From b852f2605adb7b61f2ef3229b73223566e4c6222 Mon Sep 17 00:00:00 2001 From: "xiexinchen.vendor" Date: Wed, 6 Jul 2022 06:04:40 +0000 Subject: [PATCH] [Fix] Fix RandomResize default setting --- configs/_base_/datasets/ade20k.py | 6 +++++- configs/_base_/datasets/ade20k_640x640.py | 6 +++++- configs/_base_/datasets/chase_db1.py | 6 +++++- configs/_base_/datasets/cityscapes.py | 6 +++++- configs/_base_/datasets/cityscapes_1024x1024.py | 6 +++++- configs/_base_/datasets/cityscapes_768x768.py | 6 +++++- configs/_base_/datasets/cityscapes_769x769.py | 6 +++++- configs/_base_/datasets/cityscapes_832x832.py | 6 +++++- configs/_base_/datasets/coco-stuff10k.py | 6 +++++- configs/_base_/datasets/coco-stuff164k.py | 6 +++++- configs/_base_/datasets/drive.py | 6 +++++- configs/_base_/datasets/hrf.py | 6 +++++- configs/_base_/datasets/isaid.py | 6 +++++- configs/_base_/datasets/loveda.py | 6 +++++- configs/_base_/datasets/pascal_context.py | 6 +++++- configs/_base_/datasets/pascal_context_59.py | 6 +++++- configs/_base_/datasets/pascal_voc12.py | 6 +++++- configs/_base_/datasets/pascal_voc12_aug.py | 6 +++++- configs/_base_/datasets/potsdam.py | 6 +++++- configs/_base_/datasets/stare.py | 6 +++++- configs/_base_/datasets/vaihingen.py | 6 +++++- configs/cgnet/cgnet_680x680_60k_cityscapes.py | 6 +++++- .../knet_s3_upernet_swin-l_8x2_640x640_adamw_80k_ade20k.py | 6 +++++- configs/segformer/segformer_mit-b5_640x640_160k_ade20k.py | 6 +++++- 24 files changed, 120 insertions(+), 24 deletions(-) diff --git a/configs/_base_/datasets/ade20k.py b/configs/_base_/datasets/ade20k.py index a64963fe9..4303b094c 100644 --- a/configs/_base_/datasets/ade20k.py +++ b/configs/_base_/datasets/ade20k.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(2048, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/ade20k_640x640.py b/configs/_base_/datasets/ade20k_640x640.py index 5cd6e4b91..847858591 100644 --- a/configs/_base_/datasets/ade20k_640x640.py +++ b/configs/_base_/datasets/ade20k_640x640.py @@ -5,7 +5,11 @@ crop_size = (640, 640) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(2560, 640), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2560, 640), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/chase_db1.py b/configs/_base_/datasets/chase_db1.py index 2e6093ec5..1622bec95 100644 --- a/configs/_base_/datasets/chase_db1.py +++ b/configs/_base_/datasets/chase_db1.py @@ -6,7 +6,11 @@ crop_size = (128, 128) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=img_scale, ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=img_scale, + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/cityscapes.py b/configs/_base_/datasets/cityscapes.py index 316d154ff..c2fdee473 100644 --- a/configs/_base_/datasets/cityscapes.py +++ b/configs/_base_/datasets/cityscapes.py @@ -5,7 +5,11 @@ crop_size = (512, 1024) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2048, 1024), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 1024), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/cityscapes_1024x1024.py b/configs/_base_/datasets/cityscapes_1024x1024.py index 7eab30a02..72be307b4 100644 --- a/configs/_base_/datasets/cityscapes_1024x1024.py +++ b/configs/_base_/datasets/cityscapes_1024x1024.py @@ -3,7 +3,11 @@ crop_size = (1024, 1024) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2048, 1024), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 1024), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/cityscapes_768x768.py b/configs/_base_/datasets/cityscapes_768x768.py index 09be30eca..fcee0143a 100644 --- a/configs/_base_/datasets/cityscapes_768x768.py +++ b/configs/_base_/datasets/cityscapes_768x768.py @@ -3,7 +3,11 @@ crop_size = (768, 768) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2049, 1025), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2049, 1025), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/cityscapes_769x769.py b/configs/_base_/datasets/cityscapes_769x769.py index 27a915f88..ae40ac8c5 100644 --- a/configs/_base_/datasets/cityscapes_769x769.py +++ b/configs/_base_/datasets/cityscapes_769x769.py @@ -3,7 +3,11 @@ crop_size = (769, 769) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2049, 1025), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2049, 1025), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/cityscapes_832x832.py b/configs/_base_/datasets/cityscapes_832x832.py index 783445e0f..025458035 100644 --- a/configs/_base_/datasets/cityscapes_832x832.py +++ b/configs/_base_/datasets/cityscapes_832x832.py @@ -3,7 +3,11 @@ crop_size = (832, 832) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2048, 1024), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 1024), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/coco-stuff10k.py b/configs/_base_/datasets/coco-stuff10k.py index f08772e2f..b00db2469 100644 --- a/configs/_base_/datasets/coco-stuff10k.py +++ b/configs/_base_/datasets/coco-stuff10k.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(2048, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/coco-stuff164k.py b/configs/_base_/datasets/coco-stuff164k.py index b96bd0454..e879bdb2a 100644 --- a/configs/_base_/datasets/coco-stuff164k.py +++ b/configs/_base_/datasets/coco-stuff164k.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2048, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/drive.py b/configs/_base_/datasets/drive.py index 71f9b619b..523354d05 100644 --- a/configs/_base_/datasets/drive.py +++ b/configs/_base_/datasets/drive.py @@ -6,7 +6,11 @@ crop_size = (64, 64) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=img_scale, ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=img_scale, + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/hrf.py b/configs/_base_/datasets/hrf.py index a423de889..3ca2263ab 100644 --- a/configs/_base_/datasets/hrf.py +++ b/configs/_base_/datasets/hrf.py @@ -6,7 +6,11 @@ crop_size = (256, 256) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=img_scale, ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=img_scale, + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/isaid.py b/configs/_base_/datasets/isaid.py index 6760410af..8dafae8fd 100644 --- a/configs/_base_/datasets/isaid.py +++ b/configs/_base_/datasets/isaid.py @@ -12,7 +12,11 @@ crop_size = (896, 896) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(896, 896), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(896, 896), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/loveda.py b/configs/_base_/datasets/loveda.py index 1541e171e..fcdb05865 100644 --- a/configs/_base_/datasets/loveda.py +++ b/configs/_base_/datasets/loveda.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(2048, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/pascal_context.py b/configs/_base_/datasets/pascal_context.py index 21d8ab98b..dfb1f858b 100644 --- a/configs/_base_/datasets/pascal_context.py +++ b/configs/_base_/datasets/pascal_context.py @@ -8,7 +8,11 @@ crop_size = (480, 480) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=img_scale, ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=img_scale, + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/pascal_context_59.py b/configs/_base_/datasets/pascal_context_59.py index 77b262e4d..9103fe7e3 100644 --- a/configs/_base_/datasets/pascal_context_59.py +++ b/configs/_base_/datasets/pascal_context_59.py @@ -8,7 +8,11 @@ crop_size = (480, 480) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=img_scale, ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=img_scale, + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/pascal_voc12.py b/configs/_base_/datasets/pascal_voc12.py index 2201bc8c8..aeb38d061 100644 --- a/configs/_base_/datasets/pascal_voc12.py +++ b/configs/_base_/datasets/pascal_voc12.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2048, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/pascal_voc12_aug.py b/configs/_base_/datasets/pascal_voc12_aug.py index b9401b138..cd0d3e868 100644 --- a/configs/_base_/datasets/pascal_voc12_aug.py +++ b/configs/_base_/datasets/pascal_voc12_aug.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2048, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/potsdam.py b/configs/_base_/datasets/potsdam.py index 10375076c..ef9761c76 100644 --- a/configs/_base_/datasets/potsdam.py +++ b/configs/_base_/datasets/potsdam.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(512, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(512, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/stare.py b/configs/_base_/datasets/stare.py index f41ebd835..a904fa8fd 100644 --- a/configs/_base_/datasets/stare.py +++ b/configs/_base_/datasets/stare.py @@ -6,7 +6,11 @@ crop_size = (128, 128) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=img_scale, ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=img_scale, + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/_base_/datasets/vaihingen.py b/configs/_base_/datasets/vaihingen.py index 0b4a1cc69..2b5213556 100644 --- a/configs/_base_/datasets/vaihingen.py +++ b/configs/_base_/datasets/vaihingen.py @@ -5,7 +5,11 @@ crop_size = (512, 512) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(512, 512), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(512, 512), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/cgnet/cgnet_680x680_60k_cityscapes.py b/configs/cgnet/cgnet_680x680_60k_cityscapes.py index a97af631b..93457fe51 100644 --- a/configs/cgnet/cgnet_680x680_60k_cityscapes.py +++ b/configs/cgnet/cgnet_680x680_60k_cityscapes.py @@ -30,7 +30,11 @@ model = dict(data_preprocessor=data_preprocessor) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), - dict(type='RandomResize', scale=(2048, 1024), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 1024), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size), dict(type='RandomFlip', prob=0.5), dict(type='PackSegInputs') diff --git a/configs/knet/knet_s3_upernet_swin-l_8x2_640x640_adamw_80k_ade20k.py b/configs/knet/knet_s3_upernet_swin-l_8x2_640x640_adamw_80k_ade20k.py index 98b8e2add..1dcb1d486 100644 --- a/configs/knet/knet_s3_upernet_swin-l_8x2_640x640_adamw_80k_ade20k.py +++ b/configs/knet/knet_s3_upernet_swin-l_8x2_640x640_adamw_80k_ade20k.py @@ -30,7 +30,11 @@ crop_size = (640, 640) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(2048, 640), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 640), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'), diff --git a/configs/segformer/segformer_mit-b5_640x640_160k_ade20k.py b/configs/segformer/segformer_mit-b5_640x640_160k_ade20k.py index 4a9c92099..0d13707ec 100644 --- a/configs/segformer/segformer_mit-b5_640x640_160k_ade20k.py +++ b/configs/segformer/segformer_mit-b5_640x640_160k_ade20k.py @@ -6,7 +6,11 @@ data_preprocessor = dict(size=crop_size) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=True), - dict(type='RandomResize', scale=(2048, 640), ratio_range=(0.5, 2.0)), + dict( + type='RandomResize', + scale=(2048, 640), + ratio_range=(0.5, 2.0), + keep_ratio=True), dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='RandomFlip', prob=0.5), dict(type='PhotoMetricDistortion'),