Update app.py

app.py

@@ -1,38 +1,49 @@
 import streamlit as st
 import transformers
-from transformers import pipeline, AutoTokenizer, AutoModelForSeq2SeqLM
+from transformers import pipeline
+from keybert import KeyBERT
+
+# Load summarizer and keyword model
+summarizer_bart = pipeline("summarization", model="facebook/bart-large-cnn")
+kw_model = KeyBERT()
 
 # Title
-st.markdown("<h1 style='text-align: center; color: black;'>Text Summarization App</h1>", unsafe_allow_html=True)
-st.markdown("---")
-
-# Model Selection
-model_choice = st.selectbox("Select a Summarization Model", ["BART", "T5", "PEGASUS"])
-
-# Load model and tokenizer
-@st.cache_resource
-def load_model(model_name):
-    tokenizer = AutoTokenizer.from_pretrained(model_name)
-    model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
-    return pipeline("summarization", model=model, tokenizer=tokenizer)
-
-# Map choices to model names or paths
-model_map = {
-    "BART": "facebook/bart-large-cnn",
-    "T5": "t5-small",
-    "PEGASUS": "google/pegasus-cnn_dailymail"
-}
-
-# Text Input
-text_input = st.text_area("Enter the text you want to summarize:", height=300)
-
-# Button to generate summary
-if st.button("Summarize"):
-    if not text_input.strip():
-        st.warning("Please enter some text!")
+st.markdown("<h1 style='text-align: center;'>Text Summarization App</h1>", unsafe_allow_html=True)
+
+# Modes
+mode = st.radio("Modes", ["Paragraph", "Bullet Points", "Custom"], horizontal=True)
+
+# Summary length
+length = st.slider("Summary Length", 1, 2, 1, label_visibility="collapsed")
+summary_type = "Short" if length == 1 else "Long"
+st.write(f"Summary Length: **{summary_type}**")
+
+# Layout
+col1, col2 = st.columns(2)
+
+with col1:
+    st.markdown("### Paste your text below:")
+    user_input = st.text_area("", height=300, placeholder="Paste job description or any paragraph...")
+    
+    if user_input.strip():
+        # Extract keywords using KeyBERT
+        keywords = [kw[0] for kw in kw_model.extract_keywords(user_input, top_n=5)]
+        selected_keywords = st.multiselect("Select Keywords", keywords, default=keywords)
+        st.markdown(f"**{len(user_input.split())} words**")
+
+        if st.button("Summarize", use_container_width=True):
+            summary = summarizer_bart(user_input, max_length=150 if summary_type == "Short" else 250,
+                                      min_length=40, do_sample=False)[0]['summary_text']
+            st.session_state['summary'] = summary
+    else:
+        st.warning("Please enter some text.")
+
+with col2:
+    st.markdown("### Summary")
+    if 'summary' in st.session_state:
+        st.success(st.session_state['summary'])
+        st.markdown(f"1 sentence • {len(st.session_state['summary'].split())} words")
+        st.button("Paraphrase Summary")
+        st.download_button("📥 Download Summary", st.session_state['summary'], file_name="summary.txt")
     else:
-        summarizer = load_model(model_map[model_choice])
-        summary = summarizer(text_input, max_length=150, min_length=40, do_sample=False)
-        
-        st.markdown("### Summary:")
-        st.success(summary[0]['summary_text'])
+        st.info("Your summary will appear here after clicking **Summarize**.")