]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commitdiff
models : change default encoding to utf8 (#605)
authorKamilake <redacted>
Wed, 22 Mar 2023 19:17:24 +0000 (04:17 +0900)
committerGitHub <redacted>
Wed, 22 Mar 2023 19:17:24 +0000 (21:17 +0200)
models/convert-h5-to-ggml.py

index b06ad23d328be4922ac1d95b2019c13baaf705c5..3ddee220fe1580032e79ef3cde4218979369f9d9 100644 (file)
@@ -79,11 +79,11 @@ dir_model   = sys.argv[1]
 dir_whisper = sys.argv[2]
 dir_out     = sys.argv[3]
 
-with open(dir_model + "/vocab.json", "r") as f:
+with open(dir_model + "/vocab.json", "r", encoding="utf8") as f:
     encoder = json.load(f)
-with open(dir_model + "/added_tokens.json", "r") as f:
+with open(dir_model + "/added_tokens.json", "r", encoding="utf8") as f:
     encoder_added = json.load(f)
-with open(dir_model + "/config.json", "r") as f:
+with open(dir_model + "/config.json", "r", encoding="utf8") as f:
     hparams = json.load(f)
 
 model = WhisperForConditionalGeneration.from_pretrained(dir_model)