mirror of
https://github.com/hwchase17/langchain
synced 2024-11-10 01:10:59 +00:00
cab55e9bc1
- chat vertex async - vertex stream - vertex full generation info - vertex use server-side stopping - model garden async - update docs for all the above in follow up will add [] chat vertex full generation info [] chat vertex retries [] scheduled tests
344 lines
9.0 KiB
Plaintext
344 lines
9.0 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"attachments": {},
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"# GCP Vertex AI \n",
|
|
"\n",
|
|
"Note: This is seperate from the Google PaLM integration. Google has chosen to offer an enterprise version of PaLM through GCP, and this supports the models made available through there. \n",
|
|
"\n",
|
|
"By default, Google Cloud [does not use](https://cloud.google.com/vertex-ai/docs/generative-ai/data-governance#foundation_model_development) Customer Data to train its foundation models as part of Google Cloud`s AI/ML Privacy Commitment. More details about how Google processes data can also be found in [Google's Customer Data Processing Addendum (CDPA)](https://cloud.google.com/terms/data-processing-addendum).\n",
|
|
"\n",
|
|
"To use Vertex AI PaLM you must have the `google-cloud-aiplatform` Python package installed and either:\n",
|
|
"- Have credentials configured for your environment (gcloud, workload identity, etc...)\n",
|
|
"- Store the path to a service account JSON file as the GOOGLE_APPLICATION_CREDENTIALS environment variable\n",
|
|
"\n",
|
|
"This codebase uses the `google.auth` library which first looks for the application credentials variable mentioned above, and then looks for system-level auth.\n",
|
|
"\n",
|
|
"For more information, see: \n",
|
|
"- https://cloud.google.com/docs/authentication/application-default-credentials#GAC\n",
|
|
"- https://googleapis.dev/python/google-auth/latest/reference/google.auth.html#module-google.auth\n",
|
|
"\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {
|
|
"tags": []
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"#!pip install langchain google-cloud-aiplatform"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"from langchain.chat_models import ChatVertexAI\n",
|
|
"from langchain.prompts import ChatPromptTemplate"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"chat = ChatVertexAI()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 34,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"system = \"You are a helpful assistant who translate English to French\"\n",
|
|
"human = \"Translate this sentence from English to French. I love programming.\"\n",
|
|
"prompt = ChatPromptTemplate.from_messages(\n",
|
|
" [(\"system\", system), (\"human\", human)]\n",
|
|
")\n",
|
|
"messages = prompt.format_messages()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 9,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"AIMessage(content=\" J'aime la programmation.\", additional_kwargs={}, example=False)"
|
|
]
|
|
},
|
|
"execution_count": 9,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"chat(messages)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"If we want to construct a simple chain that takes user specified parameters:"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 12,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"system = \"You are a helpful assistant that translates {input_language} to {output_language}.\"\n",
|
|
"human = \"{text}\"\n",
|
|
"prompt = ChatPromptTemplate.from_messages(\n",
|
|
" [(\"system\", system), (\"human\", human)]\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 13,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"AIMessage(content=' 私はプログラミングが大好きです。', additional_kwargs={}, example=False)"
|
|
]
|
|
},
|
|
"execution_count": 13,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"chain = prompt | chat\n",
|
|
"chain.invoke(\n",
|
|
" {\"input_language\": \"English\", \"output_language\": \"Japanese\", \"text\": \"I love programming\"}\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"attachments": {},
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"execution": {
|
|
"iopub.execute_input": "2023-06-17T21:09:25.423568Z",
|
|
"iopub.status.busy": "2023-06-17T21:09:25.423213Z",
|
|
"iopub.status.idle": "2023-06-17T21:09:25.429641Z",
|
|
"shell.execute_reply": "2023-06-17T21:09:25.429060Z",
|
|
"shell.execute_reply.started": "2023-06-17T21:09:25.423546Z"
|
|
},
|
|
"tags": []
|
|
},
|
|
"source": [
|
|
"## Code generation chat models\n",
|
|
"You can now leverage the Codey API for code chat within Vertex AI. The model name is:\n",
|
|
"- codechat-bison: for code assistance"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 18,
|
|
"metadata": {
|
|
"tags": []
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"chat = ChatVertexAI(\n",
|
|
" model_name=\"codechat-bison\",\n",
|
|
" max_output_tokens=1000,\n",
|
|
" temperature=0.5\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 20,
|
|
"metadata": {
|
|
"tags": []
|
|
},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
" ```python\n",
|
|
"def is_prime(x): \n",
|
|
" if (x <= 1): \n",
|
|
" return False\n",
|
|
" for i in range(2, x): \n",
|
|
" if (x % i == 0): \n",
|
|
" return False\n",
|
|
" return True\n",
|
|
"```\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# For simple string in string out usage, we can use the `predict` method:\n",
|
|
"print(chat.predict(\"Write a Python function to identify all prime numbers\"))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Asynchronous calls\n",
|
|
"\n",
|
|
"We can make asynchronous calls via the `agenerate` and `ainvoke` methods."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 23,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import asyncio\n",
|
|
"# import nest_asyncio\n",
|
|
"# nest_asyncio.apply()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 35,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"LLMResult(generations=[[ChatGeneration(text=\" J'aime la programmation.\", generation_info=None, message=AIMessage(content=\" J'aime la programmation.\", additional_kwargs={}, example=False))]], llm_output={}, run=[RunInfo(run_id=UUID('223599ef-38f8-4c79-ac6d-a5013060eb9d'))])"
|
|
]
|
|
},
|
|
"execution_count": 35,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"chat = ChatVertexAI(\n",
|
|
" model_name=\"chat-bison\",\n",
|
|
" max_output_tokens=1000,\n",
|
|
" temperature=0.7,\n",
|
|
" top_p=0.95,\n",
|
|
" top_k=40,\n",
|
|
")\n",
|
|
"\n",
|
|
"asyncio.run(chat.agenerate([messages]))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 36,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"AIMessage(content=' अहं प्रोग्रामिंग प्रेमामि', additional_kwargs={}, example=False)"
|
|
]
|
|
},
|
|
"execution_count": 36,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"asyncio.run(chain.ainvoke({\"input_language\": \"English\", \"output_language\": \"Sanskrit\", \"text\": \"I love programming\"}))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Streaming calls\n",
|
|
"\n",
|
|
"We can also stream outputs via the `stream` method:"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import sys"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 32,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
" 1. China (1,444,216,107)\n",
|
|
"2. India (1,393,409,038)\n",
|
|
"3. United States (332,403,650)\n",
|
|
"4. Indonesia (273,523,615)\n",
|
|
"5. Pakistan (220,892,340)\n",
|
|
"6. Brazil (212,559,409)\n",
|
|
"7. Nigeria (206,139,589)\n",
|
|
"8. Bangladesh (164,689,383)\n",
|
|
"9. Russia (145,934,462)\n",
|
|
"10. Mexico (128,932,488)\n",
|
|
"11. Japan (126,476,461)\n",
|
|
"12. Ethiopia (115,063,982)\n",
|
|
"13. Philippines (109,581,078)\n",
|
|
"14. Egypt (102,334,404)\n",
|
|
"15. Vietnam (97,338,589)"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"prompt = ChatPromptTemplate.from_messages([(\"human\", \"List out the 15 most populous countries in the world\")])\n",
|
|
"messages = prompt.format_messages()\n",
|
|
"for chunk in chat.stream(messages):\n",
|
|
" sys.stdout.write(chunk.content)\n",
|
|
" sys.stdout.flush()"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "poetry-venv",
|
|
"language": "python",
|
|
"name": "poetry-venv"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.9.1"
|
|
},
|
|
"vscode": {
|
|
"interpreter": {
|
|
"hash": "cc99336516f23363341912c6723b01ace86f02e26b4290be1efc0677e2e2ec24"
|
|
}
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 4
|
|
}
|