动手学深度学习PyTorch版-梯度消失、梯度爆炸、kaggle房价预测

最新推荐文章于 2024-05-11 13:10:08 发布

hongyesuifeng

最新推荐文章于 2024-05-11 13:10:08 发布

阅读量206

点赞数

分类专栏： python 机器学习

本文链接：https://blog.csdn.net/hongyesuifeng/article/details/104367117

版权

这篇博客探讨了深度学习中常见的梯度消失和梯度爆炸问题，并结合kaggle房价预测任务，介绍了数据预处理、模型训练、K折交叉验证以及模型选择和预测提交的全过程。

摘要由CSDN通过智能技术生成

梯度消失、梯度爆炸、kaggle房价预测

kaggle房价预测

%matplotlib inline
import torch
import torch.nn as nn
import numpy as np
import pandas as pd
import sys
sys.path.append("/home/kesci/input")
import d2lzh1981 as d2l
print(torch.__version__)
torch.set_default_tensor_type(torch.FloatTensor)

test_data = pd.read_csv("/home/kesci/input/houseprices2807/house-prices-advanced-regression-techniques/test.csv")
train_data = pd.read_csv("/home/kesci/input/houseprices2807/house-prices-advanced-regression-techniques/train.csv")

all_features = pd.concat((train_data.iloc[:, 1:-1], test_data.iloc[:, 1:]))

预处理特征

numeric_features = all_features.dtypes[all_features.dtypes != 'object'].index
all_features[numeric_features] = all_features[numeric_features].apply(
    lambda x: (x - x.mean()) / (x.std()))
# 标准化后，每个数值特征的均值变为0，所以可以直接用0来替换缺失值
all_features[numeric_features] = all_features[numeric_features].fillna(0)

# dummy_na=True将缺失值也当作合法的特征值并为其创建指示特征
all_features = pd.get_dummies(all_features, dummy_na=True)
all_features.shape

n_train = train_data.shape[0]
train_features = torch.tensor(all_features[:n_train].values, dtype=torch.float)
test_features = torch.tensor(all_features[n_train:].values