Create app.py

app.py

@@ -0,0 +1,46 @@
+import streamlit as st
+from PyPDF2 import PdfReader
+from transformers import pipeline
+
+# Zero-shot classification pipeline
+@st.cache_resource
+def load_classifier():
+    return pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
+
+# Function to extract text from PDF
+def extract_text_from_pdf(pdf_file):
+    reader = PdfReader(pdf_file)
+    text = ""
+    for page in reader.pages:
+        text += page.extract_text()
+    return text
+
+# Streamlit app UI
+def main():
+    st.title("PDF Genre Classifier")
+    st.write("Upload a PDF file, and this app will classify its genres using zero-shot classification.")
+
+    # File uploader
+    pdf_file = st.file_uploader("Upload PDF", type=["pdf"])
+    if pdf_file is not None:
+        st.write("Processing the PDF...")
+        text = extract_text_from_pdf(pdf_file)
+
+        if text.strip():
+            st.write("PDF Text Extracted. Performing Genre Classification...")
+            classifier = load_classifier()
+
+            # Define candidate genres
+            candidate_labels =["Romance", "Mystery", "Thriller", "Science Fiction", "Fantasy", "Horror", "Historical Fiction", "Crime", "Western", "Dystopian", "Biography", "Autobiography", "Memoir", "History", "Self-Help", "Travel", "Essay", "Journalism", "Sonnet", "Haiku", "Free Verse", "Narrative Poetry", "Lyric Poetry", "Tragedy", "Comedy", "Melodrama", "Farce", "Graphic Novel", "Epistolary", "Magical Realism", "Satire", Young Adult Fiction"]
+
+            # Perform zero-shot classification
+            result = classifier(text),#[:1000], candidate_labels, multi_label=True)  # Using the first 1000 characters
+            genres = sorted(zip(result["labels"], result["scores"]), key=lambda x: x[1], reverse=True)
+
+            st.subheader("Top 20 Detected Genres:")
+            top_genres = genres[:20]  # Get the top 20 genres
+            for genre, score in top_genres:
+                st.write(f"**{genre.capitalize()}**: {score:.2f}")
+        else:
+            st.error("No text could be extracted from the PDF. Please try another file.")
+