Layout: FastText
Created on February 4|Last edited on February 4
Comment
Exploring fastText models
In this quick project, I explored baselines from Facebook Research's fastText library for text representation and classification.
Optimal Learning Rate 0.5-0.6
Optimal learning rate appears to be around 0.5-0.6 (yellows). Lower values show higher loss (red) and higher values also show slightly higher loss (blue).
Experiments varying learning rate
Ngram size is most important
Word Embedding and Ngram Size
Add a comment