Upload app.py.py

app.py.py

@@ -0,0 +1,52 @@
+import streamlit as st
+from langchain_google_genai import ChatGoogleGenerativeAI  # Correct Import
+from dotenv import load_dotenv
+import os
+
+# Load API key from .env file
+load_dotenv()
+API_KEY = os.getenv("GEMINI_API_KEY")
+
+# Initialize LangChain's ChatGoogleGenerativeAI
+llm = ChatGoogleGenerativeAI(model="gemini-1.5-pro", google_api_key=API_KEY)
+
+# Function to process and format the AI output
+def format_output(text):
+    """Refine AI-generated text into a structured format."""
+    sections = text.split("**option")  # Split into different options
+    formatted_text = ""
+
+    for section in sections:
+        if section.strip():
+            formatted_text += f"🟢 **Option {section.strip()}**\n\n"  # Highlight each option
+    return formatted_text if formatted_text else "No output generated. Try again."
+
+# Function to generate AI content using LangChain
+def generate_marketing_content(prompt):
+    """Generate marketing content using LangChain's ChatGoogleGenerativeAI."""
+    try:
+        response = llm.invoke(prompt)  # Use invoke() instead of predict()
+        return format_output(response.content)  # Extract text correctly
+    except Exception as e:
+        return f"❌ Error: {str(e)}"
+
+# Streamlit UI
+st.set_page_config(page_title="AI Marketing Generator", layout="wide")
+st.title("🚀 AI Marketing Generator")
+st.write("Generate high-quality marketing slogans, ad copy, and campaign ideas instantly!")
+
+# User Inputs
+category = st.selectbox("Select Content Type", ["Slogan", "Ad Copy", "Campaign Idea"])
+tone = st.selectbox("Select Tone", ["Energetic ⚡", "Professional 👔", "Fun 🎉"])
+product = st.text_input("Enter Product Name")
+
+if st.button("🎯 Generate Marketing Content"):
+    if product:
+        prompt = f"Generate a {tone.split()[0].lower()} {category.lower()} for the following product: {product}."
+        output = generate_marketing_content(prompt)
+        
+        st.success("✅ Content Generated Successfully!")
+        st.subheader("📌 **Generated Content**:")
+        st.markdown(output, unsafe_allow_html=True)
+    else:
+        st.warning("⚠️ Please enter a product description.")