ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2024-12-24 06:46:37 +00:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

models : fix typo in convert-h5-to-ggml.py

Browse Source 
signficant -> significant
This commit is contained in:
Ikko Ashimine 2022-12-31 02:51:08 +09:00 committed by Georgi Gerganov
parent a091581eb3
commit 3467230a77
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
models/convert-h5-to-ggml.py
View File

@ -56,7 +56,7 @@ def bytes_to_unicode():
The reversible bpe codes work on unicode strings.
This means you need a large # of unicode characters in your vocab if you want to avoid UNKs.
When you're at something like a 10B token dataset you end up needing around 5K for decent coverage.
This is a signficant percentage of your normal, say, 32K bpe vocab.
This is a significant percentage of your normal, say, 32K bpe vocab.
To avoid that, we want lookup tables between utf-8 bytes and unicode strings.
And avoids mapping to whitespace/control characters the bpe code barfs on.
"""