MenghaoGuo / AutoDeeplab

Pytorch Implementation the paper Auto-DeepLab Hierarchical Neural Architecture Search for Semantic Image Segmentation
https://arxiv.org/abs/1901.02985
410 stars 97 forks source link

Not doing an architecture step until epoch > 19 #59

Open davidhuangal opened 4 years ago

davidhuangal commented 4 years ago

Why in train_autodeeplab.py do you not run self.architect.step (image_search, target_search) until epoch > 19 ?