Python中_的含义解析及OpenAI Gym step函数用_的优势探讨

Let's break down your questions clearly, starting with the general role of underscores in Python, then focusing on the specific use case in your OpenAI Gym code.

1. Common meanings of "_" in Python #

The underscore is a flexible, convention-driven character in Python with a few key uses:

• Store last result in interactive mode: In Python's shell, _ automatically holds the output of your last executed expression. For example:

  >>> 10 * 5
  50
  >>> _
  50
  

• Ignore unused values: This is the most relevant case for your code. When unpacking values (like from a function return), _ signals that we don't need that specific value. It's a widely accepted community convention that makes code instantly readable—anyone seeing _ knows that value is intentionally unused.
• Signal internal-use variables: A single underscore prefix (e.g., _internal_var) tells other developers that a variable or method is meant for internal use only. Python doesn't enforce privacy here, but it's a respected hint.
• Special method naming: Double underscores (e.g., __init__, __str__) are used for "dunder" methods (special built-in functions) and name mangling, but that's a separate topic from your question.

2. The "_" in OpenAI Gym's step() function #

First, let's clarify what env.step(action) returns: it gives a 4-value tuple:

1. observation: The environment's new state after taking the action
2. reward: The reward earned for the action
3. done: A boolean flag indicating if the episode has ended
4. info: Extra diagnostic data (like debug stats) that's often unnecessary for basic testing or training

In your example code:

import gym
env = gym.make('CartPole-v0')
for i_episode in range(20):
    observation = env.reset()
    for t in range(100):
        env.render()
        print(observation)
        action = env.action_space.sample()
        observation, reward, done, _ = env.step(action)  # The "_" here
        if done:
            print("Episode finished after {} timesteps".format(t+1))
            break

The _ is explicitly catching the info value without assigning it a meaningful name. Since you don't use info anywhere in your code, this is exactly the scenario the underscore is designed for.

3. Why use "_" instead of another name? #

• Better readability: It's a universal Python convention. Any developer reading your code will immediately understand that this value is unused, whereas a random name like temp or extra_data would require a moment to confirm it's not needed.
• Avoids IDE warnings: Most code editors (like VS Code, PyCharm) flag unused variables with warnings. Using _ tells the editor you intentionally aren't using this value, so no unnecessary alerts pop up.
• Cleaner code: It eliminates the need to invent a throwaway variable name, keeping your code concise and focused on the values you actually care about.

内容的提问来源于stack exchange，提问作者Yograj Singh Mandloi