update alfworld (ba53fcb9) · Commits · 张泽凯 / VAGEN

vagen/env/README.md

+16 −0

Original line number	Diff line number	Diff line
		@@ -70,3 +70,19 @@ alfworld-download
		python vagen/env/alfworld/startx.py 0
		python vagen/server/server.py
		```

		### ALFWorld
		```
		pip install ai2thor==2.1.0
		pip install alfworld==0.3.2
		pip3 install numpy==1.23.5
		pip3 install protobuf==3.20.3
		pip3 install pydantic==1.10.14
		pip3 install pydantic-core==2.16.3
		pip3 uninstall frozenlist gradio murmurhash preshed spacy srsly thinc weasel aiosignal annotated-types blis catalogue cloudpathlib cymem

		# Set the data path and download before running the server
		export ALFWORLD_DATA=<storage_path>
		alfworld-download
		python vagen/env/server.py
		```

+0 −9

Original line number	Diff line number	Diff line
		@@ -2,24 +2,15 @@ dataset:
		data_path: '$ALFWORLD_DATA/json_2.1.1/train'
		eval_id_data_path: '$ALFWORLD_DATA/json_2.1.1/valid_seen' # null/None to disable
		eval_ood_data_path: '$ALFWORLD_DATA/json_2.1.1/valid_unseen' # null/None to disable
		<<<<<<< HEAD
		num_train_games: 1 # max training games (<=0 indicates full dataset)
		num_eval_games: 1 # max evaluation games (<=0 indicates full dataset)
		=======
		num_train_games: 10 # max training games (<=0 indicates full dataset)
		num_eval_games: 2 # max evaluation games (<=0 indicates full dataset)
		>>>>>>> 581e528 (init alfworld)

		logic:
		domain: '$ALFWORLD_DATA/logic/alfred.pddl' # PDDL domain file that defines the world dynamics
		grammar: '$ALFWORLD_DATA/logic/alfred.twl2' # Grammar file that defines the text feedbacks

		env:
		<<<<<<< HEAD
		type: 'AlfredThorEnv' # 'AlfredTWEnv' or 'AlfredThorEnv' or 'AlfredHybrid'
		=======
		type: 'AlfredThorEnv' # 'AlfredTWEnv' or 'AlfredThorEnv' or 'AlfredHybrid'
		>>>>>>> 581e528 (init alfworld)
		regen_game_files: False # check if game is solvable by expert and save to game.tw-pddl file
		domain_randomization: False # shuffle Textworld print order and object id nums
		task_types: [1, 2, 3, 4, 5, 6] # task-type ids: 1 - Pick & Place, 2 - Examine in Light, 3 - Clean & Place, 4 - Heat & Place, 5 - Cool & Place, 6 - Pick Two & Place

+0 −4

Original line number	Diff line number	Diff line
		@@ -59,11 +59,7 @@ python3 -m vagen.trainer.main_ppo \
		trainer.critic_warmup=0 \
		trainer.logger=['console','wandb'] \
		trainer.project_name='vagen_debug' \
		<<<<<<< HEAD
		trainer.experiment_name='grpo_mask_loss_alfworld_text_debug' \
		=======
		trainer.experiment_name='grpo_mask_loss_frozenlake_vision_debug' \
		>>>>>>> 581e528 (init alfworld)
		trainer.n_gpus_per_node=1 \
		trainer.nnodes=1 \
		trainer.save_freq=100 \