I am trying this model on multiclass classification. My trained dataset contains four folders Happy, Sad, Neutral, Angry. The first problem it's reading every image from all folder when it is going to read angry or any other label image and second it's not taking the two labels only angry and sad it's taking not other two.
Can anyone help me with that?
Eagerly waiting for the solution.

I don't know why but i am getting a lot of deprecated warnings.

deleted

Can you please please help to this error

File "Bag.py", line 160, in
bov.trainModel()
File "Bag.py", line 50, in trainModel
bov_descriptor_stack = self.bov_helper.formatND(self.descriptor_list)
File "/home/ram/RAM/From 11-June-2018/1.Image_classification/Bag-of-Visual-Words-Python/helpers.py", line 87, in formatND
vStack = np.vstack((vStack, remaining))
File "/home/ram/anaconda3/envs/pytorch/lib/python2.7/site-packages/numpy/core/shape_base.py", line 234, in vstack
return _nx.concatenate([atleast_2d(_m) for _m in tup], 0)
ValueError: all the input array dimensions except for the concatenation axis must match exactly

I have read through the blog , still didn't understand why j is not used within kmeans_ret

old_count = 0
		for i in range(n_images):
			l = len(descriptor_list[i])
			for j in range(l):
				if kmeans_ret is None:
					idx = self.kmeans_ret[old_count+j]
				else:
					idx = kmeans_ret[old_count+j]
				self.mega_histogram[i][idx] += 1
			old_count += l
		print "Vocabulary Histogram Generated"

Hello!

I am a medical engineering student studying in Germany. I am trying to implement a similar problem for my master thesis and this code is almost exactly what I want to do. Is it permissible to use this code? It would be really helpful if you can let me know how I can cite you, the author for this code.

Thank you in advance.
Best Regards,
Gayathri

你好，我下载了您的代码，但是运行的时候，Bag.py
for word, imlist in self.images.items():
self.name_dict[str(label_count)] = word
print("Computing Features for ", word)
print(imlist)
for im in imlist:

这里由于python3,我改成了items（），但是imlist传进来是个空的，请问这是怎么回事呢？
我很着急，能麻烦您跟我联系一下么？

File "Bag.py", in 140, 141:
parser.add_argument('--train_path', action="store", dest="train_path", required=True)
parser.add_argument('--test_path', action="store", dest="test_path", required=True)
but in the instruction README.md file, it shows:
python Bag.py --train_set images/train/ --test_set images/test/

Need to change 'python Bag.py --train_set images/train/ --test_set images/test/' to 'python Bag.py --train_path images/train/ --test_path images/test/' in README.md to make to work properly.

Thanks.

• Use the following code in helpers.py if the system is windows
• Otherwise the "\" notation for windows will create problems

#line number 139
for each in glob("{}/*".format(path)):
	key = each.split("/")[-1]
	key = key.split("\\""")[-1]
	print(" #### Reading image category ", key, " ##### ")
	imlist[key] = []
	for imagefile in glob(path+key+"/*.jpg"):
		print("Reading file ", imagefile)
		im = cv2.imread(imagefile)
		imlist[key].append(im)
		count +=1

Hi Kushal,
I have a simliar problem (but with a completely different dataset) in my master thesis. (I'm also studying computer science here in Germany) And it would be perfect if I can quote and use some of your code ideas in my own master thesis. So I would ask, if it is possible that you add an licence (just for usage in science) in your repository?

Thank you in advance.
Best Regards,
Pablo

Always receiving this error when I run the program. Need help