Sklearn多种算法实现人脸补全的项目实践

import matplotlib.pyplot as plt
 
from sklearn.linear_model import LinearRegression,Ridge,Lasso
from sklearn.tree import DecisionTreeRegressor
from sklearn.neighbors import KNeighborsRegressor
from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestRegressor
import numpy as np

faces=datasets.fetch_olivetti_faces()
images=faces.images
display(images.shape)
 
index=np.random.randint(0,400,size=1)[0]
 
img=images[index]
plt.figure(figsize=(3,3))
plt.imshow(img,cmap=plt.cm.gray)

index=np.random.randint(0,400,size=1)[0]
up_face=images[:,:32,:]
down_face=images[:,32:,:]
 
axes=plt.subplot(1,3,1)
axes.imshow(up_face[index],cmap=plt.cm.gray)
axes=plt.subplot(1,3,2)
axes.imshow(down_face[index],cmap=plt.cm.gray)
axes=plt.subplot(1,3,3)
axes.imshow(images[index],cmap=plt.cm.gray)

X=faces.data
 
x=X[:,:2048]
y=X[:,2048:]
 
estimators={}
 
estimators['linear']=LinearRegression()
estimators['ridge']=Ridge(alpha=0.1)
estimators['lasso']=Lasso(alpha=1)
estimators['knn']=KNeighborsRegressor(n_neighbors=5)
estimators['tree']=DecisionTreeRegressor()
estimators['forest']=RandomForestRegressor()

x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.2)
result={}
print
for key,model in estimators.items():
    print(key)
    model.fit(x_train,y_train)
    y_=model.predict(x_test)
    result[key]=y_

plt.figure(figsize=(40,40))
for i in range(0,10):
    #第一列，上半张人脸
    axes=plt.subplot(10,8,8*i+1)
    up_face=x_test[i].reshape(32,64)
    axes.imshow(up_face,cmap=plt.cm.gray)
    axes.axis('off')
    if i==0:
        axes.set_title('up-face')
    
    #第8列，整张人脸
    
    axes=plt.subplot(10,8,8*i+8)
    down_face=y_test[i].reshape(32,64)
    full_face=np.concatenate([up_face,down_face])
    axes.imshow(full_face,cmap=plt.cm.gray)
    axes.axis('off')
    
    if i==0:
        axes.set_title('full-face')
    
    #绘制预测人脸
    for j,key in enumerate(result):
        axes=plt.subplot(10,8,i*8+2+j)
        y_=result[key]
        predice_face=y_[i].reshape(32,64)
        pre_face=np.concatenate([up_face,predice_face])
        axes.imshow(pre_face,cmap=plt.cm.gray)
        axes.axis('off')
        if i==0:
            axes.set_title(key)

import matplotlib.pyplot as plt
 
from sklearn.linear_model import LinearRegression,Ridge,Lasso
from sklearn.tree import DecisionTreeRegressor
from sklearn.neighbors import KNeighborsRegressor
from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestRegressor
import numpy as np
 
faces=datasets.fetch_olivetti_faces()
images=faces.images
display(images.shape)
 
index=np.random.randint(0,400,size=1)[0]
 
img=images[index]
plt.figure(figsize=(3,3))
plt.imshow(img,cmap=plt.cm.gray)
 
index=np.random.randint(0,400,size=1)[0]
up_face=images[:,:32,:]
down_face=images[:,32:,:]
 
axes=plt.subplot(1,3,1)
axes.imshow(up_face[index],cmap=plt.cm.gray)
axes=plt.subplot(1,3,2)
axes.imshow(down_face[index],cmap=plt.cm.gray)
axes=plt.subplot(1,3,3)
axes.imshow(images[index],cmap=plt.cm.gray)
 
X=faces.data
 
x=X[:,:2048]
y=X[:,2048:]
 
estimators={}
 
estimators['linear']=LinearRegression()
estimators['ridge']=Ridge(alpha=0.1)
estimators['lasso']=Lasso(alpha=1)
estimators['knn']=KNeighborsRegressor(n_neighbors=5)
estimators['tree']=DecisionTreeRegressor()
estimators['forest']=RandomForestRegressor()
x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.2)
result={}
print
for key,model in estimators.items():
    print(key)
    model.fit(x_train,y_train)
    y_=model.predict(x_test)
    result[key]=y_
 
plt.figure(figsize=(40,40))
for i in range(0,10):
    #第一列，上半张人脸
    axes=plt.subplot(10,8,8*i+1)
    up_face=x_test[i].reshape(32,64)
    axes.imshow(up_face,cmap=plt.cm.gray)
    axes.axis('off')
    if i==0:
        axes.set_title('up-face')
    
    #第8列，整张人脸
    
    axes=plt.subplot(10,8,8*i+8)
    down_face=y_test[i].reshape(32,64)
    full_face=np.concatenate([up_face,down_face])
    axes.imshow(full_face,cmap=plt.cm.gray)
    axes.axis('off')
    
    if i==0:
        axes.set_title('full-face')
    
    #绘制预测人脸
    for j,key in enumerate(result):
        axes=plt.subplot(10,8,i*8+2+j)
        y_=result[key]
        predice_face=y_[i].reshape(32,64)
        pre_face=np.concatenate([up_face,predice_face])
        axes.imshow(pre_face,cmap=plt.cm.gray)
        axes.axis('off')
        if i==0:
            axes.set_title(key)