Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
Memory
- ResNet152 inferencing requies about (a little bit less than) 2G GPU memory and training can be done with 8G GPU memory (need to set a small batch size). Deeper ResNet cannot be trained with 8G GPU memory.
Speed
- Training ResNet 152 on GeForce RTX 3070 8GB is about 15x faster than training it on Intel Core i9-10850K (5.2 GHz Turbo) (20-Thread) (10-Core) 3.6 GHz + 64G memory.