Errors with gym-unrealcv for active object tracking about gym-unrealcv HOT 6 CLOSED

It seems that the env-id you used is not correct. Please refer to this page for the correct env-id .
Specifically, the end-id of the training environment is UnrealTrack-City1StefaniPath1-DiscreteColor-v1 in the ICML 2018, and the random room is UnrealTrack-RandoomRoom-DiscreteColor-v4 .

from gym-unrealcv.

It seems that the env-id you used is not correct. Please refer to this page for the correct env-id . Specifically, the end-id of the training environment is UnrealTrack-City1StefaniPath1-DiscreteColor-v1 in the ICML 2018, and the random room is UnrealTrack-RandoomRoom-DiscreteColor-v4 .

Thank you for your reply! I make steps as you said, python train.py --num-workers=1 --env-id=UnrealTrack-City1StefaniPath1-DiscreteColor-v1 --env-val-id=UnrealTrack-City1StefaniPath1-DiscreteColor-v1 --model-id=convx2lstm --lr=0.00005 --log-dir=save/tmp --max-global-steps=250000000 --val-model-secs=-1 --max-val-episodes=40. But new questions occured in tmux 1:w-0:

ERROR:init:205:Can not connect to ('127.0.0.1', 9000)
ERROR:init:206:Error [Errno 111] Connection refused
UnrealCV server is not running. Please try again

Can you give me some advice about it?

from gym-unrealcv.

It shows that the unrealcv client in python fails to connect to the unrealcv server in UE4 environment. Do you see any new windows appearing on your desktop?

from gym-unrealcv.

It shows that the unrealcv client in python fails to connect to the unrealcv server in UE4 environment. Do you see any new windows appearing on your desktop?

Thank you for your reply! I found that the error because the server did not have a monitor. I try it on another computer with a monitor. But new errors occured:

0:ps

~/hzt/flour/active_tracking_drl$ CUDA_VISIBLE_DEVICES= /home/dell/anaconda3/envs/tracker/bin/python worker.py --log-dir save/tmp --env-id UnrealTrack-City1StefaniPath1-DiscreteColor-v1 --model-id convx2lstm --num-workers 1 --max-global-steps 250000000 --lr 5e-05 --ent-factor 0.01 --egreedy -1.0 --explore-factor 1.0 --job-name ps
/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/autobahn/wamp/auth.py:40: CryptographyDeprecationWarning: Python 3.5 support will be dropped in the next release of cryptography. Please upgrade your Python.
from cryptography.hazmat.primitives.kdf.pbkdf2 import PBKDF2HMAC
/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/universe/runtimes/init.py:7: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
spec = yaml.load(f)
[2022-04-07 20:27:39,255] Writing logs to file: /tmp/universe-739626.log
I tensorflow/core/distributed_runtime/rpc/grpc_channel.cc:197] Initialize GrpcChannelCache for job ps -> {0 -> localhost:12222}
I tensorflow/core/distributed_runtime/rpc/grpc_channel.cc:197] Initialize GrpcChannelCache for job worker -> {0 -> 127.0.0.1:12223}
I tensorflow/core/distributed_runtime/rpc/grpc_server_lib.cc:211] Started server with target: grpc://localhost:12222

1:w-0

[2022-04-07 20:27:39,262] Making new env: UnrealTrack-City1StefaniPath1-DiscreteColor-v1
Running docker-free env, pid:739730
Please wait for a while to launch env......
exec nohup /home/dell/hzt/flour/gym-unrealcv/gym_unrealcv/envs/UnrealEnv/SplineCharacterF/SplineCharacterC/Binaries/Linux/SplineCharacterC
nohup: ignoring input and appending output to 'nohup.out'
127.0.0.1
INFO:init:192:Got connection confirm: b'connected to SplineCharacterC'
Traceback (most recent call last):
File "worker.py", line 268, in
tf.app.run()
File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/tensorflow/python/platform/app.py", line 43, in run
sys.exit(main(sys.argv[:1] + flags_passthrough))
File "worker.py", line 257, in main
run_worker(args, server, cluster)
File "worker.py", line 36, in run_worker
env = create_env(args.env_id, client_id=str(args.task), remotes=args.remotes)
File "/home/dell/hzt/flour/active_tracking_drl/envs.py", line 49, in create_env
return create_unrealcv_env(env_full_name)
File "/home/dell/hzt/flour/active_tracking_drl/envs.py", line 120, in create_unrealcv_env
env = gym.make(env_full_name)
File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/envs/registration.py", line 161, in make
return registry.make(id)
File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/envs/registration.py", line 119, in make
env = spec.make()
File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/envs/registration.py", line 86, in make
env = cls(**self._kwargs)
File "/home/dell/hzt/flour/gym-unrealcv/gym_unrealcv/envs/unrealcv_tracking_spline.py", line 70, in init
self.reward_function = reward.Reward(setting)
File "/home/dell/hzt/flour/gym-unrealcv/gym_unrealcv/envs/tracking/reward.py", line 11, in init
self.dis_min = setting['min_distance']
KeyError: 'min_distance'

2:w-val

/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/autobahn/wamp/auth.py:40: CryptographyDeprecationWarning: Python 3.5 support will be dropped in the next release of cryptography. Please upgrade your Python.
from cryptography.hazmat.primitives.kdf.pbkdf2 import PBKDF2HMAC
/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/universe/runtimes/init.py:7: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
spec = yaml.load(f)
[2022-04-07 20:27:49,284] Writing logs to file: /tmp/universe-739772.log
[2022-04-07 20:27:49,286] worker_val: val_model_secs < 0, no validating, exit...

3:tb

(base) dell@dell: ~/hzt/flour/active_tracking_drl$ tensorboard --logdir save/tmp --port 12345
tensorboard: command not found

I also do not know the how to control the agent to move in a screen. The agent can only rely on the mouse to turn and can't move.
Hope for your reply!

from gym-unrealcv.

Sorry for the late reply!
I have updated the environment setting files (City1StefaniPath1.json and City2MalcomPath2.json), and the error (KeyError: 'min_distance') you reported should be fixed.

from gym-unrealcv.

Sorry for the late reply! I have updated the environment setting files (City1StefaniPath1.json and City2MalcomPath2.json), and the error (KeyError: 'min_distance') you reported should be fixed.

Thank you for helping me!

from gym-unrealcv.