Skip to content

load_gpt2() improvements
Compare
Choose a tag to compare
@minimaxir minimaxir released this 05 May 22:51
· 102 commits to master since this release
v0.4.2
873db52
  • load_gpt2() in a fresh session is much faster and uses much less memory when loaded. (for the 117M model, the system will stay under <2 GB RAM which is the critical point for cloud services)
  • start_tf_sess() now accepts a threads parameter, which is useful if you know exactly how many threads will be used.
Assets 3
Loading