sennnnn cf2cb542f7 Adjust vision transformer backbone architectures (#524) ... * Adjust vision transformer backbone architectures; * Add DropPath, trunc_normal_ for VisionTransformer implementation; * Add class token buring intermediate period and remove it during final period; * Fix some parameters loss bug; * * Store intermediate token features and impose no processes on them; * Remove class token and reshape entire token feature from NLC to NCHW; * Fix some doc error * Add a arg for VisionTransformer backbone to control if input class token into transformer; * Add stochastic depth decay rule for DropPath; * * Fix output bug when input_cls_token=False; * Add related unit test; * * Add arg: out_indices to control model output; * Add unit test for DropPath; * Apply suggestions from code review Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com>		2021-04-30 10:37:47 -07:00
..
backbones	Adjust vision transformer backbone architectures (#524)	2021-04-30 10:37:47 -07:00
decode_heads	Fix typo: upsampe_cfg -> upsample_cfg (#449)	2021-03-30 09:51:38 -07:00
losses	[Enhance] Support reading class_weight from file in loss functions to help MMDet3D (#513)	2021-04-29 01:04:15 -07:00
necks	add upsample neck (#512)	2021-04-24 21:22:09 -07:00
segmentors	[Enhance] Replace data_dict calling 'img' key to support MMDet3D (#514)	2021-04-29 01:01:34 -07:00
utils	Adjust vision transformer backbone architectures (#524)	2021-04-30 10:37:47 -07:00
__init__.py	…
builder.py	[Refactor] Use MMCV MODEL_REGISTRY (#515)	2021-04-27 23:51:09 -07:00