Update app.py

app.py

@@ -1,54 +1,57 @@
-import os
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
-from transformers import AutoProcessor, Gemma3ForConditionalGeneration, TextIteratorStreamer
-import os
-from huggingface_hub import login
-import os
 
-login(token=os.getenv("hf_token") )
-
-
-
-model_id = os.getenv("MODEL_ID", "rahul7star/gemma-3bit")
-processor = AutoProcessor.from_pretrained(model_id, padding_side="left")
-model = Gemma3ForConditionalGeneration.from_pretrained(
-    model_id, device_map="auto", torch_dtype=torch.bfloat16, attn_implementation="eager",cache_dir = "F:\\huggingface_cache"
+# Model ve tokenizer yükleme
+model_name = "rahul7star/gemma-3bit"  # veya "google/gemma-3-1b-it"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.float16,
+    device_map="auto"
 )
 
-def run_fn(message):
-    messages_list = []
-    '''
-    conversation = [
-                {
-                    "role": "user",
-                    "content": [
-                        {"type": "image", "image": "https://www.ilankelman.org/stopsigns/australia.jpg"},
-                        {"type": "text", "text": "Please describe this image in detail."},
-                    ],
-                },
-            ]
-    '''
-    messages_list.append({"role": "user", "content":[{ "type":"text","text": message}] })
+# Sistem promptu
+system_prompt = """"Enhance and expand the following prompt with more details and context:."""
 
-    inputs = processor.apply_chat_template(
-        messages_list,
-        add_generation_prompt=True,
-        tokenize=True,
-        return_dict=True,
-        return_tensors="pt",
-    ).to(device=model.device, dtype=torch.bfloat16)
-    streamer = TextIteratorStreamer(processor, timeout=30.0, skip_prompt=True, skip_special_tokens=True)
-    max_new_tokens = 100
-    generate_kwargs = dict(
+def predict(message, history):
+    # Sohbet geçmişini formatlama
+    messages = [{"role": "system", "content": system_prompt}]
+    
+    # Önceki mesajları ekleme
+    for user_msg, bot_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": bot_msg})
+    
+    # Yeni mesajı ekleme
+    messages.append({"role": "user", "content": message})
+    
+    # Tokenize etme
+    inputs = tokenizer.apply_chat_template(
+        messages, 
+        return_tensors="pt"
+    ).to(model.device)
+    
+    # Yanıt üretme
+    outputs = model.generate(
         inputs,
-        streamer=streamer,
-        max_new_tokens=max_new_tokens,
+        max_new_tokens=512,
+        temperature=0.7,
+        top_p=0.95,
+        do_sample=True
     )
-    outputs = model.generate(**generate_kwargs)
-    return outputs
-    # return None
-def greet(name):
-    return run_fn(name)
+    
+    # Yanıtı decode etme
+    response = tokenizer.decode(outputs[0][inputs.shape[1]:], skip_special_tokens=True)
+    
+    return response
+
+# Gradio arayüzü
+demo = gr.ChatInterface(
+    fn=predict,
+    title="Sağlık Danışmanı Chatbot (Gemma 3)",
+    description="Bu chatbot, sağlık, beslenme ve egzersiz konularında bilgi vermek için tasarlanmıştır. Tıbbi teşhis koymaz veya reçete önermez.",
+    theme="soft"
+)
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
 demo.launch()