从PyTorch官方的一篇教程说开去(2 - 源码)

先上图,上篇文章的运行结果,可以看到,算法在迭代了200来次左右达到人生巅峰,倒立摆金枪不倒,可以扛住连续200次操作。不幸的是,然后就出现了大幅度的回撤,每况愈下,在600次时候居然和100次的时候一个水平。

事实上,训练充满了随机性,也不乏非常漂亮的曲线,可以用来tree new bee,这也是AI领域很好水论文的体现吧。

下面两个图,分别来自,windows11本地运行 vs Colab云端运行。

呃,这个就是为啥G家主推的深度学习目前应用场景窄,被openAI狠揍的核心原因了 -

1)只能处理离散模型,数据量要求极高;

2)模型通用性差,不同的场景需要定制算法;

虽然G家多次宣称“霸权”,但事实上这个技术栈确实不适合解决通用问题。

开箱即食,以下为代码(单一python文件,windows11 + python 3.11.9 + GTX1080显卡) - 

import gymnasium as gym
import math
import random
import matplotlib
import matplotlib.pyplot as plt
from collections import namedtuple, deque
from itertools import count

import torch
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F

env = gym.make("CartPole-v1", render_mode="human")

# set up matplotlib
is_ipython = 'inline' in matplotlib.get_backend()
if is_ipython:
    from IPython import display

plt.ion()

# if GPU is to be used
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
##device = torch.device("cuda")


Transition = namedtuple('Transition',
                        ('state', 'action', 'next_state', 'reward'))


class ReplayMemory(object):

    def __init__(self, capacity):
        self.memory = deque([], maxlen=capacity)

    def push(self, *args):
        """Save a transition"""
        self.memory.append(Transition(*args))

    def sample(self, batch_size):
        return random.sample(self.memory, batch_size)

    def __len__(self):
        return len(self.memory)


class DQN(nn.Module):

    def __init__(self, n_observations, n_actions):
        super(DQN, self).__init__()
        self.layer1 = nn.Linear(n_observations, 128)
        self.layer2 = nn.Linear(128, 128)
        self.layer3 = nn.Linear(128, n_actions)

    # Called with either one element to determine next action, or a batch
    # during optimization. Returns tensor([[left0exp,right0exp]...]).
    def forward(self, x):
        x = F.relu(self.layer1(x))
        x = F.relu(self.layer2(x))
        return self.layer3(x)

# BATCH_SIZE is the number of transitions sampled from the replay buffer
# GAMMA is the discount factor as mentioned in the previous section
# EPS_START is the starting value of epsilon
# EPS_END is the final value of epsilon
# EPS_DECAY controls the rate of exponential decay of epsilon, higher means a slower decay
# TAU is the update rate of the target network
# LR is the learning rate of the ``AdamW`` optimizer
BATCH_SIZE = 128
GAMMA = 0.99
EPS_START = 0.9
EPS_END = 0.05
EPS_DECAY = 1000
TAU = 0.005
LR = 1e-4

# Get number of actions from gym action space
n_actions = env.action_space.n
# Get the number of state observations
state, info = env.reset()
n_observations = len(state)

policy_net = DQN(n_observations, n_actions).to(device)
target_net = DQN(n_observations, n_actions).to(device)
target_net.load_state_dict(policy_net.state_dict())

optimizer = optim.AdamW(policy_net.parameters(), lr=LR, amsgrad=True)
memory = ReplayMemory(10000)


steps_done = 0


def select_action(state):
    global steps_done
    sample = random.random()
    eps_threshold = EPS_END + (EPS_START - EPS_END) * \
        math.exp(-1. * steps_done / EPS_DECAY)
    steps_done += 1
    if sample > eps_threshold:
        with torch.no_grad():
            # t.max(1) will return the largest column value of each row.
            # second column on max result is index of where max element was
            # found, so we pick action with the larger expected reward.
            return policy_net(state).max(1)[1].view(1, 1)
    else:
        return torch.tensor([[env.action_space.sample()]], device=device, dtype=torch.long)


episode_durations = []


def plot_durations(show_result=False):
    plt.figure(1)
    durations_t = torch.tensor(episode_durations, dtype=torch.float)
    if show_result:
        plt.title('Result')
    else:
        plt.clf()
        plt.title('Training...')
    plt.xlabel('Episode')
    plt.ylabel('Duration')
    plt.plot(durations_t.numpy())
    # Take 100 episode averages and plot them too
    if len(durations_t) >= 100:
        means = durations_t.unfold(0, 100, 1).mean(1).view(-1)
        means = torch.cat((torch.zeros(99), means))
        plt.plot(means.numpy())

    plt.pause(0.001)  # pause a bit so that plots are updated
    if is_ipython:
        if not show_result:
            display.display(plt.gcf())
            display.clear_output(wait=True)
        else:
            display.display(plt.gcf())


def optimize_model():
    if len(memory) < BATCH_SIZE:
        return
    transitions = memory.sample(BATCH_SIZE)
    # Transpose the batch (see https://stackoverflow.com/a/19343/3343043 for
    # detailed explanation). This converts batch-array of Transitions
    # to Transition of batch-arrays.
    batch = Transition(*zip(*transitions))

    # Compute a mask of non-final states and concatenate the batch elements
    # (a final state would've been the one after which simulation ended)
    non_final_mask = torch.tensor(tuple(map(lambda s: s is not None,
                                          batch.next_state)), device=device, dtype=torch.bool)
    non_final_next_states = torch.cat([s for s in batch.next_state
                                                if s is not None])
    state_batch = torch.cat(batch.state)
    action_batch = torch.cat(batch.action)
    reward_batch = torch.cat(batch.reward)

    # Compute Q(s_t, a) - the model computes Q(s_t), then we select the
    # columns of actions taken. These are the actions which would've been taken
    # for each batch state according to policy_net
    state_action_values = policy_net(state_batch).gather(1, action_batch)

    # Compute V(s_{t+1}) for all next states.
    # Expected values of actions for non_final_next_states are computed based
    # on the "older" target_net; selecting their best reward with max(1)[0].
    # This is merged based on the mask, such that we'll have either the expected
    # state value or 0 in case the state was final.
    next_state_values = torch.zeros(BATCH_SIZE, device=device)
    with torch.no_grad():
        next_state_values[non_final_mask] = target_net(non_final_next_states).max(1)[0]
    # Compute the expected Q values
    expected_state_action_values = (next_state_values * GAMMA) + reward_batch

    # Compute Huber loss
    criterion = nn.SmoothL1Loss()
    loss = criterion(state_action_values, expected_state_action_values.unsqueeze(1))

    # Optimize the model
    optimizer.zero_grad()
    loss.backward()
    # In-place gradient clipping
    torch.nn.utils.clip_grad_value_(policy_net.parameters(), 100)
    optimizer.step()

num_episodes = 600

for i_episode in range(num_episodes):
    # Initialize the environment and get it's state
    state, info = env.reset()
    state = torch.tensor(state, dtype=torch.float32, device=device).unsqueeze(0)
    for t in count():
        action = select_action(state)
        observation, reward, terminated, truncated, _ = env.step(action.item())
        reward = torch.tensor([reward], device=device)
        done = terminated or truncated

        if terminated:
            next_state = None
        else:
            next_state = torch.tensor(observation, dtype=torch.float32, device=device).unsqueeze(0)

        # Store the transition in memory
        memory.push(state, action, next_state, reward)

        # Move to the next state
        state = next_state

        # Perform one step of the optimization (on the policy network)
        optimize_model()

        # Soft update of the target network's weights
        # θ′ ← τ θ + (1 −τ )θ′
        target_net_state_dict = target_net.state_dict()
        policy_net_state_dict = policy_net.state_dict()
        for key in policy_net_state_dict:
            target_net_state_dict[key] = policy_net_state_dict[key]*TAU + target_net_state_dict[key]*(1-TAU)
        target_net.load_state_dict(target_net_state_dict)

        if done:
            episode_durations.append(t + 1)
            plot_durations()
            break

print('Complete')
plot_durations(show_result=True)
plt.ioff()
plt.show()

迭代次数是600,如果是使用CPU而不是GPU的话,建议设置在50以内,否则你懂的... ...

前置条件是安装老黄家的Cuda,以及准备好python环境(cuda暂不支持python 3.12),安装好需要的库,需要的可以看我此前的博文。

您的进步和反馈是我最大的动力,小伙伴来个三连呗!共勉。

相关推荐

最近更新

  1. docker php8.1+nginx base 镜像 dockerfile 配置

    2024-07-22 02:48:04       95 阅读
  2. Could not load dynamic library ‘cudart64_100.dll‘

    2024-07-22 02:48:04       103 阅读
  3. 在Django里面运行非项目文件

    2024-07-22 02:48:04       84 阅读
  4. Python语言-面向对象

    2024-07-22 02:48:04       93 阅读

热门阅读

  1. Xcode代码静态分析:构建无缺陷代码的秘诀

    2024-07-22 02:48:04       25 阅读
  2. 半监督学习 原理与代码实例讲解

    2024-07-22 02:48:04       24 阅读
  3. 初学Mybatis之 CRUD 增删改查

    2024-07-22 02:48:04       22 阅读
  4. TensorFlow 的基本概念和使用场景

    2024-07-22 02:48:04       27 阅读
  5. Windows图形界面(GUI)-DLG-C/C++ - 列表视图(ListView)

    2024-07-22 02:48:04       27 阅读
  6. 大规模语言模型从理论到实践 高效模型微调

    2024-07-22 02:48:04       25 阅读
  7. [11 Qt CMake开发]

    2024-07-22 02:48:04       25 阅读
  8. 显示隐式-》初始化 & 类型转换

    2024-07-22 02:48:04       24 阅读
  9. ReentrantLock源码

    2024-07-22 02:48:04       25 阅读
  10. 向量数据库Milvus

    2024-07-22 02:48:04       30 阅读
  11. 在 ASP.NET Core Web API 中实现审计跟踪

    2024-07-22 02:48:04       21 阅读