search

chrisjbryant / lmgec-lite

A language model-based approach to Grammatical Error Correction for English that uses minimal annotated data.
49 stars 18 forks source link

Make loading huge KenLM models better when handling huge LM models #4

Closed gurunathparasaram closed 5 years ago

gurunathparasaram commented 5 years ago

Fix for issue #3: leads to better system memory efficiency and quicker loading for huge models(sizes in range of tens of GBs)

chrisjbryant commented 5 years ago

As answered in #3 , this fix is only recommended if you don't have enough resources to load the whole model into memory.