search

chenxin061 / pdarts

Codes for our paper "Progressive Differentiable Architecture Search:Bridging the Depth Gap between Search and Evaluation"
Other
359 stars 83 forks source link

use train_search , got genotype like DARTS #20

Closed ykk648 closed 4 years ago

ykk648 commented 4 years ago

by using default parameters 图片

ykk648 commented 4 years ago

One Titan X-12GB, cost 23 hours

chenxin061 commented 4 years ago
  1. About search cost: if you use a Titan X with 12GB memory, the batch size might be reduced to 64. It is strange that the search cost on 1080-ti is less than 12 hours but about 24 hours on Titan X(p). This may due to different architectures of those two GPUs.
  2. About arch: in our experiments, this happened with a very small chance, especially with small batch size. This is because our method is built on top of DARTS and sometimes we meet some unfriendly local minima.