Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/stanfordnlp/stanza.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJohn Bauer <horatio@gmail.com>2022-11-07 02:47:47 +0300
committerJohn Bauer <horatio@gmail.com>2022-11-07 02:48:43 +0300
commit28bd18b4db08e77d30edc945042416364d4209fb (patch)
tree6491bc953c1031bff1d7c5d2e281cd7fedde1dca
parent0d408676e6c6d68406ca80bc51ec614d8dbc8296 (diff)
Log the open nodes used in a model
-rw-r--r--stanza/models/constituency/trainer.py1
1 files changed, 1 insertions, 0 deletions
diff --git a/stanza/models/constituency/trainer.py b/stanza/models/constituency/trainer.py
index 07474c54..eec82619 100644
--- a/stanza/models/constituency/trainer.py
+++ b/stanza/models/constituency/trainer.py
@@ -421,6 +421,7 @@ def build_trainer(args, train_trees, dev_trees, silver_trees, foundation_cache,
# compound unary or compound open nodes which doesn't exist in the
# train set. it just means we probably won't ever get that right
open_nodes = get_open_nodes(train_trees, args)
+ logger.info("Using the following open nodes:\n %s", "\n ".join(map(str, open_nodes)))
# at this point we have:
# pretrain