Last active
November 12, 2016 09:37
-
-
Save samuelazran/cd2827e5f9bf7af0ae979fb3008a527a to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
""" | |
feature extraction for representing groups of items | |
the order of items in a group is irrelevant | |
================================================== | |
X = list of tokens: [t1,t2,t3,t4,t5] | |
Y = list of group tags: [S ,C ,C ,S ,C ] | |
F = list of features representing X: | |
-------------------------------------------------- | |
X[0] = t1 Y[0] = S F[0] = {token=t1} | |
X[1] = t2 Y[1] = C F[1] = {token=t2, group_has__t1=1} | |
X[2] = t3 Y[2] = C F[2] = {token=t3, group_has__t1=1, group_has__t2=1} | |
X[3] = t4 Y[3] = S F[3] = {token=t4, group_has__t1=1, group_has__t2=1, group_has__t3=1} | |
X[4] = t5 Y[4] = C F[4] = {token=t5, group_has__t4=1} | |
""" | |
X = ["t1","t2","t3","t4","t5"] | |
Y = ["S" ,"C" ,"C" ,"S" ,"C"] | |
F = [None] * len(X) | |
for i, x in enumerate(X): | |
if i == 0 or Y[i - 1] == "S": | |
F[i] = {} | |
else: | |
F[i] = copy.deepcopy(F[i - 1]) | |
if i > 0: | |
feature = "group_has__{}".format(F[i - 1]["token"]) | |
F[i][feature] = F[i].get(feature, 0) + 1 | |
F[i]["token"] = x |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment