Merge pull request #14 from oracle/2.6.x

Release version 2.6.6

ads/__init__.py

@@ -8,26 +8,26 @@
 import os
 import logging
 import sys
+import json
 
+__version__ = ""
+with open(
+    os.path.join(os.path.dirname(os.path.abspath(__file__)), "ads_version.json")
+) as version_file:
+    __version__ = json.load(version_file)["version"]
 import oci
 
 import matplotlib.font_manager  # causes matplotlib to regenerate its fonts
-import json
 
 import ocifs
 from ads.common.decorator.deprecate import deprecated
 from ads.common.ipython import configure_plotting, _log_traceback
 from ads.feature_engineering.accessor.series_accessor import ADSSeriesAccessor
 from ads.feature_engineering.accessor.dataframe_accessor import ADSDataFrameAccessor
-
+from ads.common import auth
 
 os.environ["GIT_PYTHON_REFRESH"] = "quiet"
 
-__version__ = ""
-with open(
-    os.path.join(os.path.dirname(os.path.abspath(__file__)), "ads_version.json")
-) as version_file:
-    __version__ = json.load(version_file)["version"]
 
 debug_mode = os.environ.get("DEBUG_MODE", False)
 documentation_mode = os.environ.get("DOCUMENTATION_MODE", "False") == "True"
@@ -64,7 +64,7 @@ def set_auth(
             oci_config_path = oci_config_location
         else:
             logging.warning(
-                f"{oci_config_location} file not exists, default value oci.config.DEFAULT_LOCATION used instead"
+                f"{oci_config_location} file does not exist, default value oci.config.DEFAULT_LOCATION used instead"
             )
             oci_config_path = oci.config.DEFAULT_LOCATION
     elif auth == "resource_principal":

ads/ads_version.json

@@ -1,3 +1,3 @@
 {
-  "version": "2.6.5"
+  "version": "2.6.6"
 }

ads/cli.py

@@ -3,13 +3,17 @@
 
 # Copyright (c) 2021, 2022 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
-
-import click
-import ads.opctl.cli
-import ads.jobs.cli
-import os
-import json
-
+try:
+    import click
+    import ads.opctl.cli
+    import ads.jobs.cli
+    import os
+    import json
+except:
+    print(
+        "Please run `pip install oracle-ads[opctl]` to install the required dependencies for ADS CLI"
+    )
+    exit()
 with open(
     os.path.join(os.path.dirname(os.path.abspath(__file__)), "ads_version.json")
 ) as version_file:

ads/common/__init__.py

@@ -7,3 +7,5 @@
 import logging
 
 logger = logging.getLogger(__name__)
+
+from ads.common import auth

ads/common/auth.py

@@ -7,6 +7,7 @@
 import oci
 import os
 from ads.common import utils
+import ads.telemetry
 
 
 def api_keys(
@@ -42,7 +43,9 @@ def api_keys(
     >>> auth = authutil.api_keys(oci_config="/home/datascience/.oci/config", profile="TEST", client_kwargs={"timeout": 6000})
     >>> oc.OCIClientFactory(**auth).object_storage # Creates Object storage client with timeout set to 6000 using API Key authentication
     """
-    configuration = oci.config.from_file(oci_config, profile)
+    configuration = ads.telemetry.update_oci_client_config(
+        oci.config.from_file(oci_config, profile)
+    )
     return {
         "config": configuration,
         "signer": oci.signer.Signer(
@@ -82,8 +85,10 @@ def resource_principal(client_kwargs=None):
     >>> oc.OCIClientFactory(**auth).object_storage # Creates Object Storage client with timeout set to 6000 seconds using resource principal authentication
     """
 
+    configuration = ads.telemetry.update_oci_client_config()
+
     return {
-        "config": {},
+        "config": configuration,
         "signer": oci.auth.signers.get_resource_principals_signer(),
         "client_kwargs": client_kwargs,
     }

ads/common/base_properties.py

@@ -7,7 +7,7 @@
 import dataclasses
 import json
 import os
-from typing import Any, Dict, Optional
+from typing import Any, Dict, Optional, Union
 
 from ads.common.config import Config, ConfigSection
 from ads.common.serializer import Serializable
@@ -42,11 +42,21 @@ def __setattr__(self, name: str, value: Any):
         if value is None:
             self.__dict__[name] = value
         elif name in self.__annotations__:
-            if self.__annotations__[name] != type(value):
+            if hasattr(self.__annotations__[name], "__origin__"):
+                if self.__annotations__[name].__origin__ is Union and not isinstance(
+                    value, self.__annotations__[name].__args__
+                ):
+                    raise TypeError(
+                        f"Field `{name}` was expected to be of type `{self.__annotations__[name].__args__}` "
+                        f"but type `{type(value)}` was provided."
+                    )
+
+            elif self.__annotations__[name] != type(value):
                 raise TypeError(
-                    f"Field `{name}` was expected to be of type `{self.__annotations__[name].__name__}` "
-                    f"but type `{type(value).__name__}` was provided."
+                    f"Field `{name}` was expected to be of type `{self.__annotations__[name]}` "
+                    f"but type `{type(value)}` was provided."
                 )
+
             self.__dict__[name] = value
 
     def with_prop(self, name: str, value: Any) -> "BaseProperties":

ads/common/data_serializer.py

@@ -0,0 +1,117 @@
+#!/usr/bin/env python
+# -*- coding: utf-8; -*-
+
+# Copyright (c) 2022 Oracle and/or its affiliates.
+# Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
+
+from __future__ import absolute_import, print_function
+
+import base64
+from io import BytesIO
+from typing import Dict, List, Union
+import requests
+import json
+import numpy as np
+import pandas as pd
+
+DEFAULT_CONTENT_TYPE_JSON = "application/json"
+DEFAULT_CONTENT_TYPE_BYTES = "application/octet-stream"
+
+
+class InputDataSerializer:
+    """[An internal class]
+    Defines the contract for input data
+
+    """
+
+    def __init__(
+        self,
+        data: Union[
+            Dict,
+            str,
+            List,
+            np.ndarray,
+            pd.core.series.Series,
+            pd.core.frame.DataFrame,
+        ],
+        data_type=None,
+    ):
+        """
+        Parameters
+        ----------
+        data: Union[Dict, str, list, numpy.ndarray, pd.core.series.Series,
+        pd.core.frame.DataFrame]
+            Data expected by the model deployment predict API.
+        data_type: Any, defaults to None.
+            Type of the data. If not provided, it will be checked against data.
+
+        """
+        if not data_type:
+            data_type = type(data)
+        if isinstance(data, np.ndarray):
+            np_bytes = BytesIO()
+            np.save(np_bytes, data, allow_pickle=True)
+            data = base64.b64encode(np_bytes.getvalue()).decode("utf-8")
+        elif isinstance(data, pd.core.series.Series):
+            data = data.tolist()
+        elif isinstance(data, pd.core.frame.DataFrame):
+            data = data.to_json()
+        elif (
+            isinstance(data, dict)
+            or isinstance(data, str)
+            or isinstance(data, list)
+            or isinstance(data, tuple)
+            or isinstance(data, bytes)
+        ):
+            pass
+        else:
+            raise TypeError(
+                "The provided data type is not json serializable. "
+                "The supported data types are Dict, str, list, "
+                "numpy.ndarray, pd.core.series.Series, "
+                "pd.core.frame.DataFrame. Please "
+                "convert to the supported data types first. "
+            )
+
+        self._data = data
+        self._data_type = str(data_type)
+
+    @property
+    def data(self):
+        return self._data
+
+    @property
+    def data_type(self):
+        return self._data_type
+
+    def to_dict(self):
+        return {
+            "data": self._data,
+            "data_type": self._data_type,
+        }
+
+    def is_bytes(self):
+        return "bytes" in self.data_type
+
+    def send(self, endpoint: str, dry_run: bool = False, **kwargs):
+        headers = dict()
+        if self.is_bytes():
+            headers["Content-Type"] = (
+                kwargs.get("content_type") or DEFAULT_CONTENT_TYPE_BYTES
+            )
+            request_kwargs = {"data": self.data}  # should pass bytes when using data
+        else:
+            headers["Content-Type"] = (
+                kwargs.get("content_type") or DEFAULT_CONTENT_TYPE_JSON
+            )
+            request_kwargs = {"json": self.to_dict()}
+        request_kwargs["headers"] = headers
+        if dry_run:
+            request_kwargs["headers"]["Accept"] = "*/*"
+            req = requests.Request("POST", endpoint, **request_kwargs).prepare()
+            if self.is_bytes():
+                return req.body
+            return json.loads(req.body)
+        else:
+            request_kwargs["auth"] = kwargs.get("signer")
+            return requests.post(endpoint, **request_kwargs).json()

ads/common/decorator/runtime_dependency.py

@@ -62,6 +62,7 @@ class OptionalDependency:
     GEO = "oracle-ads[geo]"
     ONNX = "oracle-ads[onnx]"
     OPTUNA = "oracle-ads[optuna]"
+    SPARK = "oracle-ads[spark]"
 
 
 def runtime_dependency(

ads/common/model.py

@@ -23,9 +23,9 @@
     runtime_dependency,
     OptionalDependency,
 )
+from ads.common.decorator.deprecate import deprecated
 from ads.common.utils import is_notebook
 from ads.dataset.pipeline import TransformerPipeline
-from distutils import dir_util
 from sklearn.pipeline import Pipeline
 
 
@@ -171,14 +171,18 @@ def rename(self, name):
         """
         self.name = name
 
+    @deprecated(
+        "2.6.6",
+        details="Use framework specific Model utility class for saving and deploying model saving and deploying. Check https://accelerated-data-science.readthedocs.io/en/latest/user_guide/model_registration/quick_start.html",
+    )
     def predict(self, X):
         """
         Runs the models predict function on some data
 
         Parameters
         ----------
-        X: MLData
-            A MLData object which holds the examples to be predicted on.
+        X: ADSData
+            A ADSData object which holds the examples to be predicted on.
 
         Returns
         -------
@@ -202,8 +206,8 @@ def predict_proba(self, X):
 
         Parameters
         ----------
-        X: MLData
-            A MLData object which holds the examples to be predicted on.
+        X: ADSData
+            A ADSData object which holds the examples to be predicted on.
 
         Returns
         -------
@@ -218,16 +222,20 @@ def predict_proba(self, X):
         else:
             return self.est.predict_proba(X)
 
+    @deprecated(
+        "2.6.6",
+        details="Use framework specific Model utility class for saving and deploying model saving and deploying. Check https://accelerated-data-science.readthedocs.io/en/latest/user_guide/model_registration/quick_start.html",
+    )
     def score(self, X, y_true, score_fn=None):
         """
         Scores a model according to a custom score function
 
         Parameters
         ----------
-        X: MLData
-            A MLData object which holds the examples to be predicted on.
-        y_true: MLData
-            A MLData object which holds ground truth labels for the examples which are being predicted on.
+        X: ADSData
+            A ADSData object which holds the examples to be predicted on.
+        y_true: ADSData
+            A ADSData object which holds ground truth labels for the examples which are being predicted on.
         score_fn: Scorer (callable)
             A callable object that returns a score, usually created with sklearn.metrics.make_scorer().
 
@@ -250,6 +258,10 @@ def score(self, X, y_true, score_fn=None):
             else:
                 return self.est.score(X, y_true)
 
+    @deprecated(
+        "2.6.6",
+        details="Use framework specific Model utility class for saving and deploying model saving and deploying. Check https://accelerated-data-science.readthedocs.io/en/latest/user_guide/model_registration/quick_start.html",
+    )
     def summary(self):
         """
         A summary of the ADSModel
@@ -274,14 +286,18 @@ def __repr__(self):
     def __getattr__(self, item):
         return getattr(self.est, item)
 
+    @deprecated(
+        "2.6.6",
+        details="Use framework specific Model utility class for saving and deploying model saving and deploying. Check https://accelerated-data-science.readthedocs.io/en/latest/user_guide/model_registration/quick_start.html",
+    )
     def transform(self, X):
         """
-        Process some MLData through the selected ADSModel transformers
+        Process some ADSData through the selected ADSModel transformers
 
         Parameters
         ----------
-        X: MLData
-            A MLData object which holds the examples to be transformed.
+        X: ADSData
+            A ADSData object which holds the examples to be transformed.
         """
 
         if hasattr(X, "copy"):
@@ -310,6 +326,10 @@ def is_classifier(self):
         """
         return hasattr(self, "classes_") and self.classes_ is not None
 
+    @deprecated(
+        "2.6.6",
+        details="Use framework specific Model utility class for saving and deploying model saving and deploying. Check https://accelerated-data-science.readthedocs.io/en/latest/user_guide/model_registration/quick_start.html",
+    )
     def feature_names(self, X=None):
         model_type = self._underlying_model
         if model_type == "sklearn":
@@ -367,6 +387,10 @@ def __getstate__(self):
     def __setstate__(self, state):
         self.__dict__ = state
 
+    @deprecated(
+        "2.6.6",
+        details="Use framework specific Model utility class for saving and deploying model saving and deploying. Check https://accelerated-data-science.readthedocs.io/en/latest/user_guide/model_registration/quick_start.html",
+    )
     def prepare(
         self,
         target_dir=None,