Training
GPU
A GPU is strongly recommended for training deep learning models. Training performance on a GPU can be up to 20x faster than on a CPU, and can shorten training time from weeks to hours.
Below is a list of recommended GPU options:
| Basic | Mid-Range | Performance | High-End | |
|---|---|---|---|---|
| GeForce GTX 1060 | x | |||
| GeForce GTX 1070 | x | |||
| GeForce GTX 1080 | x | |||
| GeForce RTX 2060 | x | |||
| GeForce RTX 2070 | x | |||
| GeForce GTX 1080 Ti | x | |||
| Titan Xp | x | |||
| GeForce RTX 2080 | x | |||
| GeForce RTX 2080 Ti | x | |||
| Titan RTX | x | |||
| Titan V | x | |||
| Tesla V100 | x |
Below is chart of benchmark comparisons between various cards. The red bar is most relevant to deep learning training performance.
Source: https://timdettmers.com/2019/04/03/which-gpu-for-deep-learning/
Applying
GPU
A GPU is also recommended for applying deep learning models. Application performance on a GPU can be up to 20x faster than on a CPU.
CPU
Since applying often takes seconds, compared to hours for training, a CPU can more reasonably be used if necessary. Please refer to the general System Requirements for more information on CPU requirements and recommendations.
Need more help with this?
Chat with an expert now ››
