rl-plotter 1.0.3

A plotter for reinforcement learning

rl-plotter

This is a simple tool which can plot learning curves easily for reinforcement learning.

Installation

from PIP

pip install rl_plotter

from source

python3 setup.py install

Examples

First, add a logger in your code (for example: DQN):

from rl_plotter.logger import Logger

def train(name):
    dqn = DQN()
    logger = Logger(name, env_name='PongNoFrameskip-v4', use_tensorboard=False)

    while True:
        s = env.reset()
        while True:
            total_step = logger.add_step()
            a = dqn.select_action(s, EPSILON)
            s_, r, done, info = env.step(a)

            dqn.store_transition(s, a, r, s_)
            episode_reward += r

            if dqn.replay_memory.memory_counter > REPLAY_MEMORY_SIZE:
                loss = dqn.learn()
                logger.add_loss(loss.cpu().item())
            if done:
                break
            s = s_
        logger.add_episode()
        logger.add_reward(episode_reward, freq=10)
    logger.finish()

After the training or when you are training your agent, you can plot the learning curves in this way:

python -m rl_plotter.plotter

for help use:

python -m rl_plotter.plotter --help

The learning curves looks like this:

And you can custom the style of your curves by modifiying `rl_plotter.plotter`

To Do

• reinforcement learning plot tools
• timestamp features
• history experiment data plot tools
• ~~basic data plot tools锛坕ncluding ML-Loss plot锛墌~
• dynamic plot tools