Closed woltium closed 1 year ago
Your result looks uncommon.
Basically, the Final Exp Ratio
represents the proportion of explored area of each scene.
But here they are all 1.8%. Well, maybe the mp3d dataset is broken.
Maybe you can try Gibson dataset instead to see if the final explored ratio is up to 90%.
thanks for your aggestion
i tried
python main.py --exp_name 'eval_coscan_mp3dhq0f' --scenes_file scenes/hq-a.scenes --dump_location std --num_episodes 2 --load_global train/models/ma3_history/hrl3.global
but got:
2023-04-10 11:31:36,764 Computing map for Bolton
2023-04-10 11:31:37,138 Computing map for Avonia
2023-04-10 11:31:37,194 Computing map for Applewold
2023-04-10 11:31:37,376 Computing map for Brevort
2023-04-10 11:31:38,297 Computing map for Crandon
2023-04-10 11:31:38,476 Computing map for Andover
2023-04-10 11:31:38,515 Computing map for Cooperstown
2023-04-10 11:31:38,748 Computing map for Caruthers
2023-04-10 11:31:39,658 Computing map for Adrian
Loading global train/models/ma3_history/hrl3.global
INFO:root:Time: 00d 00h 00m 06s, num timesteps 0, FPS 0,
Rewards:
Losses:
INFO:root:Time: 00d 00h 17m 52s, num timesteps 2250, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 00h 33m 49s, num timesteps 4500, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 00h 44m 49s, num timesteps 6750, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 00h 56m 07s, num timesteps 9000, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 01h 07m 09s, num timesteps 11250, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 01h 18m 15s, num timesteps 13500, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 01h 30m 08s, num timesteps 15750, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 01h 40m 16s, num timesteps 18000, FPS 2,
Rewards:
Losses:
INFO:root:Time: 00d 01h 50m 38s, num timesteps 20250, FPS 3,
Rewards:
Losses:
INFO:root:Time: 00d 02h 00m 45s, num timesteps 22500, FPS 3,
Rewards:
Losses:
INFO:root:Time: 00d 02h 11m 41s, num timesteps 24750, FPS 3,
Rewards:
Losses:
2023-04-10 13:56:18,123 Computing map for Avonia
2023-04-10 13:56:19,498 Computing map for Brevort
2023-04-10 13:56:19,550 Computing map for Crandon
2023-04-10 13:56:19,619 Computing map for Bolton
2023-04-10 13:56:20,018 Computing map for Caruthers
2023-04-10 13:56:20,129 Computing map for Cooperstown
2023-04-10 13:56:20,155 Computing map for Adrian
2023-04-10 13:56:20,255 Computing map for Andover
2023-04-10 13:56:21,056 Computing map for Applewold
INFO:root:Time: 00d 02h 24m 13s, num timesteps 27000, FPS 3,
Rewards: Global step mean/med rew: 0.0000/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 02h 36m 41s, num timesteps 29250, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 02h 50m 41s, num timesteps 31500, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 03h 04m 27s, num timesteps 33750, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 03h 18m 32s, num timesteps 36000, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 03h 30m 02s, num timesteps 38250, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 03h 42m 02s, num timesteps 40500, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 03h 55m 15s, num timesteps 42750, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 04h 08m 24s, num timesteps 45000, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 04h 21m 02s, num timesteps 47250, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 04h 34m 00s, num timesteps 49500, FPS 3,
Rewards: Global step mean/med rew: 0.0137/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
INFO:root:Time: 00d 04h 47m 03s, num timesteps 51750, FPS 3,
Rewards: Global step mean/med rew: 0.0000/0.0000, Global eps mean/med/min/max eps rew: 1.366/1.366/1.366/1.366, Global eps mean/med eps len: 2999/2999,
Losses:
2023-04-10 16:32:35,027 Computing map for Bolton
2023-04-10 16:32:35,181 Computing map for Avonia
2023-04-10 16:32:35,238 Computing map for Crandon
2023-04-10 16:32:35,543 Computing map for Applewold
2023-04-10 16:32:35,559 Computing map for Andover
2023-04-10 16:32:35,588 Computing map for Caruthers
2023-04-10 16:32:35,690 Computing map for Brevort
2023-04-10 16:32:35,734 Computing map for Cooperstown
2023-04-10 16:32:35,744 Computing map for Adrian
INFO:root:
Final Exp Area:
1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000, 1.37000,
Final Exp Ratio:
0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157, 0.04157,
Final Close Lengths:
2999.00, 2999.00, 2999.00, 2999.00, 2999.00, 2999.00, 2999.00, 2999.00, 2999.00,
Final Bump Counts:
1465.0, 5619.0, 181.0, 1694.5, 3440.5, 2320.0, 1333.0, 1484.0, 4187.0,
Final Continuous Bump Counts:
1343.5, 2429.0, 61.0, 714.5, 1976.5, 478.0, 704.5, 513.5, 1760.5,
Does this mean the gitbision dataset is also broken ? Or is the result relative to num_episodes,which is 2 in my command?
I think there must be something broken...
Were you install all packages by the script build.sh
?
(btw, use --vis_type 1
can visualize the process, which may help to find the problems)
thanks for your answer,i've rebuild the environment and i am sure all packaes are installed. and i tried the --vis_tpye command,but found there is no images. the error shows that my GLSL version is too low ,but i use 2080ti and my driver is latest
(/root/autodl-tmp/NeuralCoMapping/venv) root@autodl-container-755011a052-b98b3f7b:~/autodl-tmp/NeuralCoMapping# python main.py --exp_name 'wwstest' --scenes_file scenes/wws.scenes --dump_location std --num_episodes 1 --load_global hrl3.global --vis_type 1
/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gym/core.py:27: UserWarning: WARN: Gym minimally supports python 3.6 as the python foundation not longer supports the version, please update your version to 3.7+
"Gym minimally supports python 3.6 as the python foundation not longer supports the version, please update your version to 3.7+"
INFO:root:Importing iGibson (gibson2 module)
INFO:root:Assets path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets
INFO:root:Gibson Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/g_dataset
INFO:root:iG Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/ig_dataset
INFO:root:3D-FRONT Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/threedfront_dataset
INFO:root:CubiCasa5K Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/cubicasa_dataset
INFO:root:Example path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/examples
INFO:root:Example config path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/examples/configs
Dumping at std/models/wwstest/
INFO:root:Namespace(ablation=0, action_loss_coef=1.0, baseline='none', centralized=1, clip_param=0.2, config_file='basic.yaml', critic_lr_coef=0.05, cuda=True, depth_noise_rate=0, depth_obs=0, du_scale=2, dump_location='std', entropy_coef=0.0001, env_frame_height=192, env_frame_width=192, eps=1e-05, eval=1, eval_eps_freq=0, exp_name='wwstest', frame_height=192, frame_width=192, g_input_content='gaussian_dist_map', gamma=0.99, global_downscaling=2, global_lr=0.0005, hfov=90.0, load_global='hrl3.global', load_global_critic='0', log_interval=250, map_size_cm=4800, max_batch_size=0, max_collisions_allowed=-1, max_episode_length=3000, max_grad_norm=0.5, noise_level=0.1, noisy_actions=0, noisy_odometry=0, num_episodes=1, num_global_steps=120, num_gnn_layer=3, num_local_steps=25, num_mini_batch=1, num_processes=1, num_robots=3, object_randomization_freq=0.0, obs_threshold=1, obstacle_boundary=20, ppo_epoch=4, ppo_sample_eps=1, print_images=True, reset_floor=0, reset_orientation=1, restore_eps=0, reward_bias=4, reward_scale=0.005, robot_scale=1, rotation_augmentation=1, save_interval=10000, save_periodic=500000, scene_per_process=1, scenes_file='scenes/wws.scenes', seed=1, short_goal_dist=1, snp_location='./snp', tau=0.95, texture_randomization_freq=0.0, train_global=0, unit_size_cm=10, use_clipped_value_loss=1, use_distance_field=1, use_gae=1, use_history=1, use_single_gpu=0, value_loss_coef=3.0, vis_type=1, vision_range=32, z_offset=0.4)
INFO:root:SNP allocated 3.9 MB
Loading basic.yaml
INFO:root:Device index is larger than number of devices, falling back to use 0
--- Logging error ---
Traceback (most recent call last):
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 994, in emit
msg = self.format(record)
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 840, in format
return fmt.format(record)
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 577, in format
record.message = record.getMessage()
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 338, in getMessage
msg = msg % self.args
TypeError: not all arguments converted during string formatting
Call stack:
File "main.py", line 608, in <module>
main()
File "main.py", line 112, in main
envs = construct_envs(args, snp.dump())
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 314, in construct_envs
envs = SyncNavEnv(make_env_fns, args.device)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, in __init__
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, in <listcomp>
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 295, in make_env_fn
env = Exploration_Env(args, config_file, scene_id, device_idx, rank, snp_data)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/exploration_env.py", line 157, in __init__
super().__init__(config_file, scene_ids=scene_ids, device_idx=device_idx)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 522, in __init__
render_to_tensor=render_to_tensor)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 318, in __init__
need_rgb=need_rgb)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 212, in __init__
rendering_settings=rendering_settings
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/simulator.py", line 86, in __init__
self.load()
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 222, in load
rendering_settings=self.rendering_settings
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/render/mesh_renderer/mesh_renderer_cpu.py", line 93, in __init__
"at http://svl.stanford.edu/igibson/docs/issues.html")
Message: 'If you have trouble using EGL, please visit our trouble shooting guide'
Arguments: ('at http://svl.stanford.edu/igibson/docs/issues.html',)
--- Logging error ---
Traceback (most recent call last):
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 994, in emit
msg = self.format(record)
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 840, in format
return fmt.format(record)
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 577, in format
record.message = record.getMessage()
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 338, in getMessage
msg = msg % self.args
TypeError: not all arguments converted during string formatting
Call stack:
File "main.py", line 608, in <module>
main()
File "main.py", line 112, in main
envs = construct_envs(args, snp.dump())
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 314, in construct_envs
envs = SyncNavEnv(make_env_fns, args.device)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, in __init__
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, in <listcomp>
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 295, in make_env_fn
env = Exploration_Env(args, config_file, scene_id, device_idx, rank, snp_data)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/exploration_env.py", line 157, in __init__
super().__init__(config_file, scene_ids=scene_ids, device_idx=device_idx)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 522, in __init__
render_to_tensor=render_to_tensor)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 318, in __init__
need_rgb=need_rgb)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 212, in __init__
rendering_settings=rendering_settings
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/simulator.py", line 86, in __init__
self.load()
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 222, in load
rendering_settings=self.rendering_settings
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/render/mesh_renderer/mesh_renderer_cpu.py", line 93, in __init__
"at http://svl.stanford.edu/igibson/docs/issues.html")
Message: 'If you have trouble using EGL, please visit our trouble shooting guide'
Arguments: ('at http://svl.stanford.edu/igibson/docs/issues.html',)
ERROR::SHADER::VERTEX::COMPILATION_FAILED
0:1(10): error: GLSL 4.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::FRAGMENT::COMPILATION_FAILED
0:1(10): error: GLSL 4.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::PROGRAM::LINKING_FAILED
error: linking with uncompiled/unspecialized shadererror: linking with uncompiled/unspecialized shader
ERROR::SHADER::VERTEX::COMPILATION_FAILED
0:1(10): error: GLSL 4.10 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::FRAGMENT::COMPILATION_FAILED
0:1(10): error: GLSL 4.10 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::PROGRAM::LINKING_FAILED
error: linking with uncompiled/unspecialized shadererror: linking with uncompiled/unspecialized shader
ERROR::SHADER::VERTEX::COMPILATION_FAILED
0:1(10): error: GLSL 4.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::FRAGMENT::COMPILATION_FAILED
0:2(10): error: GLSL 4.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::PROGRAM::LINKING_FAILED
error: linking with uncompiled/unspecialized shadererror: linking with uncompiled/unspecialized shader
INFO:root:IndoorScene model: Adrian
INFO:root:StaticIndoorScene scene: Adrian
INFO:root:Loading traversable graph
INFO:root:Loading traversable graph
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/g_dataset/Adrian/mesh_z_up.obj
INFO:root:Loading robot model file: turtlebot/turtlebot.urdf
INFO:root:Loading robot model file: turtlebot/turtlebot.urdf
INFO:root:Loading robot model file: turtlebot/turtlebot.urdf
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/kobuki_description/meshes/main_body.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/kobuki_description/meshes/wheel.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/stacks/hexagons/pole_bottom.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/stacks/hexagons/plate_bottom.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/stacks/hexagons/pole_middle.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/stacks/hexagons/plate_middle.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/stacks/hexagons/pole_top.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/stacks/hexagons/pole_kinect.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/stacks/hexagons/plate_top.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/turtlebot/turtlebot_description/meshes/sensors/kinect.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
INFO:root:Loading /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets/models/mjcf_primitives/cube.obj
2023-04-17 16:56:47,805 Computing map for Adrian
Loading global hrl3.global
WARNING:matplotlib.font_manager:findfont: Font family ['Helvetica'] not found. Falling back to DejaVu Sans.
INFO:root:Time: 00d 00h 00m 01s, num timesteps 0, FPS 0,
Rewards:
Losses:
INFO:root:Time: 00d 00h 02m 34s, num timesteps 250, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 05m 05s, num timesteps 500, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 07m 32s, num timesteps 750, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 10m 02s, num timesteps 1000, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 12m 30s, num timesteps 1250, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 15m 05s, num timesteps 1500, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 17m 29s, num timesteps 1750, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 19m 54s, num timesteps 2000, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 22m 20s, num timesteps 2250, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 24m 50s, num timesteps 2500, FPS 1,
Rewards:
Losses:
INFO:root:Time: 00d 00h 27m 17s, num timesteps 2750, FPS 1,
Rewards:
Losses:
2023-04-17 17:27:00,564 Computing map for Adrian
INFO:root:
Final Exp Area:
1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000, 1.71000,
Final Exp Ratio:
0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235, 0.04235,
Final Close Lengths:
2999.00,
Final Bump Counts:
78.0,
Final Continuous Bump Counts:
11.0,
INFO:root:Device index is larger than number of devices, falling back to use 0
is uncommon. Please try python -c "from gibson2.render.mesh_renderer.get_available_devices import get_available_devices; print(get_available_devices())"
to see if there is something wrong with GPU.As for the GLSL version unsupported error, I think it may be caused by an old OS system. (I am not sure, you might try to google it.) However, just modifying shaders file may help (but I am also not sure). You can find such files:
./venv/lib/python3.6/site-packages/gibson2/render/mesh_renderer/shaders/410/skybox_fs.glsl
./venv/lib/python3.6/site-packages/gibson2/render/mesh_renderer/shaders/410/skybox_vs.glsl
./env/gibson_api/ma_env_base.py
and replace all the #version 450
and #version 410 core
with a lower one.
1、i got:
Type "help", "copyright", "credits" or "license" for more information.
>>> from gibson2.render.mesh_renderer.get_available_devices import get_available_devices
INFO:root:Importing iGibson (gibson2 module)
INFO:root:Assets path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/assets
INFO:root:Gibson Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/g_dataset
INFO:root:iG Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/ig_dataset
INFO:root:3D-FRONT Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/threedfront_dataset
INFO:root:CubiCasa5K Dataset path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/data/cubicasa_dataset
INFO:root:Example path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/examples
INFO:root:Example config path: /root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson2/examples/configs
>>> print(get_available_devices())
[]
but:
(/root/autodl-tmp/NeuralCoMapping/venv) root@autodl-container-755011a052-b98b3f7b:~/autodl-tmp/NeuralCoMapping# python
Python 3.6.13 |Anaconda, Inc.| (default, Jun 4 2021, 14:25:59)
[GCC 7.5.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> print(torch.__version__)
1.2.0
>>> torch.cuda.is_available()
True
>>> torch.cuda.device_count()
1
>>> quit()
2、 i changed 450 and 410 to 140 but still got:
Message: 'If you have trouble using EGL, please visit our trouble shooting guide '
Arguments: ('at http://svl.stanford.edu/igibson/docs/issues.html',)
--- Logging error ---
Traceback (most recent call last):
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py", line 994, in emit
msg = self.format(record)
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py" , line 840, in format
return fmt.format(record)
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py" , line 577, in format
record.message = record.getMessage()
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/logging/__init__.py" , line 338, in getMessage
msg = msg % self.args
TypeError: not all arguments converted during string formatting
Call stack:
File "main.py", line 608, in <module>
main()
File "main.py", line 112, in main
envs = construct_envs(args, snp.dump())
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 314, in construct_envs
envs = SyncNavEnv(make_env_fns, args.device)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, i n __init__
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, i n <listcomp>
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 295, in make_env_fn
env = Exploration_Env(args, config_file, scene_id, device_idx, rank, snp_dat a)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/exploration_env.py", lin e 157, in __init__
super().__init__(config_file, scene_ids=scene_ids, device_idx=device_idx)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 52 2, in __init__
render_to_tensor=render_to_tensor)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 31 8, in __init__
need_rgb=need_rgb)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 21 2, in __init__
rendering_settings=rendering_settings
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson 2/simulator.py", line 86, in __init__
self.load()
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 22 2, in load
rendering_settings=self.rendering_settings
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson 2/render/mesh_renderer/mesh_renderer_cpu.py", line 93, in __init__
"at http://svl.stanford.edu/igibson/docs/issues.html")
Message: 'If you have trouble using EGL, please visit our trouble shooting guide '
Arguments: ('at http://svl.stanford.edu/igibson/docs/issues.html',)
ERROR::SHADER::VERTEX::COMPILATION_FAILED
0:1(10): error: GLSL 4.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::FRAGMENT::COMPILATION_FAILED
0:1(10): error: GLSL 4.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::PROGRAM::LINKING_FAILED
error: linking with uncompiled/unspecialized shadererror: linking with uncompile d/unspecialized shader
ERROR::SHADER::VERTEX::COMPILATION_FAILED
0:1(10): error: illegal text following version number
ERROR::SHADER::FRAGMENT::COMPILATION_FAILED
0:1(10): error: illegal text following version number
ERROR::SHADER::PROGRAM::LINKING_FAILED
error: linking with uncompiled/unspecialized shadererror: linking with uncompile d/unspecialized shader
ERROR::SHADER::VERTEX::COMPILATION_FAILED
0:1(10): error: GLSL 4.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.40, 1.00 ES, and 3.00 ES
ERROR::SHADER::FRAGMENT::COMPILATION_FAILED
0:39(1): error: shader output explicit location requires GL_ARB_explicit_attrib_ location extension or GLSL 3.30
0:40(1): error: shader output explicit location requires GL_ARB_explicit_attrib_ location extension or GLSL 3.30
0:41(1): error: shader output explicit location requires GL_ARB_explicit_attrib_ location extension or GLSL 3.30
0:42(1): error: shader output explicit location requires GL_ARB_explicit_attrib_ location extension or GLSL 3.30
0:43(1): error: shader output explicit location requires GL_ARB
ERROR::SHADER::PROGRAM::LINKING_FAILED
error: linking with uncompiled/unspecialized shadererror: linking with uncompile d/unspecialized shader
INFO:root:IndoorScene model: Adrian
INFO:root:StaticIndoorScene scene: Adrian
^CTraceback (most recent call last):
File "main.py", line 608, in <module>
main()
File "main.py", line 112, in main
envs = construct_envs(args, snp.dump())
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 314, in construct_envs
envs = SyncNavEnv(make_env_fns, args.device)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, in __init__
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 26, i n <listcomp>
self._envs = [ctor() for ctor in env_constructors]
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/__init__.py", line 295, in make_env_fn
env = Exploration_Env(args, config_file, scene_id, device_idx, rank, snp_data)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/exploration_env.py", line 157, in __init__
super().__init__(config_file, scene_ids=scene_ids, device_idx=device_idx)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 52 2, in __init__
render_to_tensor=render_to_tensor)
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 31 9, in __init__
self.load()
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 66 5, in load
super(MAGibsonEnv, self).load()
File "/root/autodl-tmp/NeuralCoMapping/env/gibson_api/ma_env_base.py", line 389, in load
scene, load_texture=self.config.get('load_texture', True) and self.config['scene_id'] not in ['CVPR2022', 'Cross'])
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson 2/simulator.py", line 156, in wrapped_load_func
res = load_func(*args, **kwargs)
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson 2/simulator.py", line 183, in import_scene
new_object_pb_ids = scene.load()
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson 2/scenes/gibson_indoor_scene.py", line 124, in load
self.load_scene_mesh()
File "/root/autodl-tmp/NeuralCoMapping/venv/lib/python3.6/site-packages/gibson 2/scenes/gibson_indoor_scene.py", line 98, in load_scene_mesh
texture_id = p.loadTexture(texture_filename)
i ve gooled ,but got nothing,Perhaps the OpenGL version has been restricted by the server supplier. (I am renting a GPU) so sad maybe i will try another Graphics card. thanks again
btw i update my os version from 18.04 to 20.04,the opengl version is 4.5 which support the code thanks again for your answer ,means alot for me
oh, sounds great :)
The result I obtained after i use this command
python main.py --exp_name 'eval_coscan_mp3dhq0f' --scenes_file scenes/mp3d-f.scenes --dump_location std --num_episodes 1 --load_global train/models/ma3_history/hrl3.global
is :according to my understanding, this refers to the result of a single episode. How can I get the proportion of the entire map that is finally explored as in the paper (up to 90%)?