yjh0410 2259ef87f2 update		2 年之前
..
README.md	ebe1d9e27f train RTCDet-S on COCO	2 年之前
build.py	2259ef87f2 update	2 年之前
loss.py	be12f940f2 update	2 年之前
matcher.py	222a4fadd1 debug AlignedSimOTA for RTCDet-M	2 年之前
rtcdet.py	2259ef87f2 update	2 年之前
rtcdet_backbone.py	4e429e1847 train RTCDet-S on COCO	2 年之前
rtcdet_basic.py	ae03c10fc1 train RTCDet-M with 320-960	2 年之前
rtcdet_head.py	578dc295f4 redesign RTCDet	2 年之前
rtcdet_neck.py	578dc295f4 redesign RTCDet	2 年之前
rtcdet_pafpn.py	62a5b79e6a update	2 年之前
rtcdet_pred.py	f6753b0e9b train RTCDet-S	2 年之前

RTCDet: My Empirical Study of Real-Time Convolutional Object Detectors.

For training, we train my RTCDet series series with 300 epochs on COCO.
For data augmentation, we use the large scale jitter (LSJ), Mosaic augmentation and Mixup augmentation, following the setting of YOLOX, but we remove the rotation transformation which is used in YOLOX's strong augmentation.
For optimizer, we use AdamW with weight decay 0.05 and base per image lr 0.001 / 64.
For learning rate scheduler, we use linear decay scheduler.
Due to my limited computing resources, I can not train RTCDet-X with the setting of batch size=128.

Model	Scale	Batch	Weight
RTCDet-P	320	8xb16	-
RTCDet-P	416	8xb16	-
RTCDet-P	512	8xb16	-
RTCDet-P	640	8xb16	-