT License (MIT)
+* Source code: https://github.com/workhorsy/py-cpuinfo
+* Project home: https://github.com/workhorsy/py-cpuinfo
+
 pyspark
 * Copyright 2014 and onwards The Apache Software Foundation.
 * License: Apache-2.0 LICENSE

ads/common/oci_logging.py

@@ -533,7 +533,7 @@ def _search_and_format(
         time_end: datetime.datetime = None,
         limit: int = LOG_RECORDS_LIMIT,
         sort_by: str = "datetime",
-        sort_order: str = SortOrder.DESC,
+        sort_order: str = SortOrder.ASC,
         log_filter: str = None,
     ):
         """Returns the formatted log records.
@@ -550,7 +550,7 @@ def _search_and_format(
             Maximum number of records to be returned.
         sort_by: (str, optional). Defaults to "datetime"
             The field for sorting the logs.
-        sort_order: (str, optional). Defaults to "DESC".
+        sort_order: (str, optional). Defaults to "ASC".
             The sort order for the log records. Can be "ASC" or "DESC".
         log_filter : (str, optional). Defaults to None.
             Expression for filtering the logs.
@@ -574,8 +574,8 @@ def _search_and_format(
             sort_order=sort_order,
             log_filter=log_filter,
         )
-        logs = sorted((log.data for log in logs), key=lambda x: x.get("datetime"))
-        logs = [log.get("logContent", {}) for log in logs]
+
+        logs = [log.data.get("logContent", {}) for log in logs]
         return [
             {
                 "id": log.get("id"),
@@ -617,7 +617,7 @@ def tail(
         return self._search_and_format(
             source=source,
             limit=limit,
-            sort_order=SortOrder.DESC,
+            sort_order=SortOrder.ASC,
             time_start=time_start,
             log_filter=log_filter,
         )
@@ -808,7 +808,7 @@ def stream(
                 logs = self._search_and_format(
                     source=source,
                     limit=None,
-                    sort_order=SortOrder.DESC,
+                    sort_order=SortOrder.ASC,
                     time_start=time_start,
                     log_filter=log_filter,
                 )
@@ -858,7 +858,7 @@ def tail(
             self._search_and_format(
                 source=source,
                 limit=limit,
-                sort_order=SortOrder.DESC,
+                sort_order=SortOrder.ASC,
                 time_start=time_start,
                 log_filter=log_filter,
             )
@@ -927,7 +927,7 @@ def search(
         time_end: datetime.datetime = None,
         limit: int = None,
         sort_by: str = "datetime",
-        sort_order: str = SortOrder.DESC,
+        sort_order: str = SortOrder.ASC,
         log_filter: str = None,
     ) -> List[oci.loggingsearch.models.SearchResult]:
         """Searches raw logs.
@@ -951,7 +951,7 @@ def search(
             Defaults to "datetime"
         sort_order : str, optional.
             The sort order for the log records. Can be "ASC" or "DESC".
-            Defaults to "DESC".
+            Defaults to "ASC".
         log_filter : str, optional
             Expression for filtering the logs. This will be the WHERE clause of the query.
             Defaults to None.
@@ -979,7 +979,7 @@ def _search_and_format(
         time_end: datetime.datetime = None,
         limit: int = LOG_RECORDS_LIMIT,
         sort_by: str = "datetime",
-        sort_order: str = SortOrder.DESC,
+        sort_order: str = SortOrder.ASC,
         log_filter: str = None,
         need_format: bool = True,
     ) -> List[Union[oci.loggingsearch.models.SearchResult, dict]]:
@@ -1005,7 +1005,7 @@ def _search_and_format(
             Defaults to "datetime"
         sort_order : str, optional.
             The sort order for the log records. Can be "ASC" or "DESC".
-            Defaults to "DESC".
+            Defaults to "ASC".
         log_filter : str, optional
             Expression for filtering the logs. This will be the WHERE clause of the query.
             Defaults to None.
@@ -1039,12 +1039,6 @@ def _search_and_format(
                 )
             )
 
-        # _collect_logs returns a list of either dict or oci.loggingsearch.models.SearchResult
-        # objects based on `need_format` parameter, so below there are two cases for log sorting.
-        if need_format:
-            batch_logs.sort(key=lambda x: x.get("datetime"))
-        else:
-            batch_logs.sort(key=lambda x: x.data.get("datetime"))
         if limit and len(batch_logs) > limit:
             batch_logs = batch_logs[:limit]
         return batch_logs
@@ -1057,7 +1051,7 @@ def _collect_logs(
         time_end: datetime.datetime = None,
         limit: int = LOG_RECORDS_LIMIT,
         sort_by: str = "datetime",
-        sort_order: str = SortOrder.DESC,
+        sort_order: str = SortOrder.ASC,
         log_filter: str = None,
         need_format: bool = True,
     ) -> List[Union[oci.loggingsearch.models.SearchResult, dict]]:
@@ -1085,7 +1079,7 @@ def _collect_logs(
             Defaults to "datetime"
         sort_order : str, optional.
             The sort order for the log records. Can be "ASC" or "DESC".
-            Defaults to "DESC".
+            Defaults to "ASC".
         log_filter : str, optional
             Expression for filtering the logs. This will be the WHERE clause of the query.
             Defaults to None.

ads/common/utils.py

@@ -71,7 +71,7 @@
 MAX_DISPLAY_VALUES = 10
 
 # par link of the index json file.
-PAR_LINK = "https://objectstorage.us-ashburn-1.oraclecloud.com/p/Ri7zFc_h91sxMdgnza9Qnqw3Ina8hf8wzDvEpAnUXMDOnUR1U1fpsaBUjUfgPgIq/n/ociodscdev/b/service-conda-packs/o/service_pack/index.json"
+PAR_LINK = "https://objectstorage.us-ashburn-1.oraclecloud.com/p/WyjtfVIG0uda-P3-2FmAfwaLlXYQZbvPZmfX1qg0-sbkwEQO6jpwabGr2hMDBmBp/n/ociodscdev/b/service-conda-packs/o/service_pack/index.json"
 
 random_state = 42
 test_size = 0.3

ads/jobs/serializer.py

@@ -12,11 +12,9 @@
 import yaml
 from ads.common.auth import default_signer
 
+# Special type to represent the current enclosed class.
+# This type is used by factory class method or when a method returns ``self``.
 Self = TypeVar("Self", bound="Serializable")
-"""Special type to represent the current enclosed class.
-
-This type is used by factory class method or when a method returns ``self``.
-"""
 
 
 class Serializable(ABC):
@@ -72,6 +70,14 @@ def _write_to_file(s: str, uri: str, **kwargs) -> None:
                     "if you wish to overwrite."
                 )
 
+        # Add default signer if the uri is an object storage uri, and
+        # the user does not specify config or signer.
+        if (
+            uri.startswith("oci://")
+            and "config" not in kwargs
+            and "signer" not in kwargs
+        ):
+            kwargs.update(default_signer())
         with fsspec.open(uri, "w", **kwargs) as f:
             f.write(s)
 

ads/model/deployment/model_deployment.py

@@ -85,7 +85,7 @@ class LogNotConfiguredError(Exception):  # pragma: no cover
     pass
 
 
-class ModelDeploymentFailedError(Exception):  # pragma: no cover
+class ModelDeploymentPredictError(Exception):  # pragma: no cover
     pass
 
 
@@ -607,11 +607,6 @@ def deploy(
         -------
         ModelDeployment
            The instance of ModelDeployment.
-
-        Raises
-        ------
-        ModelDeploymentFailedError
-            If model deployment fails to deploy
         """
         create_model_deployment_details = (
             self._build_model_deployment_details()
@@ -626,11 +621,6 @@ def deploy(
             poll_interval=poll_interval,
         )
 
-        if response.lifecycle_state == State.FAILED.name:
-            raise ModelDeploymentFailedError(
-                f"Model deployment {response.id} failed to deploy: {response.lifecycle_details}"
-            )
-
         return self._update_from_oci_model(response)
 
     def delete(
@@ -662,6 +652,7 @@ def delete(
             max_wait_time=max_wait_time,
             poll_interval=poll_interval,
         )
+
         return self._update_from_oci_model(response)
 
     def update(
@@ -890,6 +881,12 @@ def predict(
             Prediction results.
 
         """
+        current_state = self.sync().lifecycle_state
+        if current_state != State.ACTIVE.name:
+            raise ModelDeploymentPredictError(
+                "This model deployment is not in active state, you will not be able to use predict end point. "
+                f"Current model deployment state: {current_state} "
+            )
         endpoint = f"{self.url}/predict"
         signer = authutil.default_signer()["signer"]
         header = {
@@ -953,7 +950,7 @@ def predict(
         except oci.exceptions.ServiceError as ex:
             # When bandwidth exceeds the allocated value, TooManyRequests error (429) will be raised by oci backend.
             if ex.status == 429:
-                bandwidth_mbps = self.infrastructure.bandwidth_mbps or MODEL_DEPLOYMENT_BANDWIDTH_MBPS
+                bandwidth_mbps = self.infrastructure.bandwidth_mbps or DEFAULT_BANDWIDTH_MBPS
                 utils.get_logger().warning(
                     f"Load balancer bandwidth exceeds the allocated {bandwidth_mbps} Mbps."
                     "To estimate the actual bandwidth, use formula: (payload size in KB) * (estimated requests per second) * 8 / 1024."

ads/model/generic_model.py

@@ -1575,7 +1575,10 @@ def from_model_deployment(
 
         current_state = model_deployment.state.name.upper()
         if current_state != ModelDeploymentState.ACTIVE.name:
-            raise NotActiveDeploymentError(current_state)
+            logger.warning(
+                "This model deployment is not in active state, you will not be able to use predict end point. "
+                f"Current model deployment state: `{current_state}`"
+            )
 
         model = cls.from_model_catalog(
             model_id=model_deployment.properties.model_id,

ads/model/runtime/utils.py

@@ -111,7 +111,7 @@ def _get_index_json_through_bucket(
     bucketname: str
         The Object Storage bucketname.
     auth: (Dict, optional). Defaults to None.
-        The default authetication is set using `ads.set_auth` API. If you need to override the
+        The default authentication is set using `ads.set_auth` API. If you need to override the
         default, use the `ads.common.auth.api_keys` or `ads.common.auth.resource_principal` to create appropriate
         authentication signer and kwargs required to instantiate IdentityClient object.
 
@@ -147,7 +147,7 @@ def get_service_packs(
     bucketname: str
         bucketname of the service pack.
     auth: (Dict, optional). Defaults to None.
-        The default authetication is set using `ads.set_auth` API. If you need to override the
+        The default authentication is set using `ads.set_auth` API. If you need to override the
         default, use the `ads.common.auth.api_keys` or `ads.common.auth.resource_principal` to create appropriate
         authentication signer and kwargs required to instantiate IdentityClient object.