[INFO 10:35:49] my_main Beginning training for 210000 timesteps
/data/workspace/QPLEX/pymarl-master/src/components/episode_buffer.py:104: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requiresgrad(True), rather than torch.tensor(sourceTensor).
v = th.tensor(v, dtype=dtype, device=self.device)
[INFO 10:35:49] my_main t_env: 1 / 210000
[INFO 10:35:49] my_main Estimated time left: 28 seconds. Time passed: 0 seconds
[INFO 10:35:49] my_main Saving models to results/matrix/models/qplex2021-01-20_10-35-47/1
/home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/torch/nn/functional.py:1625: UserWarning: nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.
warnings.warn("nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.")
action_pair_0_0 [0. 0.10165232] [ 0.06337885 -0.05444577] -0.44034332 8.0
action_pair_0_1 [0. 0.] [0.06337885 0.04720655] -0.4400812 -12.0
action_pair_0_2 [0. 0.16995847] [ 0.06337885 -0.12275192] -0.44051677 -12.0
action_pair_1_0 [0.07982428 0.10165232] [-0.01644544 -0.05444577] -0.4425343 -12.0
action_pair_1_1 [0.07982428 0. ] [-0.01644544 0.04720655] -0.44228593 0.0
action_pair_1_2 [0.07982428 0.16995847] [-0.01644544 -0.12275192] -0.44273216 0.0
action_pair_2_0 [0.19648269 0.10165232] [-0.13310385 -0.05444577] -0.44572845 -12.0
action_pair_2_1 [0.19648269 0. ] [-0.13310385 0.04720655] -0.44552478 0.0
action_pair_2_2 [0.19648269 0.16995847] [-0.13310385 -0.12275192] -0.44595572 0.0
[INFO 10:35:56] my_main Updated target network
/home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/jsonpickle-1.4.2-py3.6.egg/jsonpickle/ext/numpy.py:291: UserWarning: ndarray is defined by reference to an object we do not know how to serialize. A deep copy is serialized instead, breaking memory aliasing.
warnings.warn(msg)
[INFO 10:36:03] my_main Updated target network
[INFO 10:36:10] my_main Updated target network
[INFO 10:36:18] my_main Updated target network
[DEBUG 10:36:25] pymarl Stopping Heartbeat
[ERROR 10:36:25] pymarl Failed after 0:00:38!
Traceback (most recent call last):
File "/data/workspace/CollaQ/third_party/sacred/sacred/experiment.py", line 319, in run_commandline
options=args,
File "/data/workspace/CollaQ/third_party/sacred/sacred/experiment.py", line 276, in run
run()
File "/data/workspace/CollaQ/third_party/sacred/sacred/run.py", line 238, in call
self.result = self.main_function(args)
File "/data/workspace/CollaQ/third_party/sacred/sacred/config/captured_function.py", line 42, in captured_function
result = wrapped(args, **kwargs)
File "src/main.py", line 35, in my_main
run(_run, config, _log)
File "/data/workspace/QPLEX/pymarl-master/src/run.py", line 51, in run
run_sequential(args=args, logger=logger)
File "/data/workspace/QPLEX/pymarl-master/src/run.py", line 345, in run_sequential
logger.print_recent_stats()
File "/data/workspace/QPLEX/pymarl-master/src/utils/logging.py", line 48, in print_recent_stats
item = "{:.4f}".format(np.mean([x[1] for x in self.stats[k][-window:]]))
File "<__array_function internals>", line 6, in mean
File "/home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 3373, in mean
out=out, **kwargs)
File "/home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/numpy/core/_methods.py", line 170, in _mean
ret = ret.dtype.type(ret / rcount)
AttributeError: 'torch.dtype' object has no attribute 'type'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "src/main.py", line 98, in
ex.run_commandline(params)
File "/data/workspace/CollaQ/third_party/sacred/sacred/experiment.py", line 348, in run_commandline
print_filtered_stacktrace()
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 494, in print_filtered_stacktrace
print(format_filtered_stacktrace(filter_traceback), file=sys.stderr)
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 529, in format_filtered_stacktrace
return "".join(filtered_traceback_format(tb_exception))
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 571, in filtered_traceback_format
if not _is_sacred_frame(current_tb.tb_frame):
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 490, in _is_sacred_frame
return frame.f_globals["name"].split(".")[0] == "sacred"
KeyError: 'name'
Hi, this bug is caused by the incorrect version of pytorch and numpy. Make sure you install the correct version of this two packages. You can find the correct version from QPLEX/pymarl-master/requirements.txt.
when I run the following command under the directory(/data/workspace/QPLEX/pymarl-master)
occured the errors as follows: [DEBUG 10:35:47] git.cmd Popen(['git', 'version'], cwd=/data/workspace/QPLEX/pymarl-master, universal_newlines=False, shell=None, istream=None) [DEBUG 10:35:47] git.cmd Popen(['git', 'version'], cwd=/data/workspace/QPLEX/pymarl-master, universal_newlines=False, shell=None, istream=None) [DEBUG 10:35:47] git.cmd Popen(['git', 'diff', '--cached', '--abbrev=40', '--full-index', '--raw'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=None) [DEBUG 10:35:47] git.cmd Popen(['git', 'diff', '--abbrev=40', '--full-index', '--raw'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=None) [DEBUG 10:35:47] git.cmd Popen(['git', 'cat-file', '--batch-check'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=)
[DEBUG 10:35:47] git.cmd Popen(['git', 'diff', '--cached', '--abbrev=40', '--full-index', '--raw'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=None)
[DEBUG 10:35:47] git.cmd Popen(['git', 'diff', '--abbrev=40', '--full-index', '--raw'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=None)
[DEBUG 10:35:47] git.cmd Popen(['git', 'cat-file', '--batch-check'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=)
[DEBUG 10:35:47] git.cmd Popen(['git', 'diff', '--cached', '--abbrev=40', '--full-index', '--raw'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=None)
[DEBUG 10:35:47] git.cmd Popen(['git', 'diff', '--abbrev=40', '--full-index', '--raw'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=None)
[DEBUG 10:35:47] git.cmd Popen(['git', 'cat-file', '--batch-check'], cwd=/data/workspace/QPLEX, universal_newlines=False, shell=None, istream=)
src/main.py:79: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
config_dict = yaml.load(f)
src/main.py:49: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
config_dict = yaml.load(f)
[INFO 10:35:47] root Saving to FileStorageObserver in results/sacred.
{'--beat-interval': None,
'--capture': None,
'--comment': None,
'--debug': False,
'--enforce_clean': False,
'--file_storage': None,
'--force': False,
'--help': False,
'--loglevel': None,
'--mongo_db': None,
'--name': None,
'--pdb': False,
'--print-config': False,
'--priority': None,
'--queue': False,
'--s3': None,
'--sql': None,
'--tiny_db': None,
'--unobserved': False,
'COMMAND': None,
'UPDATE': ['local_results_path=results/matrix/', 'save_model=True', 'use_tensorboard=True', 'save_model_interval=200000', 't_max=210000', 'epsilon_finish=1.0'],
'help': False,
'with': True}
[DEBUG 10:35:47] pymarl Using capture mode "fd"
[INFO 10:35:47] pymarl Running command 'my_main'
[INFO 10:35:47] pymarl Started run with ID "1"
[DEBUG 10:35:47] pymarl Starting Heartbeat
[DEBUG 10:35:47] my_main Started
[INFO 10:35:47] my_main Experiment Parameters:
[INFO 10:35:47] my_main
{ 'action_selector': 'epsilon_greedy', 'adv_hypernet_embed': 64, 'adv_hypernet_layers': 3, 'agent': 'rnn', 'agent_output_type': 'q', 'batch_size': 32, 'batch_size_run': 1, 'buffer_cpu_only': True, 'buffer_size': 5000, 'burn_in_period': 32, 'checkpoint_path': '', 'critic_lr': 0.0005, 'demo_interval': 1000, 'double_q': True, 'env': 'matrix_game_2', 'env_args': { 'episode_limit': 1, 'is_print': False, 'n_actions': 3, 'n_agents': 2, 'obs_last_action': False, 'print_rew': False, 'print_steps': 1000, 'seed': 477239634, 'state_last_action': False}, 'epsilon_anneal_time': 50000, 'epsilon_finish': 1.0, 'epsilon_start': 1.0, 'evaluate': False, 'gamma': 0.99, 'grad_norm_clip': 10, 'hypernet_embed': 64, 'is_adv_attention': True, 'is_batch_rl': False, 'is_from_start': True, 'is_minus_one': True, 'is_save_buffer': False, 'is_stop_gradient': True, 'label': 'default_label', 'learner': 'dmaq_qatten_learner', 'learner_log_interval': 1000, 'load_buffer_id': 0, 'load_step': 0, 'local_results_path': 'results/matrix/', 'log_interval': 1000, 'lr': 0.0005, 'mac': 'basic_mac', 'mixer': 'dmaq', 'mixing_embed_dim': 32, 'name': 'qplex', 'num_circle': 1, 'num_kernel': 10, 'obs_agent_id': True, 'obs_last_action': True, 'optim_alpha': 0.99, 'optim_eps': 1e-05, 'repeat_id': 1, 'rnn_hidden_dim': 64, 'runner': 'episode', 'runner_log_interval': 1000, 'save_buffer_id': 0, 'save_buffer_interval': 1000, 'save_buffer_size': 10000, 'save_model': True, 'save_model_interval': 200000, 'save_replay': False, 'seed': 477239634, 't_max': 210000, 'target_update_interval': 200, 'test_greedy': True, 'test_interval': 1000, 'test_nepisode': 30, 'use_cuda': True, 'use_tensorboard': True, 'weighted_head': True}
[INFO 10:35:49] my_main Beginning training for 210000 timesteps /data/workspace/QPLEX/pymarl-master/src/components/episode_buffer.py:104: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requiresgrad(True), rather than torch.tensor(sourceTensor). v = th.tensor(v, dtype=dtype, device=self.device) [INFO 10:35:49] my_main t_env: 1 / 210000 [INFO 10:35:49] my_main Estimated time left: 28 seconds. Time passed: 0 seconds [INFO 10:35:49] my_main Saving models to results/matrix/models/qplex2021-01-20_10-35-47/1 /home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/torch/nn/functional.py:1625: UserWarning: nn.functional.sigmoid is deprecated. Use torch.sigmoid instead. warnings.warn("nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.") action_pair_0_0 [0. 0.10165232] [ 0.06337885 -0.05444577] -0.44034332 8.0 action_pair_0_1 [0. 0.] [0.06337885 0.04720655] -0.4400812 -12.0 action_pair_0_2 [0. 0.16995847] [ 0.06337885 -0.12275192] -0.44051677 -12.0 action_pair_1_0 [0.07982428 0.10165232] [-0.01644544 -0.05444577] -0.4425343 -12.0 action_pair_1_1 [0.07982428 0. ] [-0.01644544 0.04720655] -0.44228593 0.0 action_pair_1_2 [0.07982428 0.16995847] [-0.01644544 -0.12275192] -0.44273216 0.0 action_pair_2_0 [0.19648269 0.10165232] [-0.13310385 -0.05444577] -0.44572845 -12.0 action_pair_2_1 [0.19648269 0. ] [-0.13310385 0.04720655] -0.44552478 0.0 action_pair_2_2 [0.19648269 0.16995847] [-0.13310385 -0.12275192] -0.44595572 0.0 [INFO 10:35:56] my_main Updated target network /home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/jsonpickle-1.4.2-py3.6.egg/jsonpickle/ext/numpy.py:291: UserWarning: ndarray is defined by reference to an object we do not know how to serialize. A deep copy is serialized instead, breaking memory aliasing. warnings.warn(msg) [INFO 10:36:03] my_main Updated target network [INFO 10:36:10] my_main Updated target network [INFO 10:36:18] my_main Updated target network [DEBUG 10:36:25] pymarl Stopping Heartbeat [ERROR 10:36:25] pymarl Failed after 0:00:38! Traceback (most recent call last): File "/data/workspace/CollaQ/third_party/sacred/sacred/experiment.py", line 319, in run_commandline options=args, File "/data/workspace/CollaQ/third_party/sacred/sacred/experiment.py", line 276, in run run() File "/data/workspace/CollaQ/third_party/sacred/sacred/run.py", line 238, in call self.result = self.main_function(args) File "/data/workspace/CollaQ/third_party/sacred/sacred/config/captured_function.py", line 42, in captured_function result = wrapped(args, **kwargs) File "src/main.py", line 35, in my_main run(_run, config, _log) File "/data/workspace/QPLEX/pymarl-master/src/run.py", line 51, in run run_sequential(args=args, logger=logger) File "/data/workspace/QPLEX/pymarl-master/src/run.py", line 345, in run_sequential logger.print_recent_stats() File "/data/workspace/QPLEX/pymarl-master/src/utils/logging.py", line 48, in print_recent_stats item = "{:.4f}".format(np.mean([x[1] for x in self.stats[k][-window:]])) File "<__array_function internals>", line 6, in mean File "/home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 3373, in mean out=out, **kwargs) File "/home/yyf/anaconda3/envs/ma/lib/python3.6/site-packages/numpy/core/_methods.py", line 170, in _mean ret = ret.dtype.type(ret / rcount) AttributeError: 'torch.dtype' object has no attribute 'type'
During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "src/main.py", line 98, in
ex.run_commandline(params)
File "/data/workspace/CollaQ/third_party/sacred/sacred/experiment.py", line 348, in run_commandline
print_filtered_stacktrace()
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 494, in print_filtered_stacktrace
print(format_filtered_stacktrace(filter_traceback), file=sys.stderr)
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 529, in format_filtered_stacktrace
return "".join(filtered_traceback_format(tb_exception))
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 571, in filtered_traceback_format
if not _is_sacred_frame(current_tb.tb_frame):
File "/data/workspace/CollaQ/third_party/sacred/sacred/utils.py", line 490, in _is_sacred_frame
return frame.f_globals["name"].split(".")[0] == "sacred"
KeyError: 'name'
What's going on?