python - sklearn cosine similarity :: AttributeError: 'module' object has no attribute 'metrics' -
ur[5][8]= [[0,3,4,0,0,0,5,0], [4,5,0,0,0,0,3,0], [0,4,0,3,0,0,1,4], [2,0,5,0,0,0,3,0], [0,0,0,5,0,0,0,4]] 0 means movie not rated want predict rating of unrated movie of each user using cosine similarity , after calculating similarity pick k similar user , predict according that
but how calculate cosine similarity using inbuilt function in skearn library or other
code: similar=[[0] * 5 in range(5)] print similar x in range(0,5): y in range(0,5): similar[x][y] = sklearn.metrics.pairwise.cosine_similarity(ur[x],ur[y]) error : runfile('c:/users/nitin/cf/first.py', wdir='c:/users/nitin/cf') [[0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [0, 0, 0, 0, 0]] traceback (most recent call last): file "<ipython-input-18-e55296403aad>", line 1, in <module> runfile('c:/users/nitin/cf/first.py', wdir='c:/users/nitin/cf') file "c:\users\nitin\anaconda2\lib\site-packages\spyder\utils\site\sitecustomize.py", line 880, in runfile execfile(filename, namespace) file "c:\users\nitin\anaconda2\lib\site-packages\spyder\utils\site\sitecustomize.py", line 87, in execfile exec(compile(scripttext, filename, 'exec'), glob, loc) file "c:/users/nitin/cf/first.py", line 23, in <module> similar[x][y] = sklearn.metrics.pairwise.cosine_similarity(ur[x],ur[y]) attributeerror: 'module' object has no attribute 'metrics' what pass in cosine_similarity or how correctly
you need import module use it.
from sklearn.metrics.pairwise import cosine_similarity or
import sklearn # use sklearn.metrics.pairwise.cosine_similarity(ur[x],ur[y]) then use it.
from sklearn.metrics.pairwise import cosine_similarity ur = [[0,3,4,0,0,0,5,0], [4,5,0,0,0,0,3,0], [0,4,0,3,0,0,1,4], [2,0,5,0,0,0,3,0], [0,0,0,5,0,0,0,4]] similar=[[0] * 5 in range(5)] print(similar) x in range(0,5): y in range(0,5): similar[x][y] = cosine_similarity(ur[x],ur[y]) similar output:
[[array([[ 1.]]), array([[ 0.6]]), array([[ 0.37097041]]), array([[ 0.80295507]]), array([[ 0.]])],
Comments
Post a Comment