机器学习15 手写数字识别-小数据集

1.手写数字数据集

• from sklearn.datasets import load_digits
• digits = load_digits()

from sklearn.datasets import load_digits
digits = load_digits()

2.图片数据预处理

• x：归一化MinMaxScaler()
• y：独热编码OneHotEncoder()或to_categorical
• 训练集测试集划分
• 张量结构

import numpy as np
from sklearn.preprocessing import MinMaxScaler
X_data = digits.data.astype(np.float32)
scaler = MinMaxScaler()
X_data = scaler.fit_transform(X_data)
print("归一化后",X_data)
X=X_data.reshape(-1,8,8,1)
from sklearn.preprocessing import OneHotEncoder
y = digits.target.astype(np.float32).reshape(-1,1)
Y = OneHotEncoder().fit_transform(y).todense()
print("独热编码：",Y)
from sklearn.model_selection import train_test_split
X_train,X_test,y_train,y_test = train_test_split(X,Y,test_size=0.2,random_state=0,stratify=Y)
print("X_data.shape:",X_data.shape)
print("X.shape:",X.shape)

3.设计卷积神经网络结构

• 绘制模型结构图，并说明设计依据。

  from tensorflow.keras.models import Sequential
  from tensorflow.keras.layers import Dense,Dropout,Conv2D,MaxPool2D,Flatten
  #3、建立模型
  model = Sequential()
  ks = (3, 3)  # 卷积核的大小
  input_shape = X_train.shape[1:]
  # 一层卷积，padding='same',tensorflow会对输入自动补0
  model.add(Conv2D(filters=16, kernel_size=ks, padding='same', input_shape=input_shape, activation='relu'))
  # 池化层1
  model.add(MaxPool2D(pool_size=(2, 2)))
  # 防止过拟合，随机丢掉连接
  model.add(Dropout(0.25))
  # 二层卷积
  model.add(Conv2D(filters=32, kernel_size=ks, padding='same', activation='relu'))
  # 池化层2
  model.add(MaxPool2D(pool_size=(2, 2)))
  model.add(Dropout(0.25))
  # 三层卷积
  model.add(Conv2D(filters=64, kernel_size=ks, padding='same', activation='relu'))
  # 四层卷积
  model.add(Conv2D(filters=128, kernel_size=ks, padding='same', activation='relu'))
  # 池化层3
  model.add(MaxPool2D(pool_size=(2, 2)))
  model.add(Dropout(0.25))
  # 平坦层
  model.add(Flatten())
  # 全连接层
  model.add(Dense(128, activation='relu'))
  model.add(Dropout(0.25))
  # 激活函数softmax
  model.add(Dense(10, activation='softmax'))
  print(model.summary())

• 

4.模型训练

# 画Train History图
plt.rcParams['font.sans-serif'] = ['FangSong'] # 指定字体
def show_train_history(train_history, train, validation):
    plt.plot(train_history.history[train])
    plt.plot(train_history.history[validation])
    plt.ylabel('train')
    plt.xlabel('epoch')
    plt.legend(['train', 'validation'], loc='upper left')
    plt.show()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
train_history = model.fit(x=x_train,y=y_train,validation_split=0.2,batch_size=300,epochs=10,verbose=2)
show_train_history(train_history,"accuracy","val_accuracy")  # 准确率
show_train_history(train_history,"loss","val_loss")  # 损失率

 

5.模型评价

• model.evaluate()
• 交叉表与交叉矩阵
• pandas.crosstab
• seaborn.heatmap

• import pandas as pd
  import seaborn as sns
  score = model.evaluate(X_test, y_test)
  print('score：', score)
  y_pred = model.predict_classes(X_test)
  print('y_pred：', y_pred[:10])
  y_test1 = np.argmax(y_test, axis=1).reshape(-1)
  y_true = np.array(y_test1)[0]
  pd.crosstab(y_true, y_pred, rownames=['true'], colnames=['predict'])
  y_test1 = y_test1.tolist()[0]
  a = pd.crosstab(np.array(y_test1), y_pred, rownames=['Lables'], colnames=['Predict'])
  df = pd.DataFrame(a)
  sns.heatmap(df, annot=True, cmap="Blues", linewidths=0.2, linecolor='G')
  plt.show()

•