＜从零开始pytorch＞:03-torch构建模型实现鸢尾花数据分类

Mnist数据集进行分类

主要内容：
数据读取 – 路径设置、数据加载
自定义模型MyMnist_NN
自定义损失函数loss_func
自定义fit
[详情看目录]

# 1.读取Mnist数据集
%matplotlib inline
from pathlib import Path
import requests
from matplotlib import pyplot
import numpy as np

1.获取数据

• 路径设置和怎样加载数据要多回顾!!

1.1 路径设置

DATA_PATH = Path("data")
PATH = DATA_PATH / "mnist"

PATH.mkdir(parents=True, exist_ok=True)

URL = "http://deeplearning.net/data/mnist/"
FILENAME = "mnist.pkl.gz"

if not (PATH / FILENAME).exists():
        content = requests.get(URL + FILENAME).content
        (PATH / FILENAME).open("wb").write(content)

1.2 加载读取数据

import pickle
import gzip

with gzip.open((PATH / FILENAME).as_posix(), "rb") as f:
        ((x_train, y_train), (x_valid, y_valid), _) = pickle.load(f, encoding="latin-1")

1.3 展示数据

# x_train[i]是一个784维的向量,将其转换为28 * 28的二维矩阵显示出来,是图像形式
# imshow()可以传入二维矩阵(M,N) / 三维(M,N,3) /三维(M.N,4)
pyplot.imshow(x_train[1].reshape((28, 28)), cmap="gray") 
print(x_train.shape)

(50000, 784)

# 此时x_train是ndarray数组
x_train 

array([[0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.],
       ...,
       [0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.]], dtype=float32)

• 将其转换为tensor后才能进行建模训练

import torch 
x_train,y_train,x_valid,y_valid = map(
    torch.tensor,(x_train,y_train,x_valid,y_valid)
)

x_train # 此时是tensor

tensor([[0., 0., 0.,  ..., 0., 0., 0.],
        [0., 0., 0.,  ..., 0., 0., 0.],
        [0., 0., 0.,  ..., 0., 0., 0.],
        ...,
        [0., 0., 0.,  ..., 0., 0., 0.],
        [0., 0., 0.,  ..., 0., 0., 0.],
        [0., 0., 0.,  ..., 0., 0., 0.]])

# 查看下标签值的范围最大值和最小值的函数!第一次见
y_train.min(),y_train.max()

(tensor(0), tensor(9))

2.torch.nn.functional的层和函数功能展示

import torch