Issues
- 0
请教下为什么pytorch version版本的mAP只有76.5,你这里可以达到88?
#110 opened by clw5180 - 0
error when rename.py is run
#107 opened by jwnirvana - 8
YOLO-Lite哪里用到KL Loss
#76 opened by gfjiangly - 3
中断训练过程后再继续训练
#83 opened by nobody-cheng - 1
ModuleNotFoundError: No module named 'dataAug'
#103 opened by kingmacth - 0
error exits when rename.py is run
#108 opened by jwnirvana - 1
type error
#106 opened by jwnirvana - 3
是否加载预训练模型的精度差异
#105 opened by litingsjj - 3
- 2
关于focal loss 中alpha=1是相当于没有对正负样本数量进行加权是吗?
#79 opened by 123lifei - 3
self.__steps_per_period 错误, numpy.float64' object cannot be interpreted as an integer
#58 opened by tianyexing - 2
V3/NameError: name 'file' is not defined
#97 opened by MichaelCong - 7
v3 在gtx1080ti训练了多久?
#96 opened by qiqzhang - 2
关于 数据读取 的问题
#100 opened by ghoshaw - 1
方便提供yolov3和yolov3 lite的预训练权重吗?
#99 opened by sijun-zhou - 2
COCO训练需要多久呢?
#98 opened by qiqzhang - 1
readme上的params参数的大小
#92 opened by nobody-cheng - 0
- 0
data.py文件处理anchor的wh缩放问题
#94 opened by nobody-cheng - 1
About GIOU loss
#93 opened by zhangyufei1995 - 8
关于GIoU loss的疑问
#72 opened by lyx190 - 0
您好,请问你是如何使用mixup的,有增加epoch么?
#91 opened by 711e - 2
v1版本训练梯度爆炸
#80 opened by nobody-cheng - 2
多尺度训练问题
#85 opened by zhangyufei1995 - 1
预训练权重
#89 opened by zhangyufei1995 - 1
GPU利用率好低,只有百分之几 请问是为啥呢
#90 opened by zhangyufei1995 - 2
An error occurred when I ran test.py
#65 opened by caoabc - 6
v1的有关问题请教
#88 opened by zhangyufei1995 - 4
v3
#87 opened by zhangyufei1995 - 4
您好, 我想问下您的cuda是什么版本的啊,我这个一直报错
#86 opened by lukaisheng1203 - 5
有试过加入 SE 模块吗?
#62 opened by Lam1360 - 3
mobile-yolo模型图片预处理
#82 opened by lightfate - 2
生成了CKPT 文件怎么导出PB文件啊 ?
#67 opened by basser2 - 8
v1 版本梯度爆是什么原因啊
#81 opened by 123lifei - 0
cosine learning rate decay
#78 opened by Sun-Fan - 2
关于样本图片标注框转换成预测数据的问题
#77 opened by 123lifei - 1
你好,请问加载三个权重文件的时候报错怎么解决
#71 opened by noobgrow - 2
failed to run cuBLAS in RTX2080
#75 opened by HochCC - 2
test with mixup
#73 opened by pengfeidip - 4
如何使用focal loss 、soft nms等这些改进呢?
#74 opened by zychen2016 - 0
- 1
- 0
.
#70 opened by gzz1529657064 - 1
data_aug.py第41 42行max改为min是否更合适些?
#63 opened by fan0210 - 1
About GIOU code
#64 opened by Apostatee - 2
missed anchor parameters?
#61 opened by LightToYang - 2
about train log
#66 opened by 791136190 - 0
darknet cfg format
#57 opened by DrVijayK - 3
关于标签的疑问
#59 opened by StrongBirds - 2
About v3 anchor-free methods
#60 opened by Apostatee