adrianp- / gym_trading Goto Github PK

I have been playing around with implementing an updating chart with the gym.envs.classic_control.rendering object, as seen in https://github.com/openai/gym/blob/master/gym/envs/classic_control/pendulum.py

I added a variable 'self.data' to save the data, and

   def render(self, mode='human'):
        if self.viewer is None:
            print("Creating the rendering view:")
            self.viewer = rendering.Viewer(640, 480)
        fig = self.data.plot().get_figure()
        filename = './plot.png'
        fig.savefig(filename)
        plt.close('all')
        self.img = rendering.Image(filename, 640., 480.)
        self.imgtrans = rendering.Transform()
        self.img.add_attr(self.imgtrans)
        self.viewer.add_onetime(self.img)
        return self.viewer.render(return_rgb_array = mode =='rgb_array')

However only a white image window appears. The png file is fine, but it doesn't show in self.viewer. I've not been able to find much information on this. Any thoughts?

There is a case when Baseline DQN Gym_Trading Tutorial.ipynb fails with KeyError Profit. It happens when agent the whole episode choose only 0 and 3 actions. This results in that neither this line

profit = journal["Profit"].sum()

in Jupyter notebook crashes.

isse when executing this " with U.make_session(8): "

the error is:
TypeError: config must be a tf.ConfigProto, but got <class 'int'>

i am having tensorflow version 1.5
please help

Hi,
I'm using gym_trading to learn about RL for algotrading.

Compared to some simple models I did with RNNs I noticed the learning takes significantly longer.
So I switched to running it on my 2080 ti. Runs a bit faster than on CPU.
But then I notice the total GPU utilization is only about 7-10% and that's on a desktop machine.

Is there a way I can speed up the learning and use more GPU power?

Calculating max values in such way results in that Trading Duration and Open Trade dimensions have max values equal to 0.0, which is wrong.

Hi @AdrianP- !

First of all, many thanks for this repo! For me, it's the clearest and features-rich at the same time among many other trading gyms.

Sorry for using issues for asking the questing, but I haven't found any contacts on your GitHub page.

Can you please clarify the difference between reward and Profit?

Adrian is possible for you to put together a notebook to integrate it with IB for paper trading or demo trading in a simple realtime test environment

I try to create a new environnment but I have a problem :

Traceback (most recent call last):
File "test.py", line 16, in
env = gym.make('trading-v0')
File "/home/nicolas/.local/lib/python3.5/site-packages/gym/envs/registration.py", line 163, in make
return registry.make(id)
File "/home/nicolas/.local/lib/python3.5/site-packages/gym/envs/registration.py", line 121, in make
patch_deprecated_methods(env)
File "/home/nicolas/.local/lib/python3.5/site-packages/gym/envs/registration.py", line 181, in patch_deprecated_methods
env.seed = env._seed
AttributeError: 'TradingEnv' object has no attribute '_seed'

Versions :
gym (0.9.6)

Does it requires an older version of gym ?

config must be a tf.ConfigProto, but got <class 'int'>

I get this error and I wasn't able to resolve.