Better argparse description in train_mountaincar.py

[aakash94]

Improving description for argparse.ArgumentParser in stable_baselines\deepq\experiments\train_mountaincar.py

Line 34 in the file goes like

parser = argparse.ArgumentParser(description="Train DQN on cartpole")

since the file trains for the MountainCar-v0 environment as seen in line 14 env = gym.make("MountainCar-v0") and is named accordingly, a better description could be

parser = argparse.ArgumentParser(description="Train DQN on MountainCar")

System Info Describe the characteristic of your environment:

• How the library was used : from source on Windows10, libraries installed using Conda, Code ran and edited in Visual Studio Code
• GPU model and configuration : GeForce GTX 1050 mobile
• Python version : 3.7.6
• Tensorflow version : 1.14.0

[Miffyli]

Good to see I am not the only one getting into trouble copy-pasting code :D. Good catch.

I wonder if this should be accompanied with other typo/grammar checking of the documentation. I can not find the specific issue, but IIRC there was something similar suggested by @erniejunior a while back. Such cleanup could be nice for the archival of this version of SB. @araffin comments?

[araffin]

I can not find the specific issue

That was this one ;) #686

there was something similar suggested by @erniejunior a while back.

this was a bit different: #680

Such cleanup could be nice for the archival of this version of SB. @araffin comments?

not much to add, I wanted to delete all those files (including the run_atari run_mujoco) at some point as we now have the zoo to train/test the agents and store the hyperparameters.