【数据准备和特征工程】1-1感知文件中的数据

1.1.1 CSV文件

path = "../data/cities.csv" # 随便找的一个csv文件

import csv
f = open(path)
data = csv.reader(f)    # ①
for line in data:
    print(line)

['name', ' area', ' population', ' longd', ' latd']
['Nanjing', ' 6582.31', ' 8004680', ' 118.78', ' 32.04']
['Wuxi', ' 4787.61', ' 6372624', ' 120.29', ' 31.59']
['Xuzhou', ' 11764.88', ' 8580500', ' 117.2', ' 34.26']
['Changzhou', ' 4384.57', ' 4591972', ' 119.95', ' 31.79']
['Soochow', ' 8488.42', ' 10465994', ' 120.62', ' 31.32']
['Nantong', ' 8001', ' 7282835', ' 120.86', ' 32.01']
['Lianyungang', ' 7615.29', ' 4393914', ' 119.16', ' 34.59']
['Huaian', ' 9949.97', ' 4799889', ' 119.15', ' 33.5']
['Yancheng', ' 16972.42', ' 7260240', ' 120.13', ' 33.38']
['Yangzhou', ' 6591.21', ' 4459760', ' 119.42', ' 32.39']
['Zhenjiang', ' 3840.32', ' 3113384', ' 119.44', ' 32.2']
['Taizhou', ' 5787.26', ' 4618558', ' 119.9', ' 32.49']
['Suqian', ' 8555', ' 4715553', ' 118.3', ' 33.96']

import pandas as pd
# 使用pandas库读取csv文件
df = pd.read_csv(path)
df

	name	area	population	longd	latd
0	Nanjing	6582.31	8004680	118.78	32.04
1	Wuxi	4787.61	6372624	120.29	31.59
2	Xuzhou	11764.88	8580500	117.20	34.26
3	Changzhou	4384.57	4591972	119.95	31.79
4	Soochow	8488.42	10465994	120.62	31.32
5	Nantong	8001.00	7282835	120.86	32.01
6	Lianyungang	7615.29	4393914	119.16	34.59
7	Huaian	9949.97	4799889	119.15	33.50
8	Yancheng	16972.42	7260240	120.13	33.38
9	Yangzhou	6591.21	4459760	119.42	32.39
10	Zhenjiang	3840.32	3113384	119.44	32.20
11	Taizhou	5787.26	4618558	119.90	32.49
12	Suqian	8555.00	4715553	118.30	33.96

1.1.2 Excel文件

path = "../data/多分类鸢尾花.xlsx"

flower_df = pd.read_excel(path)
flower_df.head()

	萼片长	萼片宽	花瓣长	花瓣宽	种类
0	5.0	3.3	1.4	0.2	山鸢尾
1	6.7	3.1	4.4	1.4	变色鸢尾
2	6.3	2.7	4.9	1.8	维吉尼亚鸢尾
3	4.4	2.9	1.4	0.2	山鸢尾
4	7.7	2.6	6.9	2.3	维吉尼亚鸢尾

1.1.3 图形文件

# 使用pillow库
from PIL import Image    # ○16

color_image = Image.open("../data/starryNight.jpg")    # ○17
color_image

​

​

从url中读取

Image.open的参数可以是文件路径fp，也可以的字节数组bytes，下面演示从图床中读取文件并展示

import requests
from io import BytesIO
req = requests.get("https://pic-1257412153.cos.ap-nanjing.myqcloud.com/beautiful/大炎，赦封神明！_87744911.jpg")
if req.status_code != 200:
    print("图片请求错误，请尝试加入header")

byte_img = BytesIO(req.content)
img = Image.open(byte_img)
img

​

​

使用opencv读取图片

读取方式：imread(filename, flag)

cv::ImreadModes {
cv::IMREAD_UNCHANGED = -1,
cv::IMREAD_GRAYSCALE = 0,
cv::IMREAD_COLOR = 1,
cv::IMREAD_ANYDEPTH = 2,
cv::IMREAD_ANYCOLOR = 4,
cv::IMREAD_LOAD_GDAL = 8,
cv::IMREAD_REDUCED_GRAYSCALE_2 = 16,
cv::IMREAD_REDUCED_COLOR_2 = 17,
cv::IMREAD_REDUCED_GRAYSCALE_4 = 32,
cv::IMREAD_REDUCED_COLOR_4 = 33,
cv::IMREAD_REDUCED_GRAYSCALE_8 = 64,
cv::IMREAD_REDUCED_COLOR_8 = 65,
cv::IMREAD_IGNORE_ORIENTATION = 128
}

# 使用opencv库查看图片
import cv2
# 得到的是numpy数组
img = cv2.imread("../data/starryNight.jpg", -1) # 不改变
img

array([[[ 53,  90, 104],
        [ 31,  34,  39],
        [ 54,  39,  36],
        ...,
        [124, 161, 183],
        [138, 170, 189],
        [122, 150, 167]],

       [[ 65,  89, 101],
        [ 32,  20,  26],
        [ 54,  25,  21],
        ...,
        [138, 172, 195],
        [134, 166, 185],
        [132, 159, 179]],

       [[ 79,  85,  96],
        [ 53,  26,  30],
        [ 84,  37,  33],
        ...,
        [121, 156, 176],
        [131, 162, 183],
        [134, 164, 183]],

       ...,

       [[ 90, 149, 164],
        [121, 128, 148],
        [110, 129, 144],
        ...,
        [134, 171, 193],
        [116, 154, 178],
        [140, 177, 203]],

       [[ 99, 174, 183],
        [ 86, 105, 120],
        [121, 146, 156],
        ...,
        [114, 148, 171],
        [103, 139, 163],
        [144, 179, 205]],

       [[ 69, 149, 160],
        [102, 125, 140],
        [136, 159, 167],
        ...,
        [ 99, 133, 156],
        [102, 136, 159],
        [142, 176, 200]]], dtype=uint8)

import matplotlib.pyplot as plt
%matplotlib inline
# %config InlineBackend.figure_format = 'svg'

plt.imshow(img)
plt.xticks([]), plt.yticks([])

(([], []), ([], []))

​

注意到图片并不是我们想要的效果

这是因为：opencv的接口使用BGR，而matplotlib.pyplot 则是RGB模式

b,g,r = cv2.split(img)
img2 = cv2.merge([r,g,b])

plt.imshow(img2)
plt.xticks([]), plt.yticks([])
plt.show()

​