sunshineatnoon/Darknet.keras

loading weights issue

JoeHEZHAO opened this issue · 7 comments

JoeHEZHAO commented

Hi, I have downloaded yolo-tiny.weights and put it into folder 'weights'.

Then happens an ERROR:
Using Theano backend.
Traceback (most recent call last):
File "RunTinyYOLO.py", line 194, in
yoloNet = ReadTinyYOLONetWeights(os.path.join(os.getcwd(),'weights/yolo-tiny.weights'))
File "D:\DeepLearningProgram\Darknet.keras-master\utils\TinyYoloNet.py", line 80, in ReadTinyYOLONetWeights
l.biases = np.asarray(testArray[0][count])
IndexError: index 0 is out of bounds for axis 0 with size 0

And I check file 'TinyYoloNet.py', def 'ReadTinyYOLONetWeights', the 'testArray' variable is [] (null) with many many dtype.

What is the problem ??

yongwoncho commented

I have this problem. Did yo solve this problem?

Walid-Ahmed commented

I have the same problem

sunshineatnoon commented

this code is written for older version of YOLO weight, now the weight has updated to version 2.0, so the code is no longer working.

Walid-Ahmed commented
sunshineatnoon commented

@Walid-Ahmed I'm sorry but I am too busy recently, and reading weights bite by bite is a nasty job. If you need a off the shelf object detection algorithm that is implemented in popular framework, I recommend SSD, it's implemented in caffe and has a re-implement version in MXNET.

JoshGlue commented

This is an old issue, but you can download the old weights at: http://pjreddie.com/media/files/yolo-tiny.weights . For me the code works when I load that weights-file.

sunshineatnoon commented

@JoshGlue Thanks, I've been looking for this weight for a long time.