mkusner / wmd

Word Mover's Distance from Matthew J Kusner's paper "From Word Embeddings to Document Distances"
537 stars 132 forks source link

Error: emd: Signature size is limited to 100 #18

Closed icarocd closed 7 years ago

icarocd commented 7 years ago

During the wmd distance matrix computation, it occurs "emd: Signature size is limited to 100" several times. What should be done?

icarocd commented 7 years ago

The use of pre-trained word2vec based on google news provides 300-dimensional word vectors, but EMD algorithms does not seem to handle such amount. ?

ghost commented 7 years ago

you have to modify some lines in "python-emd-master\emd.h" (for python) or "emd\emd.h" (for matlab )

find definitions and edit MAX_SIG_SIZE #define MAX_SIG_SIZE 100 change 100 to bigger number (actually, the size of biggest document BOW). remember that you must re-make.