zchen/b2txt25
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

tpu

Browse Source
This commit is contained in:
Zchen
2025-10-12 20:59:55 +08:00
parent db6108f250
commit 12d571c70b
2 changed files with 1 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
model_training_nnn/rnn_args.yaml
View File

@@ -19,7 +19,7 @@ mode: train
use_amp: true # whether to use automatic mixed precision (AMP) for training
# TPU and distributed training settings
use_tpu: false # whether to use TPU for training (set to true for TPU)
use_tpu: true # whether to use TPU for training (set to true for TPU)
num_tpu_cores: 8 # number of TPU cores to use (typically 8 for v3-8 or v4-8)
gradient_accumulation_steps: 1 # number of gradient accumulation steps for distributed training
dataloader_num_workers: 0 # set to 0 for TPU to avoid multiprocessing issues