Curious about best practices.
- When should I use cuda for matrix operations and when should I not use it?
- Are cuda operations only suggested for large tensor multiplications?
- What is a reasonable size after which it is advantageous to convert to cuda tensors?
- Are there situations when one should not use cuda?
- What’s the best way to convert between cuda and standard tensors?
- Does sparsity affect the performance of cuda tensors?