Use KMeans in Sklearn to do clustering and find the nearest cluster centroid for each SIFT feature;

Use cv$2.$xfeatures$2$d$.$SIFT$\_$create$()$ to create a SIFT object;

Use cv$2.$Keypoint$()$ to generate key points;

Use sift.compute$()$ to compute SIFT descriptors given densely sampled keypoints.

Be mindful of RAM usage. Try to make the code more memory efficient, otherwise it could easily exceed RAM limits in Colab, at which point your session will crash.

If your RAM is going to run out of space, use gc$.$collect$()$ for the garbage collector to collect unused objects in memory to free some space.

Store data or features as NumPy arrays instead of lists. Computation on NumPy arrays is much more efficient than lists.

from sklearn import neighbors

np$.$random.seed$(65)$

##########$--$WRITE YOUR CODE HERE$--$##########

# The following steps are just for your reference

# You can write in your own way

#

# # densely sample keypoints

# def sample$\_$kp$($shape$,$ stride, size$)$:

# return kp

#

# # extract vocabulary of SIFT features

# def extract$\_$vocabulary$($raw$\_$data, key$\_$point$)$:

# return vocabulary

#

# # extract Bag of SIFT Representation of images

# def extract$\_$feat$($raw$\_$data, vocabulary, key$\_$point$)$:

# return feat

#

# # sample dense keypoints

# skp $=$ sample$\_$kp$(($train$\_$data$[0].$shape$[0],$train$\_$data$[0].$shape$[1]),(64,64),8)$

# vocabulary $=$ extract$\_$vocabulary$($train$\_$data, skp$)$

# train$\_$feat $=$ extract$\_$feat$($train$\_$data, vocabulary, skp$)$

# test$\_$feat $=$ extract$\_$feat$($test$\_$data, vocabulary, skp$)$

train$\_$feat $=$

test$\_$feat $=$

##########$-------$END OF CODE$-------$##########

# this block should generate

# train$\_$feat and test$\_$feat corresponding to train$\_$data and test$\_$data

In python please and thank you!