File size: 2,032 Bytes
5ece497 13c38bc 491e2bb 93ce35f 1269210 367c791 1269210 93ce35f 1269210 b30e24b 796bfa0 1269210 93ce35f 1269210 b0e5c12 1269210 75c6368 1269210 93ce35f eb04e6a 1269210 c8de58e 2aba8e9 1269210 93ce35f 1269210 8e8feea 1269210 |
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 |
import gradio as gr
from huggingface_hub import InferenceClient
import os
import pandas as pd
client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1", token=os.getenv("HF_TOKEN"))
def load_prompts():
prompts = pd.read_csv("prompts.csv")
return prompts
def respond(
message,
history,
systemmessage,
maxtokens,
temperature,
top_p,
prompts,
):
messages = [{"role": "system", "content": systemmessage}]
for val in history:
if val[0]:
messages.append({"role": "user", "content": val[0]})
if val[1]:
messages.append({"role": "assistant", "content": val[1]})
messages.append({"role": "user", "content": message})
response = ""
for message in client.chat_completion(
messages,
max_tokens=maxtokens,
stream=True,
temperature=temperature,
top_p=top_p,
):
token = message.choices[0].delta.content
response += token
yield response
prompts = load_prompts()
demo = gr.ChatInterface(
respond,
inputs=[
gr.Textbox(value="λ°λμ νκΈλ‘ λ΅λ³νλΌ. λμ μ΄λ¦μ 'νκΈλ‘'μ
λλ€. μΆλ ₯μ markdown νμμΌλ‘ μΆλ ₯νλ©° νκΈ(νκ΅μ΄)λ‘ μΆλ ₯λκ² νκ³ νμνλ©΄ μΆλ ₯λ¬Έμ νκΈλ‘ λ²μνμ¬ μΆλ ₯νλΌ. λλ νμ μΉμ νκ³ μμΈνκ² λ΅λ³μ νλΌ. λλ λν μμμ μλλ°©μ μ΄λ¦μ λ¬Όμ΄λ³΄κ³ νΈμΉμ 'μΉκ΅¬'μ μ¬μ©ν κ². λ°λμ νκΈλ‘ λ 'λ°λ§'λ‘ λ΅λ³ν κ². λλ Assistant μν μ μΆ©μ€νμ¬μΌ νλ€. λ", label="System message"),
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
gr.Slider(
minimum=0.1,
maximum=1.0,
value=0.95,
step=0.05,
label="Top-p (nucleus sampling)",
),
],
outputs="text",
)
if __name__ == "__main__":
demo.launch() |