oracle-ads 2.12.7__py3-none-any.whl → 2.12.9__py3-none-any.whl

ads/aqua/modeldeployment/deployment.py

@@ -110,6 +110,8 @@ class AquaDeploymentApp(AquaApp):
         private_endpoint_id: Optional[str] = None,
         container_image_uri: Optional[None] = None,
         cmd_var: List[str] = None,
+        freeform_tags: Optional[dict] = None,
+        defined_tags: Optional[dict] = None,
     ) -> "AquaDeployment":
         """
         Creates a new Aqua deployment
@@ -163,6 +165,10 @@ class AquaDeploymentApp(AquaApp):
             Required parameter for BYOC based deployments if this parameter was not set during model registration.
         cmd_var: List[str]
             The cmd of model deployment container runtime.
+        freeform_tags: dict
+            Freeform tags for the model deployment
+        defined_tags: dict
+            Defined tags for the model deployment
         Returns
         -------
         AquaDeployment
@@ -172,7 +178,11 @@ class AquaDeploymentApp(AquaApp):
         # TODO validate if the service model has no artifact and if it requires import step before deployment.
         # Create a model catalog entry in the user compartment
         aqua_model = AquaModelApp().create(
-            model_id=model_id, compartment_id=compartment_id, project_id=project_id
+            model_id=model_id,
+            compartment_id=compartment_id,
+            project_id=project_id,
+            freeform_tags=freeform_tags,
+            defined_tags=defined_tags,
         )
 
         tags = {}
@@ -185,7 +195,7 @@ class AquaDeploymentApp(AquaApp):
                 tags[tag] = aqua_model.freeform_tags[tag]
 
         tags.update({Tags.AQUA_MODEL_NAME_TAG: aqua_model.display_name})
-        tags.update({Tags.TASK: aqua_model.freeform_tags.get(Tags.TASK, None)})
+        tags.update({Tags.TASK: aqua_model.freeform_tags.get(Tags.TASK, UNKNOWN)})
 
         # Set up info to get deployment config
         config_source_id = model_id
@@ -418,12 +428,14 @@ class AquaDeploymentApp(AquaApp):
         if cmd_var:
             container_runtime.with_cmd(cmd_var)
 
+        tags = {**tags, **(freeform_tags or {})}
         # configure model deployment and deploy model on container runtime
         deployment = (
             ModelDeployment()
             .with_display_name(display_name)
             .with_description(description)
             .with_freeform_tags(**tags)
+            .with_defined_tags(**(defined_tags or {}))
             .with_infrastructure(infrastructure)
             .with_runtime(container_runtime)
         ).deploy(wait_for_completion=False)
@@ -533,16 +545,22 @@ class AquaDeploymentApp(AquaApp):
         return results
 
     @telemetry(entry_point="plugin=deployment&action=delete", name="aqua")
-    def delete(self,model_deployment_id:str):
-        return self.ds_client.delete_model_deployment(model_deployment_id=model_deployment_id).data
+    def delete(self, model_deployment_id: str):
+        return self.ds_client.delete_model_deployment(
+            model_deployment_id=model_deployment_id
+        ).data
 
-    @telemetry(entry_point="plugin=deployment&action=deactivate",name="aqua")
-    def deactivate(self,model_deployment_id:str):
-        return self.ds_client.deactivate_model_deployment(model_deployment_id=model_deployment_id).data
+    @telemetry(entry_point="plugin=deployment&action=deactivate", name="aqua")
+    def deactivate(self, model_deployment_id: str):
+        return self.ds_client.deactivate_model_deployment(
+            model_deployment_id=model_deployment_id
+        ).data
 
-    @telemetry(entry_point="plugin=deployment&action=activate",name="aqua")
-    def activate(self,model_deployment_id:str):
-        return self.ds_client.activate_model_deployment(model_deployment_id=model_deployment_id).data
+    @telemetry(entry_point="plugin=deployment&action=activate", name="aqua")
+    def activate(self, model_deployment_id: str):
+        return self.ds_client.activate_model_deployment(
+            model_deployment_id=model_deployment_id
+        ).data
 
     @telemetry(entry_point="plugin=deployment&action=get", name="aqua")
     def get(self, model_deployment_id: str, **kwargs) -> "AquaDeploymentDetail":

ads/aqua/modeldeployment/entities.py

@@ -98,9 +98,12 @@ class AquaDeployment(DataClassSerializable):
             ),
         )
 
-        freeform_tags = oci_model_deployment.freeform_tags or UNKNOWN_DICT
-        aqua_service_model_tag = freeform_tags.get(Tags.AQUA_SERVICE_MODEL_TAG, None)
-        aqua_model_name = freeform_tags.get(Tags.AQUA_MODEL_NAME_TAG, UNKNOWN)
+        tags = {}
+        tags.update(oci_model_deployment.freeform_tags or UNKNOWN_DICT)
+        tags.update(oci_model_deployment.defined_tags or UNKNOWN_DICT)
+
+        aqua_service_model_tag = tags.get(Tags.AQUA_SERVICE_MODEL_TAG, None)
+        aqua_model_name = tags.get(Tags.AQUA_MODEL_NAME_TAG, UNKNOWN)
         private_endpoint_id = getattr(
             instance_configuration, "private_endpoint_id", UNKNOWN
         )
@@ -125,7 +128,7 @@ class AquaDeployment(DataClassSerializable):
                 ocid=oci_model_deployment.id,
                 region=region,
             ),
-            tags=freeform_tags,
+            tags=tags,
             environment_variables=environment_variables,
             cmd=cmd,
         )

ads/aqua/ui.py

@@ -481,12 +481,12 @@ class AquaUIApp(AquaApp):
 
     @telemetry(entry_point="plugin=ui&action=list_job_shapes", name="aqua")
     def list_job_shapes(self, **kwargs) -> list:
-        """Lists all availiable job shapes for the specified compartment.
+        """Lists all available job shapes for the specified compartment.
 
         Parameters
         ----------
         **kwargs
-            Addtional arguments, such as `compartment_id`,
+            Additional arguments, such as `compartment_id`,
             for `list_job_shapes <https://docs.oracle.com/en-us/iaas/tools/python/2.122.0/api/data_science/client/oci.data_science.DataScienceClient.html#oci.data_science.DataScienceClient.list_job_shapes>`_
 
         Returns
@@ -500,6 +500,28 @@ class AquaUIApp(AquaApp):
         ).data
         return sanitize_response(oci_client=self.ds_client, response=res)
 
+    @telemetry(entry_point="plugin=ui&action=list_model_deployment_shapes", name="aqua")
+    def list_model_deployment_shapes(self, **kwargs) -> list:
+        """Lists all available shapes for model deployment in the specified compartment.
+
+        Parameters
+        ----------
+        **kwargs
+            Additional arguments, such as `compartment_id`,
+            for `list_model_deployment_shapes <https://docs.oracle.com/en-us/iaas/api/#/en/data-science/20190101/ModelDeploymentShapeSummary/ListModelDeploymentShapes>`_
+
+        Returns
+        -------
+            str has json representation of `oci.data_science.models.ModelDeploymentShapeSummary`."""
+        compartment_id = kwargs.pop("compartment_id", COMPARTMENT_OCID)
+        logger.info(
+            f"Loading model deployment shape summary from compartment: {compartment_id}"
+        )
+        res = self.ds_client.list_model_deployment_shapes(
+            compartment_id=compartment_id, **kwargs
+        ).data
+        return sanitize_response(oci_client=self.ds_client, response=res)
+
     @telemetry(entry_point="plugin=ui&action=list_vcn", name="aqua")
     def list_vcn(self, **kwargs) -> list:
         """Lists the virtual cloud networks (VCNs) in the specified compartment.

ads/llm/guardrails/base.py

@@ -1,17 +1,16 @@
 #!/usr/bin/env python
-# -*- coding: utf-8 -*--
 
-# Copyright (c) 2023 Oracle and/or its affiliates.
+# Copyright (c) 2024 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
 
 import datetime
 import functools
-import operator
 import importlib.util
+import operator
 import sys
+from typing import Any, List, Optional, Union
 
-from typing import Any, List, Dict, Tuple
 from langchain.schema.prompt import PromptValue
 from langchain.tools.base import BaseTool, ToolException
 from pydantic import BaseModel, model_validator
@@ -207,7 +206,9 @@ class Guardrail(BaseTool):
             return input.to_string()
         return str(input)
 
-    def _to_args_and_kwargs(self, tool_input: Any) -> Tuple[Tuple, Dict]:
+    def _to_args_and_kwargs(
+        self, tool_input: Union[str, dict], tool_call_id: Optional[str]
+    ) -> tuple[tuple, dict]:
         if isinstance(tool_input, dict):
             return (), tool_input
         else:

ads/llm/langchain/plugins/chat_models/oci_data_science.py

@@ -1,7 +1,6 @@
 #!/usr/bin/env python
-# -*- coding: utf-8 -*--
 
-# Copyright (c) 2023 Oracle and/or its affiliates.
+# Copyright (c) 2024 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 """Chat model for OCI data science model deployment endpoint."""
 
@@ -50,6 +49,7 @@ from ads.llm.langchain.plugins.llms.oci_data_science_model_deployment_endpoint i
 )
 
 logger = logging.getLogger(__name__)
+DEFAULT_INFERENCE_ENDPOINT_CHAT = "/v1/chat/completions"
 
 
 def _is_pydantic_class(obj: Any) -> bool:
@@ -93,6 +93,8 @@ class ChatOCIModelDeployment(BaseChatModel, BaseOCIModelDeployment):
     Key init args — client params:
         auth: dict
             ADS auth dictionary for OCI authentication.
+        default_headers: Optional[Dict]
+            The headers to be added to the Model Deployment request.
 
     Instantiate:
         .. code-block:: python
@@ -109,6 +111,10 @@ class ChatOCIModelDeployment(BaseChatModel, BaseOCIModelDeployment):
                     "temperature": 0.2,
                     # other model parameters ...
                 },
+                default_headers={
+                    "route": "/v1/chat/completions",
+                    # other request headers ...
+                },
             )
 
     Invocation:
@@ -291,6 +297,25 @@ class ChatOCIModelDeployment(BaseChatModel, BaseOCIModelDeployment):
             "stream": self.streaming,
         }
 
+    def _headers(
+        self, is_async: Optional[bool] = False, body: Optional[dict] = None
+    ) -> Dict:
+        """Construct and return the headers for a request.
+
+        Args:
+            is_async (bool, optional): Indicates if the request is asynchronous.
+                Defaults to `False`.
+            body (optional): The request body to be included in the headers if
+                the request is asynchronous.
+
+        Returns:
+            Dict: A dictionary containing the appropriate headers for the request.
+        """
+        return {
+            "route": DEFAULT_INFERENCE_ENDPOINT_CHAT,
+            **super()._headers(is_async=is_async, body=body),
+        }
+
     def _generate(
         self,
         messages: List[BaseMessage],
@@ -704,7 +729,7 @@ class ChatOCIModelDeployment(BaseChatModel, BaseOCIModelDeployment):
 
         for choice in choices:
             message = _convert_dict_to_message(choice["message"])
-            generation_info = dict(finish_reason=choice.get("finish_reason"))
+            generation_info = {"finish_reason": choice.get("finish_reason")}
             if "logprobs" in choice:
                 generation_info["logprobs"] = choice["logprobs"]
 
@@ -794,7 +819,7 @@ class ChatOCIModelDeploymentVLLM(ChatOCIModelDeployment):
     """Number of most likely tokens to consider at each step."""
 
     min_p: Optional[float] = 0.0
-    """Float that represents the minimum probability for a token to be considered. 
+    """Float that represents the minimum probability for a token to be considered.
     Must be in [0,1]. 0 to disable this."""
 
     repetition_penalty: Optional[float] = 1.0
@@ -818,7 +843,7 @@ class ChatOCIModelDeploymentVLLM(ChatOCIModelDeployment):
     the EOS token is generated."""
 
     min_tokens: Optional[int] = 0
-    """Minimum number of tokens to generate per output sequence before 
+    """Minimum number of tokens to generate per output sequence before
     EOS or stop_token_ids can be generated"""
 
     stop_token_ids: Optional[List[int]] = None
@@ -836,7 +861,7 @@ class ChatOCIModelDeploymentVLLM(ChatOCIModelDeployment):
     tool_choice: Optional[str] = None
     """Whether to use tool calling.
     Defaults to None, tool calling is disabled.
-    Tool calling requires model support and the vLLM to be configured 
+    Tool calling requires model support and the vLLM to be configured
     with `--tool-call-parser`.
     Set this to `auto` for the model to make tool calls automatically.
     Set this to `required` to force the model to always call one or more tools.
@@ -956,9 +981,9 @@ class ChatOCIModelDeploymentTGI(ChatOCIModelDeployment):
     """Total probability mass of tokens to consider at each step."""
 
     top_logprobs: Optional[int] = None
-    """An integer between 0 and 5 specifying the number of most 
-    likely tokens to return at each token position, each with an 
-    associated log probability. logprobs must be set to true if 
+    """An integer between 0 and 5 specifying the number of most
+    likely tokens to return at each token position, each with an
+    associated log probability. logprobs must be set to true if
     this parameter is used."""
 
     @property

ads/llm/langchain/plugins/llms/oci_data_science_model_deployment_endpoint.py

@@ -1,7 +1,6 @@
 #!/usr/bin/env python
-# -*- coding: utf-8 -*--
 
-# Copyright (c) 2023 Oracle and/or its affiliates.
+# Copyright (c) 2024 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
 
@@ -24,6 +23,7 @@ from typing import (
 
 import aiohttp
 import requests
+from langchain_community.utilities.requests import Requests
 from langchain_core.callbacks import (
     AsyncCallbackManagerForLLMRun,
     CallbackManagerForLLMRun,
@@ -34,14 +34,13 @@ from langchain_core.outputs import Generation, GenerationChunk, LLMResult
 from langchain_core.utils import get_from_dict_or_env
 from pydantic import Field, model_validator
 
-from langchain_community.utilities.requests import Requests
-
 logger = logging.getLogger(__name__)
 
 
 DEFAULT_TIME_OUT = 300
 DEFAULT_CONTENT_TYPE_JSON = "application/json"
 DEFAULT_MODEL_NAME = "odsc-llm"
+DEFAULT_INFERENCE_ENDPOINT = "/v1/completions"
 
 
 class TokenExpiredError(Exception):
@@ -86,6 +85,9 @@ class BaseOCIModelDeployment(Serializable):
     max_retries: int = 3
     """Maximum number of retries to make when generating."""
 
+    default_headers: Optional[Dict[str, Any]] = None
+    """The headers to be added to the Model Deployment request."""
+
     @model_validator(mode="before")
     @classmethod
     def validate_environment(cls, values: Dict) -> Dict:
@@ -101,7 +103,7 @@ class BaseOCIModelDeployment(Serializable):
                 "Please install it with `pip install oracle_ads`."
             ) from ex
 
-        if not values.get("auth", None):
+        if not values.get("auth"):
             values["auth"] = ads.common.auth.default_signer()
 
         values["endpoint"] = get_from_dict_or_env(
@@ -125,12 +127,12 @@ class BaseOCIModelDeployment(Serializable):
         Returns:
             Dict: A dictionary containing the appropriate headers for the request.
         """
+        headers = self.default_headers or {}
         if is_async:
             signer = self.auth["signer"]
             _req = requests.Request("POST", self.endpoint, json=body)
             req = _req.prepare()
             req = signer(req)
-            headers = {}
             for key, value in req.headers.items():
                 headers[key] = value
 
@@ -140,7 +142,7 @@ class BaseOCIModelDeployment(Serializable):
                 )
             return headers
 
-        return (
+        headers.update(
             {
                 "Content-Type": DEFAULT_CONTENT_TYPE_JSON,
                 "enable-streaming": "true",
@@ -152,6 +154,8 @@ class BaseOCIModelDeployment(Serializable):
             }
         )
 
+        return headers
+
     def completion_with_retry(
         self, run_manager: Optional[CallbackManagerForLLMRun] = None, **kwargs: Any
     ) -> Any:
@@ -357,7 +361,7 @@ class BaseOCIModelDeployment(Serializable):
             self.auth["signer"].refresh_security_token()
             return True
         return False
-    
+
     @classmethod
     def is_lc_serializable(cls) -> bool:
         """Return whether this model can be serialized by LangChain."""
@@ -388,6 +392,10 @@ class OCIModelDeploymentLLM(BaseLLM, BaseOCIModelDeployment):
                 model="odsc-llm",
                 streaming=True,
                 model_kwargs={"frequency_penalty": 1.0},
+                headers={
+                    "route": "/v1/completions",
+                    # other request headers ...
+                }
             )
             llm.invoke("tell me a joke.")
 
@@ -477,6 +485,25 @@ class OCIModelDeploymentLLM(BaseLLM, BaseOCIModelDeployment):
             **self._default_params,
         }
 
+    def _headers(
+        self, is_async: Optional[bool] = False, body: Optional[dict] = None
+    ) -> Dict:
+        """Construct and return the headers for a request.
+
+        Args:
+            is_async (bool, optional): Indicates if the request is asynchronous.
+                Defaults to `False`.
+            body (optional): The request body to be included in the headers if
+                the request is asynchronous.
+
+        Returns:
+            Dict: A dictionary containing the appropriate headers for the request.
+        """
+        return {
+            "route": DEFAULT_INFERENCE_ENDPOINT,
+            **super()._headers(is_async=is_async, body=body),
+        }
+
     def _generate(
         self,
         prompts: List[str],
@@ -712,9 +739,9 @@ class OCIModelDeploymentLLM(BaseLLM, BaseOCIModelDeployment):
     def _generate_info(self, choice: dict) -> Any:
         """Extracts generation info from the response."""
         gen_info = {}
-        finish_reason = choice.get("finish_reason", None)
-        logprobs = choice.get("logprobs", None)
-        index = choice.get("index", None)
+        finish_reason = choice.get("finish_reason")
+        logprobs = choice.get("logprobs")
+        index = choice.get("index")
         if finish_reason:
             gen_info.update({"finish_reason": finish_reason})
         if logprobs is not None:

ads/opctl/operator/common/utils.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# -*- coding: utf-8 -*--
 
 # Copyright (c) 2023, 2024 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
@@ -18,7 +17,6 @@ import yaml
 from cerberus import Validator
 
 from ads.opctl import logger, utils
-from ads.opctl.operator import __operators__
 
 CONTAINER_NETWORK = "CONTAINER_NETWORK"
 
@@ -26,7 +24,11 @@ CONTAINER_NETWORK = "CONTAINER_NETWORK"
 class OperatorValidator(Validator):
     """The custom validator class."""
 
-    pass
+    def validate(self, obj_dict, **kwargs):
+        # Model should be case insensitive
+        if "model" in obj_dict["spec"]:
+            obj_dict["spec"]["model"] = str(obj_dict["spec"]["model"]).lower()
+        return super().validate(obj_dict, **kwargs)
 
 
 def create_output_folder(name):
@@ -34,7 +36,7 @@ def create_output_folder(name):
     protocol = fsspec.utils.get_protocol(output_folder)
     storage_options = {}
     if protocol != "file":
-        storage_options = auth or default_signer()
+        storage_options = default_signer()
 
     fs = fsspec.filesystem(protocol, **storage_options)
     name_suffix = 1

ads/opctl/operator/lowcode/anomaly/model/base_model.py

@@ -166,9 +166,8 @@ class AnomalyOperatorBaseModel(ABC):
         yaml_appendix = rc.Yaml(self.config.to_dict())
         summary = rc.Block(
             rc.Group(
-                rc.Text(
-                    f"You selected the **`{self.spec.model}`** model.\n{model_description.text}\n"
-                ),
+                rc.Text(f"You selected the **`{self.spec.model}`** model.\n"),
+                model_description,
                 rc.Text(
                     "Based on your dataset, you could have also selected "
                     f"any of the models: `{'`, `'.join(SupportedModels.keys() if self.spec.datetime_column else NonTimeADSupportedModels.keys())}`."

ads/opctl/operator/lowcode/anomaly/model/factory.py

@@ -26,9 +26,9 @@ class UnSupportedModelError(Exception):
 
     def __init__(self, operator_config: AnomalyOperatorConfig, model_type: str):
         supported_models = (
-            SupportedModels.values
+            SupportedModels.values()
             if operator_config.spec.datetime_column
-            else NonTimeADSupportedModels.values
+            else NonTimeADSupportedModels.values()
         )
         message = (
             f"Model: `{model_type}` is not supported. "

ads/opctl/operator/lowcode/common/transformations.py

@@ -1,18 +1,19 @@
 #!/usr/bin/env python
-# -*- coding: utf-8 -*--
 
-# Copyright (c) 2023 Oracle and/or its affiliates.
+# Copyright (c) 2023, 2024 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
+from abc import ABC
+
+import pandas as pd
+
 from ads.opctl import logger
+from ads.opctl.operator.lowcode.common.const import DataColumns
 from ads.opctl.operator.lowcode.common.errors import (
-    InvalidParameterError,
     DataMismatchError,
+    InvalidParameterError,
 )
-from ads.opctl.operator.lowcode.common.const import DataColumns
 from ads.opctl.operator.lowcode.common.utils import merge_category_columns
-import pandas as pd
-from abc import ABC
 
 
 class Transformations(ABC):
@@ -58,6 +59,7 @@ class Transformations(ABC):
 
         """
         clean_df = self._remove_trailing_whitespace(data)
+        # clean_df = self._normalize_column_names(clean_df)
         if self.name == "historical_data":
             self._check_historical_dataset(clean_df)
         clean_df = self._set_series_id_column(clean_df)
@@ -95,8 +97,11 @@ class Transformations(ABC):
     def _remove_trailing_whitespace(self, df):
         return df.apply(lambda x: x.str.strip() if x.dtype == "object" else x)
 
+    # def _normalize_column_names(self, df):
+    #     return df.rename(columns=lambda x: re.sub("[^A-Za-z0-9_]+", "", x))
+
     def _set_series_id_column(self, df):
-        self._target_category_columns_map = dict()
+        self._target_category_columns_map = {}
         if not self.target_category_columns:
             df[DataColumns.Series] = "Series 1"
             self.has_artificial_series = True
@@ -125,10 +130,10 @@ class Transformations(ABC):
             df[self.dt_column_name] = pd.to_datetime(
                 df[self.dt_column_name], format=self.dt_column_format
             )
-        except:
+        except Exception as ee:
             raise InvalidParameterError(
                 f"Unable to determine the datetime type for column: {self.dt_column_name} in dataset: {self.name}. Please specify the format explicitly. (For example adding 'format: %d/%m/%Y' underneath 'name: {self.dt_column_name}' in the datetime_column section of the yaml file if you haven't already. For reference, here is the first datetime given: {df[self.dt_column_name].values[0]}"
-            )
+            ) from ee
         return df
 
     def _set_multi_index(self, df):
@@ -242,7 +247,6 @@ class Transformations(ABC):
                 "Class": "A",
                 "Num": 2
             },
-            
         }
     """