9f01a37ad4
Refactor ViTDet backbone and simple feature pyramid ( #177 )
...
1. The vitdet backbone implemented by d2 is about 20% faster than the vitdet backbone originally reproduced by easycv.
2. 50.57 -> 50.65
2022-09-16 11:03:53 +08:00
0cb91de0cb
add DeiT III ( #171 )
...
1.Add a backbone: deitiii.
2.Add an optimizer: lamb.
3.Add a sampler: RASampler.
4.Add a lr update hook: CosineAnnealingWarmupByEpochLrUpdaterHook.
5.In easycv/models/classification/classification.py, I remove the default mixup_cfg to keep the classification.py clean.
2022-09-14 15:24:54 +08:00
b198c5a81f
support DINO algo ( #144 )
...
dino_4sc_r50_12e：48.71
dino_4sc_r50_24e：50.53
dino_4sc_r50_36e：50.69
dino_4sc_swinl_12e: 56.86
dino_4sc_swinl_36e: 58.04
dino_5sc_swinl_36e: 58.47
2022-08-31 15:18:11 +08:00
e3678fbbfa
FCOS update torch_style ( #170 )
...
fcos update torch style
2022-08-29 16:21:14 +08:00
6821cdc9db
refactor detection predictor ( #158 )
...
The use of unified DetrPredictor for detection test and predictor.
2022-08-26 10:28:17 +08:00
9aaa600f79
Yolox improve with REPConv/ASFF/TOOD ( #154 )
...
* add attention layer and more loss function
* add attention layer and various loss functions
* add siou loss
* add tah,various attention layers, and different loss functions
* add asff sim, gsconv
* blade utils fit faster
* blade optimize for yolox static & fp16
* decode output for yolox control by cfg
* add reparameterize_models for export
* e2e trt_nms plugin export support and numeric test
* split preprocess from end2end+blade, speedup from 17ms->7.2ms
Co-authored-by: zouxinyi0625 <zouxinyi.zxy@alibaba-inc.com>
2022-08-24 18:11:15 +08:00
9517bb80ff
add dn_detr algo ( #119 )
...
support dn-detr-r50: 44.39
refactor set_criterion
2022-07-27 15:06:06 +08:00
8c93caa2d9
support Efficientformer algo ( #128 )
...
* add efficientformer
2022-07-27 11:36:03 +08:00
eeabb4c6ab
refactor det ( #124 )
2022-07-13 12:11:40 +08:00
5fb0a86f69
support fcos ( #100 )
...
support fcos(38.57)
2022-07-12 18:07:02 +08:00
6b8b04db72
add upernet algo ( #118 )
...
* add upernet algo
* fix import onnx bug
2022-07-12 11:14:09 +08:00
0b472bacf1
refactor detection, support detr and dab-detr ( #104 )
...
refactor detection, support detr and dab-detr
2022-07-11 14:52:49 +08:00
e61488d319
Support ViTDet algo ( #35 )
...
* adapt mmlab modules
* add vitdet
* support conv aggregation
* modify vitdet load pretrained
* support fp16
* modify defaultformatbundle
* modify aug
* bugfix sampler
* bugfix mmresize
* bugfix fp16&nonetype
* bugfix filterannotation
* support dlc
* bugfix dist
* bugfix detsourcecoco
* smodify mmdet_parse_losses
* bugfix nan
* bugfix eval
* bugfix data=nonetype
* modify resize_embed
* support vitdet_conv
* add vitdet_conv init_weight
* add test_vitdet
* uniform rand_another
* uniform use fp16 method
* add test_fp16
Co-authored-by: jiangnana.jnn <jiangnana.jnn@alibaba-inc.com>
2022-06-10 21:49:32 +08:00
3ee118f065
Expand cls model zoo ( #55 )
...
1. expand cls model zoo
2. uniform load_pretrained
2022-06-01 11:01:29 +08:00
4c1b3145e6
[feature] add segmentation modules ( #71 )
...
* add segmentation module & add ResNetV1c & ResNetV1d & add benchmark
2022-05-31 20:19:17 +08:00
055deba53e
initial commit
2022-04-02 20:01:06 +08:00
tuofeilun
zzoneee
wuziheng
yhq
Cathy0908
wenmeng.zwm