Commit aa1e3c09 authored by jameskrw's avatar jameskrw
Browse files

added gae mask in scripts

parent 4028cc0f
Loading
Loading
Loading
Loading
+1 −0
Original line number Diff line number Diff line
@@ -76,6 +76,7 @@ python3 -m vagen.trainer.main_ppo \
    rollout_manager.window_size=5 \
    rollout_manager.use_multi_turn_reward=False \
    rollout_manager.use_loss_mask=False \
    rollout_manager.use_gae_mask=False \
    trainer.val_before_train=True \
    trainer.val_generations_to_log_to_wandb=8 \
    rollout_manager.n_trajectory=1 \
+1 −0
Original line number Diff line number Diff line
@@ -76,6 +76,7 @@ python3 -m vagen.trainer.main_ppo \
    rollout_manager.window_size=5 \
    rollout_manager.use_multi_turn_reward=False \
    rollout_manager.use_loss_mask=True \
    rollout_manager.use_gae_mask=True \
    trainer.val_before_train=True \
    trainer.val_generations_to_log_to_wandb=8 \
    rollout_manager.n_trajectory=1 \
+1 −0
Original line number Diff line number Diff line
@@ -76,6 +76,7 @@ python3 -m vagen.trainer.main_ppo \
    rollout_manager.window_size=5 \
    rollout_manager.use_multi_turn_reward=False \
    rollout_manager.use_loss_mask=True \
    rollout_manager.use_gae_mask=True \
    trainer.val_before_train=True \
    trainer.val_generations_to_log_to_wandb=8 \
    rollout_manager.n_trajectory=1 \
+1 −0
Original line number Diff line number Diff line
@@ -77,6 +77,7 @@ python3 -m vagen.trainer.main_ppo \
    rollout_manager.window_size=5 \
    rollout_manager.use_multi_turn_reward=True \
    rollout_manager.use_loss_mask=True \
    rollout_manager.use_gae_mask=True \
    trainer.val_before_train=True \
    trainer.val_generations_to_log_to_wandb=8 \
    rollout_manager.n_trajectory=1 \
+1 −0
Original line number Diff line number Diff line
@@ -76,6 +76,7 @@ python3 -m vagen.trainer.main_ppo \
    rollout_manager.window_size=5 \
    rollout_manager.use_multi_turn_reward=True \
    rollout_manager.use_loss_mask=True \
    rollout_manager.use_gae_mask=True \
    trainer.val_before_train=True \
    trainer.val_generations_to_log_to_wandb=8 \
    rollout_manager.n_trajectory=1 \
Loading