File size: 3,034 Bytes
f95dea8 c167fb9 898c536 f95dea8 c167fb9 f95dea8 c167fb9 f95dea8 c167fb9 f95dea8 c167fb9 f95dea8 |
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 |
import gradio as gr
from huggingface_hub import InferenceClient
import streamlit as st
from transformers import pipeline
from datasets import load_dataset
force_download=True
# Initialize the Hugging Face InferenceClient
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
# Initialize text-generation pipeline with the model for Streamlit
model_name = "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF"
pipe = pipeline("text-generation", model=model_name)
# Load the dataset
ds = load_dataset("refugee-law-lab/canadian-legal-data", "default", split="train")
# Gradio Function
def respond(
message,
history: list[tuple[str, str]],
system_message,
max_tokens,
temperature,
top_p,
):
messages = [{"role": "system", "content": system_message}]
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=max_tokens,
stream=True,
temperature=temperature,
top_p=top_p,
):
token = message.choices[0].delta.content
response += token
yield response
# Gradio interface setup
demo = gr.ChatInterface(
respond,
additional_inputs=[
gr.Textbox(value="You are a friendly Chatbot.", 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)",
),
],
)
# Streamlit interface setup
def streamlit_interface():
st.title("Canadian Legal Text Generator")
st.write("Enter a prompt related to Canadian legal data and generate text using Llama-3.1.")
# Show dataset sample
st.subheader("Sample Data from Canadian Legal Dataset:")
st.write(ds[:5]) # Displaying the first 5 rows of the dataset
# Prompt input
prompt = st.text_area("Enter your prompt:", placeholder="Type something...")
if st.button("Generate Response"):
if prompt:
# Generate text based on the prompt
with st.spinner("Generating response..."):
generated_text = pipe(prompt, max_length=100, do_sample=True, temperature=0.7)[0]["generated_text"]
st.write("**Generated Text:**")
st.write(generated_text)
else:
st.write("Please enter a prompt to generate a response.")
# Running Gradio and Streamlit interfaces
if __name__ == "__main__":
st.sidebar.title("Choose an Interface")
interface = st.sidebar.radio("Select", ("Streamlit", "Gradio"))
if interface == "Streamlit":
streamlit_interface()
else:
demo.launch()
|