doesn't work

[ssaru]

return empty list from np.fromfile at TinyYoloNet.py, ReadTinyYOLONetWeights. line 76

how to fix it?

[sunshineatnoon]

Maybe #5 can help you. This repo based on the old YOLO weight, now YOLO has updated to 2.0. So you might need the old weight here.

[ssaru]

it's look like read weights file!

but... i getting another error!

Traceback (most recent call last):
File "RunTinyYOLO.py", line 209, in
model = SimpleNet(yoloNet)
File "RunTinyYOLO.py", line 38, in SimpleNet
model.add(Convolution2D(16, 3, 3, weights=[yoloNet.layers[1].weights,yoloNet.layers[1].biases],border_mode='valid',subsample=(1,1)))
File "/home/martin/Documents/enumnet/goverment_project/Deeplearning/Study/keras-darknet/env/lib/python3.5/site-packages/keras/models.py", line 476, in add
output_tensor = layer(self.outputs[0])
File "/home/martin/Documents/enumnet/goverment_project/Deeplearning/Study/keras-darknet/env/lib/python3.5/site-packages/keras/engine/topology.py", line 576, in call
self.set_weights(self._initial_weights)
File "/home/martin/Documents/enumnet/goverment_project/Deeplearning/Study/keras-darknet/env/lib/python3.5/site-packages/keras/engine/topology.py", line 1190, in set_weights
'provided weight shape ' + str(w.shape))
ValueError: Layer weight shape (3, 3, 448, 16) not compatible with provided weight shape (16, 3, 3, 3)

[Abduoit]

Hi, I tried with the old weight but still getting error

abdulrahman@abdulrahman-ThinkPad-X230-Tablet:~/Darknet.keras$ python RunTinyYOLO.py
Traceback (most recent call last):
File "RunTinyYOLO.py", line 6, in
from utils.crop import crop
File "/home/abdulrahman/Darknet.keras/utils/crop.py", line 3, in
from scipy import misc
ImportError: No module named scipy

[ssaru]

@Abduoit
I think that you should install scipy module

pip install scipy
Or
pip3 install scipy

[Abduoit]

Thx @ssaru
I did
Now I get the following error

abdulrahman@abdulrahman-ThinkPad-X230-Tablet:~/Darknet.keras$ python RunTinyYOLO.py
Traceback (most recent call last):
File "RunTinyYOLO.py", line 8, in
from keras.models import Sequential
ImportError: No module named keras.models

[ssaru]

You installed keras? With tensorflow

1. You should install tensorflow
   And after
   pip install keras

[Abduoit]

@ssaru
I do have tensorflow and keras and theano
now I am getting the following error

abdulrahman@abdulrahman-ThinkPad-X230-Tablet:~/Darknet.keras$ python RunTinyYOLO.py
Using TensorFlow backend.
RunTinyYOLO.py:38: UserWarning: Update your Conv2D call to the Keras 2 API: Conv2D(16, (3, 3), padding="valid", strides=(1, 1), weights=[array([[[...)
model.add(Convolution2D(16, 3, 3, weights=[yoloNet.layers[1].weights,yoloNet.layers[1].biases],border_mode='valid',subsample=(1,1)))
Traceback (most recent call last):
File "RunTinyYOLO.py", line 209, in
model = SimpleNet(yoloNet)
File "RunTinyYOLO.py", line 38, in SimpleNet
model.add(Convolution2D(16, 3, 3, weights=[yoloNet.layers[1].weights,yoloNet.layers[1].biases],border_mode='valid',subsample=(1,1)))
File "/usr/local/lib/python2.7/dist-packages/keras/models.py", line 469, in add
output_tensor = layer(self.outputs[0])
File "/usr/local/lib/python2.7/dist-packages/keras/engine/topology.py", line 576, in call
self.set_weights(self._initial_weights)
File "/usr/local/lib/python2.7/dist-packages/keras/engine/topology.py", line 1190, in set_weights
'provided weight shape ' + str(w.shape))
ValueError: Layer weight shape (3, 3, 448, 16) not compatible with provided weight shape (16, 3, 3, 3)

[ssaru]

Yes good!
Now you got error messagr that same with me
Im waiting author's answer :)

[sunshineatnoon]

@ssaru I don't have access to a machine with GPU, so probably will not fix this bug. You might want to try other YOLO implementations: Tensorflow Version or Pytorch Version(YOLO2)

[ssaru]

Ok! i will try!
thanks :) !!