openai / gpt-2

Code for the paper "Language Models are Unsupervised Multitask Learners"
https://openai.com/blog/better-language-models/
Other
22.57k stars 5.53k forks source link

batch size redudency #197

Open glsscnnn opened 5 years ago

glsscnnn commented 5 years ago
if batch_size is None:
    batch_size = 1

Unnecessary batch_size = 1 is set as default what's the point of this? Why not just catch error or something seems more useful.