From fac15fe09857c9c63403f446475ee16ad37bb66d Mon Sep 17 00:00:00 2001 From: kaiyangzhou Date: Sun, 4 Apr 2021 10:09:15 +0800 Subject: [PATCH] update comments in the model zoo --- docs/MODEL_ZOO.md | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/MODEL_ZOO.md b/docs/MODEL_ZOO.md index ee98be3..6020911 100644 --- a/docs/MODEL_ZOO.md +++ b/docs/MODEL_ZOO.md @@ -4,6 +4,7 @@ - When computing model size and FLOPs, only layers that are used at test time are considered (see `torchreid.utils.compute_model_complexity`). - Asterisk (\*) means the model is trained from scratch. - `combineall=True` means all images in the dataset are used for model training. +- Why not use heavy data augmentation like [random erasing](https://arxiv.org/abs/1708.04896) for model training? It's because heavy data augmentation might harm the cross-dataset generalization performance (see [this paper](https://arxiv.org/abs/1708.04896)). ## ImageNet pretrained models