docs: add vectorstore docs #22
Conversation
product-auto-label bot added the api: cloudsql-postgres 
There was a problem hiding this comment.
@averikitsch I'm working on docs for MySQL and MSSQL so just leaving some drive by comments 😄
Co-authored-by: Jack Wotherspoon <jackwoth@google.com>
Co-authored-by: Jack Wotherspoon <jackwoth@google.com>
Co-authored-by: Jack Wotherspoon <jackwoth@google.com>
Co-authored-by: Jack Wotherspoon <jackwoth@google.com>
docs/vector_store.ipynb Outdated
| }, | ||
| "source": [ | ||
| "### 🦜🔗 Library Installation\n", | ||
| "The integration lives in its own `langchain-google-cloud-sql-pg` package, so we need to install it." |
There was a problem hiding this comment.
maybe call out that we are pip installing vertex ai for emebddings as well since you are calling this out and the next line installs two things
docs/vector_store.ipynb Outdated
| "source": [ | ||
| "from langchain_google_cloud_sql_pg import PostgreSQLEngine\n", | ||
| "\n", | ||
| "engine.run_as_sync( # Use run_as_sync() method to run without async/await\n", |
There was a problem hiding this comment.
maybe move the comment "# Use run_as_sync() method to run without async/await" to the comment section above? and maybe explain why they might want to use this option
| "outputs": [], | ||
| "source": [ | ||
| "from langchain_google_vertexai import VertexAIEmbeddings\n", | ||
| "\n", |
There was a problem hiding this comment.
add a text block above saying, here we choose the textembedding-gecko model, and choose the latest version. nit we should fix a version and not recommend latest.
docs/vector_store.ipynb Outdated
| "source": [ | ||
| "from langchain_google_cloud_sql_pg import CloudSQLVectorStore\n", | ||
| "\n", | ||
| "store = CloudSQLVectorStore.create_sync(\n", |
There was a problem hiding this comment.
is create_sync always required?
| "\n", | ||
| "all_texts = [\"Apples and oranges\", \"Cars and airplanes\", \"Pineapple\", \"Train\", \"Banana\"]\n", | ||
| "metadatas = [{\"len\": len(t)} for t in all_texts]\n", | ||
| "ids = [str(uuid.uuid4()) for _ in all_texts]\n", |
There was a problem hiding this comment.
we made it so the user has to add a id and we dont do it for them right? maybe we could a comment saying that, so that they know id will be a means they have to refer to the texts
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "from langchain_google_cloud_sql_pg.indexes import IVFFlatIndex\n", |
There was a problem hiding this comment.
maybe add a text block, "you can create an index to speed up vector search querires. read more about indexes here"
Co-authored-by: Jack Wotherspoon <jackwoth@google.com>
Co-authored-by: Jack Wotherspoon <jackwoth@google.com>
4 participants
Thank you for opening a Pull Request! Before submitting your PR, there are a few things you can do to make sure it goes smoothly:
Fixes #<issue_number_goes_here> 🦕