Update app.py

app.py

@@ -1,33 +1,27 @@
 from transformers import AutoTokenizer, AutoModelForSequenceClassification
-import torch
 
-# Define the model path
 MODEL_NAME = "RPW/NIH-1.2_Llama-3.2-11B-Vision-Instruct"
-device = "cuda" if torch.cuda.is_available() else "cpu"
 
-instruction = "You are an expert radiographer. Describe accurately what you see in this image."
-
-# Load the tokenizer and model
+# โหลด tokenizer
 tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
-model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME).to(device)
 
-# Example of messages
-messages = [{"role": "system", "content": instruction}]
+# ตั้งค่า special tokens
+special_tokens = {
+    "bos_token": "<|begin_of_text|>",
+    "eos_token": "<|end_of_text|>",
+    "additional_special_tokens": [
+        "<|reserved_special_token_0|>",
+        "<|reserved_special_token_1|>",
+        "<|finetune_right_pad_id|>",
+        "<|step_id|>",
+        "<|start_header_id|>",
+        "<|end_header_id|>"
+    ]
+}
+tokenizer.add_special_tokens(special_tokens)
 
-input_text = tokenizer.apply_chat_template(messages, add_generation_prompt=True)
-inputs = tokenizer(
-    image,
-    input_text,
-    add_special_tokens=False,
-    return_tensors="pt"
-).to(device)
+# โหลด model
+model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME)
 
-text_streamer = TextStreamer(tokenizer, skip_prompt=True)
-_ = model.generate(
-    **inputs,
-    streamer=text_streamer,
-    max_new_tokens=128,
-    use_cache=True,
-    temperature=1.5,
-    min_p=0.1
-)
+# ปรับขนาด embedding เพื่อรองรับ special tokens
+model.resize_token_embeddings(len(tokenizer))