Spaces:
Sleeping
Sleeping
File size: 2,512 Bytes
d8b357d 0e99818 d8b357d |
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 |
import streamlit as st
from huggingface_hub import InferenceClient
import os
client = InferenceClient("mistralai/Mistral-7B-Instruct-v0.3")
# Setze einen Standardwert, falls SECRET_PROMPT nicht gesetzt ist
secret_prompt = os.getenv("SECRET_PROMPT", "Default prompt: ")
def format_prompt(message, history):
prompt = secret_prompt
for user_prompt, bot_response in history:
prompt += f"[INST] {user_prompt} [/INST]"
prompt += f" {bot_response}</s> "
prompt += f"[INST] {message} [/INST]"
return prompt
def generate(prompt, history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0):
temperature = float(temperature)
if temperature < 1e-2:
temperature = 1e-2
top_p = float(top_p)
generate_kwargs = dict(
temperature=temperature,
max_new_tokens=max_new_tokens,
top_p=top_p,
repetition_penalty=repetition_penalty,
do_sample=True,
seed=42,
)
formatted_prompt = format_prompt(prompt, history)
stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
output = ""
for response in stream:
output += response.token.text
yield output
return output
st.title("Einfach.Mistral 7B v0.3")
history = []
with st.sidebar:
temperature = st.slider(
"Temperature",
value=0.9,
min_value=0.0,
max_value=1.0,
step=0.05,
help="Higher values produce more diverse outputs",
)
max_new_tokens = st.slider(
"Max new tokens",
value=256,
min_value=0,
max_value=1048,
step=64,
help="The maximum numbers of new tokens",
)
top_p = st.slider(
"Top-p (nucleus sampling)",
value=0.90,
min_value=0.0,
max_value=1.0,
step=0.05,
help="Higher values sample more low-probability tokens",
)
repetition_penalty = st.slider(
"Repetition penalty",
value=1.2,
min_value=1.0,
max_value=2.0,
step=0.05,
help="Penalize repeated tokens",
)
message = st.text_input("Your message:", "")
if st.button("Generate"):
if message:
for output in generate(message, history, temperature, max_new_tokens, top_p, repetition_penalty):
st.text_area("Generated Text", value=output, height=400)
history.append((message, output))
else:
st.warning("Please enter a message.")
|