Skip to content

Commit 62dc4da

Browse files
committed
Creating a modelpack
1 parent a7937f9 commit 62dc4da

File tree

1 file changed

+35
-0
lines changed

1 file changed

+35
-0
lines changed
Lines changed: 35 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,35 @@
1+
from medcat.vocab import Vocab
2+
from medcat.cdb import CDB
3+
from medcat.cat import CAT
4+
5+
6+
cdb_model_path = "data/medcat_models/cdb"
7+
cdb_model = cdb_model_path + "<NAME OF MODEL HERE>.dat" # Change to specific cdb of interest
8+
9+
vocab_model_path = "data/medcat_models/vocab"
10+
modelpack_path = "data/medcat_models/modelpack"
11+
modelpack_name = "<NAME OF output MODELPACK HERE>.dat" # Change to the name of your model
12+
13+
# Load cdb
14+
cdb = CDB.load(cdb_model)
15+
16+
# Set cdb configuration
17+
# technically we already created this during the cdb creation
18+
cdb.config.ner['min_name_len'] = 2
19+
cdb.config.ner['upper_case_limit_len'] = 3
20+
cdb.config.general['spell_check'] = True
21+
cdb.config.linking['train_count_threshold'] = 10
22+
cdb.config.linking['similarity_threshold'] = 0.3
23+
cdb.config.linking['train'] = True
24+
cdb.config.linking['disamb_length_limit'] = 4
25+
cdb.config.general['full_unlink'] = True
26+
27+
# Load vocab
28+
vocab = Vocab.load(vocab_model_path)
29+
30+
# Initialise the model
31+
cat = CAT(cdb=cdb, config=cdb.config, vocab=vocab)
32+
33+
# Create and save model pack
34+
cat.create_model_pack(save_dir_path=modelpack_path, model_pack_name=modelpack_name)
35+

0 commit comments

Comments
 (0)