zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-06-25 01:41:19 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

chore: fix the load quantization model for deepseek coder (#203)

Browse Source 
* chore: fix the load quantization model

* change to explicitly check for quantization config
This commit is contained in:
Anchen 2023-12-30 00:25:38 +11:00 committed by GitHub
parent 31ddbd7806
commit 1cdbf9e886
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
llms/deepseek-coder/deepseek_coder.py
View File

@ -244,7 +244,7 @@ def load_model(model_path: str):
model = DeepseekCoder(model_args) model = DeepseekCoder(model_args)
weights = mx.load(str(model_path / "weights.npz")) weights = mx.load(str(model_path / "weights.npz"))
if quantization := config.get("quantization", False): if quantization is not None:
nn.QuantizedLinear.quantize_module(model, **quantization) nn.QuantizedLinear.quantize_module(model, **quantization)
model.update(tree_unflatten(list(weights.items()))) model.update(tree_unflatten(list(weights.items())))