search

ml-explore / mlx-examples

Examples in the MLX framework
MIT License
5.92k stars 838 forks source link

lora resume error #841

Closed l0d0v1c closed 3 months ago

l0d0v1c commented 3 months ago

command:

python -m mlx_lm.lora --model ./Phi-3-mini-128k-instruct --adapter-path adapters-pluc100 --resume-adapter-file adapters-pluc100 --data ./data-pluc100 --train --iters 200 --lora-layers 8

worked in previous version (0.14) but now it gives with 0.14.3

File "/Users/pro/mambaforge/envs/torch/lib/python3.11/site-packages/mlx_lm/lora.py", line 173, in train_model print(f"Loading pretrained adapters from {resume_adapter_file}") ^^^^^^^^^^^^^^^^^^^ NameError: name 'resume_adapter_file' is not defined

shenshuihai01 commented 3 months ago

me too

awni commented 3 months ago

This is fixed on main. We will do a release soon with the update.