Loading scripts/exps/state_reward/frozenlake/grounding_worldmodeling/run.sh +3 −0 Original line number Original line Diff line number Diff line Loading @@ -78,4 +78,7 @@ python3 -m vagen.trainer.main_ppo \ trainer.val_before_train=True \ trainer.val_before_train=True \ trainer.val_generations_to_log_to_wandb=8 \ trainer.val_generations_to_log_to_wandb=8 \ rollout_manager.n_trajectory=2 \ rollout_manager.n_trajectory=2 \ rollout_manager.use_service=True \ rollout_manager.timeout=300 \ rollout_manager.base_url="http://localhost:$PORT" \ 2>&1 | tee $EXPERIMENT_NAME.log 2>&1 | tee $EXPERIMENT_NAME.log scripts/exps/state_reward/frozenlake/grounding_worldmodeling/run_tmux.sh +3 −0 Original line number Original line Diff line number Diff line Loading @@ -129,6 +129,9 @@ tmux send-keys -t "$TRAIN_SESSION" "python3 -m vagen.trainer.main_ppo \\ trainer.val_before_train=True \\ trainer.val_before_train=True \\ trainer.val_generations_to_log_to_wandb=8 \\ trainer.val_generations_to_log_to_wandb=8 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.use_service=True \\ rollout_manager.timeout=300 \\ rollout_manager.base_url=\"http://localhost:$PORT\" \\ 2>&1 | tee $EXPERIMENT_NAME.log" C-m 2>&1 | tee $EXPERIMENT_NAME.log" C-m echo "-----------------------------------------" echo "-----------------------------------------" Loading scripts/exps/state_reward/maniskill/grounding_worldmodeling/run.sh +2 −1 Original line number Original line Diff line number Diff line Loading @@ -86,6 +86,7 @@ python3 -m vagen.trainer.main_ppo \ trainer.val_generations_to_log_to_wandb=8 \ trainer.val_generations_to_log_to_wandb=8 \ rollout_manager.n_trajectory=2 \ rollout_manager.n_trajectory=2 \ rollout_manager.use_service=True \ rollout_manager.use_service=True \ rollout_manager.timeout=240 \ rollout_manager.timeout=300 \ rollout_manager.base_url="http://localhost:5001" \ rollout_manager.base_url="http://localhost:5001" \ 2>&1 | tee $EXPERIMENT_NAME.log 2>&1 | tee $EXPERIMENT_NAME.log scripts/exps/state_reward/maniskill/grounding_worldmodeling/run_tmux.sh +1 −1 Original line number Original line Diff line number Diff line Loading @@ -130,7 +130,7 @@ tmux send-keys -t "$TRAIN_SESSION" "python3 -m vagen.trainer.main_ppo \\ trainer.val_generations_to_log_to_wandb=8 \\ trainer.val_generations_to_log_to_wandb=8 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.use_service=True \\ rollout_manager.use_service=True \\ rollout_manager.timeout=240 \\ rollout_manager.timeout=300 \\ rollout_manager.base_url=\"http://localhost:$PORT\" \\ rollout_manager.base_url=\"http://localhost:$PORT\" \\ 2>&1 | tee $EXPERIMENT_NAME.log" C-m 2>&1 | tee $EXPERIMENT_NAME.log" C-m Loading scripts/exps/state_reward/navigation/grounding_worldmodeling/run.sh +1 −1 Original line number Original line Diff line number Diff line Loading @@ -82,7 +82,7 @@ python3 -m vagen.trainer.main_ppo \ trainer.val_generations_to_log_to_wandb=8 \ trainer.val_generations_to_log_to_wandb=8 \ rollout_manager.n_trajectory=1 \ rollout_manager.n_trajectory=1 \ rollout_manager.use_service=True \ rollout_manager.use_service=True \ rollout_manager.timeout=240 \ rollout_manager.timeout=300 \ +rollout_manager.mini_batch_size=64 \ +rollout_manager.mini_batch_size=64 \ rollout_manager.base_url="http://localhost:5000" \ rollout_manager.base_url="http://localhost:5000" \ 2>&1 | tee $EXPERIMENT_NAME.log 2>&1 | tee $EXPERIMENT_NAME.log Loading
scripts/exps/state_reward/frozenlake/grounding_worldmodeling/run.sh +3 −0 Original line number Original line Diff line number Diff line Loading @@ -78,4 +78,7 @@ python3 -m vagen.trainer.main_ppo \ trainer.val_before_train=True \ trainer.val_before_train=True \ trainer.val_generations_to_log_to_wandb=8 \ trainer.val_generations_to_log_to_wandb=8 \ rollout_manager.n_trajectory=2 \ rollout_manager.n_trajectory=2 \ rollout_manager.use_service=True \ rollout_manager.timeout=300 \ rollout_manager.base_url="http://localhost:$PORT" \ 2>&1 | tee $EXPERIMENT_NAME.log 2>&1 | tee $EXPERIMENT_NAME.log
scripts/exps/state_reward/frozenlake/grounding_worldmodeling/run_tmux.sh +3 −0 Original line number Original line Diff line number Diff line Loading @@ -129,6 +129,9 @@ tmux send-keys -t "$TRAIN_SESSION" "python3 -m vagen.trainer.main_ppo \\ trainer.val_before_train=True \\ trainer.val_before_train=True \\ trainer.val_generations_to_log_to_wandb=8 \\ trainer.val_generations_to_log_to_wandb=8 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.use_service=True \\ rollout_manager.timeout=300 \\ rollout_manager.base_url=\"http://localhost:$PORT\" \\ 2>&1 | tee $EXPERIMENT_NAME.log" C-m 2>&1 | tee $EXPERIMENT_NAME.log" C-m echo "-----------------------------------------" echo "-----------------------------------------" Loading
scripts/exps/state_reward/maniskill/grounding_worldmodeling/run.sh +2 −1 Original line number Original line Diff line number Diff line Loading @@ -86,6 +86,7 @@ python3 -m vagen.trainer.main_ppo \ trainer.val_generations_to_log_to_wandb=8 \ trainer.val_generations_to_log_to_wandb=8 \ rollout_manager.n_trajectory=2 \ rollout_manager.n_trajectory=2 \ rollout_manager.use_service=True \ rollout_manager.use_service=True \ rollout_manager.timeout=240 \ rollout_manager.timeout=300 \ rollout_manager.base_url="http://localhost:5001" \ rollout_manager.base_url="http://localhost:5001" \ 2>&1 | tee $EXPERIMENT_NAME.log 2>&1 | tee $EXPERIMENT_NAME.log
scripts/exps/state_reward/maniskill/grounding_worldmodeling/run_tmux.sh +1 −1 Original line number Original line Diff line number Diff line Loading @@ -130,7 +130,7 @@ tmux send-keys -t "$TRAIN_SESSION" "python3 -m vagen.trainer.main_ppo \\ trainer.val_generations_to_log_to_wandb=8 \\ trainer.val_generations_to_log_to_wandb=8 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.n_trajectory=2 \\ rollout_manager.use_service=True \\ rollout_manager.use_service=True \\ rollout_manager.timeout=240 \\ rollout_manager.timeout=300 \\ rollout_manager.base_url=\"http://localhost:$PORT\" \\ rollout_manager.base_url=\"http://localhost:$PORT\" \\ 2>&1 | tee $EXPERIMENT_NAME.log" C-m 2>&1 | tee $EXPERIMENT_NAME.log" C-m Loading
scripts/exps/state_reward/navigation/grounding_worldmodeling/run.sh +1 −1 Original line number Original line Diff line number Diff line Loading @@ -82,7 +82,7 @@ python3 -m vagen.trainer.main_ppo \ trainer.val_generations_to_log_to_wandb=8 \ trainer.val_generations_to_log_to_wandb=8 \ rollout_manager.n_trajectory=1 \ rollout_manager.n_trajectory=1 \ rollout_manager.use_service=True \ rollout_manager.use_service=True \ rollout_manager.timeout=240 \ rollout_manager.timeout=300 \ +rollout_manager.mini_batch_size=64 \ +rollout_manager.mini_batch_size=64 \ rollout_manager.base_url="http://localhost:5000" \ rollout_manager.base_url="http://localhost:5000" \ 2>&1 | tee $EXPERIMENT_NAME.log 2>&1 | tee $EXPERIMENT_NAME.log
