Update app.py

app.py

@@ -1,64 +1,51 @@
-import gradio as gr
-from huggingface_hub import InferenceClient
+import warnings
+warnings.filterwarnings("ignore")
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+import logging
+logging.getLogger("streamlit").setLevel(logging.ERROR)
 
+import streamlit as st
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+@st.cache_resource
+def load_model():
+    model_name = "radlab/polish-gpt2-small-v2"
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    model = AutoModelForCausalLM.from_pretrained(model_name)
+    return tokenizer, model
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+tokenizer, model = load_model()
 
-    messages.append({"role": "user", "content": message})
+st.set_page_config(page_title="Polski Chatbot AI", page_icon="🤖")
+st.title("🤖 Polski Chatbot AI")
+st.caption("Model: radlab/polish-gpt2-small-v2")
 
-    response = ""
+if "history" not in st.session_state:
+    st.session_state.history = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+user_input = st.text_input("Wpisz wiadomość:", "")
 
-        response += token
-        yield response
+if st.button("Wyślij") and user_input.strip() != "":
+    st.session_state.history += f"Użytkownik: {user_input}\nAI:"
 
+    input_ids = tokenizer.encode(st.session_state.history, return_tensors="pt", truncation=True, max_length=1024)
+    output = model.generate(
+        input_ids,
+        max_length=input_ids.shape[1] + 80,
+        pad_token_id=tokenizer.eos_token_id,
+        do_sample=True,
+        top_k=50,
+        top_p=0.95,
+        temperature=0.7
+    )
+    output_text = tokenizer.decode(output[0], skip_special_tokens=True)
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+    model_reply = output_text[len(st.session_state.history):].split("Użytkownik:")[0].strip()
+    st.session_state.history += f" {model_reply}\n"
 
+st.subheader("🗨️ Historia rozmów")
+st.text_area("📖", st.session_state.history.strip(), height=300)
 
-if __name__ == "__main__":
-    demo.launch()
+if st.button("🧹 Wyczyść historię"):
+    st.session_state.history = ""