zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-06-26 02:33:23 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

clean up

Browse Source
This commit is contained in:
Goekdeniz-Guelmez 2025-01-31 21:37:15 +01:00
parent 5998272ec2
commit a03d434bb9
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
llms/mlx_lm/lora.py
View File

@ -324,7 +324,7 @@ def evaluate_model(args, model: nn.Module, tokenizer: TokenizerWrapper, test_set
delta=args.delta,
loss_type=args.dpo_loss_type,
)
print(f"Test loss {test_loss:.8f}, Rewards: {test_rewards[0]:.8f}, {test_rewards[1]:.8f}")
print(f"Test loss {test_loss:.3f}, Rewards: {test_rewards[0]:.3f}, {test_rewards[1]:.3f}")
else:
test_loss = evaluate(
model=model,

4
llms/mlx_lm/tuner/dpo_trainer.py
View File

@ -327,7 +327,7 @@ def train_dpo(
if rank == 0:
print(
f"Iter {it}: "
f"Val loss {val_loss:.8f}, "
f"Val loss {val_loss:.3f}, "
f"Val chosen reward {val_rewards[0]:.3f}, "
f"Val rejected reward {val_rewards[1]:.3f}, "
f"Val accuracy {val_metrics['accuracies']:.3f}, "
@ -375,7 +375,7 @@ def train_dpo(
if rank == 0:
print(
f"Iter {it}: Train loss {train_loss:.8f}, "
f"Iter {it}: Train loss {train_loss:.3f}, "
f"Chosen reward {train_rewards[0]:.3f}, "
f"Rejected reward {train_rewards[1]:.3f}, "
f"Accuracy {avg_metrics['accuracies']:.3f}, "