zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-07-01 05:41:12 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

removing tokenizer and updates

Browse Source
This commit is contained in:
Goekdeniz-Guelmez 2025-01-26 15:25:27 +01:00
parent 4d0e52f7c8
commit 557649d8da
1 changed files with 0 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
llms/mlx_lm/tuner/dpo_trainer.py
View File

@ -128,7 +128,6 @@ def dpo_loss(
'chosen_logits_mean': mx.mean(policy_chosen_score) 'chosen_logits_mean': mx.mean(policy_chosen_score)
} }
return mx.mean(losses), reward, num_tokens, metrics return mx.mean(losses), reward, num_tokens, metrics
@ -180,7 +179,6 @@ def evaluate_dpo(
model, model,
reference_model, reference_model,
dataset, dataset,
tokenizer,
batch_size, batch_size,
num_batches, num_batches,
beta: float, beta: float,
@ -328,7 +326,6 @@ def train_dpo(
model=model, model=model,
reference_model=reference_model, reference_model=reference_model,
dataset=val_dataset, dataset=val_dataset,
tokenizer=tokenizer,
batch_size=args.batch_size, batch_size=args.batch_size,
num_batches=args.val_batches, num_batches=args.val_batches,
max_seq_length=args.max_seq_length, max_seq_length=args.max_seq_length,