nelson1425/EfficientAD
Unofficial implementation of EfficientAD https://arxiv.org/abs/2303.14535
PythonApache-2.0
Issues
- 24
缺陷检测交流微信群,互相交流学习进步
#32 opened by RuojiWang - 5
The image_size
#24 opened by Sencc - 2
Do I need to re-pretrain a teacher network when I train my own dataset?
#39 opened by genzhengmiaohong - 3
Get error when evaluate breakfast_box's output with LOCO's official evaluation code
#10 opened by cinout - 1
- 0
- 0
Is it possible to change the input image size?
#48 opened by Itaru-aso - 0
i got bad result compare 90.1 (mvtec loco)
#47 opened by parkkyungjun - 0
Pixelwise_AUC isn't evaluated?
#23 opened by alext1995 - 0
Wide-ResNet101_2
#46 opened by A0401A - 0
- 0
how to test one img?
#44 opened by czy5638222 - 1
Error when evaluate with MVTec code
#42 opened by winstonchu498 - 1
I encountered an error while evaluating
#37 opened by zty292 - 0
- 0
- 0
IndexError: list index out of range
#38 opened by noivan0 - 0
result of inference on broken large bottle
#36 opened by aria-23 - 4
- 0
- 1
- 1
[BUG] In some cases the quantile may lead to NAN exception or divide zero exception .
#34 opened by limaodaxia - 2
Purpose of padding
#12 opened by AlessandroPolidori - 0
How inteprete the hyper parameter table?
#33 opened by limaodaxia - 2
Distilled teacher model
#30 opened by Doan-IT - 1
The output size of the network is inconsistent with the description in the paper
#28 opened by limaodaxia - 0
If padding=True, worse results but good segmentation precision. Why?
#29 opened by AlessandroPolidori - 0
- 2
good job bro!
#2 opened by nicebro123 - 1
- 0
The feature size not the same with the papar
#25 opened by Apostatee - 1
Asking about deployment of onnx format models
#17 opened by Kyle-fang - 2
The source of teacher weights
#22 opened by WS-W - 0
- 4
- 0
Pretrained weights
#21 opened by nhjcqu - 4
Can the pre-trained model of the teacher network 'teacher_small.pth' be trained by myself?
#15 opened by William9Baker - 0
- 0
Train on ViSA dataset
#14 opened by Ehteshamciitwah - 1
A minor typo in distillation training
#13 opened by wangh09 - 1
- 2
how to inference
#7 opened by syha2994 - 1
- 2
- 1
- 2
Per category result for MVTec AD
#5 opened by alexriedel1 - 1
mvtec_ad_evaluation.tar.xz
#3 opened by Jypeng-64 - 1
How to obtain --imagenet_ train_ path
#1 opened by Leonardo0325