alita-sdk 0.3.238__tar.gz → 0.3.239__tar.gz

{alita_sdk-0.3.238 → alita_sdk-0.3.239}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.238
+Version: 0.3.239
 Summary: SDK for building langchain agents using resources from Alita
 Author-email: Artem Rozumenko <artyom.rozumenko@gmail.com>, Mikalai Biazruchka <mikalai_biazruchka@epam.com>, Roman Mitusov <roman_mitusov@epam.com>, Ivan Krakhmaliuk <lifedjik@gmail.com>, Artem Dubrovskiy <ad13box@gmail.com>
 License-Expression: Apache-2.0

{alita_sdk-0.3.238 → alita_sdk-0.3.239}/alita_sdk/runtime/clients/client.py

@@ -156,6 +156,20 @@ class AlitaClient:
             return resp.json()
         return []
 
+    def get_embeddings(self, embedding_model: str) -> OpenAIEmbeddings:
+        """
+        Get an instance of OpenAIEmbeddings configured with the project ID and auth token.
+
+        Returns:
+            An instance of OpenAIEmbeddings configured for the project.
+        """
+        return OpenAIEmbeddings(
+            base_url=f"{self.base_url}{self.llm_path}",
+            model=embedding_model,
+            api_key=self.auth_token,
+            openai_organization=str(self.project_id),
+        )
+
     def get_llm(self, model_name: str, model_config: dict) -> ChatOpenAI:
         """
         Get a ChatOpenAI model instance based on the model name and configuration.

{alita_sdk-0.3.238 → alita_sdk-0.3.239}/alita_sdk/runtime/langchain/interfaces/llm_processor.py

@@ -29,7 +29,6 @@ from langchain.prompts import PromptTemplate  # pylint: disable=E0401
 
 from langchain.schema import HumanMessage, SystemMessage
 
-from ...clients import AlitaClient
 from ...llms.preloaded import PreloadedEmbeddings, PreloadedChatModel  # pylint: disable=E0401
 from ..retrievers.AlitaRetriever import AlitaRetriever
 from ..tools.log import print_log
@@ -60,29 +59,29 @@ def get_model(model_type: str, model_params: dict):
     raise RuntimeError(f"Unknown model type: {model_type}")
 
 
-def get_embeddings(elitea_client: AlitaClient, embeddings_model: str, embeddings_params: dict):
+# TODO: review usage of this function
+def get_embeddings(embeddings_model: str, embeddings_params: dict):
     """ Get *Embeddings """
-    # if embeddings_model is None:
-    #     return None
-    # if "." in embeddings_model:
-    #     target_pkg, target_name = embeddings_model.rsplit(".", 1)
-    #     target_cls = getattr(
-    #         importlib.import_module(target_pkg),
-    #         target_name
-    #     )
-    #     return target_cls(**embeddings_params)
-    # if embeddings_model == "PreloadedEmbeddings":
-    #     return PreloadedEmbeddings(**embeddings_params)
-    # if embeddings_model == "Chroma":
-    #     from langchain_chroma import Chroma
-    #     return Chroma(**embeddings_params)
-    # if embeddings_model in embeddings:
-    #     model = getattr(
-    #         __import__("langchain_community.embeddings", fromlist=[embeddings_model]),
-    #         embeddings_model
-    #     )
-    #     return model(**embeddings_params)
-    elitea_client.get_llm(embeddings_model, embeddings_params)
+    if embeddings_model is None:
+        return None
+    if "." in embeddings_model:
+        target_pkg, target_name = embeddings_model.rsplit(".", 1)
+        target_cls = getattr(
+            importlib.import_module(target_pkg),
+            target_name
+        )
+        return target_cls(**embeddings_params)
+    if embeddings_model == "PreloadedEmbeddings":
+        return PreloadedEmbeddings(**embeddings_params)
+    if embeddings_model == "Chroma":
+        from langchain_chroma import Chroma
+        return Chroma(**embeddings_params)
+    if embeddings_model in embeddings:
+        model = getattr(
+            __import__("langchain_community.embeddings", fromlist=[embeddings_model]),
+            embeddings_model
+        )
+        return model(**embeddings_params)
     raise RuntimeError(f"Unknown Embedding type: {embeddings_model}")
 
 

{alita_sdk-0.3.238 → alita_sdk-0.3.239}/alita_sdk/tools/elitea_base.py

@@ -9,7 +9,7 @@ from langchain_core.documents import Document
 from langchain_core.tools import ToolException
 from pydantic import BaseModel, create_model, Field, SecretStr
 
-from alita_sdk.runtime.langchain.interfaces.llm_processor import get_embeddings
+# from alita_sdk.runtime.langchain.interfaces.llm_processor import get_embeddings
 from .chunkers import markdown_chunker
 from .utils import TOOLKIT_SPLITTER
 from .vector_adapters.VectorStoreAdapter import VectorStoreAdapterFactory
@@ -222,11 +222,14 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
     embedding_model_params: Optional[Dict[str, Any]] = {"model_name": "sentence-transformers/all-MiniLM-L6-v2"}
     vectorstore_type: Optional[str] = "PGVector"
     _vector_store: Optional[Any] = None
+    _embedding: Optional[Any] = None
     alita: Any = None # Elitea client, if available
 
     def __init__(self, **kwargs):
         super().__init__(**kwargs)
         self._adapter = VectorStoreAdapterFactory.create_adapter(self.vectorstore_type)
+        if self.alita and self.embedding_model:
+            self._embedding = self.alita.get_embeddings(self.embedding_model)
 
     def _index_tool_params(self, **kwargs) -> dict[str, tuple[type, Field]]:
         """
@@ -239,9 +242,7 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
         return markdown_chunker
 
     def _get_dependencies_chunker_config(self, document: Optional[Document] = None):
-        embedding = get_embeddings(self.embedding_model, self.embedding_model_params)
-        #
-        return {'embedding': embedding, 'llm': self.llm}
+        return {'embedding': self._embedding, 'llm': self.llm}
 
     def _base_loader(self, **kwargs) -> Generator[Document, None, None]:
         """ Loads documents from a source, processes them,
@@ -275,7 +276,6 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
     def index_data(self, **kwargs):
         from alita_sdk.tools.chunkers import __confluence_chunkers__ as chunkers, __confluence_models__ as models
         docs = self._base_loader(**kwargs)
-        embedding = get_embeddings(self.alita, self.embedding_model, self.embedding_model_params)
         chunking_tool = kwargs.get("chunking_tool")
         if chunking_tool:
             # Resolve chunker from the provided chunking_tool name
@@ -284,7 +284,7 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
             base_chunking_config = kwargs.get("chunking_config", {})
             config_model = models.get(chunking_tool)
             # Set required fields that should come from the instance (and Fallback for chunkers without models)
-            base_chunking_config['embedding'] = embedding
+            base_chunking_config['embedding'] = self._embedding
             base_chunking_config['llm'] = self.llm
             #
             if config_model:
@@ -301,7 +301,7 @@ class BaseVectorStoreToolApiWrapper(BaseToolApiWrapper):
         collection_suffix = kwargs.get("collection_suffix")
         progress_step = kwargs.get("progress_step")
         clean_index = kwargs.get("clean_index")
-        vs = self._init_vector_store(embeddings=embedding)
+        vs = self._init_vector_store(embeddings=self._embedding)
         #
         return vs.index_documents(docs, collection_suffix=collection_suffix, progress_step=progress_step, clean_index=clean_index)
 

{alita_sdk-0.3.238 → alita_sdk-0.3.239}/alita_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.238
+Version: 0.3.239
 Summary: SDK for building langchain agents using resources from Alita
 Author-email: Artem Rozumenko <artyom.rozumenko@gmail.com>, Mikalai Biazruchka <mikalai_biazruchka@epam.com>, Roman Mitusov <roman_mitusov@epam.com>, Ivan Krakhmaliuk <lifedjik@gmail.com>, Artem Dubrovskiy <ad13box@gmail.com>
 License-Expression: Apache-2.0

{alita_sdk-0.3.238 → alita_sdk-0.3.239}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "alita_sdk"
-version = "0.3.238"
+version = "0.3.239"
 description = "SDK for building langchain agents using resources from Alita"
 readme = "README.md"
 requires-python = ">=3.10"