???????? ?????? ????

????????? ?????? ????????

?????: ?????? ???????? ???? ????????? ??????????? ??????. ?????? ?????? ???? ?????? ?? ??????? ???????? ?????? ??????? 8 ?????? ?????? ???????. ????? ????????? ?????????? ??????? ?????? ????????? ?????? ?????? ????????. ????????? ????????? ??????? ??????????????? ??? ?????????????? knn ????????? ??? ????? ????? ?????? ?????? ??????? ????????? ??????? ????????? ?????????. ????????? ?????? ???? ?????? 8 ????? ?????? ????????? ???????? ?????? ?????? ??????.

??????? ??????????:

Anaconda3

???? ?????? ???????????: 0-9 ?? ?????? ?????????? ???????? ????????? ???? ?????????? ????????? 5000 ???? ?????.

1. ????????? ??????? ?????? ????????? ????? ???????? ??????? ??????????? ????????? ?????????.

1 import matplotlib.pyplot as plt
2 import numpy as np
3 import pandas as pd

2.????? ???? ???????? ??????? ???? ???????. 

???????? ?????????????? ??????? ?????? ???? ???????? ??? ?????? 28 ??????? ?????? ??? ???? ????? ??? ??????? numpy ?????? ??????? ?????? ????????????? ?????????.

1 img_arr = plt.imread('./data/3/3_33.bmp')
2 img_arr.shape
3 plt.imshow(img_arr)

1- ?????

3. ???????? ??????????.

3.1 ????? ???? ???????????? ???????????.

feature = [] # ??? ??? ???? ?????? ???????
target = [] # ??? ??? ???? ???????? ????

????? ??? ?????? ???? ???? ?????? ???????????? feature ???? ???????? ?????? ?????? ???????????? target ???? ???????? ????????.

???????? 10 ????? ???????? ?????? ????? ?????????? ??????????????? for ???????? ?????? ????? ?????? ????? ???????????? feature ????????? ?????? ?????? ?????? ???????????? target ???? ???????? ???????. ?????? ??? ??????????? ??? ???? ????????? ????.

 1 feature = [] # ??? ??? ???? numpy ???????
 2 target = [] # ??? ??? ???? ???????? ????
 3 
 4 for i in range(10):
 5     for j in range(1,501):
 6         imgPath = './data/'+str(i)+'/'+str(i)+'_'+ str(j)+ '.bmp'
 7         img_arr = plt.imread(imgPath)
 8         feature.append(img_arr)
 9         target.append(i)

??????? ???????????? ????? ??????????? ??? ???????? ?????? ???????? ????????. ??????? feature ???? ????????? ?????? 5000 ???? ???????? ??? ??????? numpy ??????? ????????? ?????? ?? target ???? ???????? ??? ?????????? ??????????? ?????? ????????? ?????????? ?????? ??? ???????. ????? ??????????? ??? ?????? ??????????? ????? ?????? ??????. ?????? ???????? ????? 0 ????? ??????? ???????? ?????? ?????? ?????????? ???????? ????? 0 ????? ???? ?????? ????? ????.

?????????? ????????????? ????????? ????????? ??????? ????? ???? ?????? ??? ??????? ?????????? ?????? ???????? ???????. ????? shape. ?????? ???????? ???????.

feature = np.array(feature)
target = np.array(target)
feature.shape 

feature.shape   ??? ???????? (5000, 28, 28) ????? ?????. ????? ???? ??? ??????? ?????? ??????? ?????? ?????????. ??????? nkk ??????????? ??????????? ??? ??????? ?????? ??????? ????? ??????? ????????? ??? ??????? ?????? ???????? ??? ??????? ?????? ????????? ????????????. ?????? feature ???? ?????????? ??? ???? ???? ???????? ??? ??????? ?????? ????????? ???? ??????? ???????? ????????????.

1 feature = feature.reshape((5000),784)

???? feature.shape ?????? ??? ???????? ??????? ?????? ???????? ?????? ??????? ???????? ???????????? ????????. ????????? ???????? (5000, 784) ??? ?????????. ????? ??? ??????? ???????? ?????? ???????? ?????. (5000 ????? ??? ??? 784 ?????? ????? ? ?? ?????? ????? ???????????? ?????? ????? ?????? (5000, 28, 28) ????? ??? ??????? ?????????? 5000 ??????? ??? ???? ???????? 28 ??????? 28 ??????? 784 ??? ???? ???????? ??????? ???? ??????? ??? ??????? ??????? ??????????? ??? ??????? ??????? ?????????? ???????????. ????? ???? ???? ????????? ??????? ?????? ?????? ??????? ??????????.)

??????? ???? ???????????? ???????? ?????? ??????????.

3.2 ???? ?????? ???????????? ??? ????? ?????? ???? ???????? ???????? ???????????? ???? ???? ??????? ?????????? ????????? ??????? ????????????. 

???? ?????? ??????????? ??? ?????? ?????? ???????????? ????-???? ??? ????? ??????? ????? ??????? ??? ???????? ????????? ???? ??????. ?????? ?????? ??? ??????? ???????? ?????? ???????? ??????? ??????? ??????? ????? ????????? ????????. ??? ?????? ??????? ????? ????????? ????????? ??? ?????? ???? ?????? ???? ???? ?????? ????? ?????? ???? ????? ???????? ??????. ??? ????????? ????????? ???????? ??? ???????? ????????? ????????? ??? ????????? ???????????? ?????????? ?????? ??????????.

1 np.random.seed(10) # ????? ????????????
2 np.random.shuffle(feature)
3 np.random.seed(10) # ????? ?????????????? ???????? ?????????
4 np.random.shuffle(target)

??? ????????? ??? ???????? ??????????? ????? ??????????? ??? ?????? ?????????????? ?????? ??????????? ?????? ???? ????? ??? ????? ?? ??????? ?????????????? ????????????. ??????? ??? ??????? ???????????? ????? ??? ????? ?????? ?????????? ??????????. ?????? ???????? ?????????? 777 - ?????? 5 ????? ??????? ?????? ?????? ?????? ????????? 777 - ????????? ??? 5 ????? ???? ????? ????.

4. ??? ???????? ???????? ??? ??? ???????? ??? ????? ???????.

??????? 5000 ??? ???? ?????????? 4980 ??????? ???????? ???????????? ??????????? ?? ?????? 20 ??????? ??????? ????????????. ?????? ??? ??? ?????????? ???????? ?????? ??? ????? ???????.

1 # ????????? ??????? ??????? ????? ??????.
2 x_train = feature[:4980]
3 y_train = target[:4980]
4 # ??????? ??????? ??????? ????? ??????.
5 x_test = feature[4980:]
6 y_test = target[4980:]
7 
8 x_train.shape # ????????? ??????? ?????? ????????????? ???? 4980 ????
9 x_test.shape # ??????? ??????? ?????? ????????????? ???? ???? 20 ???? ?????? ????.

3.3 ???????? ??????????

1 from sklearn.neighbors import KNeighborsClassifier # ??????? ??????? ??????
2 knn = KNeighborsClassifier(n_neighbors=17) # ?????? ????????????? ?????? ???????? ???? ??????? ??????????? 17 ??? ????????.
3 knn.fit(x_train,y_train) # ???????? ??????????

????????? ????????????? ????? ???????????? ?????? ????????. ??? ???????? ????????????? ???????? 

KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',
metric_params=None, n_jobs=None, n_neighbors=17, p=2,
weights='uniform')

3.4 ????????? ???????? ????????

???? ????? ??????????? ?????? ????? ????? ?????? ?????????????? ????????? ???????? ????????? ????????? ???????? ???????.

1 print('???????? ????????:',knn.predict(x_test))
2 print('?????? ???? :',y_test)

?????????? ?????? ???????? ????????????

????????? ????????: [6 6 4 4 0 8 5 8 8 2 4 3 3 9 4 1 1 2 9 2]
?????? ???? : [6 6 4 4 0 8 5 8 2 2 4 3 3 9 4 2 6 2 9 2]

???????? ???????? ??? ?????? ?????? ????????? ?????? ?????????????? ??? ????????? ??? ????? ????????? ????? ????????????? ?????? ????????? ????? ???? ????????? ?????????.

2-?????

4. ????????? ???????? ??????

???????? ??????????? ???????? ?????????? ?????? ??????? ?? ????????? ????? ??????????? ???? ??? ????? ??????????? ????? ???????? ???????? ?????? ??? ??????? ???????????? ????? ?????????? ?????? ???? ???? ??????????????? ??????.

1 from sklearn.externals import joblib # ???????? ???????? ??????? ?????? ??????
2 joblib.dump(knn,'./knn_dwdar.m') # ???????? ??????? ??????

5. ???????? ?????????

???? ?????????????? ??????? ?????????? ???????? ?????? ??????? ???????. ????????? ????? ????????? ???????? ???????? ??????? ?????.

5.1 ?? ????? ?? ??????? ???????? ???????? ????????? ??????? ?????? ????????? ????????? ?????? ?????? ???????.

3- ?????

???? ??????? python ??? ????? ???????? ?????????? ?????? ??????? ??????.

1 ex_img_arr = plt.imread('./數字.jpg')
2 plt.imshow(ex_img_arr)
3 
4 # ????????? ?????? ??????? ??????????.
5 ex_img_arr.shape
6 # ???????? (241, 257, 3) ?????? ?????? ????? ??? ???????? ??? ??????? ???? ?????? ?????????????? ???????. 

5.2 ????????? 2 ????? ?????? ????? ??????

1 img_two_arr = ex_img_arr[0:75,130:185,:]
2 plt.imshow(img_two_arr) # ??????? ???? ????
3 
4 # ????????? ?????? ??????? ?????????.
5 img_two_arr.shape
6 # ?????????? ????????? ??????: (75, 55, 3) 

5.3 ????????? ???????? ????????? ??? ??????? ?????? ????????? ??? ??????? ?????? ????????? ????????????. ?????? ???? ???????????? ?????? ??? ??????? ????? ??????????? ??????? ??????????????.

 1 # ??????? ???????? ??? ???? ??? ???????????? ????????
 2 import scipy.ndimage as ndimage
 3 img_two_arr = ndimage.zoom(img_two_arr,zoom = (28/75,28/55))
 4 # ??????????? ??????? ????? ????????? ???????.
 5 img_two_arr.shape
 6 # ????????: (28, 28) 
 7 
 8 #  ?????????????? ??????? ??????? ???? ???????.
 9 plt.imshow(img_two_arr)
10 # ???????? ????????? ???????? ??????? ???????????? ???????.
11 
12 # ???? ??? ???? ???? ??? ???????? ???? ????????? ????? ????? ????????.
13 
14 img_two_arr = img_two_arr.reshape((1,-1)) 

???????? ??????? ???????? 2 ????? ??????? ??????? ???????? ??????????? ?????????.

6. ??????? ??????? ??????? ??????? ???????

1 knn.predict(img_two_arr)
2 # ????????: array([2])

???????? ?????? ?????? ????????? ???? 2 ?????? ????????? ??????? ???????.

標籤：其他

上一篇：關于選用ccflow你所擔心的問題都在這里為您解答

下一篇：Markdown格式及語法小結