Pytorch-cyclegan-and-pix2pix: Load on laptop CPUs
I have an HP Notebook - 15-ac026tx with Intel i5 (5th gen), 4GB RAM, 2GB AMD Radeon graphics card.Full Specifications
I want to run your code on my laptop. Won't it harm my laptop battery and its hardware?
I read somewhere on the Internet that training high computation models can severely harm laptops?
I just want to know is it okay if I train it on my laptop?
divyanshj16
All 6 comments
I am not sure if PyTorch supports AMD GPUs. You can train your model on the Intel CPUs, But it will be quite slow. I haven't trained any model on an everyday use laptop by myself.
junyanz
on 13 Jun 2017
How much time does it take on your computer?
divyanshj16
on 13 Jun 2017
It takes 3-4 hours for pix2pix training on facades datasets with a Nvidia GTX 1080 GPU.
junyanz
on 13 Jun 2017
Haven't you saved the model? so that we can use it without training?
divyanshj16
on 13 Jun 2017
There are a few pix2pix and cyclegan models available. See this post on pre-trained models for more details.
junyanz
on 30 Dec 2017
Add a hint for those who searched 'time' to get here:
'It takes 3-4 hours for pix2pix training on facades datasets with a Nvidia GTX 1080 GPU.'
which means that if you use a desktop nvidia gpu, like GTX 745 I am using, It has a 4GB vram (50% of the GTX1080), then it shall use 6-8 hours IF ONLY CONSIDERS THE BATCH SIZE LOSS
And If you are using cpu, consider a desktop i7 with enough RAM , it is about 48-64 hours in total (800% time consuming)
for a laptop 4GB RAM, you need to reduce the batch size, which is ... even slower, and if you are using i5-laptop-version, please save your time and don't even try
lxj616
on 15 Jan 2018
Related issues
AlexanLee
路
4Comments
John1231983
路
3Comments
John1231983
路
3Comments
Overtown
路
3Comments
shivom9713
路
4Comments
Most helpful comment
Add a hint for those who searched 'time' to get here:
'It takes 3-4 hours for pix2pix training on facades datasets with a Nvidia GTX 1080 GPU.'
which means that if you use a desktop nvidia gpu, like GTX 745 I am using, It has a 4GB vram (50% of the GTX1080), then it shall use 6-8 hours IF ONLY CONSIDERS THE BATCH SIZE LOSS
And If you are using cpu, consider a desktop i7 with enough RAM , it is about 48-64 hours in total (800% time consuming)
for a laptop 4GB RAM, you need to reduce the batch size, which is ... even slower, and if you are using i5-laptop-version, please save your time and don't even try