quark0 / darts

Differentiable architecture search for convolutional and recurrent networks
https://arxiv.org/abs/1806.09055
Apache License 2.0
3.92k stars 843 forks source link

ImageNet training time #135

Open thinkInJava33 opened 4 years ago

thinkInJava33 commented 4 years ago

I trained the model on imagenet with 2 1080ti, it took me 3 hours per epoch, which means i need about a month to finish the training, it is much longer than it mentioned in the paper,which is 12days with a single 1080ti. @quark0 @cclauss

Liu-pf commented 3 years ago

Hello, I also have questions about this and I want to know if you have solved this problem? Can you explain how you used the Imagenet dataset to validate the model?Thank you!

I trained the model on imagenet with 2 1080ti, it took me 3 hours per epoch, which means i need about a month to finish the training, it is much longer than it mentioned in the paper,which is 12days with a single 1080ti. @quark0 @cclauss