Spaces:
Sleeping
Sleeping
File size: 1,997 Bytes
ac69024 9c3d50e ac69024 9a5450f 9c3d50e 9a5450f ac69024 9a5450f ac69024 9a5450f ac69024 9a5450f ac69024 9a5450f ac69024 9a5450f ac69024 9a5450f |
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 |
import streamlit as st
from transformers import BartTokenizerFast, BartForConditionalGeneration
@st.cache_resource
def load_model():
model_path = "bart_small_samsum" # model path
tokenizer = BartTokenizerFast.from_pretrained(model_path)
model = BartForConditionalGeneration.from_pretrained(model_path)
return tokenizer, model
# Set maximum lengths for input and target sequences
max_input_length = 128
max_target_length = 64
def summarize(input_text, tokenizer, model):
# Tokenize input text
inputs = tokenizer(input_text, return_tensors="pt", max_length=max_input_length, truncation=True)
# Generate summary
summary_ids = model.generate(
inputs["input_ids"],
max_length=max_target_length,
min_length=30,
length_penalty=2.0,
num_beams=4,
early_stopping=True
)
# Decode the generated summary
summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
return summary
# Streamlit app
st.title("Summarization Tool Using Bart-small Finetuned on Small sized Samsum Dataset")
# Load model
tokenizer, model = load_model()
# Text input
input_text = st.text_area("Enter your dialogue here:", height=200)
if st.button("Summarize"):
if input_text:
with st.spinner("Generating summary..."):
summary = summarize(input_text, tokenizer, model)
st.subheader("Summary:")
st.write(summary)
else:
st.warning("Please enter some text to summarize.")
# Add some information about the model
st.sidebar.header("About")
st.sidebar.info(
"This app uses a fine-tuned BART-Small model to summarize dialogues. "
"Enter your dialogue in the text area and click 'Summarize' to generate a summary."
)
# You can add more information or customization in the sidebar
st.sidebar.header("Model Details")
st.sidebar.text("Model: BART-small")
st.sidebar.text("Max Input Length: 128 tokens")
st.sidebar.text("Max Summary Length: 64 tokens") |