File size: 5,209 Bytes
cba278b
f5e9a40
cba278b
 
 
 
 
 
 
 
16da4c9
cba278b
 
 
 
16da4c9
cba278b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f5e9a40
cba278b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16da4c9
cba278b
 
 
 
 
 
f5e9a40
cba278b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
import os
import gradio as gr
import torch
from langchain_huggingface import HuggingFaceEmbeddings
from langchain_community.document_loaders import TextLoader
from langchain_community.vectorstores import FAISS
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.chains import RetrievalQA
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
from langchain_huggingface import HuggingFacePipeline

# Configure GPU settings
os.environ["CUDA_VISIBLE_DEVICES"] = "0"
device = "cuda" if torch.cuda.is_available() else "cpu"
print(f"Using device: {device}")

class CPSLChatbot:
    def __init__(self):
        self.initialize_components()

    def initialize_components(self):
        try:
            # Load and process document
            doc_loader = TextLoader("dataset.txt")
            docs = doc_loader.load()
            text_splitter = RecursiveCharacterTextSplitter(
                chunk_size=1000, 
                chunk_overlap=100
            )
            split_docs = text_splitter.split_documents(docs)

            # Initialize embeddings and vector store
            self.embeddings = HuggingFaceEmbeddings(
                model_name="all-MiniLM-L6-v2",
                model_kwargs={'device': device}
            )
            self.vectordb = FAISS.from_documents(split_docs, self.embeddings)

            # Load model and tokenizer
            model_name = "01-ai/Yi-Coder-9B-Chat"
            self.tokenizer = AutoTokenizer.from_pretrained(model_name)
            self.model = AutoModelForCausalLM.from_pretrained(
                model_name,
                device_map="auto",
                torch_dtype=torch.float16 if device == "cuda" else torch.float32,
                trust_remote_code=True
            )

            # Set up QA pipeline
            self.qa_pipeline = pipeline(
                "text-generation",
                model=self.model,
                tokenizer=self.tokenizer,
                max_new_tokens=750,
                pad_token_id=self.tokenizer.eos_token_id,
                device=0 if device == "cuda" else -1
            )

            # Initialize LangChain components
            llm = HuggingFacePipeline(pipeline=self.qa_pipeline)
            retriever = self.vectordb.as_retriever(search_kwargs={"k": 5})
            self.qa_chain = RetrievalQA.from_chain_type(
                retriever=retriever,
                chain_type="stuff",
                llm=llm,
                return_source_documents=False
            )
            print("Initialization completed successfully")

        except Exception as e:
            print(f"Initialization error: {str(e)}")
            raise

    def preprocess_query(self, query):
        if "script" in query.lower() or "code" in query.lower():
            return f"Write a CPSL script: {query}"
        return query

    def clean_response(self, response):
        result = response.get("result", "")
        if "Answer:" in result:
            return result.split("Answer:")[1].strip()
        return result.strip()

    def get_response(self, user_input):
        try:
            processed_query = self.preprocess_query(user_input)
            raw_response = self.qa_chain.invoke({"query": processed_query})
            return self.clean_response(raw_response)
        except Exception as e:
            return f"Error processing query: {str(e)}"

def create_gradio_interface():
    chatbot = CPSLChatbot()
    
    with gr.Blocks(title="CPSL Chatbot") as chat_interface:
        gr.Markdown("# CPSL Chatbot with GPU Support")
        gr.Markdown("Using Yi-Coder-9B-Chat model for CPSL script generation and queries")
        
        chat_history = gr.Chatbot(
            value=[], 
            elem_id="chatbot",
            height=600
        )
        
        with gr.Row():
            user_input = gr.Textbox(
                label="Your Message:",
                placeholder="Type your message here...",
                show_label=True,
                elem_id="user-input"
            )
            send_button = gr.Button("Send", variant="primary")

        def chat_response(user_message, history):
            if not user_message:
                return history, history
            
            bot_response = chatbot.get_response(user_message)
            history.append((user_message, bot_response))
            return history, history

        send_button.click(
            chat_response,
            inputs=[user_input, chat_history],
            outputs=[chat_history, chat_history],
            api_name="chat"
        )
        
        # Clear the input textbox after sending
        send_button.click(lambda: "", None, user_input)
        
        # Also allow Enter key to send message
        user_input.submit(
            chat_response,
            inputs=[user_input, chat_history],
            outputs=[chat_history, chat_history],
        )
        user_input.submit(lambda: "", None, user_input)

    return chat_interface

if __name__ == "__main__":
    interface = create_gradio_interface()
    interface.launch(
        server_name="0.0.0.0",
        server_port=7860,
        share=True,
        enable_queue=True
    )