PramuPerera/DeepOneClass

Question about the dataset

quocnhat opened this issue · 3 comments

quocnhat commented

Hi, Thanks for your works.
Do you think the type of reference dataset needs to be similar to the target one? I am implementing Pytorch on my custom dataset (scene dataset), Do I need to find another scene dataset to be the reference? Because Imagenet dataset ( original paper) focuses on the specific object which is similar to the target object (chair)

vanchung1995 commented

Hello @quocnhat , I tried to run this project but it doesnot work.
My steps is:

  • clone repo https://github.com/BVLC/caffe
  • clone this repo to sub folder examples of caffe repo
  • Download dataset for abnormal detection from https://sites.google.com/view/babaksaleh/home and move it to sub folder data of caffe
  • Run python examples/DeepOneClass/run.py --dataset data/ --backbone VGG --cafferoot ~/caffe/ --nclass 6 --noneclass 1 --task abnormal --type feature

But it show the error bellow.
Can you help me, please!

image

quocnhat commented

Sorry I am not familiar with caffe. I used pytorch instead. Btw, the script shows that it is missing file when debugging ( File not found error), you need to specify the dataset first

vanchung1995 commented

Thanks for your answer. I wonder how to specify my data. For example, I want to detect a image is a dog or not, so how my data folder should be structure, and the command to run?
In my case, I put my data folder tree be like:

data
|__Normal_Object_Dataset
|____1
|______dog1.png
|______dog2.png
.....

I am not clear about the data.
Can you explain it in detail for ma, thank you very much