From 624f74d8b3e84d60e542766d60018510d3af248d Mon Sep 17 00:00:00 2001 From: Paul Wise Date: Sat, 26 Feb 2022 16:07:29 +0800 Subject: [PATCH 3/4] Only load data for train_gensim when it is being used by tests Otherwise it will get loaded even when tests that use it aren't being run. --- gensim/test/test_fasttext.py | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/gensim/test/test_fasttext.py b/gensim/test/test_fasttext.py index ecc44a30..be831819 100644 --- a/gensim/test/test_fasttext.py +++ b/gensim/test/test_fasttext.py @@ -916,14 +916,18 @@ def test_sg_hs_training_fromfile(shrink_windows): assert overlap_count >= 2, message -with open(datapath('toy-data.txt')) as fin: - TOY_SENTENCES = [fin.read().strip().split(' ')] +TOY_SENTENCES = None def train_gensim(bucket=100, min_count=5): # # Set parameters to match those in the load_native function # + # Global toy data so there is one instance for all tests + global TOY_SENTENCES + if TOY_SENTENCES is None: + with open(datapath('toy-data.txt')) as fin: + TOY_SENTENCES = [fin.read().strip().split(' ')] model = FT_gensim(bucket=bucket, vector_size=5, alpha=0.05, workers=1, sample=0.0001, min_count=min_count) model.build_vocab(TOY_SENTENCES) model.train(TOY_SENTENCES, total_examples=len(TOY_SENTENCES), epochs=model.epochs) -- 2.36.0