summaryrefslogtreecommitdiffstats
path: root/linguistics-components/src/test/java
diff options
context:
space:
mode:
authorBjørn Christian Seime <bjorncs@yahooinc.com>2023-05-11 15:41:00 +0200
committerBjørn Christian Seime <bjorncs@yahooinc.com>2023-05-11 16:41:54 +0200
commitfe63824738fc1892221311e7ddd777efcb209f5b (patch)
treedc7d3ce16c4e56ab7cbbc941f2cb9f162d6dacb2 /linguistics-components/src/test/java
parentae700d12753e1a81de4def087d2f64607f0361df (diff)
Disable special tokens by default
Diffstat (limited to 'linguistics-components/src/test/java')
-rw-r--r--linguistics-components/src/test/java/com/yahoo/language/huggingface/HuggingFaceTokenizerTest.java1
1 files changed, 1 insertions, 0 deletions
diff --git a/linguistics-components/src/test/java/com/yahoo/language/huggingface/HuggingFaceTokenizerTest.java b/linguistics-components/src/test/java/com/yahoo/language/huggingface/HuggingFaceTokenizerTest.java
index f9fa0ef2afe..c79ecbfbfbe 100644
--- a/linguistics-components/src/test/java/com/yahoo/language/huggingface/HuggingFaceTokenizerTest.java
+++ b/linguistics-components/src/test/java/com/yahoo/language/huggingface/HuggingFaceTokenizerTest.java
@@ -71,6 +71,7 @@ class HuggingFaceTokenizerTest {
private static HuggingFaceTokenizer createTokenizer(Path tmp, String model) throws IOException {
return new HuggingFaceTokenizer.Builder()
+ .addSpecialTokens(false)
.addDefaultModel(decompressModelFile(tmp, Paths.get("src/test/models/huggingface/%s.json.gz".formatted(model))))
.build();
}