Journal of Big Data

Table 4 Results for 1-of-70, 1-of-256, log-70 and log-256

From: Improving deep neural network design with new text data representations

Embedding	Alphabet size	Time per Epoch	Accuracy
log-m	70	13.5s	73.106
log-m	256	13.5s	74.945
1-of-m	70	65.5s	69.845
1-of-m	256	77.7s	69.748

Best results in italic

Back to article page