Deploying Ml Models On Gpu Temperature