TypeError: __init__() got an unexpected keyword argument 'initial_amount'

[shintafiaa]

I got a bug on cell number [18]. Line code number 14 is red underlined.

here is the code which is red underlined:
e_train_gym = StockTradingEnv(df = train, **env_kwargs)

here is the error warning:
TypeError Traceback (most recent call last)
in ()
12 }
13
---> 14 e_train_gym = StockTradingEnv(df = train, **env_kwargs)

TypeError: init() got an unexpected keyword argument 'initial_amount'

How to solve this bug? thank's before.

[ObaidaAlhaasan]

Hi team,

I got the same error and when I commented out the initial_amount line of code I got

TypeError                                 Traceback (most recent call last)
[<ipython-input-52-6f99a2a2ef8c>](https://localhost:8080/#) in <module>()
     12 }
     13 
---> 14 e_train_gym = StockTradingEnv(df = train, **env_kwargs)

TypeError: __init__() missing 1 required positional argument: 'initial_list'

And in the source code, I saw that the Initial_amount is commented out, is it as simple as un-comment that line? If so I would be happy to submit a PR
Commented out initial_amount line of code

[ObaidaAlhaasan]

I see, there is a PR to revert the change that commented initial_amount
PR LINK

[NorthKoreaBestKorea]

Trying to get started and cant run the initial notebook because of this error.

[mkabjian]

I ran into the same problem. It seems 'initial_list' is now used instead of 'initial_amount'. 'buy_cost_pct' and 'sell_cost_pct' are now lists instead of floats. I'm guessing this was done so that initial amounts and slippage could be varied by instrument.

Not sure if this the correct solution but the following code seemed to work:

Create the lists:
init_list = [1000000] * 1 + [0] * stock_dimension
buy_cost_list = sell_cost_list = [0.001] * stock_dimension

Update the environment args:
env_kwargs = { "hmax": 100, "initial_list": init_list, "buy_cost_pct": buy_cost_list, "sell_cost_pct": sell_cost_list, "state_space": state_space, "stock_dim": stock_dimension, "tech_indicator_list": config.TECHNICAL_INDICATORS_LIST, "action_space": stock_dimension, "reward_scaling": 1e-4 }

Again, I have no idea if this is the correct solution but everything seemed to run after these fixes.

[dpaiton]

This was also discussed a while back here: #495

It was closed, but there is still a problem to be addressed.

[ytliu74]

I have submit a PR which has been merged and fix this bug in tutorials/1-Introduction/FinRL_StockTrading_NeurIPS_2018.ipynb

[EvanHong99]

I ran into the same problem. It seems 'initial_list' is now used instead of 'initial_amount'. 'buy_cost_pct' and 'sell_cost_pct' are now lists instead of floats. I'm guessing this was done so that initial amounts and slippage could be varied by instrument.

Not sure if this the correct solution but the following code seemed to work:

Create the lists: init_list = [1000000] * 1 + [0] * stock_dimension buy_cost_list = sell_cost_list = [0.001] * stock_dimension

Update the environment args: env_kwargs = { "hmax": 100, "initial_list": init_list, "buy_cost_pct": buy_cost_list, "sell_cost_pct": sell_cost_list, "state_space": state_space, "stock_dim": stock_dimension, "tech_indicator_list": config.TECHNICAL_INDICATORS_LIST, "action_space": stock_dimension, "reward_scaling": 1e-4 }

Again, I have no idea if this is the correct solution but everything seemed to run after these fixes.

It works! But you need to change TECHNICAL_INDICATORS_LIST into INDICATORS.

I can't fix this bug because I couldn't understand the operation in the source code.

With your help I commented the meaning of self.state:

    # for multiple stock
    # 将所有东西转换成一个list表示的state
    state = (
        [self.initial_amount] # 初始空余资金
        + self.data.close.values.tolist() #data[self.day]当天股票价格
        + self.initial_list[1:] #初始各股票持有share
        + sum(
            [
                self.data[tech].values.tolist()
                for tech in self.tech_indicator_list
            ], #当天各股票因子list
            [],
        )
    ) # append initial stocks_share to initial state, instead of all zero 

[zhumingpassional]

developpers updated the code today: initial_amount exists in args, and add the num_stock_shares, which should be initialized as [0] * stock_dimension

[PierrickLozach]

I updated the code to this:

init_list = [1000000] * 1 + [0] * stock_dimension
buy_cost_list = sell_cost_list = [0.001] * stock_dimension
num_stock_shares = [0] * stock_dimension

env_kwargs = { 
    "hmax": 100, 
    "initial_amount": init_list, 
    "num_stock_shares": num_stock_shares,
    "buy_cost_pct": buy_cost_list, 
    "sell_cost_pct": sell_cost_list, 
    "state_space": state_space, 
    "stock_dim": stock_dimension, 
    "tech_indicator_list": config.INDICATORS, 
    "action_space": stock_dimension, 
    "reward_scaling": 1e-4 
}
e_train_gym = StockTradingEnv(df = train, **env_kwargs)

It runs fine however the next step (Environment for Training) now returns this error:

ValueError                                Traceback (most recent call last)
[<ipython-input-27-575def09184a>](https://localhost:8080/#) in <module>()
----> 1 env_train, _ = e_train_gym.get_sb_env()
      2 print(type(env_train))

2 frames
[/usr/local/lib/python3.7/dist-packages/finrl/finrl_meta/env_stock_trading/env_stocktrading.py](https://localhost:8080/#) in get_sb_env(self)
    504     def get_sb_env(self):
    505         e = DummyVecEnv([lambda: self])
--> 506         obs = e.reset()
    507         return e, obs

[/usr/local/lib/python3.7/dist-packages/stable_baselines3/common/vec_env/dummy_vec_env.py](https://localhost:8080/#) in reset(self)
     60         for env_idx in range(self.num_envs):
     61             obs = self.envs[env_idx].reset()
---> 62             self._save_obs(env_idx, obs)
     63         return self._obs_from_buf()
     64 

[/usr/local/lib/python3.7/dist-packages/stable_baselines3/common/vec_env/dummy_vec_env.py](https://localhost:8080/#) in _save_obs(self, env_idx, obs)
     90         for key in self.keys:
     91             if key is None:
---> 92                 self.buf_obs[key][env_idx] = obs
     93             else:
     94                 self.buf_obs[key][env_idx] = obs[key]

ValueError: setting an array element with a sequence. The requested array would exceed the maximum number of dimension of 1.

[zhumingpassional]

@PierrickI3 "initial_amount" should be set 1000000

[PierrickLozach]

That works. Thanks!

[zhumingpassional]

@PierrickI3 Could you please submit a PR for this problem if you have enough time?

[PierrickLozach]

Not sure how I can do that.

Here is the final version of my Trade Data Split step which works:

buy_cost_list = sell_cost_list = [0.001] * stock_dimension
num_stock_shares = [0] * stock_dimension

env_kwargs = { 
    "hmax": 100, 
    "initial_amount": 1000000, 
    "num_stock_shares": num_stock_shares,
    "buy_cost_pct": buy_cost_list, 
    "sell_cost_pct": sell_cost_list, 
    "state_space": state_space, 
    "stock_dim": stock_dimension, 
    "tech_indicator_list": config.INDICATORS, 
    "action_space": stock_dimension, 
    "reward_scaling": 1e-4 
}
e_train_gym = StockTradingEnv(df = train, **env_kwargs)

[zhumingpassional]

@PierrickI3 Fixed. Thanks.

[flyingsandwich1]

Not sure how I can do that.

Here is the final version of my Trade Data Split step which works:

buy_cost_list = sell_cost_list = [0.001] * stock_dimension
num_stock_shares = [0] * stock_dimension

env_kwargs = { 
    "hmax": 100, 
    "initial_amount": 1000000, 
    "num_stock_shares": num_stock_shares,
    "buy_cost_pct": buy_cost_list, 
    "sell_cost_pct": sell_cost_list, 
    "state_space": state_space, 
    "stock_dim": stock_dimension, 
    "tech_indicator_list": config.INDICATORS, 
    "action_space": stock_dimension, 
    "reward_scaling": 1e-4 
}
e_train_gym = StockTradingEnv(df = train, **env_kwargs)

Where did you put this in the code? I'm having issues where it says:

Traceback (most recent call last):
File "C:\Users\User\PycharmProjects\Trader\MultipleStockTrader.py", line 453, in
buy_cost_list = sell_cost_list = [0.001] * stock_dim
NameError: name 'stock_dim' is not defined

Would really appreciate the help! I'm a beginner :)

[zhumingpassional]

@flyingsandwich1 which notebook did you run? Have you revised anything?