Commit Graph

2340 Commits

Author SHA1 Message Date
Glenn Jocher d79c3bd076 parameterize augment scales 2020-04-07 14:19:43 -07:00
Glenn Jocher b9b14bef59 scale_img() bug fix 2020-04-07 13:35:47 -07:00
Glenn Jocher 067ee264c0 scale_img() bug fix 2020-04-07 12:59:52 -07:00
Glenn Jocher 68f58f4dec scale_img() update 2020-04-07 12:51:52 -07:00
Glenn Jocher 1a3c77df95 Merge remote-tracking branch 'origin/master' 2020-04-07 12:27:56 -07:00
Glenn Jocher 1a511d2906 updated --augment sizes and results 2020-04-07 12:27:49 -07:00
Wang Xinyu b20e3c4c40
Update README, add link to yolov3-spp-tensorrt (#1017) 2020-04-06 22:56:39 -07:00
Glenn Jocher 4fa3fd2df3 tensorboard notice 2020-04-06 16:33:23 -07:00
Glenn Jocher 05ae6e8499 tensorboard/focal loss reporting update 2020-04-06 15:45:18 -07:00
Glenn Jocher c7f93bae40 update coco-tuned hyp['cls'] to current dataset 2020-04-06 10:58:07 -07:00
Glenn Jocher 4fc0012829 initial batchnorm to 0.03 momentum 2020-04-05 18:03:49 -07:00
Glenn Jocher 26fc4fb018 dataloader default color to imagenet mean 114 2020-04-05 18:02:41 -07:00
Glenn Jocher 2baf4e3f93 imagenet normalization on layer 0 batchnorm2d() 2020-04-05 17:33:06 -07:00
Glenn Jocher b70cfa9a29 mAP updates for rect inference 64 commit 2020-04-05 17:18:24 -07:00
Glenn Jocher c6d4e80335 move inference augmentation to model.forward() 2020-04-05 17:14:26 -07:00
Glenn Jocher 4da5c6c114 rect padding to 64, mAP increase 42.7 to 42.9 2020-04-05 16:06:27 -07:00
Glenn Jocher bb59ffe68f model forward() zip() removal 2020-04-05 15:22:32 -07:00
Glenn Jocher a657345b45 add FeatureConcat() module 2020-04-05 14:47:41 -07:00
Glenn Jocher 968b2ec004 .fuse() after .eval() 2020-04-05 14:05:12 -07:00
Glenn Jocher d04738a27c forward updated if-else 2020-04-05 13:49:13 -07:00
Glenn Jocher e81a152a92 tensorboard notice and model verbose option 2020-04-05 13:35:58 -07:00
Glenn Jocher a19b1a3b94 line thickness 2020-04-05 11:10:05 -07:00
Glenn Jocher 6203340888 detect.py multi_label default False 2020-04-05 11:05:49 -07:00
Glenn Jocher 41246aa042 tensorboard updates 2020-04-04 19:34:39 -07:00
Glenn Jocher 00c1fdd805 add MixConv2d() layer 2020-04-03 20:03:44 -07:00
Glenn Jocher eb9fb245aa add support for standalone BatchNorm2d() 2020-04-03 14:21:47 -07:00
Glenn Jocher 682c2b27e7 smart bias bug fix 2020-04-03 12:42:09 -07:00
Glenn Jocher 41a002e798 grid.float() 2020-04-03 12:38:08 -07:00
Glenn Jocher 93055a9d58 create_grids() to YOLOLayer method 2020-04-02 20:23:55 -07:00
Glenn Jocher 91f563c2a2 create_grids() to YOLOLayer method 2020-04-02 19:10:51 -07:00
Glenn Jocher 207c6fcff9 merge NMS full matrix 2020-04-02 18:53:40 -07:00
Glenn Jocher aa4591d7e9 batchnorm momentum to 0.03 2020-04-02 14:10:45 -07:00
Glenn Jocher 9155ef3f4f burnin merged with prebias 2020-04-02 14:08:21 -07:00
Glenn Jocher 27c7334e81 new layers.py file 2020-04-02 12:22:15 -07:00
Glenn Jocher 4ac60018f6 FLOPS report 2020-04-01 14:05:41 -07:00
Glenn Jocher ea80ba65af documentation updates 2020-04-01 12:16:37 -07:00
Glenn Jocher 765b8f3a3b documentation update 2020-04-01 12:12:14 -07:00
Glenn Jocher 5b322b6038 nvcr update 2020-04-01 09:57:00 -07:00
Glenn Jocher 300e9a7ad6 merge NMS full matrix 2020-03-31 21:31:09 -07:00
Glenn Jocher 8d788e10c4 mAP updates 2020-03-31 19:07:41 -07:00
Glenn Jocher 02802e67f2 merge NMS full matrix 2020-03-31 18:18:08 -07:00
Glenn Jocher f4eecef700 merge NMS speed/memory improvements 2020-03-31 15:37:23 -07:00
Glenn Jocher 992e0d7cb4 default test --conf to 0.001 2020-03-31 14:36:25 -07:00
Glenn Jocher 98271eb6ed remove deprecated models 2020-03-31 14:27:10 -07:00
Glenn Jocher 16862ea846 update 'reproduce our results' 2020-03-30 21:21:45 -07:00
Glenn Jocher b2d9f1898f burnin lr ramp 300 iterations 2020-03-30 19:27:42 -07:00
Glenn Jocher ac2aa56e0a feature fusion update 2020-03-30 17:53:17 -07:00
Glenn Jocher 108334db29 FLOPs update 2020-03-30 16:04:08 -07:00
Glenn Jocher 105882b3c6 GFLOPs correction 2020-03-30 15:30:53 -07:00
Glenn Jocher de52a008a5 default --img-size to 512 2020-03-30 11:46:20 -07:00