Can someone update this project for non-Python coder?
For example, support Python3.x, support newer gym version?

python gym_http_client.py

Traceback (most recent call last):
File "gym_http_client.py", line 22, in parse_server_error_or_raise_for_status
j = resp.json()
File "C:\Miniconda3\lib\site-packages\requests\models.py", line 897, in json
return complexjson.loads(self.text, **kwargs)
File "C:\Miniconda3\lib\json_init.py", line 348, in loads
return _default_decoder.decode(s)
File "C:\Miniconda3\lib\json\decoder.py", line 337, in decode
obj, end = self.raw_decode(s, idx=_w(s, 0).end())
File "C:\Miniconda3\lib\json\decoder.py", line 355, in raw_decode
raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "gym_http_client.py", line 154, in
obs_info = client.env_observation_space_info(instance_id)
File "gym_http_client.py", line 95, in env_observation_space_info
resp = self._get_request(route)
File "gym_http_client.py", line 44, in _get_request
return self._parse_server_error_or_raise_for_status(resp)
File "gym_http_client.py", line 27, in _parse_server_error_or_raise_for_status
resp.raise_for_status()
File "C:\Miniconda3\lib\site-packages\requests\models.py", line 940, in raise_for_status
raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 500 Server Error: INTERNAL SERVER ERROR for url: http://127.0.0.1:5000/v1/envs/da42e4bc/observation_space/

File "gym_http_server.py", line 282, in env_action_space_sample
return jsonify(action = action)

Error thrown if the action space is an array, need to convert to list.

Fixed with the following code, which is actually in the gym_http_server.py but was overwritten on a recent push:

    def get_action_space_sample(self, instance_id):
        env = self._lookup_env(instance_id)
        action = env.action_space.sample()
        if isinstance(action, (list, tuple)) or ('numpy' in str(type(action))):
            try:
                action = action.tolist()
            except TypeError:
                print(type(action))
                print('TypeError')
        return action

Fixed with 1b42f25

Can you buy this chicken for $2000

Minimal Scala code request

private val contentType = ContentTypes.`application/json` //Data Content Type

  def requestToApi(request: gymApi): HttpResponse = {
  val url = "http://127.0.0.1" + ":" + 5000 + request.url
  val httpRequest = HttpRequest(uri = url).withMethod(request.method).withEntity(HttpEntity(contentType, """{}"""))
  val responseFuture: Future[HttpResponse] = Http().singleRequest(httpRequest)
  Await.result(responseFuture, timeout.second)
}

response match {
  case HttpResponse(StatusCodes.OK, headers, entity, _) => println(entity.toString())
}

Error:

  [2019-03-05 22:25:49,624] ERROR in app: Exception on /v1/envs/5d5aee73/observation_space/ [GET]
Traceback (most recent call last):
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/app.py", line 1988, in wsgi_app
    response = self.full_dispatch_request()
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/app.py", line 1641, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/app.py", line 1544, in handle_user_exception
    reraise(exc_type, exc_value, tb)
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/app.py", line 1639, in full_dispatch_request
    rv = self.dispatch_request()
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/app.py", line 1625, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "gym_http_server.py", line 340, in env_observation_space_info
    return jsonify(info = info)
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/json.py", line 266, in jsonify
    (dumps(data, indent=indent, separators=separators), '\n'),
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/json.py", line 126, in dumps
    rv = _json.dumps(obj, **kwargs)
  File "/usr/lib/python2.7/json/__init__.py", line 251, in dumps
    sort_keys=sort_keys, **kw).encode(obj)
  File "/usr/lib/python2.7/json/encoder.py", line 209, in encode
    chunks = list(chunks)
  File "/usr/lib/python2.7/json/encoder.py", line 434, in _iterencode
    for chunk in _iterencode_dict(o, _current_indent_level):
  File "/usr/lib/python2.7/json/encoder.py", line 408, in _iterencode_dict
    for chunk in chunks:
  File "/usr/lib/python2.7/json/encoder.py", line 408, in _iterencode_dict
    for chunk in chunks:
  File "/usr/lib/python2.7/json/encoder.py", line 332, in _iterencode_list
    for chunk in chunks:
  File "/usr/lib/python2.7/json/encoder.py", line 442, in _iterencode
    o = _default(o)
  File "/home/rigas/.local/lib/python2.7/site-packages/flask/json.py", line 83, in default
    return _json.JSONEncoder.default(self, o)
  File "/usr/lib/python2.7/json/encoder.py", line 184, in default
    raise TypeError(repr(o) + " is not JSON serializable")
TypeError: 4.8000002 is not JSON serializable

Any ideas what problem might be ?

Using the latest gym version environments will terminate after 200 time steps. In python I can write

env = gym.make("MountainCar-v0").env to change this behaviour. See also: http://stackoverflow.com/questions/42787924/why-is-episode-done-after-200-time-steps-gym-environment-mountaincar

But how could I get this working using the gym API in R? In the example_agent.R I tried

env_id = "MountainCar-v0.env"
instance_id = env_create(client, env_id)

but this returns an error.

I ran travis-CI at my forked repos. every python2/3 seem to be broken https://travis-ci.org/ShigekiKarita/gym-http-api

I start the gym_http_server,
when I run example_agent.py,I get the error message on server console

Server starting at: http://127.0.0.1:5000

• Serving Flask app "gym_http_server" (lazy loading)
• Environment: production
  WARNING: Do not use the development server in a production environment.
  Use a production WSGI server instead.
• Debug mode: off
  �[33mWARN: gym.spaces.Box autodetected dtype as <type 'numpy.float32'>. Please provide explicit dtype.�[0m
  [2018-08-25 19:59:14,161] ERROR in app: Exception on /v1/envs/f8c9a1cd/step/ [POST]
  Traceback (most recent call last):
  File "C:\Anaconda2\lib\site-packages\flask\app.py", line 2292, in wsgi_app
  response = self.full_dispatch_request()
  File "C:\Anaconda2\lib\site-packages\flask\app.py", line 1815, in full_dispatch_request
  rv = self.handle_user_exception(e)
  File "C:\Anaconda2\lib\site-packages\flask\app.py", line 1718, in handle_user_exception
  reraise(exc_type, exc_value, tb)
  File "C:\Anaconda2\lib\site-packages\flask\app.py", line 1813, in full_dispatch_request
  rv = self.dispatch_request()
  File "C:\Anaconda2\lib\site-packages\flask\app.py", line 1799, in dispatch_request
  return self.view_functionsrule.endpoint
  File "gym_http_server.py", line 258, in env_step
  [obs_jsonable, reward, done, info] = envs.step(instance_id, action, render)
  File "gym_http_server.py", line 71, in step
  env.render()
  File "C:\Anaconda2\lib\site-packages\gym\core.py", line 284, in render
  return self.env.render(mode)
  File "C:\Anaconda2\lib\site-packages\gym\core.py", line 284, in render
  return self.env.render(mode)
  File "C:\Anaconda2\lib\site-packages\gym\envs\classic_control\cartpole.py", line 137, in render
  return self.viewer.render(return_rgb_array = mode=='rgb_array')
  File "C:\Anaconda2\lib\site-packages\gym\envs\classic_control\rendering.py", line 82, in render
  glClearColor(1,1,1,1)
  File "C:\Anaconda2\lib\site-packages\pyglet\gl\lib.py", line 105, in errcheck
  raise GLException(msg)

Hey @catherio, thanks for all the work you've put into this. I'm wondering if you could tell me how actively OpenAI plans to maintain this repo. If you are all lacking dev-time maybe putting this project on TravisCI and adding a merge-bot, if a current contributor +1s the PR, could help speed things up? (Seeing as there are currently 13 language bindings, I'd guess that this repo requires more than the current number of developers to ensure stability, anyhow).

Thanks for the nice api. I managed to run CartPole-v0 using python example_agent.py command. Now I need to run HandManipulateBlock-v0. But this environment is not available in gym 0.7.4 (specified in requirements.txt). So I changed the versions of numpy, gym and requests in requirements.txt to the following:
numpy==1.15.0,
gym==0.10.5,
requests==2.19.1
Now gym-http-api recognizes HandManipulateBlock-v0 environment. But when I try to run the environment using python example_agent.py, the execution halts with the followint exception:

Traceback (most recent call last):
File "/home/navid/gym-http-api/gym_http_client.py", line 22, in _parse_server_error_or_raise_for_status
j = resp.json()
File "/home/navid/my-gym-http-api/lib/python3.5/site-packages/requests/models.py", line 896, in json
return complexjson.loads(self.text, **kwargs)
File "/usr/lib/python3.5/json/init.py", line 319, in loads
return _default_decoder.decode(s)
File "/usr/lib/python3.5/json/decoder.py", line 339, in decode
obj, end = self.raw_decode(s, idx=_w(s, 0).end())
File "/usr/lib/python3.5/json/decoder.py", line 357, in raw_decode
raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

Could anyone give me some hint about how to fix this? Thanks.
Best regards,
Navid

I'm using the gym API in R running the example_agent.R file.

When I use the CartPole-v0 environment this works fine, but then I try the "FrozenLake-v0" I get an error: Error: API did not return json when I run

for (i in 1:episode_count) { ob <- env_reset(client, instance_id) for (i in 1:max_steps) { action <- env_action_space_sample(client, instance_id) results <- env_step(client, instance_id, action, render = TRUE) if (results[["done"]]) break } }

and in my WinPython Command Prompt (Python 3.6.0.1Qt5) I get the following error:

TypeError: 0 is not JSON serializable

Hope, someone can help :)

For the /v1/upload API, the request must have a training_dir key. However, the python client erroneously uses a key called directory instead:

data = {'directory': directory,
        'force': force,
        'resume': resume,
        'video_callable': video_callable}
self._post_request(route, data)

The upload() method was probably just never tested before.

Fixing this might save someone a bit of debugging time if they use the Python client as a guide for their own client.

Hello,
using ML ver 8.5 (R2015a), the required python packages in requirements.txt (python 2.7), there are errors in few calls:
(1) client.env_monitor_start(instance_id, outdir, true);
(2) client.env_monitor_close(instance_id);
(3) client.upload
(see bellow for details).
Could you please provide indications or update the code, also to work with the latest python packages.

I think there is some incompatibility with the "webwrite" matlab function; by now I "resolved" this with catching and ignoring the error in the outermost script (example_agent), and it seems to work, but I'm not sure this is a good solution.

Thanks, Ivilin

-- (1)
Error using readContentFromWebService (line 37)
The server returned the message: "NO CONTENT" for URL, 'http://127.0.0.1:5000/v1/envs/6ecaff44/monitor/start/' (with HTTP response code 204).
Error in webwrite (line 126)
[varargout{1:nargout}] = readContentFromWebService(connection, options);
Error in gym_http_client/post_request (line 25)
resp_data = webwrite(url, req_data, obj.webopt);
Error in gym_http_client/env_monitor_start (line 115)
obj.post_request(route, req_data);
Error in example_agent (line 21)
client.env_monitor_start(instance_id, outdir, true);

-- (2)
Error using readContentFromWebService (line 37)
The server returned the message: "NO CONTENT" for URL, 'http://127.0.0.1:5000/v1/envs/d8d7a341/monitor/close/' (with HTTP response code 204).
Error in webwrite (line 126)
[varargout{1:nargout}] = readContentFromWebService(connection, options);
Error in gym_http_client/post_request (line 25)
resp_data = webwrite(url, req_data, obj.webopt);
Error in gym_http_client/env_monitor_close (line 120)
obj.post_request(route, []);
Error in example_agent (line 46)
client.env_monitor_close(instance_id);

-- (3)
Error using readContentFromWebService (line 37)
The server returned the message: "BAD REQUEST" for URL, 'http://127.0.0.1:5000/v1/upload/' (with HTTP response code 400).
Error in webwrite (line 126)
[varargout{1:nargout}] = readContentFromWebService(connection, options);
Error in gym_http_client/post_request (line 25)
resp_data = webwrite(url, req_data, obj.webopt);
Error in gym_http_client/upload (line 141)
obj.post_request(route, req_data);
Error in example_agent (line 55)
client.upload(outdir);

Step to reproduce:
python2 gym_http_server.py
and in another terminal
python2 gym_http_client.py
or
python2 example_agent.py

server terminal output:

[2016-07-02 19:05:15,513] Making new env: CartPole-v0
[2016-07-02 19:05:15,527] Creating monitor directory /tmp/random-agent-results
[2016-07-02 19:05:15,530] Starting new video recorder writing to /tmp/random-agent-results/openaigym.video.0.17805.video000000.mp4
[2016-07-02 19:05:15,918] ERROR in app: Exception on /v1/envs/5bd5d0fb/reset/ [POST]
Traceback (most recent call last):
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1988, in wsgi_app
    response = self.full_dispatch_request()
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1641, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1544, in handle_user_exception
    reraise(exc_type, exc_value, tb)
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1639, in full_dispatch_request
    rv = self.dispatch_request()
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1625, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "gym_http_server.py", line 189, in env_reset
    observation = envs.reset(instance_id)
  File "gym_http_server.py", line 54, in reset
    obs = env.reset()
  File "/usr/lib/python2.7/site-packages/gym/core.py", line 124, in reset
    self.monitor._after_reset(observation)
  File "/usr/lib/python2.7/site-packages/gym/monitoring/monitor.py", line 248, in _after_reset
    self.video_recorder.capture_frame()
  File "/usr/lib/python2.7/site-packages/gym/monitoring/video_recorder.py", line 105, in capture_frame
    frame = self.env.render(mode=render_mode)
  File "/usr/lib/python2.7/site-packages/gym/core.py", line 175, in render
    return self._render(mode=mode, close=close)
  File "/usr/lib/python2.7/site-packages/gym/envs/classic_control/cartpole.py", line 141, in _render
    return self.viewer.get_array()
  File "/usr/lib/python2.7/site-packages/gym/envs/classic_control/rendering.py", line 107, in get_array
    arr = arr.reshape(self.height, self.width, 4)
ValueError: total size of new array must be unchanged
[2016-07-02 19:05:15,918] Exception on /v1/envs/5bd5d0fb/reset/ [POST]
Traceback (most recent call last):
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1988, in wsgi_app
    response = self.full_dispatch_request()
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1641, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1544, in handle_user_exception
    reraise(exc_type, exc_value, tb)
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1639, in full_dispatch_request
    rv = self.dispatch_request()
  File "/usr/lib/python2.7/site-packages/flask/app.py", line 1625, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "gym_http_server.py", line 189, in env_reset
    observation = envs.reset(instance_id)
  File "gym_http_server.py", line 54, in reset
    obs = env.reset()
  File "/usr/lib/python2.7/site-packages/gym/core.py", line 124, in reset
    self.monitor._after_reset(observation)
  File "/usr/lib/python2.7/site-packages/gym/monitoring/monitor.py", line 248, in _after_reset
    self.video_recorder.capture_frame()
  File "/usr/lib/python2.7/site-packages/gym/monitoring/video_recorder.py", line 105, in capture_frame
    frame = self.env.render(mode=render_mode)
  File "/usr/lib/python2.7/site-packages/gym/core.py", line 175, in render
    return self._render(mode=mode, close=close)
  File "/usr/lib/python2.7/site-packages/gym/envs/classic_control/cartpole.py", line 141, in _render
    return self.viewer.get_array()
  File "/usr/lib/python2.7/site-packages/gym/envs/classic_control/rendering.py", line 107, in get_array
    arr = arr.reshape(self.height, self.width, 4)
ValueError: total size of new array must be unchanged
^C[2016-07-02 19:05:22,624] Finished writing results. You can upload them to the scoreboard via gym.upload(u'/tmp/random-agent-results')

I only have to check that my understanding of the task is correct before submitting a pull request.
As far as I can see, the picking of a random action from the action space is done by the example agent and that shouldn't be the case. Instead, this should be done on the "server side" by actually using the sample() method available in the action space of the specific environment, assuming the additional endpoint is provided.

when: running the example python agent
where: in a virtual environment with Python 3.5.1, since

The urlparse module is renamed to urllib.parse in Python 3.

Measure latency/performance. How slow is HTTP, Flask? Under what use patterns is HTTP sufficiently performative, and under what use patterns would a user want a faster implementation, for example using ZeroMQ or similar?

The function get_action_space_info does not return the action space when the action space contains a Tuple. It will just return {'name': 'Tuple'}. Tested on Copy-v0.

@catherio: Can you shed some light on which of these might be outstanding? If we can translate some of these into issues and sort them, myself and other community members might be able to start to tackle them.

Looks done (confirmation would be helpful)

• Implement 'sample' (and test it)
• Implement 'contains' (and test it)

TODO

• Make the directory structure better conform to standard python package structures (http://www.kennethreitz.org/essays/repository-structure-and-python, http://peterdowns.com/posts/first-time-with-pypi.html)
• Docker integration
• Handle ResetNeeded while monitor is active
• Handle APIConnectionError: Unexpected error communicating with OpenAI Gym
• Check: was anything improved by adding session / socket reuse?
• Reports of "broken pipe" errors: reproduce and investigate
• Measure latency/performance. How slow is HTTP, Flask? What is the use case for this implementation, versus a potential future faster ZeroMQ implementation?
• Make remote environments have the same interface as non-remote environments
• Get Travis CI working
• Test all possible environments in integration tests
• Handle the error thrown if the directory isn't cleared for the monitor
• Document the fact that two-monitors-open will cause a problem; be clear that this is meant to be one-client

Can you confirm that the code in this repository is under the MIT license like the gym repo?

A commit of a LICENSE file would be great.

Thanks!

Hey GUYS,

I am trying to render atari game the video game has some issues with the graphics, On classic control are pretty nice I have a problem only with atari does anyone knows why?

I get the following error while running a custom made environment or any predefined environment like CartPole-v0 (I am trying to use the CPP binding, but also I get a similar error with "python example_agent.py"):-

./test1

ERROR: bad HTTP response code, and also cannot parse server message:

Internal Server Error

The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.

python gym_http_server.py

Server starting at: http://127.0.0.1:5000
[2021-01-21 15:31:17,337] ERROR in app: Exception on /v1/envs/ff0e588b/action_space/ [GET]
Traceback (most recent call last):
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/app.py", line 1988, in wsgi_app
response = self.full_dispatch_request()
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/app.py", line 1641, in full_dispatch_request
rv = self.handle_user_exception(e)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/app.py", line 1544, in handle_user_exception
reraise(exc_type, exc_value, tb)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/_compat.py", line 33, in reraise
raise value
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/app.py", line 1639, in full_dispatch_request
rv = self.dispatch_request()
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/app.py", line 1625, in dispatch_request
return self.view_functionsrule.endpoint
File "gym_http_server.py", line 278, in env_action_space_info
return jsonify(info = info)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/json.py", line 266, in jsonify
(dumps(data, indent=indent, separators=separators), '\n'),
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/json.py", line 126, in dumps
rv = _json.dumps(obj, **kwargs)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/json/init.py", line 238, in dumps
**kw).encode(obj)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/json/encoder.py", line 199, in encode
chunks = self.iterencode(o, _one_shot=True)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/json/encoder.py", line 257, in iterencode
return _iterencode(o, 0)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/site-packages/flask/json.py", line 83, in default
return _json.JSONEncoder.default(self, o)
File "/home/amit/miniconda3/envs/cobel-spike-http-v6/lib/python3.7/json/encoder.py", line 179, in default
raise TypeError(f'Object of type {o.class.name} '
TypeError: Object of type float32 is not JSON serializable

Kindly help me understand the problem and how to solve it.

I have -
Flask 0.11.1
Flask-Compress 1.4.0
Flask-SocketIO 2.9.6
gym 0.12.4
numpy 1.16.3
pip 19.1.1
pytest 6.2.1
requests 2.25.0

I have recently been getting the following error. I am pretty sure that the gym changed the monitoring interface of late, however, I've also been changing workstations, so I'm not sure if this is a bug or if I installed things incorrectly.

The error:

[2017-03-02 15:36:32,956] Exception on /v1/envs/be06fa08/monitor/start/ [POST]
Traceback (most recent call last):
  File "/home/stites/.pyenv/versions/gym-http-api/lib/python3.6/site-packages/flask/app.py", line 1988, in wsgi_app
    response = self.full_dispatch_request()
  File "/home/stites/.pyenv/versions/gym-http-api/lib/python3.6/site-packages/flask/app.py", line 1641, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/home/stites/.pyenv/versions/gym-http-api/lib/python3.6/site-packages/flask/app.py", line 1544, in handle_user_exception
    reraise(exc_type, exc_value, tb)
  File "/home/stites/.pyenv/versions/gym-http-api/lib/python3.6/site-packages/flask/_compat.py", line 33, in reraise
    raise value
  File "/home/stites/.pyenv/versions/gym-http-api/lib/python3.6/site-packages/flask/app.py", line 1639, in full_dispatch_request
    rv = self.dispatch_request()
  File "/home/stites/.pyenv/versions/gym-http-api/lib/python3.6/site-packages/flask/app.py", line 1625, in dispatch_request
    return self.view_functions[rule.endpoint](**req.view_args)
  File "gym_http_server.py", line 337, in env_monitor_start
    envs.monitor_start(instance_id, directory, force, resume, video_callable)
  File "gym_http_server.py", line 126, in monitor_start
    self.envs[instance_id] = wrappers.Monitor(env, directory, force=force, resume=resume, video_callable=v_c)
AttributeError: module 'gym.wrappers' has no attribute 'Monitor'

and here's my pip freeze with python 3.6.0:

appdirs==1.4.2
click==6.7
Flask==0.11.1
gym==0.1.7
itsdangerous==0.24
Jinja2==2.9.5
MarkupSafe==0.23
nose2==0.6.5
numpy==1.11.1rc1
packaging==16.8
pyglet==1.2.4
pyparsing==2.1.10
requests==2.11.0
six==1.10.0
Werkzeug==0.11.15

Hello,

Currently, the API doesn't seem to be compatible with environments like Doom or ATARI.

Hi. Thanks for making gym available to other languages. I'm currently seeing an error when trying to reset the Blackjack-v0 environment. Here's a reproducible example using curl

# start the server
python gym_http_server.py

# create blackjack instance
curl -d '{"env_id":"Blackjack-v0"}' -H "Content-Type: application/json" -X POST http://127.0.0.1:5000/v1/envs/

# reset instance
curl -X POST http://127.0.0.1:5000/v1/envs/<instance_id>/reset/

The issue I believe is caused when calling env.observation_space.to_jsonable(obs) in reset(). The to_jsonable method expects a list of tuples for Blackjack, not a single tuple.

My current workaround is to check to see if the observation is a tuple and if so, wrap it a list first. This works fine for Blackjack but I haven't tested the other environments.

def reset(self, instance_id):
    env = self._lookup_env(instance_id)
    obs = env.reset()
    if (isinstance(obs, tuple)):
        obs = [obs]
    return env.observation_space.to_jsonable(obs)

can this be merged yet ?
#48

On linux systems that are using openssl-1.1, compiling the Rust bindings with the dependency versions in Cargo.toml will fail because the hyper version pulls in an old version of openssl bindings that's incompatible with openssl-1.1.

After installing all the prerequisites and trying to run the random_agent.cpp in
cpp bindings, it threw an Error, saying it cannot convert a certain value to float.
Changing from Breakout-v0 to CartPole-v0 made it work for the time being.