- Do you really see a cost advantage by using GPUs (over CPUs) for predictions
- which cloud provider is your preferred provider and why, specifically relating to predictions using multiple ML models (conv. nets, xgboost models etc.)
- which cloud provider is your preferred provider and why, specifically relating to predictions using multiple ML models (conv. nets, xgboost models etc.)