github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-29 01:48:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

google-genai[patch]: release 0.0.9, safety settings docs (#17432)

Browse Source
This commit is contained in:
Erick Friis 2024-02-13 10:01:25 -08:00 committed by GitHub
parent db6f266d97
commit 065cde69b1
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 69 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
docs/docs/integrations/chat/google_generative_ai.ipynb
View File

@ -320,20 +320,51 @@
"4. Message may be blocked if they violate the safety checks of the LLM. In this case, the model will return an empty response."
]
},
{
"cell_type": "markdown",
"id": "54793b9e",
"metadata": {},
"source": [
"### Safety Settings\n",
"\n",
"Gemini models have default safety settings that can be overridden. If you are receiving lots of \"Safety Warnings\" from your models, you can try tweaking the `safety_settings` attribute of the model. For example, to turn off safety blocking for dangerous content, you can construct your LLM as follows:"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "75fdfad6",
"metadata": {},
"outputs": [],
"source": []
"source": [
"from langchain_google_genai import (\n",
" ChatGoogleGenerativeAI,\n",
" HarmBlockThreshold,\n",
" HarmCategory,\n",
")\n",
"\n",
"llm = ChatGoogleGenerativeAI(\n",
" model=\"gemini-pro\",\n",
" safety_settings={\n",
" HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT: HarmBlockThreshold.BLOCK_NONE,\n",
" },\n",
")"
]
},
{
"cell_type": "markdown",
"id": "e68e203d",
"metadata": {},
"source": [
"For an enumeration of the categories and thresholds available, see Google's [safety setting types](https://ai.google.dev/api/python/google/generativeai/types/SafetySettingDict)."
]
},
{
"cell_type": "markdown",
"id": "92b5aca5",
"metadata": {},
"source": [
"## Additional Configuraation\n",
"## Additional Configuration\n",
"\n",
"You can pass the following parameters to ChatGoogleGenerativeAI in order to customize the SDK's behavior:\n",
"\n",

34
docs/docs/integrations/llms/google_ai.ipynb
View File

@ -265,12 +265,40 @@
]
},
{
"cell_type": "code",
"execution_count": null,
"cell_type": "markdown",
"id": "aefe6df7",
"metadata": {},
"source": [
"### Safety Settings\n",
"\n",
"Gemini models have default safety settings that can be overridden. If you are receiving lots of \"Safety Warnings\" from your models, you can try tweaking the `safety_settings` attribute of the model. For example, to turn off safety blocking for dangerous content, you can construct your LLM as follows:"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "7e2682e6",
"metadata": {},
"outputs": [],
"source": []
"source": [
"from langchain_google_genai import GoogleGenerativeAI, HarmBlockThreshold, HarmCategory\n",
"\n",
"llm = GoogleGenerativeAI(\n",
" model=\"gemini-pro\",\n",
" google_api_key=api_key,\n",
" safety_settings={\n",
" HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT: HarmBlockThreshold.BLOCK_NONE,\n",
" },\n",
")"
]
},
{
"cell_type": "markdown",
"id": "e8d0ee0b",
"metadata": {},
"source": [
"For an enumeration of the categories and thresholds available, see Google's [safety setting types](https://ai.google.dev/api/python/google/generativeai/types/SafetySettingDict)."
]
}
],
"metadata": {

2
docs/docs/integrations/text_embedding/google_generative_ai.ipynb
View File

@ -200,7 +200,7 @@
"id": "2e7857e5",
"metadata": {},
"source": [
"## Additional Configuraation\n",
"## Additional Configuration\n",
"\n",
"You can pass the following parameters to ChatGoogleGenerativeAI in order to customize the SDK's behavior:\n",
"\n",

8
libs/partners/google-genai/pyproject.toml
View File

@ -1,6 +1,6 @@
[tool.poetry]
name = "langchain-google-genai"
version = "0.0.8"
version = "0.0.9"
description = "An integration package connecting Google's genai package and LangChain"
authors = []
readme = "README.md"
@ -70,9 +70,9 @@ types-google-cloud-ndb = "^2.2.0.1"
[tool.ruff.lint]
select = [
"E", # pycodestyle
"F", # pyflakes
"I", # isort
"E", # pycodestyle
"F", # pyflakes
"I", # isort
"T201", # print
]