Performance issue in the difinition of get_cache_fn, server/embedding_as_service/text/xlnet/models/train.py(P1)

amansrivastava17 / embedding-as-service

One-Stop Solution to encode sentence to fixed length vectors from various embedding techniques

MIT License

204 stars 29 forks source link

Performance issue in the difinition of get_cache_fn, server/embedding_as_service/text/xlnet/models/train.py(P1) #66

Closed DLPerf closed 3 years ago

DLPerf commented 3 years ago

Hello, I found a performance issue in the definition of get_cache_fn, server/embedding_as_service/text/xlnet/models/train.py, tf.zeros will be created repeatedly during program execution, resulting in reduced efficiency. I think it should be created before the loop in cache_fn.

Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.

amansrivastava17 commented 3 years ago

Hi DLperf, Thank you for pointing the issue, will be glad if you can create a PR. Will be happy to review.