support to export gguf q4_0 and q4_1 format #393
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: n1ck-guo <heng.guo@intel.com>
for more information, see https://pre-commit.ci
Signed-off-by: n1ck-guo <heng.guo@intel.com>
for more information, see https://pre-commit.ci
Signed-off-by: n1ck-guo <heng.guo@intel.com>
for more information, see https://pre-commit.ci
wenhuach21
reviewed
Dec 27, 2024
wenhuach21
reviewed
Dec 27, 2024
wenhuach21
reviewed
Dec 30, 2024
wenhuach21
reviewed
Dec 30, 2024
Signed-off-by: n1ck-guo <heng.guo@intel.com>
wenhuach21
reviewed
Dec 30, 2024
wenhuach21
reviewed
Dec 30, 2024
wenhuach21
reviewed
Dec 30, 2024
wenhuach21
reviewed
Dec 30, 2024
wenhuach21
reviewed
Dec 30, 2024
wenhuach21
reviewed
Dec 31, 2024
Signed-off-by: n1ck-guo <heng.guo@intel.com>
wenhuach21
reviewed
Dec 31, 2024
Signed-off-by: n1ck-guo <heng.guo@intel.com>
wenhuach21
reviewed
Dec 31, 2024
wenhuach21
reviewed
Dec 31, 2024
| endianness: gguf.GGUFEndian | ||
| use_temp_file: bool | ||
| lazy: bool | ||
| part_names: list[str] |
There was a problem hiding this comment.
why not calling gguf code directly
There was a problem hiding this comment.
cannot. Different model series use different class to write gguf file.
Signed-off-by: n1ck-guo <heng.guo@intel.com>
Signed-off-by: n1ck-guo <heng.guo@intel.com>
|
remember to add checker in save_quantized and add warning when combined with fp_layers later |
wenhuach21
approved these changes
Jan 7, 2025
wenhuach21
reviewed
Jan 7, 2025
wenhuach21
reviewed
Jan 7, 2025
| for format in formats: | ||
| if format not in supported_formats: | ||
| raise ValueError(f"{format} is not supported, we only support {supported_formats}") | ||
| if format in ["gguf:q4_0", "gguf:q4_1"]: |
There was a problem hiding this comment.
support gguf later if we could inference the exact type by the quantization config
wenhuach21
reviewed
Jan 7, 2025
wenhuach21
requested changes
Jan 7, 2025
wenhuach21
reviewed
Jan 7, 2025
| @@ -1267,6 +1267,14 @@ def save_quantized(self, output_dir=None, format="auto_round", inplace=True, **k | |||
| if processor is not None: | |||
| processor.save_pretrained(output_dir) | |||
| return | |||
| if format in ["gguf:q4_0", "gguf:q4_1"]: | |||
| if self.group_size != 32: | |||
wenhuach21
approved these changes
Jan 7, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
using llama.cpp(llama-cli) test q4_0 and q4_1 quantized file, work well.
#288