eli6 commented 3 weeks ago

Prerequisites

Please answer the following questions for yourself before submitting an issue.

[x] I am running the latest code. Development is very rapid so there are no tagged versions as of now.
[x] I carefully followed the README.md.
[x] I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
[x] I reviewed the Discussions, and have a new bug or useful enhancement to share.

Expected Behavior

I am following the functions guide here and am running the first example the "basic demo": https://github.com/abetlen/llama-cpp-python/blob/main/examples/notebooks/Functions.ipynb

I expect to get the reply and result of the function call like in the notebook.

Current behaviour

Instead I get this error when running it, since the "name" of the function is not in the expected format and contains a lot more than just the name (the part that starts with "functions.get_current...."} string. I expect the function_name to be only get_current_weather but when running it, the name is: functions.get_current_weather\n{"location": "San Francisco"}<|reserved_special_token_249|>get_current_weather\n{"location": "Tokyo"}<|reserved_special_token_249|>get_current_weather\n{"location": "Paris"}'), type='function').

  File "...", line 84, in <module>
    print(run_conversation())
          ^^^^^^^^^^^^^^^^^^
  File "...", line 65, in run_conversation
    function_to_call = available_functions[function_name]
                       ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^
KeyError: 'functions.get_current_weather\n{"location": "San Francisco"}<|reserved_special_token_249|>get_current_weather\n{"location": "Tokyo"}<|reserved_special_token_249|>get_current_weather\n{"location": "Paris"}'

Environment and Context

Please provide detailed information about your computer setup. This is important in case the issue is not reproducible except for under certain specific conditions.

Physical (or virtual) hardware you are using, e.g. for Linux:

Running a mac with M1 chip
Operating System, e.g. for Linux:

MacOS 14.13.1 Sonoma

SDK version, e.g. for Linux:

Python 3.12.4

Failure Information (for bugs)

the tool_call response from the server looks like this, the name is in a weird format and there are no arguments:

ChatCompletionMessageToolCall(id='call_3VnQzKr1UZsaG2IchpR3Y7ai', function=Function(arguments='{}', name='functions.get_current_weather\n{"location": "San Francisco"}<|reserved_special_token_249|>get_current_weather\n{"location": "Tokyo"}<|reserved_special_token_249|>get_current_weather\n{"location": "Paris"}'), type='function')

Steps to Reproduce

Please provide detailed steps for reproducing the issue. We are not sitting in front of your screen, so the more detail the better.

start the server like this: python -m llama_cpp.server --model /some-path/functionary-small-v2.5.Q4_0.gguf --chat_format functionary-v2 --hf_pretrained_model_name_or_path meetkai/functionary-small-v2.5-GGUF
step 2

run the first example from the functions notebook against that server: https://github.com/abetlen/llama-cpp-python/blob/main/examples/notebooks/Functions.ipynb

step 3

the output is this error message:

  File "...", line 84, in <module>
    print(run_conversation())
          ^^^^^^^^^^^^^^^^^^
  File "...", line 65, in run_conversation
    function_to_call = available_functions[function_name]
                       ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^
KeyError: 'functions.get_current_weather\n{"location": "San Francisco"}<|reserved_special_token_249|>get_current_weather\n{"location": "Tokyo"}<|reserved_special_token_249|>get_current_weather\n{"location": "Paris"}'

I am not sure if the following about llama.cpp applies to this case. Please tell me if it does.

Note: Many issues seem to be regarding functional or performance issues / differences with llama.cpp. In these cases we need to confirm that you're comparing against the version of llama.cpp that was built with your python package, and which parameters you're passing to the context.

Try the following:

git clone https://github.com/abetlen/llama-cpp-python
cd llama-cpp-python
rm -rf _skbuild/ # delete any old builds
python -m pip install .
cd ./vendor/llama.cpp
Follow llama.cpp's instructions to cmake llama.cpp
Run llama.cpp's ./main with the same arguments you previously passed to llama-cpp-python and see if you can reproduce the issue. If you can, log an issue with llama.cpp

Failure Logs

Please include any relevant log snippets or files. If it works under one configuration but not under another, please provide logs for both configurations and their corresponding outputs so it is easy to see where behavior changes.

Also, please try to avoid using screenshots if at all possible. Instead, copy/paste the console output and use Github's markdown to cleanly format your logs for easy readability.

Example environment info:

llama-cpp-python$ git log | head -1
commit 47b0aa6e957b93dbe2c29d53af16fbae2dd628f2

llama-cpp-python$ python3 --version
Python 3.10.10

llama-cpp-python$ pip list | egrep "uvicorn|fastapi|sse-starlette|numpy"
fastapi                  0.95.0
numpy                    1.24.3
sse-starlette            1.3.3
uvicorn                  0.21.1

llama-cpp-python/vendor/llama.cpp$ git log | head -3
commit 66874d4fbcc7866377246efbcee938e8cc9c7d76
Author: Kerfuffle <44031344+KerfuffleV2@users.noreply.github.com>
Date:   Thu May 25 20:18:01 2023 -0600

EkChinHui commented 2 weeks ago

Hey there, I ran into the same problem. It doesn't work with functionaryv2.5 presumably because there is a change in formatting. Try functionaryv2 instead.

allenporter commented 1 week ago

Yes, the format is changed and Functionary v2.5 support is being added here https://github.com/abetlen/llama-cpp-python/pull/1509

abetlen / llama-cpp-python

tool_call function "name" has incorrect format so function calling does not work for functionary #1560