Add enconding parameter to vocab.json opening to fix errors

This commit is contained in:
Joonas Pihlajamaa 2022-10-23 11:55:01 +03:00
parent 4196856c7b
commit 4e887dc350

View File

@ -234,7 +234,7 @@ dir_tokenizer = tokenizer.name_or_path
# output in the same directory as the model
fname_out = dir_out + "/ggml-model.bin"
with open(dir_tokenizer + "/vocab.json", "r") as f:
with open(dir_tokenizer + "/vocab.json", "r", encoding="utf8") as f:
tokens = json.load(f)
# use 16-bit or 32-bit floats