File size: 4,535 Bytes
cdf6cb1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
498ddeb
 
 
790c7af
 
 
 
 
 
 
 
 
 
 
 
 
 
498ddeb
 
ab440a4
 
498ddeb
dbb6c49
790c7af
 
 
 
 
ab440a4
790c7af
 
 
 
 
 
 
 
 
 
 
 
 
 
498ddeb
790c7af
22c11b2
 
 
 
 
 
 
 
 
3c0fc42
 
 
cdf6cb1
3c0fc42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
878d57a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# import dependencies
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, pipeline
import os
import gradio as gr
#from google.colab import drive
import chromadb
from langchain.llms import HuggingFacePipeline
from langchain.document_loaders import TextLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import Chroma
from langchain import HuggingFacePipeline
from langchain.document_loaders import PyPDFDirectoryLoader
from langchain.chains import ConversationalRetrievalChain
from langchain.memory import ConversationBufferMemory

#import locale
#locale.getpreferredencoding = lambda: "UTF-8"

# specify model huggingface mode name
model_name = "anakin87/zephyr-7b-alpha-sharded"
#https://huggingface.co/anakin87/zephyr-7b-alpha-sharded

#HuggingFaceH4/zephyr-7b-alpha
#https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha

# function for loading 4-bit quantized model
def load_quantized_model(model_name: str):
    """
    :param model_name: Name or path of the model to be loaded.
    :return: Loaded quantized model.
    """
    bnb_config = BitsAndBytesConfig(
        #load_in_4bit=True,
        load_in_4bit=False,
        #bnb_4bit_use_double_quant=True,
        bnb_4bit_use_double_quant=False,
        bnb_4bit_quant_type="nf4"
        #bnb_4bit_compute_dtype=torch.bfloat16
    )

    model = AutoModelForCausalLM.from_pretrained(
        model_name,
        load_in_4bit=True,
        #torch_dtype=torch.bfloat16,
        quantization_config=bnb_config
    )
    return model

# fucntion for initializing tokenizer
def initialize_tokenizer(model_name: str):
    """
    Initialize the tokenizer with the specified model_name.

    :param model_name: Name or path of the model for tokenizer initialization.
    :return: Initialized tokenizer.
    """
    tokenizer = AutoTokenizer.from_pretrained(model_name)
    tokenizer.bos_token_id = 1  # Set beginning of sentence token id
    return tokenizer    

# load model
model = load_quantized_model(model_name)

# initialize tokenizer
tokenizer = initialize_tokenizer(model_name)

# specify stop token ids
stop_token_ids = [0]    

# load pdf files
loader = PyPDFDirectoryLoader(pdf_files)
documents = loader.load()

# split the documents in small chunks
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100) #Chage the chunk_size and chunk_overlap as needed
all_splits = text_splitter.split_documents(documents)

# specify embedding model (using huggingface sentence transformer)
embedding_model_name = "sentence-transformers/all-mpnet-base-v2"
#model_kwargs = {"device": "cuda"}
#embeddings = HuggingFaceEmbeddings(model_name=embedding_model_name, model_kwargs=model_kwargs)
embeddings = HuggingFaceEmbeddings(model_name=embedding_model_name)

#embed document chunks
vectordb = Chroma.from_documents(documents=all_splits, embedding=embeddings, persist_directory="chroma_db")

# specify the retriever
retriever = vectordb.as_retriever()

# build huggingface pipeline for using zephyr-7b-alpha
pipeline = pipeline(
        "text-generation",
        model=model,
        tokenizer=tokenizer,
        use_cache=True,
        device_map="auto",
        max_length=2048,
        do_sample=True,
        top_k=5,
        num_return_sequences=1,
        eos_token_id=tokenizer.eos_token_id,
        pad_token_id=tokenizer.eos_token_id,
)

# specify the llm
llm = HuggingFacePipeline(pipeline=pipeline)

# build conversational retrieval chain with memory (rag) using langchain
def create_conversation(query: str, chat_history: list) -> tuple:
    try:

        memory = ConversationBufferMemory(
            memory_key='chat_history',
            return_messages=False
        )
        qa_chain = ConversationalRetrievalChain.from_llm(
            llm=llm,
            retriever=retriever,
            memory=memory,
            get_chat_history=lambda h: h,
        )

        result = qa_chain({'question': query, 'chat_history': chat_history})
        chat_history.append((query, result['answer']))
        return '', chat_history

    except Exception as e:
        chat_history.append((query, e))
        return '', chat_history

# build gradio ui
with gr.Blocks() as demo:
    
    chatbot = gr.Chatbot(label='Chat with your data (Zephyr 7B Alpha)')
    msg = gr.Textbox()
    clear = gr.ClearButton([msg, chatbot])

    msg.submit(create_conversation, [msg, chatbot], [msg, chatbot])

demo.launch()