Spaces:
Sleeping
Sleeping
File size: 2,913 Bytes
4eeec68 146db4f 4eeec68 146db4f 9eb3567 6919ed6 146db4f 4eeec68 146db4f 4eeec68 9eb3567 4eeec68 9eb3567 4eeec68 9eb3567 4eeec68 9eb3567 4eeec68 9eb3567 4eeec68 9eb3567 4eeec68 9eb3567 4eeec68 9eb3567 4eeec68 |
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 |
# import gradio as gr
# from langchain_community.llms import LlamaCpp
# from langchain.prompts import PromptTemplate
# from langchain.chains import LLMChain
# from langchain_core.callbacks import StreamingStdOutCallbackHandler
# from langchain.retrievers import TFIDFRetriever
# from langchain.chains import RetrievalQA
# from langchain.memory import ConversationBufferMemory
# from langchain_community.chat_models import ChatLlamaCpp
# callbacks = [StreamingStdOutCallbackHandler()]
# print("creating ll started")
# llm = ChatLlamaCpp(
# model_path="finbro-v0.1.0-llama-3-8B-instruct-1m.gguf",
# n_batch=8,
# temperature=0.85,
# max_tokens=256,
# top_p=0.95,
# top_k = 10,
# callback_manager=callbacks,
# n_ctx=2048,
# verbose=True, # Verbose is required to pass to the callback manager
# )
# print("creating llm ended")
# def greet(question, model_type):
# print(f"question is {question}")
# out_gen = "testsetestestetsetsets"
# return out_gen
# demo = gr.Interface(fn=greet, inputs=["text", gr.Dropdown(
# ["With memory", "Without memory"], label="Memory status", info="With using memory, the output will be slow but strong"
# ),], outputs="text")
# demo.launch(debug=True, share=True)
import gradio as gr
from langchain_community.llms import LlamaCpp
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain
from langchain_core.callbacks import StreamingStdOutCallbackHandler
from langchain.retrievers import TFIDFRetriever
from langchain.chains import RetrievalQA
from langchain.memory import ConversationBufferMemory
from langchain_community.chat_models import ChatLlamaCpp
callbacks = [StreamingStdOutCallbackHandler()]
print("creating ll started")
M_NAME = "finbro-v0.1.0-llama-3-8B-instruct-1m.gguf"
llm = ChatLlamaCpp(
model_path=M_NAME,
n_batch=8,
temperature=0.85,
max_tokens=256,
top_p=0.95,
top_k = 10,
callback_manager=callbacks,
n_ctx=2048,
verbose=True, # Verbose is required to pass to the callback manager
)
# print("creating ll ended")
def greet(question, model_type):
print("prompt started ")
print(f"question is {question}")
template = """You are the Finiantial expert:
### Instruction:
{question}
### Input:
### Response:
"""
print("test1")
prompt = PromptTemplate(template=template, input_variables=["question"])
print("test2")
llm_chain_model = LLMChain(prompt=prompt, llm=llm)
print("test3")
out_gen = llm_chain_model.run(question)
print("test4")
print(f"out is: {out_gen}")
return out_gen
demo = gr.Interface(fn=greet, inputs=["text", gr.Dropdown(
["Without memory", "With memory"], label="Memory status", info="With using memory, the output will be slow but strong"
),], outputs="text")
demo.launch(debug=True, share=True) |