Find answers from the community

Updated 11 months ago

Embeddings

I am using the EmbeddedTablesUnstructuredRetrieverPack like this

EmbeddedTablesUnstructuredRetrieverPack = download_llama_pack(
"EmbeddedTablesUnstructuredRetrieverPack",
"./embedded_tables_unstructured_pack",
)

and then

embedded_tables_unstructured_pack = EmbeddedTablesUnstructuredRetrieverPack(
"quarterly-nvidia/quarterly-nvidia.html",
nodes_save_path="quarterly-nvidia.pkl")

However, I get this message: "Embeddings have been explicitly disabled. Using MockEmbedding.".

How to fix this?

8 comments

LLogan M

That's not an error, there's nothing to fix

LLogan M

There's a step under the hood where the llm is used to generate summaries of tables. That's where the message comes from

NNeza

I tried to find where the message is, in order to understand what's happenning under the hood
but I could not find anything,
how is it from your side @Sayan ?
do you have more info @Logan M ?

NNeza

Also, I have seen an execution on a tutorial once, and they did not have this message displayed,I am not sure why

LLogan M

Probably just an older saved version of the execution

LLogan M

The "under the hood" code in question is this line of code

https://github.com/run-llama/llama_index/blob/47eb96157cf50e39dc6f3239c1140792a8d2863e/llama_index/node_parser/relational/base_element.py#L133

NNeza

Thank you !
It was so hard to find this piece of code in my editor for some reason !
I'd follow on the code to find what is calling what, but I couldn't find it

LLogan M

No worries! I realize the code is slightly nested lol

Add a reply