search

shuaijiang / Whisper-Finetune

Fine-tune the Whisper speech recognition model to support training without timestamp data, training with timestamp data, and training without speech data. Accelerate inference and support Web deployment, Windows desktop deployment, and Android deployment
Apache License 2.0
164 stars 8 forks source link

多卡训练，现存需要多少 #1

Open huacilang opened 2 months ago

huacilang commented 2 months ago

你好，用V100多卡训练aishell数据集，从2张到4张到8张，都报torch.cuda.OutOfMemoryError: CUDA out of memory，是哪里需要调整吗

daiqizhi commented 1 month ago

--per_device_train_batch_size 4 --per_device_eval_batch_size 4