create app.py

app.py

@@ -0,0 +1,46 @@
+import streamlit as st
+from transformers import T5Tokenizer, T5ForConditionalGeneration
+import huggingface_hub
+import torch
+
+# Login to Hugging Face Hub
+# huggingface_hub.login("")
+
+# Load model and tokenizer
+repo_id = "Darshan03/t5-model-small"  # Replace with your hub username and model name
+tokenizer = T5Tokenizer.from_pretrained(repo_id)
+model = T5ForConditionalGeneration.from_pretrained(repo_id)
+
+# Function to generate headline
+def generate_headline(text, model, tokenizer):
+    MAX_TOKEN_LEN = 256
+    device = "cuda" if torch.cuda.is_available() else "cpu"  # Check for GPU availability
+    model = model.to(device)  # Move model to the device
+
+    input_ids = tokenizer(
+        text, return_tensors="pt", padding="max_length", truncation=True, max_length=MAX_TOKEN_LEN
+    ).input_ids.to(device)  # Move input ids to the device
+
+    with torch.no_grad():  # Inference doesn't need gradients
+        outputs = model.generate(input_ids, max_length=128, num_beams=4, early_stopping=True)
+
+    headline = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return headline
+
+# Streamlit UI
+st.title("T5 Headline Generator")
+
+# Input text box
+input_text = st.text_area("Enter your text:")
+
+if st.button("Generate Headline"):
+    if input_text:
+        headline = generate_headline(input_text, model, tokenizer)
+        st.subheader("Generated Headline:")
+        st.write(headline)
+    else:
+        st.error("Please enter some text.")
+
+# Refresh button
+if st.button("Refresh"):
+    st.experimental_rerun()