CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-11 10:36:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

convert : better logging of partially reflinkable tensors

Browse Source
This commit is contained in:
Francis Couture-Harpin
2025-09-09 11:00:00 -04:00
parent 2499e47cfd
commit 34f37c283b
2 changed files with 12 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
gguf-py/gguf/gguf_writer.py
View File

@@ -201,7 +201,7 @@ class GGUFWriter:
assert len(filenames) == len(self.tensors)
for name, tensors in zip(filenames, self.tensors):
total_size = sum(ti.nbytes for ti in tensors.values())
reflinkable_size = count_reflinkable_size(ti.tensor for ti in tensors.values()) if self.use_reflinks else 0
reflinkable_size = count_reflinkable_size((name, ti.tensor) for name, ti in tensors.items()) if self.use_reflinks else 0
logger.info(f"{name}: n_tensors = {len(tensors)}, total_size = {GGUFWriter.format_n_bytes_to_str(total_size)}{', reflinked = ' + GGUFWriter.format_n_bytes_to_str(total_size - reflinkable_size) if self.use_reflinks else ''}")
if self.dry_run: