Easy training on custom dataset. Various backends (MobileNet and SqueezeNet) supported. A YOLO demo to detect raccoon run entirely in brower is accessible at https://git.io/vF7vI (not on Windows).
I got this error while training and I don't know what is the reason of this error.
I checked my data and I found all my images has its own annotation file.
But inside this function aug_image(self, train_instance, jitter)some images becomes 'None' so when it reaches to this line
h, w, c = image.shape
it gets this error
'NoneType' object has no attribute 'shape'
so why this occurs? is this image file corrupted ?
and how i know these images that have None type and I can delete them from my dataset?
I got this error while training and I don't know what is the reason of this error. I checked my data and I found all my images has its own annotation file. But inside this function
aug_image(self, train_instance, jitter)
some images becomes 'None
' so when it reaches to this lineh, w, c = image.shape
it gets this error'NoneType' object has no attribute 'shape'
so why this occurs? is this image file corrupted ? and how i know these images that have
None
type and I can delete them from my dataset?please any help would be appreciated.