app.py

app.py

@@ -0,0 +1,70 @@
+import os
+import streamlit as st
+from PyPDF2 import PdfReader
+from langchain.text_splitter import CharacterTextSplitter
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.chains.question_answering import load_qa_chain
+import random
+from langchain import HuggingFaceHub
+from langchain.callbacks import get_openai_callback
+
+def main():
+    # ---------------------------- created personal API -----------------------------
+    os.environ["HUGGINGFACEHUB_API_TOKEN"] = "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"
+
+    # ------------------ Designing Page ---------------
+    st.set_page_config(page_title="Ask Your PDF")
+    st.header("Ask your PDF :")
+
+    pdf = st.file_uploader("Upload your File here", type="pdf")
+
+    # Check Pdf
+    if pdf is not None:
+        pdf_reader = PdfReader(pdf)
+
+        text = ""
+
+        # Extract pages from pdf
+        for page in pdf_reader.pages:
+            text += page.extract_text()
+
+        # split into chunks
+        text_spliter = CharacterTextSplitter(
+            separator="\n",
+            chunk_size=1000,
+            chunk_overlap=0,
+            length_function=len
+        )
+
+        chunks = text_spliter.split_text(text)
+
+        # create embeddings
+        embedding = HuggingFaceEmbeddings()
+        knowledge_base = FAISS.from_texts(chunks, embedding)
+
+        user_questions = st.text_input("Ask a Question from PDF : ")
+        if user_questions:
+
+            greeting = ["hy", 'hello', 'hey', "hi"]
+            greet_msg = ["Hello Dear!", 'Hey!', 'Hey Friend!']
+
+            if user_questions in greeting:
+                response = random.choice(greet_msg)
+            elif user_questions == "by" or user_questions == "bye":
+                response = "GoodBye Sir!, Have a Nice Day....."
+            else:
+
+                docs = knowledge_base.similarity_search(user_questions)
+
+                chain = load_qa_chain(HuggingFaceHub(repo_id="google/flan-t5-xxl", model_kwargs={"temperature":0.1, "max_length":512}), chain_type="stuff")
+                with get_openai_callback() as cb:
+
+                    response = chain.run(input_documents=docs, question=user_questions)
+                    print(cb)
+
+            st.write(response)
+
+
+if __name__ == "__main__":
+    main()