大模型

Llama/LlamaDemo3.py

@@ -0,0 +1,52 @@
+import os
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from sentence_transformers import SentenceTransformer, util
+import torch
+
+# 加载预训练模型的分词器和模型
+tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
+model = AutoModelForCausalLM.from_pretrained(
+    "meta-llama/Llama-2-7b-chat-hf",
+    device_map='auto')
+
+# 加载HuggingFace的Embeddings模型
+embedder = SentenceTransformer('all-MiniLM-L6-v2')
+
+# 指定知识库文件目录
+knowledge_base_dir = "path/to/your/knowledge/base"
+
+# 读取目录中的所有文件，并将其内容作为知识库条目
+knowledge_base = []
+for filename in os.listdir(knowledge_base_dir):
+    filepath = os.path.join(knowledge_base_dir, filename)
+    if os.path.isfile(filepath):
+        with open(filepath, 'r', encoding='utf-8') as file:
+            knowledge_base.append(file.read())
+
+# 将知识库文本转换为embeddings
+knowledge_embeddings = embedder.encode(knowledge_base, convert_to_tensor=True)
+
+# 定义一个用户的问题或提示
+prompt = "请给我讲个玫瑰的爱情故事?"
+
+# 将用户的问题转换为embedding
+question_embedding = embedder.encode(prompt, convert_to_tensor=True)
+
+# 在知识库中寻找最相似的文本
+cos_scores = util.pytorch_cos_sim(question_embedding, knowledge_embeddings)[0]
+top_result = torch.topk(cos_scores, k=1)
+
+# 使用找到的最相关的文本作为模型的输入
+input_text = knowledge_base[top_result.indices[0]]
+
+# 使用分词器将找到的最相关的文本转化为模型可以理解的格式
+inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
+
+# 使用模型生成文本
+outputs = model.generate(inputs["input_ids"], max_new_tokens=2000)
+
+# 将生成的令牌解码成文本
+response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+# 打印生成的响应
+print(response)

Llama/LlamaDemo4.py

@@ -0,0 +1,62 @@
+from openai import OpenAI
+from typing_extensions import override
+from openai import AssistantEventHandler
+
+
+client = OpenAI(api_key="")
+
+
+assistant = client.beta.assistants.create(
+    name="Math Tutor",
+    instructions="You are a personal math tutor. Write and run code to answer math questions.",
+    tools=[{"type": "code_interpreter"}],
+    model="gpt-4-turbo-preview",
+)
+
+thread = client.beta.threads.create()
+
+message = client.beta.threads.messages.create(
+    thread_id=thread.id,
+    role="user",
+    content="I need to solve the equation `3x + 11 = 14`. Can you help me?"
+)
+
+
+
+# First, we create a EventHandler class to define
+# how we want to handle the events in the response stream.
+
+class EventHandler(AssistantEventHandler):
+    @override
+    def on_text_created(self, text) -> None:
+        print(f"\nassistant > ", end="", flush=True)
+
+    @override
+    def on_text_delta(self, delta, snapshot):
+        print(delta.value, end="", flush=True)
+
+    def on_tool_call_created(self, tool_call):
+        print(f"\nassistant > {tool_call.type}\n", flush=True)
+
+    def on_tool_call_delta(self, delta, snapshot):
+        if delta.type == 'code_interpreter':
+            if delta.code_interpreter.input:
+                print(delta.code_interpreter.input, end="", flush=True)
+            if delta.code_interpreter.outputs:
+                print(f"\n\noutput >", flush=True)
+                for output in delta.code_interpreter.outputs:
+                    if output.type == "logs":
+                        print(f"\n{output.logs}", flush=True)
+
+
+# Then, we use the `create_and_stream` SDK helper
+# with the `EventHandler` class to create the Run
+# and stream the response.
+
+with client.beta.threads.runs.create_and_stream(
+        thread_id=thread.id,
+        assistant_id=assistant.id,
+        instructions="Please address the user as Jane Doe. The user has a premium account.",
+        event_handler=EventHandler(),
+) as stream:
+    stream.until_done()

ai-generated-daily/2024-03-11.md

@@ -0,0 +1,29 @@
+1.prompt策略遇上分治算法
+
+论文地址：https://arxiv.org/pdf/2402.05359.pdf
+
+2.不依赖token，字节级模型，直接处理二进制
+
+论文：https://arxiv.org/abs/2402.19155
+
+代码：https://github.com/sanderwood/bgpt
+
+模型：https://huggingface.co/sander-wood/bgpt
+
+项目主页：https://byte-gpt.github.io
+
+3.马斯克宣布本周开源Grok
+
+https://x.com/elonmusk/status/1767108624038449405?s=20
+
+4.RAG搜索复杂PDF文档
+
+https://x.com/llama_index/status/1766862130761125958?s=20
+
+5.几行代码微调LLM Agent
+
+https://x.com/khoomeik/status/1766805213644800011?s=20
+
+6.使用LoRA技术微调Gemma-2B模型
+
+https://huggingface.co/blog/gemma-peft

ollama/demo.py

@@ -0,0 +1,36 @@
+from vanna.ollama import Ollama
+from vanna.chromadb.chromadb_vector import ChromaDB_VectorStore
+
+class MyVanna(ChromaDB_VectorStore, Ollama):
+    def __init__(self, config=None):
+        ChromaDB_VectorStore.__init__(self, config=config)
+        Ollama.__init__(self, config=config)
+
+
+vn = MyVanna(config={'model': 'gemma:7b'})
+
+vn.connect_to_sqlite('my-database.sqlite')
+
+df_ddl = vn.run_sql("SELECT type, sql FROM sqlite_master WHERE sql is not null")
+
+for ddl in df_ddl['sql'].to_list():
+    vn.train(ddl=ddl)
+
+vn.train(ddl="""
+    CREATE TABLE IF NOT EXISTS my-table (
+        id INT PRIMARY KEY,
+        name VARCHAR(100),
+        age INT
+    )
+""")
+
+vn.train(
+    documentation="Our business defines OTIF score as the percentage of orders that are delivered on time and in full")
+
+vn.train(sql="SELECT * FROM my-table WHERE name = 'John Doe'")
+
+training_data = vn.get_training_data()
+
+from vanna.flask import VannaFlaskApp
+
+VannaFlaskApp(vn).run()

ollama/demo2.py

@@ -0,0 +1,2 @@
+import vanna
+print(dir(vanna))

openai/demo.py

@@ -0,0 +1,41 @@
+from openai import OpenAI
+
+client = OpenAI(
+    # defaults to os.environ.get("OPENAI_API_KEY")
+    api_key="",
+    base_url="https://api.chatanywhere.tech/v1"
+)
+
+
+
+# 非流式响应
+def gpt_35_api(messages: list):
+    """为提供的对话消息创建新的回答
+
+    Args:
+        messages (list): 完整的对话消息
+    """
+    completion = client.chat.completions.create(model="gpt-3.5-turbo", messages=messages)
+    print(completion.choices[0].message.content)
+
+def gpt_35_api_stream(messages: list):
+    """为提供的对话消息创建新的回答 (流式传输)
+
+    Args:
+        messages (list): 完整的对话消息
+    """
+    stream = client.chat.completions.create(
+        model='gpt-3.5-turbo',
+        messages=messages,
+        stream=True,
+    )
+    for chunk in stream:
+        if chunk.choices[0].delta.content is not None:
+            print(chunk.choices[0].delta.content, end="")
+
+if __name__ == '__main__':
+    messages = [{'role': 'user','content': '鲁迅和周树人的关系'},]
+    # 非流式调用
+    # gpt_35_api(messages)
+    # 流式调用
+    gpt_35_api_stream(messages)