Skip to content

Add tokenizer.ggml.pre to gguf-new-metadata.py #7627

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
May 30, 2024
Merged

Add tokenizer.ggml.pre to gguf-new-metadata.py #7627

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
95a5896
Add tokenizer.ggml.pre to gguf-new-metadata.py
Galunid May 29, 2024
62820ee
Fix accidental print
Galunid May 29, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions gguf-py/scripts/gguf-new-metadata.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,6 +144,7 @@ def main() -> None:
parser.add_argument("--general-description", type=str, help="The models general.description", metavar='"Description ..."')
parser.add_argument("--chat-template", type=str, help="Chat template string (or JSON string containing templates)", metavar='"{% ... %} ..."')
parser.add_argument("--chat-template-config", type=Path, help="Config file containing chat template(s)", metavar='tokenizer_config.json')
parser.add_argument("--pre-tokenizer", type=str, help="The models tokenizer.ggml.pre", metavar='"pre tokenizer"')
parser.add_argument("--remove-metadata", action="append", type=str, help="Remove metadata (by key name) from output model", metavar='general.url')
parser.add_argument("--special-token", action="append", type=str, help="Special token by value", nargs=2, metavar=(' | '.join(token_names.keys()), '"<token>"'))
parser.add_argument("--special-token-by-id", action="append", type=str, help="Special token by id", nargs=2, metavar=(' | '.join(token_names.keys()), '0'))
Expand Down Expand Up @@ -172,6 +173,9 @@ def main() -> None:
if template:
new_metadata[gguf.Keys.Tokenizer.CHAT_TEMPLATE] = MetadataDetails(gguf.GGUFValueType.STRING, template)

if args.pre_tokenizer:
new_metadata[gguf.Keys.Tokenizer.PRE] = MetadataDetails(gguf.GGUFValueType.STRING, args.pre_tokenizer)

if remove_metadata:
logger.warning('*** Warning *** Warning *** Warning **')
logger.warning('* Most metadata is required for a fully functional GGUF file,')
Expand Down
Loading