Spaces:
Sleeping
Sleeping
File size: 4,711 Bytes
9614f6b d6f366b 8f6f616 d6f366b f6d009c d6f366b a11cce5 a37d6d4 a11cce5 d6f366b 4ecbcef d6f366b 7d2deb9 d6f366b 078dec5 d6f366b a11cce5 d6f366b |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 |
print(9)
import argparse
# from dataclasses import dataclass
from langchain.prompts import ChatPromptTemplate
try:
from langchain_community.vectorstores import Chroma
except:
from langchain_community.vectorstores import Chroma
# from langchain.document_loaders import DirectoryLoader
from langchain_community.document_loaders import DirectoryLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.schema import Document
# from langchain.embeddings import OpenAIEmbeddings
#from langchain_openai import OpenAIEmbeddings
from langchain_community.vectorstores import Chroma
import openai
from dotenv import load_dotenv
import os
import shutil
import torch
from langchain_experimental.text_splitter import SemanticChunker
from typing import List
import re
import warnings
from typing import List
import torch
from langchain import PromptTemplate
from langchain.chains import ConversationChain
from langchain.chains.conversation.memory import ConversationBufferWindowMemory
from langchain.llms import HuggingFacePipeline
from langchain.schema import BaseOutputParser
from transformers import (
AutoModelForCausalLM,
AutoTokenizer,
StoppingCriteria,
StoppingCriteriaList,
pipeline,
)
import subprocess
import sys
def install(package):
subprocess.check_call([sys.executable, "-m", "pip", "install", package])
install('accelerate')
MODEL_NAME = "tiiuae/falcon-7b-instruct"
llama_pipeline = pipeline(
"text-generation",
model=MODEL_NAME,
torch_dtype=torch.float16,
device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
from transformers import AutoModel,AutoTokenizer
model2 = AutoModel.from_pretrained("sentence-transformers/all-MiniLM-L6-v2")
tokenizer2 = AutoTokenizer.from_pretrained("sentence-transformers/all-MiniLM-L6-v2")
# this shoub be used when we can not use sentence_transformers (which reqiures transformers==4.39. we cannot use
# this version since causes using large amount of RAm when loading falcon model)
# a custom embedding
#from sentence_transformers import SentenceTransformer
warnings.filterwarnings("ignore", category=UserWarning)
class MyEmbeddings:
def __init__(self):
#self.model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
self.model=model2
def embed_documents(self, texts: List[str]) -> List[List[float]]:
inputs = tokenizer2(texts, padding=True, truncation=True, return_tensors="pt")
# Get the model outputs
with torch.no_grad():
outputs = self.model(**inputs)
# Mean pooling to get sentence embeddings
embeddings = outputs.last_hidden_state.mean(dim=1)
return [embeddings[i].tolist() for i, sentence in enumerate(texts)]
def embed_query(self, query: str) -> List[float]:
inputs = tokenizer2(query, padding=True, truncation=True, return_tensors="pt")
# Get the model outputs
with torch.no_grad():
outputs = self.model(**inputs)
# Mean pooling to get sentence embeddings
embeddings = outputs.last_hidden_state.mean(dim=1)
return embeddings[0].tolist()
embeddings = MyEmbeddings()
splitter = SemanticChunker(embeddings)
CHROMA_PATH = "chroma8"
# call the chroma generated in a directory
db = Chroma(persist_directory=CHROMA_PATH, embedding_function=embeddings)
prompt = """
The following is a friendly conversation between a human and an AI. The AI is talkative and provides lots of specific details from its context.
Current conversation:
Human: Who is Dwight K Schrute?
AI:
""".strip()
template = """
The following
Current conversation:
{history}
Human: {input}
AI:""".strip()
def get_llama_response(message: str, history: list) -> str:
query_text = message
results = db.similarity_search_with_relevance_scores(query_text, k=3)
if len(results) == 0 or results[0][1] < 0.5:
print(f"Unable to find matching results.")
context_text = "\n\n---\n\n".join([doc.page_content for doc, _score in results])
query = """
Answer the question based only on the following context. Dont provide any information out of the context:
{context}
---
Answer the question based on the above context: {question}
"""
query=query.format(context=context_text,question=message)
sequences = llama_pipeline(
query,
do_sample=True,
top_k=10,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
max_length=1024,
)
generated_text = sequences[0]['generated_text']
response = generated_text[len(query):]
return response.strip()
import gradio as gr
gr.ChatInterface(get_llama_response).launch() |