Closed weka511 closed 1 year ago
Mikolov et al, Distributed Representations of Words and Phrases and their Compositionality, recommend creating negative samples, k for each central word, where k is 5-20 for small datasets, 2-5 for large.
Fixed for word2vec2.py, still need to do word2vec.py
Sort out terminology, NCE vs NEG
Working
Mikolov et al, Distributed Representations of Words and Phrases and their Compositionality, recommend creating negative samples, k for each central word, where k is 5-20 for small datasets, 2-5 for large.