Upload app.py

app.py

@@ -0,0 +1,88 @@
+
+import fitz  # PyMuPDF
+import os
+import numpy as np
+import faiss
+from sentence_transformers import SentenceTransformer
+import gradio as gr
+
+def load_pdf(file_path):
+    doc = fitz.open(file_path)
+    return "\n".join(page.get_text() for page in doc)
+
+def split_into_chunks(text, chunk_size=500):
+    words = text.split()
+    return [" ".join(words[i:i+chunk_size]) for i in range(0, len(words), chunk_size)]
+
+def create_index(chunks):
+    model = SentenceTransformer('all-MiniLM-L6-v2')
+    embeddings = model.encode(chunks)
+    index = faiss.IndexFlatL2(embeddings.shape[1])
+    index.add(np.array(embeddings))
+    return index, embeddings, chunks, model
+
+def ask_question(query, index, embeddings, chunks, model, top_k=3, context_window=1):
+    query_emb = model.encode([query])
+    distances, indices = index.search(np.array(query_emb), top_k)
+
+    context_chunks = set()
+    for i in indices[0]:
+        for j in range(max(0, i - context_window), min(len(chunks), i + context_window + 1)):
+            context_chunks.add(j)
+
+    sorted_indices = sorted(context_chunks)
+    return "\n\n".join([chunks[i] for i in sorted_indices])
+
+def generate_simple_flashcards(chunks, max_flashcards=10):
+    flashcards = []
+    for chunk in chunks:
+        sentences = chunk.split('. ')
+        for s in sentences:
+            if " is " in s and len(s.split()) < 25:
+                parts = s.split(" is ", 1)
+                q = f"What is {parts[0].strip()}?"
+                a = parts[1].strip().rstrip('.')
+                flashcards.append((q, a))
+                if len(flashcards) >= max_flashcards:
+                    return flashcards
+    return flashcards
+
+state = {"index": None, "embeddings": None, "chunks": [], "model": None}
+
+def process_pdfs(files):
+    all_chunks = []
+    for file in files:
+        text = load_pdf(file.name)
+        chunks = split_into_chunks(text)
+        all_chunks.extend(chunks)
+    if all_chunks:
+        index, emb, chks, model = create_index(all_chunks)
+        state.update({"index": index, "embeddings": emb, "chunks": chks, "model": model})
+        return f"Processed {len(files)} PDF(s)."
+    else:
+        return "No text extracted."
+
+def ask(query):
+    return ask_question(query, state["index"], state["embeddings"], state["chunks"], state["model"])
+
+def flashcards():
+    pairs = generate_simple_flashcards(state["chunks"])
+    return "\n\n".join([f"Q: {q}\nA: {a}" for q, a in pairs])
+
+with gr.Blocks() as demo:
+    gr.Markdown("## 📘 AI Revision Assistant (Free Version) – Ask, Review, Quiz Yourself")
+    with gr.Row():
+        file_input = gr.File(label="Upload PDFs", file_count="multiple")
+        status = gr.Textbox(label="Status")
+    file_input.change(process_pdfs, inputs=file_input, outputs=status)
+
+    with gr.Tab("Ask a Question"):
+        qbox = gr.Textbox(label="Your Question")
+        abox = gr.Textbox(label="Answer")
+        qbox.submit(ask, inputs=qbox, outputs=abox)
+
+    with gr.Tab("Simple Flashcards"):
+        fbox = gr.Textbox(label="Auto-Generated Flashcards")
+        gr.Button("Generate").click(flashcards, outputs=fbox)
+
+demo.launch()