add llamaindex + groq cookbook #503
Conversation
|
Thanks @jerryjliu for the PR can we pls add it here, under this folder, https://github.com/meta-llama/llama-recipes/tree/main/recipes/llama_api_providers |
lol my bad! moving it now |
|
@HamidShojanazeri check now |
There was a problem hiding this comment.
Thanks Jerry. Found a couple of errors while running the notebook. @jerryjliu
| "source": [ | ||
| "from llama_index.llms.groq import Groq\n", | ||
| "\n", | ||
| "llm = Groq(model=\"llama3-8b-8192\")\n", |
There was a problem hiding this comment.
should add
os.environ["GROQ_API_KEY"] = GROQ_API_KEY
or change to
llm = Groq(model="llama3-8b-8192", api_key=GROQ_API_TOKEN)
| "source": [ | ||
| "### Load Data\n", | ||
| "\n", | ||
| "We load data using LlamaParse by default, but you can also choose to opt for our free pypdf reader (in SimpleDirectoryReader by default) if you don't have an account." |
There was a problem hiding this comment.
Please add a note saying how to sign up to get a key. or uncomment the free one and comment the LlamaParse one. Also, is pypdf good enough? When should one use it and when LlamaParse? @jerryjliu
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "# from llama_index.core.query_engine import SubQuestionQueryEngine\n", |
There was a problem hiding this comment.
remove these commented lines? or explain or uncomment? @jerryjliu I uncommented and running it got "ValueError: No valid JSON found in output"
There was a problem hiding this comment.
@jeffxtang just tested again. for some reason i can't get sub-question to work on groq 70b, so i took the sub-question section out
| "outputs": [], | ||
| "source": [ | ||
| "drake_tool = QueryEngineTool(\n", | ||
| " drake_index.as_query_engine(),\n", |
same as #502 but with groq