Resolved invalid API version number and recv failed

Hi, first i'm sorry for my poor english.
I tried to follow the micro-game kart tutorial to learn how Ml-agents works; first i tried with tensorflow cpu but my intel proc does'nt support avx/avx2 , so i followed a tutorial to install tensorflow gpu.

I use anaconda in a virtual environment ,i'm using python version 3.7.9 and tensorflow-gpuv2.1.0 ; ml-agents v0.19.0

Steps i did in Unity: i downladed the micro-game kart project ; opened the ml agent training scene

Steps i did in python: with anaconda3

-> create a virtual environment (python 3.7.9)
->install in that environment tensorflow-gpu, after installing cudatoolkit & cudnn ,
->install mlagents from the latestbranch ( with pip install -e . in each directory , mlagents-enves and mlagents
->run the ml agnts learn command from the kart microgame folder

i runned first the python command and then pressed play . I got the error below .
thanks in advance for your help .

Code (CSharp):

1. 2020-09-17 15:22:32.529850: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2. WARNING:tensorflow:From C:\Users\-\.conda\envs\chameleon\lib\site-packages\tensorflow_core\python\compat\v2_compat.py:88: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
3. Instructions for updating:
4. non-resource variables are not supported in the long term
5.  
6.  
7.                         ▄▄▄▓▓▓▓
8.                    ╓▓▓▓▓▓▓█▓▓▓▓▓
9.               ,▄▄▄m▀▀▀'  ,▓▓▓▀▓▓▄                           ▓▓▓  ▓▓▌
10.            ▄▓▓▓▀'      ▄▓▓▀  ▓▓▓      ▄▄     ▄▄ ,▄▄ ▄▄▄▄   ,▄▄ ▄▓▓▌▄ ▄▄▄    ,▄▄
11.           ▄▓▓▓▀        ▄▓▓▀   ▐▓▓▌     ▓▓▌   ▐▓▓ ▐▓▓▓▀▀▀▓▓▌ ▓▓▓ ▀▓▓▌▀ ^▓▓▌  ╒▓▓▌
12.         ▄▓▓▓▓▓▄▄▄▄▄▄▄▄▓▓▓      ▓▀      ▓▓▌   ▐▓▓ ▐▓▓    ▓▓▓ ▓▓▓  ▓▓▌   ▐▓▓▄ ▓▓▌
13.         ▀▓▓▓▓▀▀▀▀▀▀▀▀▀▀▓▓▄     ▓▓      ▓▓▌   ▐▓▓ ▐▓▓    ▓▓▓ ▓▓▓  ▓▓▌    ▐▓▓▐▓▓
14.           ^█▓▓▓        ▀▓▓▄   ▐▓▓▌     ▓▓▓▓▄▓▓▓▓ ▐▓▓    ▓▓▓ ▓▓▓  ▓▓▓▄    ▓▓▓▓`
15.             '▀▓▓▓▄      ^▓▓▓  ▓▓▓       └▀▀▀▀ ▀▀ ^▀▀    `▀▀ `▀▀   '▀▀    ▐▓▓▌
16.                ▀▀▀▀▓▄▄▄   ▓▓▓▓▓▓,                                      ▓▓▓▓▀
17.                    `▀█▓▓▓▓▓▓▓▓▓▌
18.                         ¬`▀▀▀█▓
19.  
20.  
21. Version information:
22.   ml-agents: 0.19.0,
23.   ml-agents-envs: 0.19.0,
24.   Communicator API: 1.0.0,
25.   TensorFlow: 2.1.0
26. 2020-09-17 15:22:38 WARNING [learn.py:256] The --train option has been deprecated. Train mode is now the default. Use --inference to run in inference mode.
27. 2020-09-17 15:22:38 INFO [learn.py:271] run_seed set to 2703
28. 2020-09-17 15:22:39.414271: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
29. WARNING:tensorflow:From C:\Users\-\.conda\envs\chameleon\lib\site-packages\tensorflow_core\python\compat\v2_compat.py:88: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
30. Instructions for updating:
31. non-resource variables are not supported in the long term
32. 2020-09-17 15:22:41 INFO [environment.py:199] Listening on port 5004. Start training by pressing the Play button in the Unity Editor.
33. Process Process-1:
34. Traceback (most recent call last):
35.   File "C:\Users\-\.conda\envs\chameleon\lib\multiprocessing\process.py", line 297, in _bootstrap
36.     self.run()
37.   File "C:\Users\-\.conda\envs\chameleon\lib\multiprocessing\process.py", line 99, in run
38.     self._target(*self._args, **self._kwargs)
39.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\subprocess_env_manager.py", line 139, in worker
40.     worker_id, [env_parameters, engine_configuration_channel, stats_channel]
41.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\learn.py", line 207, in create_unity_environment
42.     log_folder=log_folder,
43.   File "h:\tutos\ml-agents-latest_release\ml-agents-envs\mlagents_envs\environment.py", line 220, in __init__
44.     aca_params.package_version,
45.   File "h:\tutos\ml-agents-latest_release\ml-agents-envs\mlagents_envs\environment.py", line 85, in _check_communication_compatibility
46.     unity_communicator_version = StrictVersion(unity_com_ver)
47.   File "C:\Users\-\.conda\envs\chameleon\lib\distutils\version.py", line 40, in __init__
48.     self.parse(vstring)
49.   File "C:\Users\-\.conda\envs\chameleon\lib\distutils\version.py", line 137, in parse
50.     raise ValueError("invalid version number '%s'" % vstring)
51. ValueError: invalid version number 'API-13'
52. 2020-09-17 15:22:57 INFO [trainer_controller.py:192] Learning was interrupted. Please wait while the graph is generated.
53. 2020-09-17 15:22:57 INFO [trainer_controller.py:76] Saved Model
54. Traceback (most recent call last):
55.   File "C:\Users\-\.conda\envs\chameleon\lib\multiprocessing\connection.py", line 312, in _recv_bytes
56.     nread, err = ov.GetOverlappedResult(True)
57. BrokenPipeError: [WinError 109] Le canal de communication a été fermé
58.  
59. During handling of the above exception, another exception occurred:
60.  
61. Traceback (most recent call last):
62.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\subprocess_env_manager.py", line 88, in recv
63.     response: EnvironmentResponse = self.conn.recv()
64.   File "C:\Users\-\.conda\envs\chameleon\lib\multiprocessing\connection.py", line 250, in recv
65.     buf = self._recv_bytes()
66.   File "C:\Users\-\.conda\envs\chameleon\lib\multiprocessing\connection.py", line 321, in _recv_bytes
67.     raise EOFError
68. EOFError
69.  
70. During handling of the above exception, another exception occurred:
71.  
72. Traceback (most recent call last):
73.   File "C:\Users\-\.conda\envs\chameleon\Scripts\mlagents-learn-script.py", line 33, in <module>
74.     sys.exit(load_entry_point('mlagents', 'console_scripts', 'mlagents-learn')())
75.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\learn.py", line 276, in main
76.     run_cli(parse_command_line())
77.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\learn.py", line 272, in run_cli
78.     run_training(run_seed, options)
79.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\learn.py", line 149, in run_training
80.     tc.start_learning(env_manager)
81.   File "h:\tutos\ml-agents-latest_release\ml-agents-envs\mlagents_envs\timers.py", line 305, in wrapped
82.     return func(*args, **kwargs)
83.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\trainer_controller.py", line 201, in start_learning
84.     raise ex
85.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\trainer_controller.py", line 177, in start_learning
86.     self._reset_env(env_manager)
87.   File "h:\tutos\ml-agents-latest_release\ml-agents-envs\mlagents_envs\timers.py", line 305, in wrapped
88.     return func(*args, **kwargs)
89.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\trainer_controller.py", line 113, in _reset_env
90.     env_manager.reset(config=new_config)
91.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\env_manager.py", line 66, in reset
92.     self.first_step_infos = self._reset_env(config)
93.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\subprocess_env_manager.py", line 290, in _reset_env
94.     ew.previous_step = EnvironmentStep(ew.recv().payload, ew.worker_id, {}, {})
95.   File "h:\tutos\ml-agents-latest_release\ml-agents\mlagents\trainers\subprocess_env_manager.py", line 94, in recv
96.     raise UnityCommunicationException("UnityEnvironment worker: recv failed.")
97. mlagents_envs.exception.UnityCommunicationException: UnityEnvironment worker: recv failed.

 

Hi again,
I've tried to test the ml-agents examples( 3DBall test ) with a new conda virtual env with ml-agents 0.19.0 , ml-agents-envs 0.19.0 tensorflow2.1.0 and tensorflow-gpu 2.3.0 and IT WORKED !!!
I could make a training session, get the nn file and inference then.

But when i came back with the same env for the kart project , same error :

Code (CSharp):

1.  Version information:
2.   ml-agents: 0.19.0,
3.   ml-agents-envs: 0.19.0,
4.   Communicator API: 1.0.0,
5.   TensorFlow: 2.1.0
6. 2020-09-18 15:23:11 INFO [learn.py:271] run_seed set to 3872
7. 2020-09-18 15:23:11.999721: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
8. WARNING:tensorflow:From c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\tensorflow_core\python\compat\v2_compat.py:88: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
9. Instructions for updating:
10. non-resource variables are not supported in the long term
11. 2020-09-18 15:23:14 INFO [environment.py:199] Listening on port 5004. Start training by pressing the Play button in the Unity Editor.
12. Process Process-1:
13. Traceback (most recent call last):
14.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\multiprocessing\process.py", line 297, in _bootstrap
15.     self.run()
16.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\multiprocessing\process.py", line 99, in run
17.     self._target(*self._args, **self._kwargs)
18.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\subprocess_env_manager.py", line 139, in worker
19.     worker_id, [env_parameters, engine_configuration_channel, stats_channel]
20.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\learn.py", line 207, in create_unity_environment
21.     log_folder=log_folder,
22.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents_envs\environment.py", line 220, in __init__
23.     aca_params.package_version,
24.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents_envs\environment.py", line 85, in _check_communication_compatibility
25.     unity_communicator_version = StrictVersion(unity_com_ver)
26.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\distutils\version.py", line 40, in __init__
27.     self.parse(vstring)
28.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\distutils\version.py", line 137, in parse
29.     raise ValueError("invalid version number '%s'" % vstring)
30. ValueError: invalid version number 'API-13'
31. 2020-09-18 15:23:25 INFO [trainer_controller.py:192] Learning was interrupted. Please wait while the graph is generated.
32. 2020-09-18 15:23:25 INFO [trainer_controller.py:76] Saved Model
33. Traceback (most recent call last):
34.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\multiprocessing\connection.py", line 312, in _recv_bytes
35.     nread, err = ov.GetOverlappedResult(True)
36. BrokenPipeError: [WinError 109] Le canal de communication a été fermé
37.  
38. During handling of the above exception, another exception occurred:
39.  
40. Traceback (most recent call last):
41.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\subprocess_env_manager.py", line 88, in recv
42.     response: EnvironmentResponse = self.conn.recv()
43.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\multiprocessing\connection.py", line 250, in recv
44.     buf = self._recv_bytes()
45.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\multiprocessing\connection.py", line 321, in _recv_bytes
46.     raise EOFError
47. EOFError
48.  
49. During handling of the above exception, another exception occurred:
50.  
51. Traceback (most recent call last):
52.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\runpy.py", line 193, in _run_module_as_main
53.     "__main__", mod_spec)
54.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\runpy.py", line 85, in _run_code
55.     exec(code, run_globals)
56.   File "C:\Users\-\.conda\envs\MLAgent_training_env\Scripts\mlagents-learn.exe\__main__.py", line 7, in <module>
57.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\learn.py", line 276, in main
58.     run_cli(parse_command_line())
59.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\learn.py", line 272, in run_cli
60.     run_training(run_seed, options)
61.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\learn.py", line 149, in run_training
62.     tc.start_learning(env_manager)
63.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents_envs\timers.py", line 305, in wrapped
64.     return func(*args, **kwargs)
65.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\trainer_controller.py", line 201, in start_learning
66.     raise ex
67.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\trainer_controller.py", line 177, in start_learning
68.     self._reset_env(env_manager)
69.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents_envs\timers.py", line 305, in wrapped
70.     return func(*args, **kwargs)
71.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\trainer_controller.py", line 113, in _reset_env
72.     env_manager.reset(config=new_config)
73.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\env_manager.py", line 66, in reset
74.     self.first_step_infos = self._reset_env(config)
75.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\subprocess_env_manager.py", line 290, in _reset_env
76.     ew.previous_step = EnvironmentStep(ew.recv().payload, ew.worker_id, {}, {})
77.   File "c:\users\-\.conda\envs\mlagent_training_env\lib\site-packages\mlagents\trainers\subprocess_env_manager.py", line 94, in recv
78.     raise UnityCommunicationException("UnityEnvironment worker: recv failed.")
79. mlagents_envs.exception.UnityCommunicationException: UnityEnvironment worker: recv failed.

I saw in different threads that recv failed could be an issue with http_ proxy but i don't have any proxy , do i have to check something inside Unity ?
Why could i have an invalid version number with kart project and not with 3DBall project ? Do i have to downgrade the ml-agent within unity package manager?
I'm sorry i'm a total beginner and sorry for my grammatical mistakes if i made !
And of course thanks in advance for your answers

 

And also i got this error in unity console :

Couldn't connect to trainer on port 5004 using API version API-13. Will perform inference instead.
UnityEngine.Debug:Log(Object)
MLAgents.Academy:InitializeEnvironment() (at Assets/Karting/ML_Agents/Scripts/Academy.cs:228)
MLAgents.Academy:LazyInitialization() (at Assets/Karting/ML_Agents/Scripts/Academy.cs:147)
MLAgents.Agent:OnEnable() (at Assets/Karting/ML_Agents/Scripts/Agent.cs:255)

i guess it's because python API is 14 and unity mlagents is 13 but if i upgrade the mlagents it doesn't work

 

Having similar issues as you; both the API-13 and recv failed errors. I'm on TF 2.3, Python 3.7.4 and Unity 19.4..Same agents version. Like you, training works fine in the ml agents examples (3D Ball, Roller Ball), but not in the Microgame. The main difference is that the kart game doesn't use the ml agents package and its dependencies. Playing around with that breaks a whole lot of stuff.

 

OK , problem solved : after a couple of days trying to figure it out , i solved it doing this :
if you are using tensorflow gpu and were in this case, you have to do the following in this order :

-create a new environment with python 3.7.9
-install mlagents0.13.1 with development method i mean :
- download or clone the repo from github release ( be careful and take version 0.13 cause the latest release doesn't match the API )
- in your conda prompt access the folder of the downloaded release cd + the path to your folder )
-type "cd .\ml-agents-env " then type "pip install -e ."
-type "cd .."
-type "cd .\ml-agents" then type again "pip install -e ."

-install tensorflow-gpu with the command line : "conda install -c conda-forge tensorflow-gpu"

-type " conda remove tensorflow"
-type " conda install tensorflow-estimator=2.1.0"
-type " conda install tensorflow-gpu=2.1.0"

I've done alle these steps in that order and now i can train and inference my karts as it's explained in the tutorial ; nevertheless i regret that the tutorial isn't updated for the latest release and that we can't find good explanations for training with a gpu tensorflow ( even less in french, which is my native language)

By the way, thank you for all the courses