Rotated-RetinaNet

Rotated-RetinaNet copied to clipboard

当我运行eval.py，HRSC2016数据d的测试集时显示 no labels found in test/ground truth !!，这里需要如何准备标签

ZehaoZhang-Uestc

Can't find demo weight file

1

Hello, i'm really appreciate your work. But when i try to run a demo.py file, it said i need to have a weights/last.pth file. So where can i download this...

chauthehan

when I want to train on DOTA dataset , I got this error : cls_targets[positive_indices, assigned_annotations[positive_indices, -1].long()] = 1 RuntimeError: CUDA error: device-side assert triggered

fhaji75

tuple index out of range

1

大佬这是为啥 我用的是DOTA数据机 数据格式是这样的 ![Uploading image.png…]()

shenhaibb

图片的大小设置一样，但是实验效果达不到表中提供的数据，请问ucas_aod的图片是800*800,keep_ratio=False 吗

ming-eng

模型训练

7

你好，我训练自己的数据集（类似于DOTA数据集），但是训练出的模型检测不出任何东西（cls scores全部为0），请问一下可能是哪里出了问题，或者能提供一个你在DOTA数据集下的训练的模型吗？

PressEtoRace

low map

3

dear ming:why my map is lower than your result?i follow your steps to operate on HRSC2016 dataset.what's the problem with my result? can you direct ?thanks. 

yzk-lab

你好，请问你的YOLOV3和retinanet的参数量和FLOPs是多少呢

1

18804601171

Help please

1

hi, will you please add a Jupiter notebook to execute the demo for DOTA or UCAD. thanks regards

UserrNN

Conversion to a TFlite model for performing on-device inference

1

Hi, Do you think it would be possible to convert this model or the yolov3 rotated to ONNX-TensorFlow-TFlite and use it on an android device?

RomStriker