explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-12-26 18:06:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

Reduce batch size during pretrain

Browse Source
This commit is contained in:
Matthew Honnibal 2018-12-10 15:30:23 +00:00
parent 83ac227bd3
commit df15279e88
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
spacy/cli/pretrain.py
View File

@ -110,7 +110,7 @@ def pretrain(
msg.row(("#", "# Words", "Total Loss", "Loss", "w/s"), **row_settings) msg.row(("#", "# Words", "Total Loss", "Loss", "w/s"), **row_settings)
for epoch in range(nr_iter): for epoch in range(nr_iter):
for batch in util.minibatch_by_words( for batch in util.minibatch_by_words(
((text, None) for text in texts), size=5000 ((text, None) for text in texts), size=3000
): ):
docs = make_docs(nlp, [text for (text, _) in batch]) docs = make_docs(nlp, [text for (text, _) in batch])
loss = make_update(model, docs, optimizer, drop=dropout) loss = make_update(model, docs, optimizer, drop=dropout)