[LangChain] Add example programs and notebooks#85
Conversation
| "from pprint import pprint\n", | ||
| "\n", | ||
| "CONNECTION_STRING = \"crate://crate@localhost/?schema=notebook\"\n", | ||
| "\n", |
There was a problem hiding this comment.
Do we need the schema in the connection string? For me, it causes a failure on the SELECT statement below, because mlb_teams_2012? is apparently created in the doc` schema.
There was a problem hiding this comment.
6779392 should fix that flaw. I would like to use it, in order to educate people about it.
framework/langchain/readme.md
Outdated
|
|
||
| ### CrateDB Cloud | ||
|
|
||
| Todo. |
There was a problem hiding this comment.
| Todo. | |
| Sign up or log in to [CrateDB Cloud](https://console.cratedb.cloud) and create a free tier cluster. Within just a few minutes, you will have a cloud-based development environment. As soon as your project scales, you can easily move to a different cluster tier or scale horizontally. |
cc2943c to
1965489
Compare
8fd2d10 to
dbf3dab
Compare
|
Within the README, there are now badges to open the Jupyter Notebooks on GitHub, on Binder, and on Google Colab. See LangChain and CrateDB » What's inside. |
| { | ||
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "### Overwriting a vector store\n", | ||
| "\n", | ||
| "If you have an existing collection, you can overwrite it by using `from_documents`,\n", | ||
| "aad setting `pre_delete_collection = True`." | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "db = CrateDBVectorSearch.from_documents(\n", | ||
| " documents=docs,\n", | ||
| " embedding=embeddings,\n", | ||
| " collection_name=COLLECTION_NAME,\n", | ||
| " connection_string=CONNECTION_STRING,\n", | ||
| " pre_delete_collection=True,\n", | ||
| ")" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "docs_with_score = db.similarity_search_with_score(\"foo\")" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "docs_with_score[0]" | ||
| ] | ||
| }, |
There was a problem hiding this comment.
I am observing problems here. It is probably related to the other report by @ckurze.
There was a problem hiding this comment.
Adding software tests will help to keep our sanity.
7cf13da to
cd92557
Compare
f43c9f5 to
f5546bf
Compare
cd92557 to
cd9b202
Compare
About
We have been looking into making LangChain work together with CrateDB. This patch adds corresponding example programs and Jupyter Notebooks.
Preview
Readme, rendered by GitHub: LangChain and CrateDB
Backlog
both the basic example programs, and the Jupyter Notebooks.
/cc @marijaselakovic, @hammerhead, @seut, @ckurze