TypeError: 'NoneType' object is not iterable about pic-numero

[oduwa]

Thanks for letting me know about this. I've now fixed the issue but realised that this was written with the most current version of tensorflow in 2016 so you will need tensorflow version r0.7, I think. This is because the newer definitions of skflow.DNNClassifier() changed to require the feature_column field in the constructor, which is not provided in the project (as it didn't exist when it was written)

[sepidehhosseinzadeh]

Thanks for your response. Now I have the following error:

Traceback (most recent call last):
File "test.py", line 4, in
count = PicNumero.run_with_cnn(imagePath)
File "/home/ec2-user/SageMaker/research/Pic-Numero/PicNumero/PicNumero.py", line 161, in run_with_cnn
r = CNN.classify(img_data[0], model_file=None,featureRepresentation='glcm', shouldSaveResult=True)
File "/home/ec2-user/SageMaker/research/Pic-Numero/PicNumero/CNN.py", line 262, in classify
clf = get_model(model_file)
File "/home/ec2-user/SageMaker/research/Pic-Numero/PicNumero/CNN.py", line 112, in get_model
classifier = skflow.DNNClassifier(feature_engineering_fn=conv_model, n_classes=2)
TypeError: init() takes at least 3 arguments (3 given)

[oduwa]

Yes that's the error i was referring to in my earlier comment about skflow.DNNClassifier(). Are you using the august 2016 version of tensorflow (which i think is r0.7)?