use adaptive_avg_pool2d instead of avg_pool2d

gpleiss / efficient_densenet_pytorch

A memory-efficient implementation of DenseNets

MIT License

1.51k stars 329 forks source link

use adaptive_avg_pool2d instead of avg_pool2d #64

Closed aneeshnema closed 4 years ago

aneeshnema commented 4 years ago

Closes #56 removed self.avgpool_size as no longer needed. moved scheduler.step() below train_epoch to remove pytorch warning "UserWarning: Detected call of lr_scheduler.step() before optimizer.step()" when using pytorch version > 1.1.0.

gpleiss commented 4 years ago

Thanks @aneeshnema !