3 ms·
> Since it uses a deep neural network, I don't think it's "thousands a second". The generator is like 150MB. The forward pass is <0.1s. Hypothetically you shou
by gwern 8y ago
> Since it uses a deep neural network, I don't think it's "thousands a second".
The generator is like 150MB. The forward pass is <0.1s. Hypothetically you should be able to generate on a decent GPU like a 1080ti with 11GB VRAM at full utilization <730 images per second. Use a few GPUs and you're at thousands per second.
- gambler 8y agoWhere is this data from? I see 300MB model on their Google Drive. And if I understand correctly, you also need source and destination images to transfer styles from and to, so it's not like the model generates photos out of thin air.
- gwern 8y agoThe 300MB model covers both the G and D. You only need G to generate. The style transfer is just noise. And I time my own 512px anime StyleGANs at ~21 images per second per model; half that throughput to account for the increased model size and depth of a 1024px. No matter how you tweak the numbers - halve it again if you wish! - it's clear that thousands per second is entirely attainable with a few GPUs at low cost. (For comparison, 8 V100s is ~$7/hr on AWS; 10x1080ti is ~$1.3/hr on Vast.ai.)