I encountered an error when running the humanoid_amp.py example using the following command.
python train.py task=HumanoidAMP
(isaacgym) jiao@jiao-Predator-PHN16-71:~/isaacgym/IsaacGymEnvs/isaacgymenvs$ python train.py task=HumanoidAMP num_envs=1024
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/hydra.py:119: UserWarning: Future Hydra versions will no longer change working directory at job runtime by default.
See https://hydra.cc/docs/1.2/upgrades/1.1_to_1.2/changes_to_job_working_dir/ for more information.
ret = run_job(
Importing module 'gym_38' (/home/jiao/isaacgym/python/isaacgym/_bindings/linux-x86_64/gym_38.so)
Setting GYM_USD_PLUG_INFO_PATH to /home/jiao/isaacgym/python/isaacgym/_bindings/linux-x86_64/usd/plugInfo.json
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/torch/utils/cpp_extension.py:25: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
from pkg_resources import packaging # type: ignore[attr-defined]
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/pkg_resources/__init__.py:2871: DeprecationWarning: Deprecated call to `pkg_resources.declare_namespace('mpl_toolkits')`.
Implementing implicit namespace packages (as specified in PEP 420) is preferred to `pkg_resources.declare_namespace`. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages
declare_namespace(pkg)
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/pkg_resources/__init__.py:2871: DeprecationWarning: Deprecated call to `pkg_resources.declare_namespace('google')`.
Implementing implicit namespace packages (as specified in PEP 420) is preferred to `pkg_resources.declare_namespace`. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages
declare_namespace(pkg)
PyTorch version 1.13.1+cu117
Device count 1
/home/jiao/isaacgym/python/isaacgym/_bindings/src/gymtorch
Using /home/jiao/.cache/torch_extensions/py38_cu117 as PyTorch extensions root...
Emitting ninja build file /home/jiao/.cache/torch_extensions/py38_cu117/gymtorch/build.ninja...
Building extension module gymtorch...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
ninja: no work to do.
Loading extension module gymtorch...
2024-03-28 20:46:46,636 - INFO - logger - logger initialized
<unknown>:3: DeprecationWarning: invalid escape sequence \*
Error: FBX library failed to load - importing FBX data will not succeed. Message: No module named 'fbx'
FBX tools must be installed from https://help.autodesk.com/view/FBX/2020/ENU/?guid=FBX_Developer_Help_scripting_with_python_fbx_installing_python_fbx_html
/home/jiao/isaacgym/python/isaacgym/torch_utils.py:135: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
def get_axis_params(value, axis_idx, x_value=0., dtype=np.float, n_dims=3):
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/networkx/classes/graph.py:23: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3, and in 3.10 it will stop working
from collections import Mapping
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/networkx/classes/reportviews.py:95: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3, and in 3.10 it will stop working
from collections import Mapping, Set, Iterable
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/networkx/readwrite/graphml.py:346: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
(np.int, "int"), (np.int8, "int"),
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/torch/utils/tensorboard/__init__.py:4: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
if not hasattr(tensorboard, "__version__") or LooseVersion(
task:
name: HumanoidAMP
physics_engine: physx
env:
numEnvs: 1024
envSpacing: 5
episodeLength: 300
cameraFollow: True
enableDebugVis: False
pdControl: True
powerScale: 1.0
controlFrequencyInv: 2
stateInit: Random
hybridInitProb: 0.5
numAMPObsSteps: 2
localRootObs: False
contactBodies: ['right_foot', 'left_foot']
terminationHeight: 0.5
enableEarlyTermination: True
motion_file: amp_humanoid_run.npy
asset:
assetFileName: mjcf/amp_humanoid.xml
plane:
staticFriction: 1.0
dynamicFriction: 1.0
restitution: 0.0
sim:
dt: 0.0166
substeps: 2
up_axis: z
use_gpu_pipeline: True
gravity: [0.0, 0.0, -9.81]
physx:
num_threads: 4
solver_type: 1
use_gpu: True
num_position_iterations: 4
num_velocity_iterations: 0
contact_offset: 0.02
rest_offset: 0.0
bounce_threshold_velocity: 0.2
max_depenetration_velocity: 10.0
default_buffer_size_multiplier: 5.0
max_gpu_contact_pairs: 8388608
num_subscenes: 4
contact_collection: 2
task:
randomize: False
randomization_params:
frequency: 600
observations:
range: [0, 0.002]
operation: additive
distribution: gaussian
actions:
range: [0.0, 0.02]
operation: additive
distribution: gaussian
sim_params:
gravity:
range: [0, 0.4]
operation: additive
distribution: gaussian
schedule: linear
schedule_steps: 3000
actor_params:
humanoid:
color: True
rigid_body_properties:
mass:
range: [0.5, 1.5]
operation: scaling
distribution: uniform
setup_only: True
schedule: linear
schedule_steps: 3000
rigid_shape_properties:
friction:
num_buckets: 500
range: [0.7, 1.3]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
restitution:
range: [0.0, 0.7]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
dof_properties:
damping:
range: [0.5, 1.5]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
stiffness:
range: [0.5, 1.5]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
lower:
range: [0, 0.01]
operation: additive
distribution: gaussian
schedule: linear
schedule_steps: 3000
upper:
range: [0, 0.01]
operation: additive
distribution: gaussian
schedule: linear
schedule_steps: 3000
train:
params:
seed: 42
algo:
name: amp_continuous
model:
name: continuous_amp
network:
name: amp
separate: True
space:
continuous:
mu_activation: None
sigma_activation: None
mu_init:
name: default
sigma_init:
name: const_initializer
val: -2.9
fixed_sigma: True
learn_sigma: False
mlp:
units: [1024, 512]
activation: relu
d2rl: False
initializer:
name: default
regularizer:
name: None
disc:
units: [1024, 512]
activation: relu
initializer:
name: default
load_checkpoint: False
load_path:
config:
name: HumanoidAMP
full_experiment_name: HumanoidAMP
env_name: rlgpu
ppo: True
multi_gpu: False
mixed_precision: False
normalize_input: True
normalize_value: True
value_bootstrap: True
num_actors: 1024
reward_shaper:
scale_value: 1
normalize_advantage: True
gamma: 0.99
tau: 0.95
learning_rate: 5e-05
lr_schedule: constant
kl_threshold: 0.008
score_to_win: 20000
max_epochs: 5000
save_best_after: 100
save_frequency: 50
print_stats: True
grad_norm: 1.0
entropy_coef: 0.0
truncate_grads: False
e_clip: 0.2
horizon_length: 16
minibatch_size: 32768
mini_epochs: 6
critic_coef: 5
clip_value: False
seq_len: 4
bounds_loss_coef: 10
amp_obs_demo_buffer_size: 200000
amp_replay_buffer_size: 1000000
amp_replay_keep_prob: 0.01
amp_batch_size: 512
amp_minibatch_size: 4096
disc_coef: 5
disc_logit_reg: 0.05
disc_grad_penalty: 5
disc_reward_scale: 2
disc_weight_decay: 0.0001
normalize_amp_input: True
task_reward_w: 0.0
disc_reward_w: 1.0
pbt:
enabled: False
task_name: HumanoidAMP
experiment:
num_envs: 1024
seed: 42
torch_deterministic: False
max_iterations:
physics_engine: physx
pipeline: gpu
sim_device: cuda:0
rl_device: cuda:0
graphics_device_id: 0
num_threads: 4
solver_type: 1
num_subscenes: 4
test: False
checkpoint:
sigma:
multi_gpu: False
wandb_activate: False
wandb_group:
wandb_name: HumanoidAMP
wandb_entity:
wandb_project: isaacgymenvs
wandb_tags: []
wandb_logcode_dir:
capture_video: False
capture_video_freq: 1464
capture_video_len: 100
force_render: True
headless: False
Setting seed: 42
Using rl_device: cuda:0
Using sim_device: cuda:0
{'name': 'HumanoidAMP', 'full_experiment_name': None, 'env_name': 'rlgpu', 'ppo': True, 'multi_gpu': False, 'mixed_precision': False, 'normalize_input': True, 'normalize_value': True, 'value_bootstrap': True, 'num_actors': 1024, 'reward_shaper': {'scale_value': 1}, 'normalize_advantage': True, 'gamma': 0.99, 'tau': 0.95, 'learning_rate': 5e-05, 'lr_schedule': 'constant', 'kl_threshold': 0.008, 'score_to_win': 20000, 'max_epochs': 5000, 'save_best_after': 100, 'save_frequency': 50, 'print_stats': True, 'grad_norm': 1.0, 'entropy_coef': 0.0, 'truncate_grads': False, 'e_clip': 0.2, 'horizon_length': 16, 'minibatch_size': 32768, 'mini_epochs': 6, 'critic_coef': 5, 'clip_value': False, 'seq_len': 4, 'bounds_loss_coef': 10, 'amp_obs_demo_buffer_size': 200000, 'amp_replay_buffer_size': 1000000, 'amp_replay_keep_prob': 0.01, 'amp_batch_size': 512, 'amp_minibatch_size': 4096, 'disc_coef': 5, 'disc_logit_reg': 0.05, 'disc_grad_penalty': 5, 'disc_reward_scale': 2, 'disc_weight_decay': 0.0001, 'normalize_amp_input': True, 'task_reward_w': 0.0, 'disc_reward_w': 1.0, 'device': 'cuda:0', 'population_based_training': False, 'pbt_idx': None}
self.seed = 42
Started to train
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/gym/spaces/box.py:84: UserWarning: WARN: Box bound precision lowered by casting to float32
logger.warn(f"Box bound precision lowered by casting to {self.dtype}")
[Warning] [carb.gym.plugin] useGpu is set, forcing single scene (0 subscenes)
Not connected to PVD
+++ Using GPU PhysX
Physics Engine: PhysX
Physics Device: cuda:0
GPU Pipeline: enabled
/home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/tasks/amp/humanoid_amp_base.py:186: DeprecationWarning: an integer is required (got type isaacgym._bindings.linux-x86_64.gym_38.DofDriveMode). Implicit conversion to integers using __int__ is deprecated, and may be removed in a future version of Python.
asset_options.default_dof_drive_mode = gymapi.DOF_MODE_NONE
Loading 1/1 motion files: /home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/tasks/../../assets/amp/motions/amp_humanoid_run.npy
Loaded 1 motions with a total length of 1.350s.
Box(-1.0, 1.0, (28,), float32) Box(-inf, inf, (105,), float32)
WARNING: seq_len is deprecated, use seq_length instead
seq_length: 4
current training device: cuda:0
Error executing job with overrides: ['task=HumanoidAMP', 'num_envs=1024']
Traceback (most recent call last):
File "train.py", line 210, in launch_rlg_hydra
runner.run({
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/rl_games/torch_runner.py", line 133, in run
self.run_train(args)
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/rl_games/torch_runner.py", line 113, in run_train
agent = self.algo_factory.create(self.algo_name, base_name='run', params=self.params)
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/rl_games/common/object_factory.py", line 15, in create
return builder(**kwargs)
File "train.py", line 188, in <lambda>
runner.algo_factory.register_builder('amp_continuous', lambda **kwargs : amp_continuous.AMPAgent(**kwargs))
File "/home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/learning/amp_continuous.py", line 53, in __init__
super().__init__(base_name, params)
File "/home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/learning/common_agent.py", line 58, in __init__
a2c_common.A2CBase.__init__(self, base_name, params)
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/rl_games/common/a2c_common.py", line 249, in __init__
assert(self.batch_size % self.minibatch_size == 0)
AssertionError
Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
Then I used the following command to resolve the issue.
export HYDRA_FULL_ERROR=1
Then I continued running the example program
AttributeError: ‘AMPAgent’ object has no attribute ‘seq_len’
(isaacgym) jiao@jiao-Predator-PHN16-71:~/isaacgym/IsaacGymEnvs/isaacgymenvs$ python train.py task=HumanoidAMP
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/hydra.py:119: UserWarning: Future Hydra versions will no longer change working directory at job runtime by default.
See https://hydra.cc/docs/1.2/upgrades/1.1_to_1.2/changes_to_job_working_dir/ for more information.
ret = run_job(
Importing module 'gym_38' (/home/jiao/isaacgym/python/isaacgym/_bindings/linux-x86_64/gym_38.so)
Setting GYM_USD_PLUG_INFO_PATH to /home/jiao/isaacgym/python/isaacgym/_bindings/linux-x86_64/usd/plugInfo.json
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/torch/utils/cpp_extension.py:25: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
from pkg_resources import packaging # type: ignore[attr-defined]
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/pkg_resources/__init__.py:2871: DeprecationWarning: Deprecated call to `pkg_resources.declare_namespace('mpl_toolkits')`.
Implementing implicit namespace packages (as specified in PEP 420) is preferred to `pkg_resources.declare_namespace`. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages
declare_namespace(pkg)
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/pkg_resources/__init__.py:2871: DeprecationWarning: Deprecated call to `pkg_resources.declare_namespace('google')`.
Implementing implicit namespace packages (as specified in PEP 420) is preferred to `pkg_resources.declare_namespace`. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages
declare_namespace(pkg)
PyTorch version 1.13.1+cu117
Device count 1
/home/jiao/isaacgym/python/isaacgym/_bindings/src/gymtorch
Using /home/jiao/.cache/torch_extensions/py38_cu117 as PyTorch extensions root...
Emitting ninja build file /home/jiao/.cache/torch_extensions/py38_cu117/gymtorch/build.ninja...
Building extension module gymtorch...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
ninja: no work to do.
Loading extension module gymtorch...
2024-03-28 20:49:28,190 - INFO - logger - logger initialized
<unknown>:3: DeprecationWarning: invalid escape sequence \*
Error: FBX library failed to load - importing FBX data will not succeed. Message: No module named 'fbx'
FBX tools must be installed from https://help.autodesk.com/view/FBX/2020/ENU/?guid=FBX_Developer_Help_scripting_with_python_fbx_installing_python_fbx_html
/home/jiao/isaacgym/python/isaacgym/torch_utils.py:135: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
def get_axis_params(value, axis_idx, x_value=0., dtype=np.float, n_dims=3):
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/networkx/classes/graph.py:23: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3, and in 3.10 it will stop working
from collections import Mapping
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/networkx/classes/reportviews.py:95: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3, and in 3.10 it will stop working
from collections import Mapping, Set, Iterable
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/networkx/readwrite/graphml.py:346: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
(np.int, "int"), (np.int8, "int"),
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/torch/utils/tensorboard/__init__.py:4: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
if not hasattr(tensorboard, "__version__") or LooseVersion(
task:
name: HumanoidAMP
physics_engine: physx
env:
numEnvs: 4096
envSpacing: 5
episodeLength: 300
cameraFollow: True
enableDebugVis: False
pdControl: True
powerScale: 1.0
controlFrequencyInv: 2
stateInit: Random
hybridInitProb: 0.5
numAMPObsSteps: 2
localRootObs: False
contactBodies: ['right_foot', 'left_foot']
terminationHeight: 0.5
enableEarlyTermination: True
motion_file: amp_humanoid_run.npy
asset:
assetFileName: mjcf/amp_humanoid.xml
plane:
staticFriction: 1.0
dynamicFriction: 1.0
restitution: 0.0
sim:
dt: 0.0166
substeps: 2
up_axis: z
use_gpu_pipeline: True
gravity: [0.0, 0.0, -9.81]
physx:
num_threads: 4
solver_type: 1
use_gpu: True
num_position_iterations: 4
num_velocity_iterations: 0
contact_offset: 0.02
rest_offset: 0.0
bounce_threshold_velocity: 0.2
max_depenetration_velocity: 10.0
default_buffer_size_multiplier: 5.0
max_gpu_contact_pairs: 8388608
num_subscenes: 4
contact_collection: 2
task:
randomize: False
randomization_params:
frequency: 600
observations:
range: [0, 0.002]
operation: additive
distribution: gaussian
actions:
range: [0.0, 0.02]
operation: additive
distribution: gaussian
sim_params:
gravity:
range: [0, 0.4]
operation: additive
distribution: gaussian
schedule: linear
schedule_steps: 3000
actor_params:
humanoid:
color: True
rigid_body_properties:
mass:
range: [0.5, 1.5]
operation: scaling
distribution: uniform
setup_only: True
schedule: linear
schedule_steps: 3000
rigid_shape_properties:
friction:
num_buckets: 500
range: [0.7, 1.3]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
restitution:
range: [0.0, 0.7]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
dof_properties:
damping:
range: [0.5, 1.5]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
stiffness:
range: [0.5, 1.5]
operation: scaling
distribution: uniform
schedule: linear
schedule_steps: 3000
lower:
range: [0, 0.01]
operation: additive
distribution: gaussian
schedule: linear
schedule_steps: 3000
upper:
range: [0, 0.01]
operation: additive
distribution: gaussian
schedule: linear
schedule_steps: 3000
train:
params:
seed: 42
algo:
name: amp_continuous
model:
name: continuous_amp
network:
name: amp
separate: True
space:
continuous:
mu_activation: None
sigma_activation: None
mu_init:
name: default
sigma_init:
name: const_initializer
val: -2.9
fixed_sigma: True
learn_sigma: False
mlp:
units: [1024, 512]
activation: relu
d2rl: False
initializer:
name: default
regularizer:
name: None
disc:
units: [1024, 512]
activation: relu
initializer:
name: default
load_checkpoint: False
load_path:
config:
name: HumanoidAMP
full_experiment_name: HumanoidAMP
env_name: rlgpu
ppo: True
multi_gpu: False
mixed_precision: False
normalize_input: True
normalize_value: True
value_bootstrap: True
num_actors: 4096
reward_shaper:
scale_value: 1
normalize_advantage: True
gamma: 0.99
tau: 0.95
learning_rate: 5e-05
lr_schedule: constant
kl_threshold: 0.008
score_to_win: 20000
max_epochs: 5000
save_best_after: 100
save_frequency: 50
print_stats: True
grad_norm: 1.0
entropy_coef: 0.0
truncate_grads: False
e_clip: 0.2
horizon_length: 16
minibatch_size: 32768
mini_epochs: 6
critic_coef: 5
clip_value: False
seq_len: 4
bounds_loss_coef: 10
amp_obs_demo_buffer_size: 200000
amp_replay_buffer_size: 1000000
amp_replay_keep_prob: 0.01
amp_batch_size: 512
amp_minibatch_size: 4096
disc_coef: 5
disc_logit_reg: 0.05
disc_grad_penalty: 5
disc_reward_scale: 2
disc_weight_decay: 0.0001
normalize_amp_input: True
task_reward_w: 0.0
disc_reward_w: 1.0
pbt:
enabled: False
task_name: HumanoidAMP
experiment:
num_envs:
seed: 42
torch_deterministic: False
max_iterations:
physics_engine: physx
pipeline: gpu
sim_device: cuda:0
rl_device: cuda:0
graphics_device_id: 0
num_threads: 4
solver_type: 1
num_subscenes: 4
test: False
checkpoint:
sigma:
multi_gpu: False
wandb_activate: False
wandb_group:
wandb_name: HumanoidAMP
wandb_entity:
wandb_project: isaacgymenvs
wandb_tags: []
wandb_logcode_dir:
capture_video: False
capture_video_freq: 1464
capture_video_len: 100
force_render: True
headless: False
Setting seed: 42
Using rl_device: cuda:0
Using sim_device: cuda:0
{'name': 'HumanoidAMP', 'full_experiment_name': None, 'env_name': 'rlgpu', 'ppo': True, 'multi_gpu': False, 'mixed_precision': False, 'normalize_input': True, 'normalize_value': True, 'value_bootstrap': True, 'num_actors': 4096, 'reward_shaper': {'scale_value': 1}, 'normalize_advantage': True, 'gamma': 0.99, 'tau': 0.95, 'learning_rate': 5e-05, 'lr_schedule': 'constant', 'kl_threshold': 0.008, 'score_to_win': 20000, 'max_epochs': 5000, 'save_best_after': 100, 'save_frequency': 50, 'print_stats': True, 'grad_norm': 1.0, 'entropy_coef': 0.0, 'truncate_grads': False, 'e_clip': 0.2, 'horizon_length': 16, 'minibatch_size': 32768, 'mini_epochs': 6, 'critic_coef': 5, 'clip_value': False, 'seq_len': 4, 'bounds_loss_coef': 10, 'amp_obs_demo_buffer_size': 200000, 'amp_replay_buffer_size': 1000000, 'amp_replay_keep_prob': 0.01, 'amp_batch_size': 512, 'amp_minibatch_size': 4096, 'disc_coef': 5, 'disc_logit_reg': 0.05, 'disc_grad_penalty': 5, 'disc_reward_scale': 2, 'disc_weight_decay': 0.0001, 'normalize_amp_input': True, 'task_reward_w': 0.0, 'disc_reward_w': 1.0, 'device': 'cuda:0', 'population_based_training': False, 'pbt_idx': None}
self.seed = 42
Started to train
/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/gym/spaces/box.py:84: UserWarning: WARN: Box bound precision lowered by casting to float32
logger.warn(f"Box bound precision lowered by casting to {self.dtype}")
[Warning] [carb.gym.plugin] useGpu is set, forcing single scene (0 subscenes)
Not connected to PVD
+++ Using GPU PhysX
Physics Engine: PhysX
Physics Device: cuda:0
GPU Pipeline: enabled
/home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/tasks/amp/humanoid_amp_base.py:186: DeprecationWarning: an integer is required (got type isaacgym._bindings.linux-x86_64.gym_38.DofDriveMode). Implicit conversion to integers using __int__ is deprecated, and may be removed in a future version of Python.
asset_options.default_dof_drive_mode = gymapi.DOF_MODE_NONE
Loading 1/1 motion files: /home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/tasks/../../assets/amp/motions/amp_humanoid_run.npy
Loaded 1 motions with a total length of 1.350s.
Box(-1.0, 1.0, (28,), float32) Box(-inf, inf, (105,), float32)
WARNING: seq_len is deprecated, use seq_length instead
seq_length: 4
current training device: cuda:0
build mlp: 105
build mlp: 105
build mlp: 210
sigma
actor_mlp.0.weight
actor_mlp.0.bias
actor_mlp.2.weight
actor_mlp.2.bias
critic_mlp.0.weight
critic_mlp.0.bias
critic_mlp.2.weight
critic_mlp.2.bias
value.weight
value.bias
mu.weight
mu.bias
_disc_mlp.0.weight
_disc_mlp.0.bias
_disc_mlp.2.weight
_disc_mlp.2.bias
_disc_logits.weight
_disc_logits.bias
RunningMeanStd: (1,)
RunningMeanStd: (105,)
Error executing job with overrides: ['task=HumanoidAMP']
Traceback (most recent call last):
File "train.py", line 219, in <module>
launch_rlg_hydra()
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/main.py", line 94, in decorated_main
_run_hydra(
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
_run_app(
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/utils.py", line 457, in _run_app
run_and_report(
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
raise ex
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
return func()
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/utils.py", line 458, in <lambda>
lambda: hydra.run(
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 132, in run
_ = ret.return_value
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/core/utils.py", line 260, in return_value
raise self._return_value
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/hydra/core/utils.py", line 186, in run_job
ret.return_value = task_function(task_cfg)
File "train.py", line 210, in launch_rlg_hydra
runner.run({
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/rl_games/torch_runner.py", line 133, in run
self.run_train(args)
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/rl_games/torch_runner.py", line 113, in run_train
agent = self.algo_factory.create(self.algo_name, base_name='run', params=self.params)
File "/home/jiao/anaconda3/envs/isaacgym/lib/python3.8/site-packages/rl_games/common/object_factory.py", line 15, in create
return builder(**kwargs)
File "train.py", line 188, in <lambda>
runner.algo_factory.register_builder('amp_continuous', lambda **kwargs : amp_continuous.AMPAgent(**kwargs))
File "/home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/learning/amp_continuous.py", line 53, in __init__
super().__init__(base_name, params)
File "/home/jiao/isaacgym/IsaacGymEnvs/isaacgymenvs/learning/common_agent.py", line 98, in __init__
self.dataset = amp_datasets.AMPDataset(self.batch_size, self.minibatch_size, self.is_discrete, self.is_rnn, self.ppo_device, self.seq_len)
AttributeError: 'AMPAgent' object has no attribute 'seq_len'
But when I ran the Ant example program, it worked fine
python train.py task=Ant
Please help！！！！！