ConvLSTM的原理说明可参见另一博客-【ConvLSTM第一期】ConvLSTM原理。

1 准备工作：python库包安装

1.1 安装必要库

pip install torch torchvision matplotlib numpy

案例说明：模拟视频帧的时序建模

🎯 目标：给定一个人工生成的动态图像序列（例如移动的方块），使用 ConvLSTM 对其进行建模，输出预测结果，并查看输出的维度和特征变化。

ConvLSTM概述

ConvLSTM 的基本结构，包括：

• ConvLSTMCell：实现了一个时间步的 ConvLSTM 单元，类似于一个“时刻”的神经元。
• ConvLSTM：实现了多层ConvLSTM结构，能够处理一整个时间序列的视频帧数据。

损失函数说明

MSE（均方误差） 衡量预测值和真实值之间的平均平方差。

关于训练终止条件：
可以根据 MSE是否达到某个阈值（如 < 0.001）提前终止训练，这是所谓的 “Early Stopping（提前停止）策略”。

（python全代码）

MSE损失函数曲线如下：可知MSE一直在下降，虽然存在振荡

前9帧图像及预测的第十帧图像得到的动图如下：

python完整代码如下：

import os
import torch
import torch.nn as nn
import torch.optim as optim
import matplotlib.pyplot as plt
import numpy as np
from PIL import Image# 设置字体
plt.rcParams['font.family'] = 'Times New Roman'# 创建保存图像目录
os.makedirs("./Figures", exist_ok=True)# 设置设备
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")# ====================================
# 一、ConvLSTM 模型结构
# ====================================class ConvLSTMCell(nn.Module):def __init__(self, input_channels, hidden_channels, kernel_size, bias=True):super(ConvLSTMCell, self).__init__()padding = kernel_size // 2self.input_channels = input_channelsself.hidden_channels = hidden_channelsself.conv = nn.Conv2d(input_channels + hidden_channels, 4 * hidden_channels, kernel_size, padding=padding, bias=bias)def forward(self, x, h_prev, c_prev):combined = torch.cat([x, h_prev], dim=1)conv_output = self.conv(combined)cc_i, cc_f, cc_o, cc_g = torch.chunk(conv_output, 4, dim=1)i = torch.sigmoid(cc_i)f = torch.sigmoid(cc_f)o = torch.sigmoid(cc_o)g = torch.tanh(cc_g)c = f * c_prev + i * gh = o * torch.tanh(c)return h, cclass ConvLSTM(nn.Module):def __init__(self, input_channels, hidden_channels, kernel_size, num_layers):super(ConvLSTM, self).__init__()self.num_layers = num_layerslayers = []for i in range(num_layers):in_channels = input_channels if i == 0 else hidden_channelslayers.append(ConvLSTMCell(in_channels, hidden_channels, kernel_size))self.layers = nn.ModuleList(layers)def forward(self, input_seq):b, t, c, h, w = input_seq.size()h_t = [torch.zeros(b, layer.hidden_channels, h, w).to(input_seq.device) for layer in self.layers]c_t = [torch.zeros(b, layer.hidden_channels, h, w).to(input_seq.device) for layer in self.layers]for time in range(t):x = input_seq[:, time]for i, layer in enumerate(self.layers):h_t[i], c_t[i] = layer(x, h_t[i], c_t[i])x = h_t[i]return h_t[-1]  # 返回最后一层最后一帧的隐藏状态# ====================================
# 二、生成移动方块序列数据
# ====================================def generate_moving_square_sequence(batch_size, time_steps, height, width):data = torch.zeros((batch_size, time_steps, 1, height, width))for b in range(batch_size):dx = np.random.randint(1, 3)dy = np.random.randint(1, 3)x = np.random.randint(0, width - 6)y = np.random.randint(0, height - 6)for t in range(time_steps):data[b, t, 0, y:y+5, x:x+5] = 1.0x = (x + dx) % (width - 5)y = (y + dy) % (height - 5)return data# ====================================
# 三、模型、损失、优化器
# ====================================class ConvLSTM_Predictor(nn.Module):def __init__(self):super().__init__()self.convlstm = ConvLSTM(input_channels=1, hidden_channels=16, kernel_size=3, num_layers=1)self.decoder = nn.Conv2d(in_channels=16, out_channels=1, kernel_size=1)def forward(self, input_seq):hidden = self.convlstm(input_seq)pred = self.decoder(hidden)return predmodel = ConvLSTM_Predictor().to(device)
criterion = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr=1e-3)# ====================================
# 四、训练过程
# ====================================mse_list = []
max_epochs = 100
mse_threshold = 0.001
height, width = 64, 64for epoch in range(max_epochs):model.train()seq = generate_moving_square_sequence(8, 10, height, width).to(device)input_seq = seq[:, :9]target_frame = seq[:, 9, 0].unsqueeze(1)optimizer.zero_grad()output = model(input_seq)loss = criterion(output, target_frame)loss.backward()optimizer.step()mse = loss.item()mse_list.append(mse)print(f"Epoch {epoch+1}/{max_epochs}, MSE: {mse:.6f}")# 提前停止条件if mse < mse_threshold:print(f"✅ 提前停止：MSE 已达到阈值 {mse_threshold}")break# ====================================
# 五、测试与可视化结果
# ====================================model.eval()
with torch.no_grad():test_seq = generate_moving_square_sequence(1, 10, height, width).to(device)input_seq = test_seq[:, :9]true_frame = test_seq[:, 9, 0]pred_frame = model(input_seq)[0, 0].cpu().numpy()# 保存输入帧
for t in range(9):frame = input_seq[0, t, 0].cpu().numpy()plt.imshow(frame, cmap='gray')plt.title(f"Input Frame t={t}")plt.colorbar()plt.savefig(f"./Figures/input_frame_{t}.png")plt.close()# 保存 Ground Truth
plt.imshow(true_frame[0].cpu().numpy(), cmap='gray')
plt.title("Ground Truth Frame t=9")
plt.colorbar()
plt.savefig("./Figures/ground_truth_t9.png")
plt.close()# 保存预测帧
plt.imshow(pred_frame, cmap='gray')
plt.title("Predicted Frame t=9")
plt.colorbar()
plt.savefig("./Figures/predicted_t9.png")
plt.close()# 保存 MSE 曲线图
plt.plot(mse_list)
plt.title("Training MSE Loss")
plt.xlabel("Epoch")
plt.ylabel("MSE")
plt.grid(True)
plt.savefig("./Figures/mse_curve.png")
plt.close()# ---------------- 生成动图 ----------------frames = []# 添加前9帧输入
for t in range(9):img = Image.open(f"./Figures/input_frame_{t}.png")frames.append(img.copy())# 添加预测帧
img = Image.open("./Figures/predicted_t9.png")
frames.append(img.copy())# 保存动图
frames[0].save("./Figures/sequence.gif", save_all=True, append_images=frames[1:], duration=500, loop=0)
print("✅ 所有图像和动图已保存至 ./Figures 文件夹")

参考