fixed Encoder(When using vocab_expansion) and GensenSingle for CPU

Maluuba / gensen

Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning

Other

311 stars 43 forks source link

Open najafmurtaza opened 4 years ago

najafmurtaza commented 4 years ago

GensenSingle was loading model directly to GPU instead of checking with self.cuda().

When using vocab_expansion
1. Encoder had bug with try,except where it was moving all tensors to GPU by default. Now added cuda attribute, which moves the data using self.cuda() checks.

msftgits commented 4 years ago

All CLA requirements met.