langchain-google-genai 0.0.2__tar.gz → 0.0.4__tar.gz

{langchain_google_genai-0.0.2 → langchain_google_genai-0.0.4}/PKG-INFO

@@ -1,14 +1,18 @@
 Metadata-Version: 2.1
 Name: langchain-google-genai
-Version: 0.0.2
+Version: 0.0.4
 Summary: An integration package connecting Google's genai package and LangChain
+Home-page: https://github.com/langchain-ai/langchain/blob/master/libs/partners/google-genai
 Requires-Python: >=3.9,<4.0
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
+Provides-Extra: images
 Requires-Dist: google-generativeai (>=0.3.1,<0.4.0)
 Requires-Dist: langchain-core (>=0.1,<0.2)
+Requires-Dist: pillow (>=10.1.0,<11.0.0) ; extra == "images"
+Project-URL: Repository, https://github.com/langchain-ai/langchain/blob/master/libs/partners/google-genai
 Description-Content-Type: text/markdown
 
 # langchain-google-genai
@@ -17,10 +21,17 @@ This package contains the LangChain integrations for Gemini through their genera
 
 ## Installation
 
-```python
+```bash
 pip install -U langchain-google-genai
 ```
 
+### Image utilities
+To use image utility methods, like loading images from GCS urls, install with extras group 'images':
+
+```bash
+pip install -e "langchain-google-genai[images]"
+```
+
 ## Chat Models
 
 This package contains the `ChatGoogleGenerativeAI` class, which is the recommended way to interface with the Google Gemini series of models.
@@ -70,3 +81,15 @@ The value of `image_url` can be any of the following:
 - A base64 encoded image (e.g., `data:image/png;base64,abcd124`)
 - A PIL image
 
+
+
+## Embeddings
+
+This package also adds support for google's embeddings models.
+
+```
+from langchain_google_genai import GoogleGenerativeAIEmbeddings
+
+embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+embeddings.embed_query("hello, world!")
+```

{langchain_google_genai-0.0.2 → langchain_google_genai-0.0.4}/README.md

@@ -4,10 +4,17 @@ This package contains the LangChain integrations for Gemini through their genera
 
 ## Installation
 
-```python
+```bash
 pip install -U langchain-google-genai
 ```
 
+### Image utilities
+To use image utility methods, like loading images from GCS urls, install with extras group 'images':
+
+```bash
+pip install -e "langchain-google-genai[images]"
+```
+
 ## Chat Models
 
 This package contains the `ChatGoogleGenerativeAI` class, which is the recommended way to interface with the Google Gemini series of models.
@@ -56,3 +63,16 @@ The value of `image_url` can be any of the following:
 - A local file path
 - A base64 encoded image (e.g., `data:image/png;base64,abcd124`)
 - A PIL image
+
+
+
+## Embeddings
+
+This package also adds support for google's embeddings models.
+
+```
+from langchain_google_genai import GoogleGenerativeAIEmbeddings
+
+embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+embeddings.embed_query("hello, world!")
+```

langchain_google_genai-0.0.4/langchain_google_genai/__init__.py

@@ -0,0 +1,46 @@
+"""**LangChain Google Generative AI Integration**
+
+This module integrates Google's Generative AI models, specifically the Gemini series, with the LangChain framework. It provides classes for interacting with chat models and generating embeddings, leveraging Google's advanced AI capabilities.
+
+**Chat Models**
+
+The `ChatGoogleGenerativeAI` class is the primary interface for interacting with Google's Gemini chat models. It allows users to send and receive messages using a specified Gemini model, suitable for various conversational AI applications.
+
+**Embeddings**
+
+The `GoogleGenerativeAIEmbeddings` class provides functionalities to generate embeddings using Google's models.
+These embeddings can be used for a range of NLP tasks, including semantic analysis, similarity comparisons, and more.
+
+**Installation**
+
+To install the package, use pip:
+
+```python
+pip install -U langchain-google-genai
+```
+## Using Chat Models
+
+After setting up your environment with the required API key, you can interact with the Google Gemini models.
+
+```python
+from langchain_google_genai import ChatGoogleGenerativeAI
+
+llm = ChatGoogleGenerativeAI(model="gemini-pro")
+llm.invoke("Sing a ballad of LangChain.")
+```
+
+## Embedding Generation
+
+The package also supports creating embeddings with Google's models, useful for textual similarity and other NLP applications.
+
+```python
+from langchain_google_genai import GoogleGenerativeAIEmbeddings
+
+embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+embeddings.embed_query("hello, world!")
+```
+"""  # noqa: E501
+from langchain_google_genai.chat_models import ChatGoogleGenerativeAI
+from langchain_google_genai.embeddings import GoogleGenerativeAIEmbeddings
+
+__all__ = ["ChatGoogleGenerativeAI", "GoogleGenerativeAIEmbeddings"]

langchain_google_genai-0.0.4/langchain_google_genai/_common.py

@@ -0,0 +1,4 @@
+class GoogleGenerativeAIError(Exception):
+    """
+    Custom exception class for errors associated with the `Google GenAI` API.
+    """

{langchain_google_genai-0.0.2 → langchain_google_genai-0.0.4}/langchain_google_genai/chat_models.py

@@ -5,7 +5,6 @@ import logging
 import os
 from io import BytesIO
 from typing import (
-    TYPE_CHECKING,
     Any,
     AsyncIterator,
     Callable,
@@ -22,6 +21,8 @@ from typing import (
 )
 from urllib.parse import urlparse
 
+# TODO: remove ignore once the google package is published with types
+import google.generativeai as genai  # type: ignore[import]
 import requests
 from langchain_core.callbacks.manager import (
     AsyncCallbackManagerForLLMRun,
@@ -38,7 +39,7 @@ from langchain_core.messages import (
     HumanMessageChunk,
 )
 from langchain_core.outputs import ChatGeneration, ChatGenerationChunk, ChatResult
-from langchain_core.pydantic_v1 import Field, root_validator
+from langchain_core.pydantic_v1 import Field, SecretStr, root_validator
 from langchain_core.utils import get_from_dict_or_env
 from tenacity import (
     before_sleep_log,
@@ -48,11 +49,8 @@ from tenacity import (
     wait_exponential,
 )
 
-logger = logging.getLogger(__name__)
+from langchain_google_genai._common import GoogleGenerativeAIError
 
-if TYPE_CHECKING:
-    # TODO: remove ignore once the google package is published with types
-    import google.generativeai as genai  # type: ignore[import]
 IMAGE_TYPES: Tuple = ()
 try:
     import PIL
@@ -63,8 +61,10 @@ except ImportError:
     PIL = None  # type: ignore
     Image = None  # type: ignore
 
+logger = logging.getLogger(__name__)
+
 
-class ChatGoogleGenerativeAIError(Exception):
+class ChatGoogleGenerativeAIError(GoogleGenerativeAIError):
     """
     Custom exception class for errors associated with the `Google GenAI` API.
 
@@ -106,7 +106,7 @@ def _create_retry_decorator() -> Callable[[Any], Any]:
     )
 
 
-def chat_with_retry(*, generation_method: Callable, **kwargs: Any) -> Any:
+def _chat_with_retry(*, generation_method: Callable, **kwargs: Any) -> Any:
     """
     Executes a chat generation method with retry logic using tenacity.
 
@@ -139,7 +139,7 @@ def chat_with_retry(*, generation_method: Callable, **kwargs: Any) -> Any:
     return _chat_with_retry(**kwargs)
 
 
-async def achat_with_retry(*, generation_method: Callable, **kwargs: Any) -> Any:
+async def _achat_with_retry(*, generation_method: Callable, **kwargs: Any) -> Any:
     """
     Executes a chat generation method with retry logic using tenacity.
 
@@ -269,8 +269,6 @@ def _convert_to_parts(
     content: Sequence[Union[str, dict]],
 ) -> List[genai.types.PartType]:
     """Converts a list of LangChain messages into a google parts."""
-    import google.generativeai as genai
-
     parts = []
     for part in content:
         if isinstance(part, str):
@@ -410,8 +408,7 @@ def _response_to_result(
 class ChatGoogleGenerativeAI(BaseChatModel):
     """`Google Generative AI` Chat models API.
 
-    To use you must have the google.generativeai Python package installed and
-    either:
+    To use, you must have either:
 
         1. The ``GOOGLE_API_KEY``` environment variable set with your API key, or
         2. Pass your API key using the google_api_key kwarg to the ChatGoogle
@@ -435,7 +432,7 @@ Supported examples:
     max_output_tokens: int = Field(default=None, description="Max output tokens")
 
     client: Any  #: :meta private:
-    google_api_key: Optional[str] = None
+    google_api_key: Optional[SecretStr] = None
     temperature: Optional[float] = None
     """Run inference with this temperature. Must by in the closed
        interval [0.0, 1.0]."""
@@ -487,17 +484,9 @@ Supported examples:
         google_api_key = get_from_dict_or_env(
             values, "google_api_key", "GOOGLE_API_KEY"
         )
-        try:
-            import google.generativeai as genai
-
-            genai.configure(api_key=google_api_key)
-        except ImportError:
-            raise ChatGoogleGenerativeAIError(
-                "Could not import google.generativeai python package. "
-                "Please install it with `pip install google-generativeai`"
-            )
-
-        values["client"] = genai
+        if isinstance(google_api_key, SecretStr):
+            google_api_key = google_api_key.get_secret_value()
+        genai.configure(api_key=google_api_key)
         if (
             values.get("temperature") is not None
             and not 0 <= values["temperature"] <= 1
@@ -560,7 +549,7 @@ Supported examples:
         **kwargs: Any,
     ) -> ChatResult:
         params = self._prepare_params(messages, stop, **kwargs)
-        response: genai.types.GenerateContentResponse = chat_with_retry(
+        response: genai.types.GenerateContentResponse = _chat_with_retry(
             **params,
             generation_method=self._generation_method,
         )
@@ -574,7 +563,7 @@ Supported examples:
         **kwargs: Any,
     ) -> ChatResult:
         params = self._prepare_params(messages, stop, **kwargs)
-        response: genai.types.GenerateContentResponse = await achat_with_retry(
+        response: genai.types.GenerateContentResponse = await _achat_with_retry(
             **params,
             generation_method=self._async_generation_method,
         )
@@ -588,7 +577,7 @@ Supported examples:
         **kwargs: Any,
     ) -> Iterator[ChatGenerationChunk]:
         params = self._prepare_params(messages, stop, **kwargs)
-        response: genai.types.GenerateContentResponse = chat_with_retry(
+        response: genai.types.GenerateContentResponse = _chat_with_retry(
             **params,
             generation_method=self._generation_method,
             stream=True,
@@ -614,7 +603,7 @@ Supported examples:
         **kwargs: Any,
     ) -> AsyncIterator[ChatGenerationChunk]:
         params = self._prepare_params(messages, stop, **kwargs)
-        async for chunk in await achat_with_retry(
+        async for chunk in await _achat_with_retry(
             **params,
             generation_method=self._async_generation_method,
             stream=True,

langchain_google_genai-0.0.4/langchain_google_genai/embeddings.py

@@ -0,0 +1,99 @@
+from typing import Dict, List, Optional
+
+# TODO: remove ignore once the google package is published with types
+import google.generativeai as genai  # type: ignore[import]
+from langchain_core.embeddings import Embeddings
+from langchain_core.pydantic_v1 import BaseModel, Field, SecretStr, root_validator
+from langchain_core.utils import get_from_dict_or_env
+
+from langchain_google_genai._common import GoogleGenerativeAIError
+
+
+class GoogleGenerativeAIEmbeddings(BaseModel, Embeddings):
+    """`Google Generative AI Embeddings`.
+
+    To use, you must have either:
+
+        1. The ``GOOGLE_API_KEY``` environment variable set with your API key, or
+        2. Pass your API key using the google_api_key kwarg to the ChatGoogle
+           constructor.
+
+    Example:
+        .. code-block:: python
+
+            from langchain_google_genai import GoogleGenerativeAIEmbeddings
+
+            embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+            embeddings.embed_query("What's our Q1 revenue?")
+    """
+
+    model: str = Field(
+        ...,
+        description="The name of the embedding model to use. "
+        "Example: models/embedding-001",
+    )
+    task_type: Optional[str] = Field(
+        None,
+        description="The task type. Valid options include: "
+        "task_type_unspecified, retrieval_query, retrieval_document, "
+        "semantic_similarity, classification, and clustering",
+    )
+    google_api_key: Optional[SecretStr] = Field(
+        None,
+        description="The Google API key to use. If not provided, "
+        "the GOOGLE_API_KEY environment variable will be used.",
+    )
+
+    @root_validator()
+    def validate_environment(cls, values: Dict) -> Dict:
+        """Validates that the python package exists in environment."""
+        google_api_key = get_from_dict_or_env(
+            values, "google_api_key", "GOOGLE_API_KEY"
+        )
+        if isinstance(google_api_key, SecretStr):
+            google_api_key = google_api_key.get_secret_value()
+        genai.configure(api_key=google_api_key)
+        return values
+
+    def _embed(
+        self, texts: List[str], task_type: str, title: Optional[str] = None
+    ) -> List[List[float]]:
+        task_type = self.task_type or "retrieval_document"
+        try:
+            result = genai.embed_content(
+                model=self.model,
+                content=texts,
+                task_type=task_type,
+                title=title,
+            )
+        except Exception as e:
+            raise GoogleGenerativeAIError(f"Error embedding content: {e}") from e
+        return result["embedding"]
+
+    def embed_documents(
+        self, texts: List[str], batch_size: int = 5
+    ) -> List[List[float]]:
+        """Embed a list of strings. Vertex AI currently
+        sets a max batch size of 5 strings.
+
+        Args:
+            texts: List[str] The list of strings to embed.
+            batch_size: [int] The batch size of embeddings to send to the model
+
+        Returns:
+            List of embeddings, one for each text.
+        """
+        task_type = self.task_type or "retrieval_document"
+        return self._embed(texts, task_type=task_type)
+
+    def embed_query(self, text: str) -> List[float]:
+        """Embed a text.
+
+        Args:
+            text: The text to embed.
+
+        Returns:
+            Embedding for the text.
+        """
+        task_type = self.task_type or "retrieval_query"
+        return self._embed([text], task_type=task_type)[0]

{langchain_google_genai-0.0.2 → langchain_google_genai-0.0.4}/pyproject.toml

@@ -1,14 +1,19 @@
 [tool.poetry]
 name = "langchain-google-genai"
-version = "0.0.2"
+version = "0.0.4"
 description = "An integration package connecting Google's genai package and LangChain"
 authors = []
 readme = "README.md"
+repository = "https://github.com/langchain-ai/langchain/blob/master/libs/partners/google-genai"
 
 [tool.poetry.dependencies]
 python = ">=3.9,<4.0"
 langchain-core = "^0.1"
 google-generativeai = "^0.3.1"
+pillow = { version = "^10.1.0", optional = true }
+
+[tool.poetry.extras]
+images = ["pillow"]
 
 [tool.poetry.group.test]
 optional = true
@@ -16,11 +21,12 @@ optional = true
 [tool.poetry.group.test.dependencies]
 pytest = "^7.3.0"
 freezegun = "^1.2.2"
-pytest-mock  = "^3.10.0"
+pytest-mock = "^3.10.0"
 syrupy = "^4.0.2"
 pytest-watcher = "^0.3.4"
 pytest-asyncio = "^0.21.1"
-langchain-core = {path = "../../core", develop = true}
+langchain-core = { path = "../../core", develop = true }
+numpy = "^1.26.2"
 
 [tool.poetry.group.codespell]
 optional = true
@@ -32,6 +38,8 @@ codespell = "^2.2.0"
 optional = true
 
 [tool.poetry.group.test_integration.dependencies]
+pillow = "^10.1.0"
+
 
 [tool.poetry.group.lint]
 optional = true
@@ -41,7 +49,7 @@ ruff = "^0.1.5"
 
 [tool.poetry.group.typing.dependencies]
 mypy = "^0.991"
-langchain-core = {path = "../../core", develop = true}
+langchain-core = { path = "../../core", develop = true }
 types-requests = "^2.28.11.5"
 types-google-cloud-ndb = "^2.2.0.1"
 types-pillow = "^10.1.0.2"
@@ -50,7 +58,7 @@ types-pillow = "^10.1.0.2"
 optional = true
 
 [tool.poetry.group.dev.dependencies]
-langchain-core = {path = "../../core", develop = true}
+langchain-core = { path = "../../core", develop = true }
 pillow = "^10.1.0"
 types-requests = "^2.31.0.10"
 types-pillow = "^10.1.0.2"
@@ -58,19 +66,16 @@ types-google-cloud-ndb = "^2.2.0.1"
 
 [tool.ruff]
 select = [
-  "E",  # pycodestyle
-  "F",  # pyflakes
-  "I",  # isort
+  "E", # pycodestyle
+  "F", # pyflakes
+  "I", # isort
 ]
 
 [tool.mypy]
 disallow_untyped_defs = "True"
-exclude = ["notebooks", "examples", "example_data", "langchain_core/pydantic"]
 
 [tool.coverage.run]
-omit = [
-    "tests/*",
-]
+omit = ["tests/*"]
 
 [build-system]
 requires = ["poetry-core>=1.0.0"]

langchain_google_genai-0.0.2/langchain_google_genai/__init__.py

@@ -1,3 +0,0 @@
-from langchain_google_genai.chat_models import ChatGoogleGenerativeAI
-
-__all__ = ["ChatGoogleGenerativeAI"]