was gonna try chain-of.tables pack, but

At a glance

The community member was trying to use the "chain-of.tables" pack, but the page they were directed to gave a 404 error. Another community member suggested trying a different link, but the first community member encountered issues when trying to install the pack using the CLI command or Python code provided.

The issues included a FileNotFoundError when trying to download the pack, and the community members discussed potential causes like path issues or the original tutorial page being unavailable. They suggested trying a Git install of the package, but were unsure if that would work. Overall, there does not appear to be an explicitly marked answer in the comments.

Useful resources

LLuke Skywalker

was gonna try chain-of.tables pack, but the page gives me a 404. Yesterday it was still up.
https://llamahub.ai/l/llama_packs-tables-chain_of_table?from=all

13 comments

TTeemu

Have you tried this link? https://llamahub.ai/l/llama-packs/llama-index-packs-tables?from=all

LLuke Skywalker

hmmm thats an interesting way of installing on top of llama-index with cli command or the python code provided? 🤔
First does not work when building an Dockerimage.

Plain Text

8.227     download_llama_pack(
8.227   File "/opt/conda/lib/python3.10/site-packages/llama_index/core/llama_pack/download.py", line 56, in download_llama_pack
8.227     pack_cls = download_llama_pack_template(
8.227   File "/opt/conda/lib/python3.10/site-packages/llama_index/core/download/pack.py", line 126, in download_llama_pack_template
8.227     spec.loader.exec_module(module)  # type: ignore
8.227   File "<frozen importlib._bootstrap_external>", line 879, in exec_module
8.227   File "<frozen importlib._bootstrap_external>", line 1016, in get_code
8.227   File "<frozen importlib._bootstrap_external>", line 1073, in get_data
8.227 FileNotFoundError: [Errno 2] No such file or directory: '/app/chain_of_table_pack/llama_index/packs/tables/base.py'

Second method I would like to avoid.

LLuke Skywalker

@AlphaAtlas

AAlphaAtlas

One thing, be sure to build the docker container from scratch

AAlphaAtlas

If its trying to re-use any old python environment, it will not work 😅

AAlphaAtlas

The whole python environment needs to be nuked, it doesn't like having any old llama index packages

LLuke Skywalker

I did. Issue is the following: The dependencies get installed fine with both approaches. However the actual python code is not downloaded or copied to the download folder ....

AAlphaAtlas

I suspect it some kind of path issue, poking through the function itself in a sec

AAlphaAtlas

Yeah I was going to suggest a git install with some debug messages around here to see what the path is: https://github.com/run-llama/llama_index/blob/a900d5e67424c2b2c46b0aa9ef62502af556d449/llama-index-core/llama_index/core/download/pack.py#L110-L117

AAlphaAtlas

But how would a git install of the package even work now?

LLuke Skywalker

🤷‍♂️ Its an enigma 😅

I have the feeling its connected to the original tutorial page I was on yesterday had a 404. Both this and the download script seem to reference a non-existent branch 🤔

LLuke Skywalker

https://github.com/run-llama/llama_index/issues/10777

LLuke Skywalker

Attachment

Add a reply

Find answers from the community

was gonna try chain-of.tables pack, but