symhsym 07e526a231 Update train_voxcpm_finetune.py ...

修改了issue#185中提到的问题，在训练时进行validate会对原模型执行to(torch.bfloat16)然后to(torch.float32)的操作，这样可能导致模型数值浮动，因此这个修改让validate步骤保留原模型数值

2026-02-11 11:17:47 +08:00

..

test_voxcpm_ft_infer.py

Print all log messages to stderr instead of stdout

2026-01-12 15:30:45 +08:00

test_voxcpm_lora_infer.py

Print all log messages to stderr instead of stdout

2026-01-12 15:30:45 +08:00

train_voxcpm_finetune.py

Update train_voxcpm_finetune.py

2026-02-11 11:17:47 +08:00