Update Dockerfile

Dockerfile

@@ -131,41 +131,4 @@ ENV MALLOC_TRIM_THRESHOLD_=0
 
 # Command to run the app
 CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "5000", "--timeout-keep-alive", "300"]
-EOF
-
-And here's the updated main.py initialization code:
-
-<antArtifact identifier="updated-init" type="application/vnd.ant.code" language="python" title="Updated PlantChatbot initialization">
-def __init__(self, preprocessed_data_path: str):
-    self.device = "cuda" if torch.cuda.is_available() else "cpu"
-    
-    print("Loading sentence transformer model...")
-    self.embedding_model = SentenceTransformer(
-        model_name_or_path="/app/models/sentence_transformer",
-        device=self.device
-    )
-    
-    print("Loading data...")
-    self.load_data(preprocessed_data_path)
-    
-    print("Loading LLM tokenizer...")
-    self.tokenizer = AutoTokenizer.from_pretrained(
-        "/app/models/llm",
-        trust_remote_code=True,
-        local_files_only=True
-    )
-    
-    print("Loading LLM model...")
-    self.model = AutoModelForCausalLM.from_pretrained(
-        "/app/models/llm",
-        device_map="auto" if torch.cuda.is_available() else None,
-        trust_remote_code=True,
-        torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-        local_files_only=True,
-        low_cpu_mem_usage=True
-    )
-    
-    if self.tokenizer.pad_token is None:
-        self.tokenizer.pad_token = self.tokenizer.eos_token
-    
-    print("Initialization complete!")
+EOF