发布日期: 2024-02-02

文章字数: 236

阅读时长: 1 分

阅读次数:

1. Introduction

1. About the LlamaIndex course

2. Basic Setup

pip install llama-index openai

3. Help & resources

2. Introduction to LlamaIndex and LLM applications

1. Intro to section and LLM applications

2. Train ChatGPT (LLMs) on custom data - RAG

微调-训练-还是RAG

3. The difference between LlamaIndex and LangChain

4. LLMs and data privacy

6. How LlamaIndex works

# %% import
import logging
import sys
import os

logging.basicConfig(level=logging.INFO, stream=sys.stdout)
logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))
# %%
import os

os.environ['NUMEXPR_MAX_THREADS'] = '4'
os.environ['NUMEXPR_NUM_THREADS'] = '2'
import numexpr as ne
# %% pip install nltk
from dotenv import load_dotenv

load_dotenv()
import openai
# openai.api_key=''

# %%
from llama_index import VectorStoreIndex, SimpleDirectoryReader

documents = SimpleDirectoryReader('assets/AndrewHuberman/sleep').load_data()
index = VectorStoreIndex.from_documents(documents)

# %%
query_engine = index.as_query_engine()

response = query_engine.query('what can i do to sleep better?')
print(response)

7. How to use LLMs with LlamaIndex

from dotenv import load_dotenv

load_dotenv()
import os

os.environ['NUMEXPR_MAX_THREADS'] = '4'
os.environ['NUMEXPR_NUM_THREADS'] = '2'
import numexpr as ne

# %%
from llama_index.llms import OpenAI

llm = OpenAI(temperature=0, model='gpt-4', max_tokens=250)
response = llm.complete('What is AI?')
print(response)
# print(response.raw)

# %%
from llama_index.llms import OpenAI, ChatMessage

llm = OpenAI(temperature=0, model='gpt-4', max_tokens=250)
messages = [
    ChatMessage(role='system', content='Talk like a hippie'),
    ChatMessage(role='user', content='Tell me about AI'),
]
response = llm.chat(messages)
print(response)