![](https://img-blog.csdnimg.cn/20201014180756927.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
Pytorch框架教程
liuxiao789
快不了,那就慢一点,稳一点
展开
-
模型参数统计和可视化工具
原创 2022-04-01 10:04:25 · 253 阅读 · 0 评论 -
PyTorch最基本的模型框架
import torch import torch.nn as nn import matplotlib.pyplot as plt import numpy as np torch.manual_seed(10) # ============================ step 1/5 生成数据 ============================ sample_nums = 100 mean_value = 1.7 bias = 1 n_data = torch.ones(sample_.原创 2021-04-27 11:12:15 · 354 阅读 · 0 评论 -
2-1 transform1--数据增强
import os import numpy as np import torch import random from torch.utils.data import DataLoader import torchvision.transforms as transforms from tools.my_dataset import RMBDataset from PIL import Image from matplotlib import pyplot as plt def set_seed(s.原创 2020-10-21 21:28:07 · 564 阅读 · 0 评论 -
Anaconda安装命令
创建anaconda虚拟环境,一个项目对应一个虚拟环境。其中base是真实的环境。 1、Anaconda 安装成功。 接下来需要设置 Anaconda 仓库镜像,因为默认连接的是国外镜像地址,下载速度比较慢, 我们把镜像地址改为清华大学开源软件镜像站,Anaconda Prompt 窗口输入: conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ conda config --set sho原创 2020-10-21 10:48:22 · 1459 阅读 · 1 评论 -
3.4 池化、线性
import os import torch import random import numpy as np import torchvision import torch.nn as nn from torchvision import transforms from matplotlib import pyplot as plt from PIL import Image from tools.common_tools import transform_invert, set_seed set_s.原创 2020-10-13 21:28:55 · 185 阅读 · 0 评论 -
3.3 卷积层实现
import os import torch.nn as nn from PIL import Image from torchvision import transforms from matplotlib import pyplot as plt from tools.common_tools import transform_invert, set_seed set_seed(2) # 设置随机种子---》权值初始化1,2,3 ''' eg: 3个不同的卷积核权值:不同的特征检测器--》出现3个.原创 2020-10-13 21:10:11 · 575 阅读 · 0 评论 -
3.2 模型容器与AlexNet构建
# -*- coding: utf-8 -*- import torch import torchvision import torch.nn as nn from collections import OrderedDict # ============================ 【1】Sequential class LeNetSequential(nn.Module): def __init__(self, classes): super(LeNetSequent.原创 2020-10-13 20:27:39 · 97 阅读 · 0 评论 -
3.1模型创建步骤与nn.Module
import os import random import numpy as np import torch import torch.nn as nn from torch.utils.data import DataLoader import torchvision.transforms as transforms import torch.optim as optim from matplotlib import pyplot as plt from model.lenet import LeNe.原创 2020-10-12 18:11:36 · 343 阅读 · 0 评论 -
GPU的使用
# -*- coding: utf-8 -*- """ # @file name : cuda_methods.py # @author : TingsongYu https://github.com/TingsongYu # @date : 2019-11-11 # @brief : 数据迁移至cuda的方法 """ import torch import torch.nn as nn device = torch.device("cuda:0" if torch.cud.原创 2020-10-10 19:02:47 · 491 阅读 · 1 评论 -
模型加载与保存 和 迁移学习
# -*- coding: utf-8 -*- """ # @file name : model_save.py # @author : TingsongYu https://github.com/TingsongYu # @date : 2019-11-04 # @brief : 模型的保存 """ import torch import numpy as np import torch.nn as nn from tools.common_tools import se.原创 2020-10-10 18:59:58 · 536 阅读 · 0 评论 -
Tensorboard2 使用
1、SummaryWriter(类) import numpy as np import matplotlib.pyplot as plt from torch.utils.tensorboard import SummaryWriter from tools.common_tools import set_seed set_seed(1) # 设置随机种子 # ----------------------------------- 0 SummaryWriter ----------------原创 2020-09-28 15:16:05 · 180 阅读 · 0 评论 -
TensorBoard1 相关知识
一、安装(可视化训练过程,用来监控训练是否状态良好)----可视化需要借助Tensorboard {支持标量、图像、文本、音频、视频、Eenbedding等可视化} 1、安装Tensorboard,同时运行时还提示没有模块past,需要安装future 2、在pycharm---》terminal---》cd 进入当前文件夹: Tensorboard --logdir=./runs---》里面有envents文件 TensorBoard 2.3.0 at http://localhost:6006/ .原创 2020-09-28 14:21:53 · 108 阅读 · 0 评论 -
3、DataSet 与 DataLoader
# -*- coding: utf-8 -*- """ # @file name : train_lenet.py # @author : tingsongyu # @date : 2019-09-07 10:08:00 # @brief : 人民币分类模型训练 """ import os import random import numpy as np import torch import torch.nn as nn from torch.utils.data imp.原创 2020-09-16 15:58:56 · 580 阅读 · 0 评论 -
2、逻辑回归
# -*- coding: utf-8 -*- import torch import torch.nn as nn import matplotlib.pyplot as plt import numpy as np torch.manual_seed(10) # ============================ step 1/5 生成数据 ============================ sample_nums = 100 mean_value = 1.7 bias = 1 n_d.原创 2020-09-10 11:49:32 · 99 阅读 · 0 评论 -
1、PyTorch张量-线性回归-计算图-autograd
1 创建张量的三种方法 # -*- coding:utf-8 -*- import torch import numpy as np torch.manual_seed(1) # =============================== exmaple 1 =============================== # 【1】通过torch.tensor创建张量 # # flag = True flag = False if flag: arr = np.ones((3, 3))原创 2020-09-07 19:21:39 · 229 阅读 · 0 评论