Createapp.py

app.py

@@ -0,0 +1,38 @@
+import torch
+import streamlit as st
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Load the model and tokenizer
+model_name = "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF"
+model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+# Set up the Streamlit app interface
+st.title("LLM Text Classifier")
+st.write("Enter text below, and the language model will classify or respond to it.")
+
+# Take user input
+prompt = st.text_input("Enter your text here:")
+
+if st.button("Generate Response"):
+    # Check if input is provided
+    if prompt:
+        with st.spinner("Generating response..."):
+            # Tokenize input
+            tokenized_prompt = tokenizer(prompt, return_tensors="pt")
+            input_ids = tokenized_prompt['input_ids'].cuda() if torch.cuda.is_available() else tokenized_prompt['input_ids']
+            attention_mask = tokenized_prompt['attention_mask'].cuda() if torch.cuda.is_available() else tokenized_prompt['attention_mask']
+
+            # Generate the model's response
+            response_token_ids = model.generate(
+                input_ids,
+                attention_mask=attention_mask,
+                max_new_tokens=50,
+                pad_token_id=tokenizer.eos_token_id
+            )
+
+            # Decode and display the generated text
+            generated_text = tokenizer.decode(response_token_ids[0], skip_special_tokens=True)
+            st.write("**Model's Response:**", generated_text)
+    else:
+        st.warning("Please enter some text before generating a response.")