Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,7 +1,37 @@
+import os
+
 os.environ["MPLCONFIGDIR"] = "/tmp"
 os.environ["STREAMLIT_BROWSER_GATHER_USAGE_STATS"] = "false"
 os.environ["STREAMLIT_SERVER_HEADLESS"] = "true"
 os.environ["STREAMLIT_SERVER_ENABLE_FILE_WATCHER"] = "false"
 os.environ["STREAMLIT_CONFIG_DIR"] = "/tmp/.streamlit"
 os.environ["HF_HOME"] = "/tmp/huggingface"
-os.environ["TRANSFORMERS_CACHE"] = "/tmp/huggingface"
+os.environ["TRANSFORMERS_CACHE"] = "/tmp/huggingface"
+
+import streamlit as st
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# App config and title
+st.set_page_config(page_title="DeepSeek-R1 Chatbot", page_icon="🤖")
+st.title("🧠 DeepSeek-R1 CPU Chatbot")
+st.caption("Running entirely on CPU using Hugging Face Transformers")
+
+@st.cache_resource
+def load_model():
+    tokenizer = AutoTokenizer.from_pretrained("deepseek-ai/DeepSeek-Coder-1.3B-base")
+    model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-Coder-1.3B-base")
+    return tokenizer, model
+
+tokenizer, model = load_model()
+
+user_input = st.text_area("📥 Enter your prompt here:", "Explain what a neural network is.")
+
+if st.button("🧠 Generate Response"):
+    with st.spinner("Thinking..."):
+        inputs = tokenizer(user_input, return_tensors="pt")
+        outputs = model.generate(**inputs, max_new_tokens=100)
+        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+        st.markdown("### 🤖 Response:")
+        st.write(response)