推荐系统召回双塔模型模型训练耗时对比测试

Thomas2143

于 2025-07-30 00:44:26 发布

阅读量38

点赞数

CC 4.0 BY-SA版权

分类专栏：总结文章标签： python

本文链接：https://blog.csdn.net/qq_35515661/article/details/149761459

总结专栏收录该内容

793 篇文章 ¥99.90 ¥299.90

订阅专栏

结果

代码

import torch
import torch.nn as nn
import torch.optim as optim
import numpy as np
import time
start_time = time.time()  # 添加：记录开始时间

# 检查CUDA
print(f"CUDA可用: {torch.cuda.is_available()}")
# device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device = "cpu"
print(f"使用设备: {device}")

# 简单双塔模型
class SimpleTwoTower(nn.Module):
    def __init__(self, feature_dim=10, embedding_dim=32):
        super().__init__()
        self.user_tower = nn.Sequential(
            nn.Linear(feature_dim, 64),
            nn.ReLU(),
            nn.Linear(64, embedding_dim)
        )
        self.item_tower =

了解本专栏