ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update save.py

Browse source
This commit is contained in:
Daniel Han 2024-10-10 23:22:17 -07:00
commit 71ac721cec
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
unsloth/save.py
View file

@ -1063,7 +1063,7 @@ def save_to_gguf(
n_cpus *= 2
# Concurrency from https://rentry.org/llama-cpp-conversions#merging-loras-into-a-model
final_location = (Path(model_directory) / f"unsloth.{first_conversion.upper()}.gguf").absolute()
final_location = str((Path(model_directory) / f"unsloth.{first_conversion.upper()}.gguf").absolute())
print(f"Unsloth: [1] Converting model at {model_directory} into {first_conversion} GGUF format.\n"\
f"The output location will be {final_location}\n"\
@ -1132,7 +1132,7 @@ def save_to_gguf(
for quant_method in quantization_method:
if quant_method != first_conversion:
print(f"Unsloth: [2] Converting GGUF 16bit into {quant_method}. This will take 20 minutes...")
final_location = (Path(model_directory) / f"unsloth.{quant_method.upper()}.gguf").absolute()
final_location = str((Path(model_directory) / f"unsloth.{quant_method.upper()}.gguf").absolute())
command = f"./{quantize_location} {full_precision_location} "\
f"{final_location} {quant_method} {n_cpus}"