Comments (9)
No concern.
from h2ogpt.
If you want to just fix your already-present env, do:
git pull
pip install langchain==0.0.354 langchain_experimental==0.0.47 langchain-community==0.0.8 langsmith==0.0.77 langchain-core==0.1.6
from h2ogpt.
Additionally on the note about the Windows docs:
Looking at a 3month old YT video, I can see the windows docs did used to have a RUN section.
https://www.youtube.com/watch?v=aE8rxzTLld8
It's no longer there, but the hyperlink at the top seems to imply it should be.
Just mentioning this in case there is an additional step or few we should have done?
Final note: I followed the instructions on the page, but did compare back to that YT vid.
But I took the Win documentation page as the steps and order - not the out-of-date vid, in case that question came up.
from h2ogpt.
One more note, the URL issue I mentioned above is also discussed by the OP of this issue:#1248
So possibly my installation is flawed, since I did exactly what the docs said to do, but this user is saying the quote marks are in the wrong place for that URL to properly apply during install?
from h2ogpt.
Update
Due the semantics error detailed in 1248, I deleted the conda env and tried again from scratch.
This time, I got zero red or yellow text (errors).
On the other hand, basically everything it said was using cached version.
So what I can't work out is, if the first time around the extra index had characters out of place, so it DIDN'T look at the correct index, it got whatever, then 2nd time around despite clearing the env, it's still using the wrong things, because it was using the stuff from first time around from the cache.
Because
I get the same error after getting to the end of the Windows/GPU doc, then running
python generate.py --base_model=TheBloke/zephyr-7B-beta-GGUF --prompt_type=zephyr --max_seq_len=4096
Error is:
(h2ogpt) C:\AI\Text\h2ogpt>python generate.py --base_model=TheBloke/zephyr-7B-beta-GGUF --prompt_type=zephyr --max_seq_len=4096
Using Model llama
Traceback (most recent call last):
File "C:\AI\Text\h2ogpt\generate.py", line 16, in <module>
entrypoint_main()
File "C:\AI\Text\h2ogpt\generate.py", line 12, in entrypoint_main
H2O_Fire(main)
File "C:\AI\Text\h2ogpt\src\utils.py", line 65, in H2O_Fire
fire.Fire(component=component, command=args)
File "C:\Users\MyCompy\miniconda3\envs\h2ogpt\lib\site-packages\fire\core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "C:\Users\MyCompy\miniconda3\envs\h2ogpt\lib\site-packages\fire\core.py", line 475, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "C:\Users\MyCompy\miniconda3\envs\h2ogpt\lib\site-packages\fire\core.py", line 691, in _CallAndUpdateTrace
component = fn(*varargs, **kwargs)
File "C:\AI\Text\h2ogpt\src\gen.py", line 1620, in main
from src.gpt_langchain import get_embedding
File "C:\AI\Text\h2ogpt\src\gpt_langchain.py", line 34, in <module>
from langchain.callbacks import streaming_stdout
File "C:\Users\MyCompy\miniconda3\envs\h2ogpt\lib\site-packages\langchain\callbacks\__init__.py", line 45, in <module>
from langchain_core.tracers.context import (
ImportError: cannot import name 'tracing_enabled' from 'langchain_core.tracers.context' (C:\Users\MyCompy\miniconda3\envs\h2ogpt\lib\site-packages\langchain_core\tracers\context.py)
I even put this into ChatGPT and it said I should try uninstalling/reinstalling Langchain Core, and/or upgrading it to latest version, both of which do nothing to stop the error.
Should I try a third time, with caches cleared, and the correctly formatted extra-index line?
from h2ogpt.
Langchain broke their package updates with incorrect versioning of packages. I've adding pinnings so things don't drift. So fixed in main, docker image being built.
from h2ogpt.
from h2ogpt.
Awesome, thank you so much!
from h2ogpt.
Sorry to add to this, just one last question.
The fix worked, gradio UI is up. Thankyou @pseudotensor!
Just wanted to check if these errors should be any concern?
C:\Users\REDACTED\miniconda3\envs\h2ogpt\lib\site-packages\pydantic\_internal\_fields.py:149: UserWarning: Field "model_name" has conflict with protected namespace "model_".
You may be able to resolve this warning by setting `model_config['protected_namespaces'] = ()`.
warnings.warn(
C:\Users\REDACTED\miniconda3\envs\h2ogpt\lib\site-packages\pydantic\_internal\_fields.py:149: UserWarning: Field "model_names" has conflict with protected namespace "model_".
You may be able to resolve this warning by setting `model_config['protected_namespaces'] = ()`.
warnings.warn(
Cheers
from h2ogpt.
Related Issues (20)
- Don't do reduction in summary when first run is already based on the whole text HOT 1
- Starting get_model: llama Cannot connect HOT 3
- MACBook Pro Intel 32GB: Is this suitable to run h2gpt: Loading model error. HOT 2
- generation hangs initially when querying data HOT 7
- ggml_metal_get_buffer HOT 3
- LangChain v0.1.0 Update
- windows version doesn't start because of missing path HOT 3
- What is the solution to the problem? HOT 1
- ingestion takes forever HOT 2
- Windows installer - h2oGPT wont load HOT 2
- Error fetching or processing https://en.m.wikipedia.org/wiki/Yemen, exception: Invalid file. The FileType.UNK file type is not supported in partition. HOT 2
- How to retrieve H2OGpt from websites without internet? HOT 1
- Why is there no response quickly? HOT 1
- Why is there no response quickly? HOT 1
- h2ogpt to support imagePullSecret and custom image tags
- Memory to store models of closed browser sessions persisting. HOT 9
- How do I upload images that H2ogpt recognizes and import information from them? Calculate the question? HOT 1
- How does h2ogpt respond to images? HOT 1
- Yi thinks its only 4k, h2ogpt or transformers not computing max_seq_len right -- also check parallel calls for summarization HOT 1
- Contents in angle brackets are removed from responses and can break WebUI. HOT 7
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from h2ogpt.