cove9988 / tradinggym Goto Github PK

Hi! I'm South Korean and I want to help you translate README.md to Korean.
Is it okay to translate your README.md?
Thank you.

找不到csv数据,不知道数据的格式,能不能发一份数据上来.

Hi, I can't download. I use a mac. I need your help. here is my error message:

Pauls-MacBook-Pro:~ paul$ pip3 install https://github.com/cove9988/TradingGym.git
Collecting https://github.com/cove9988/TradingGym.git
Downloading https://github.com/cove9988/TradingGym.git
/ 81kB 467kB/s
Cannot unpack file /private/var/folders/bd/d0vxwn3j5rnfjwnsgy0wp0300000gn/T/pip-wtngrslj-unpack/TradingGym.git (downloaded from /private/var/folders/bd/d0vxwn3j5rnfjwnsgy0wp0300000gn/T/pip-6v70569e-build, content-type: text/html; charset=utf-8); cannot detect archive format
Cannot determine archive format of /private/var/folders/bd/d0vxwn3j5rnfjwnsgy0wp0300000gn/T/pip-6v70569e-build

感谢repo

python setup.py install
https://github.com/cove9988/TradingGym/blob/master/envs/trading.py#L240

这一行好像错了哦
上面一行还是for呢, 这一行就直接是class了

另外有群加吗, 群的二维码过期了

谢谢!

dqn_agent.py and data_reader.py reference "price_usdeur.csv" but this file does not exist. data_reader.py tries to read in "ue2.csv" but this file does not exist.

CSVStreamer.py is commented to indicate that the .csv file to be fed into _generator should have no index column, but it not quite clear what the initial .csv data should look like beyond this. Is it just two columns, one with the bid and the other with the ask price of an asset?

Using TensorFlow backend.
Traceback (most recent call last):
File "dqn_agent.py", line 15, in
from tgym.envs.trading_spread import SpreadTrading
File "/home/smilewater/anaconda3/lib/python3.6/site-packages/tgym/envs/init.py", line 1, in
from trading import SpreadTrading
ModuleNotFoundError: No module named 'trading'

Not sure if this has to do with the format of the .csv file I created or not, but when I set "gen_type = 'C'" in "examples/dqn_agent.py" to get data from "./examples/price_2.csv" (which I created), it runs for 2 sessions and then I get the following error:

End of data reached, rewinding.
Traceback (most recent call last):
  File "C:\Users\...\AppData\Local\Programs\Python\Python36\lib\site-packages\tgym-0.1.15-py3.6.egg\envs\trading_tick.py", line 130, in step
    self._tick_buy,self._tick_sell  = next(self._data_generator)
  File "C:\Users\...\AppData\Local\Programs\Python\Python36\lib\site-packages\tgym-0.1.15-py3.6.egg\tgym\core.py", line 74, in __next__
    raise(e)
  File "C:\Users\...\AppData\Local\Programs\Python\Python36\lib\site-packages\tgym-0.1.15-py3.6.egg\tgym\core.py", line 71, in __next__
    return next(self.generator)
StopIteration

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "examples/dqn_agent.py", line 189, in <module>
    next_state, reward, done, _ = environment.step(action)
  File "C:\Users\...\AppData\Local\Programs\Python\Python36\lib\site-packages\tgym-0.1.15-py3.6.egg\envs\trading_tick.py", line 134, in step
    info['status'] = 'No more data.'
UnboundLocalError: local variable 'info' referenced before assignment

My .CSV, I have 4,699 rows in total. In column A, I have bid prices (3.4509...) and in column B I have ask prices (3.4511...). Does this have something to do with it? When I run dqn_agent.py using gen_type = 'R' for a random generator, it works fine.

I'm thinking it has something to do with the values in this section here:

` episodes = 50
episode_length = 10000
trading_fee = 0.05
time_fee = 0
history_length = 50

# history_length number of historical states in the observation vector.
profit_taken = 10
stop_loss = -5    # default -5
render_show = False
trading_type = 'T'
if trading_type == 'T':
    environment = TickTrading(data_generator=generator, trading_fee=trading_fee, time_fee=time_fee, history_length=history_length, episode_length=episode_length, profit_taken=profit_taken, stop_loss=stop_loss)
    action_size = len(TickTrading._actions)
else:
    environment = SpreadTrading(spread_coefficients=[1],
                                data_generator=generator,
                                trading_fee=trading_fee,
                                time_fee=time_fee,
                                history_length=history_length,
                                episode_length=episode_length)
    action_size = len(SpreadTrading._actions)
state = environment.reset()`

...but can't seem to find a working combination. Thoughts? Thanks!

hi
when I run this code "python C:\Users\jalal\TradingGym\examples\dqn_agent.py" I got below error
how can I fix it?
"C:\Anaconda3\lib\site-packages\h5py_init_.py:36: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type.
from ._conv import register_converters as register_converters
Using TensorFlow backend.
Traceback (most recent call last):
File "C:\Users\jalal\TradingGym\examples\dqn_agent.py", line 15, in
from tgym.envs.trading_spread import SpreadTrading
File "C:\Anaconda3\lib\site-packages\tgym\envs_init.py", line 1, in
from trading_spread import SpreadTrading
ModuleNotFoundError: No module named 'trading_spread'"

The wechat qr code is outdated. can renew it, or can you add me in necmetttinkarakaya

Hi, we are working on a similar project. Would you like to collab on one project together?

Let me know