is there a way to do it? Or should i just move this attribute accessing inside my train_model task?

just seeing this, I responded above - yes this is what you have to do, as all computation is deferred