zjunlp / easyinstruct Goto Github PK

1、目前从代码看，KG2Instruct方法输出是信息抽取后的三元组，从三元组如何产生指令微调训练样本，能否可以详细描述，并开放相关代码？
2、Evol-Instruct方法中Elimination Evolving过滤指令原则有：
1）与原始指令相比，进化后的指令没有提供任何信息增益；
2）进化出的指令使得LLMs很难产生回复；
......
请问这些过滤指令的原则具体是如何实现的？是否可以详细描述下，或公开相关代码？

谢谢！期盼回复~

Building wheel for llama-cpp-python (pyproject.toml) ... error
error: subprocess-exited-with-error

Not searching for unused variables given on the command line.
-- The C compiler identification is unknown
CMake Error at CMakeLists.txt:3 (ENABLE_LANGUAGE):
No CMAKE_C_COMPILER could be found.

    Tell CMake where to find the compiler by setting either the environment
    variable "CC" or the CMake cache entry CMAKE_C_COMPILER to the full path to
    the compiler, or to the compiler name if it is in the PATH.


  -- Configuring incomplete, errors occurred!

ERROR: Failed building wheel for llama-cpp-python
Failed to build llama-cpp-python
ERROR: Could not build wheels for llama-cpp-python, which is required to install pyproject.toml-based projects

使用setup.py安装
import easyinstruct 无法读取 easyinstruct文件夹下面的engines文件夹，报错：

import easyinstruct
Traceback (most recent call last):
File "", line 1, in
File "d:\tianchi\easyinstruct\easyinstruct_init_.py", line 1, in
from .prompts import *
File "d:\tianchi\easyinstruct\easyinstruct\prompts_init_.py", line 1, in
from .base_prompt import BasePrompt
File "d:\tianchi\easyinstruct\easyinstruct\prompts\base_prompt.py", line 7, in
from engines import llama_engine
ModuleNotFoundError: No module named 'engines'

ModuleNotFoundError: No module named 'llama_cpp'

OpenAI API has been updated to version 1. openai.ChatCompletion module is no longer supported.

In base_prompt.py,

response = openai.ChatCompletion.create(
                model=engine,
                messages=messages,
                temperature=temperature,
                max_tokens=max_tokens,
                top_p=top_p,
                n=n,
                frequency_penalty=frequency_penalty,
                presence_penalty=presence_penalty,
            )

should be changed to

response = client.chat.completions.create(
                model=engine,
                messages=messages,
                temperature=temperature,
                max_tokens=max_tokens,
                top_p=top_p,
                n=n,
                frequency_penalty=frequency_penalty,
                presence_penalty=presence_penalty,
            )

在运行example/llm/run.py文件时，在已经安装easyinstruct==0.0.3的情况下，发生报错：
Traceback (most recent call last):
File "/home/ubuntu/MyFiles/DeepKE-main/example/llm/run.py",line 6,in
from easyinstruct.prompts import IEPrompt
File "/home/ubuntu/miniconda3/envs/py39-cul16/lib/python3.9/site-packages/easyinstruct/init.py",line 1,in
from .prompts import
File "/home/ubuntu/miniconda3/envs/py39-cull6/lib/python3.9/site-packages/easyinstruct/prompts/init.py",line 1,in
from base prompt import BasePrompt
File "/home/ubuntu/miniconda3/envs/py39-cul16/lib/python3.9/site-packages/easyinstruct/prompts/base prompt.py",line 5,in
from easyinstruct.utils.api import API_NAME_DICT
ModuleNotFoundError: No module named 'easyinstruct.utils'
Process finished with exit code 1

请问这个该如何解决呢？

可能是版本更迭中有疏忽？漏了一句self.examples=examples

Anthropic changed their python sdk - making this code line outdated.

Would love to know if this might help - https://github.com/BerriAI/litellm

~Simple I/O library, that standardizes all the llm api calls to the OpenAI call

from litellm import completion

## set ENV variables
# ENV variables can be set in .env file, too. Example in .env.example
os.environ["OPENAI_API_KEY"] = "openai key"
os.environ["ANTHROPIC_API_KEY"] = "anthropic key"

messages = [{ "content": "Hello, how are you?","role": "user"}]

# openai call
response = completion(model="gpt-3.5-turbo", messages=messages)

# anthropic call
response = completion("claude-v-2", messages)

Hello, I have found a misspelling error in ie_prompt.py lines 16 and 17: "extraion" (instead of "extraction").

I just read the paper When Do Program-of-Thought Works for Reasoning. The result is pretty interesting and I would like to use CIRS metric to test with some other public code datasets. It is said in the paper that the code will be integrated into this project so I come to ask for help. Thank you :)