tangger/Search-R1
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

add grpo script

Browse Source
This commit is contained in:
PeterGriffinJin
2025-03-12 15:16:03 +00:00
parent 1bd9cf1749
commit c4e0269cfc
2 changed files with 84 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
train_ppo.sh
View File

@@ -77,11 +77,11 @@ PYTHONUNBUFFERED=1 python3 -m verl.trainer.main_ppo \
trainer.n_gpus_per_node=8 \
trainer.nnodes=1 \
trainer.save_freq=100 \
trainer.test_freq=100 \
trainer.test_freq=50 \
trainer.project_name=$WAND_PROJECT \
trainer.experiment_name=$EXPERIMENT_NAME \
trainer.total_epochs=15 \
trainer.total_training_steps=300 \
trainer.total_training_steps=305 \
trainer.default_hdfs_dir=null \
trainer.default_local_dir=verl_checkpoints/$EXPERIMENT_NAME \
max_turns=2 \