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() #查看数据集的数字图片 import matplotlib.pyplot as plt plt.imshow(digits.images[5]) plt.show()
查看数据集的一张数字:
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
#对X进行归一化 from sklearn.preprocessing import MinMaxScaler import numpy as np X_data = digits.data.astype(np.float32) scaler = MinMaxScaler() X_data = scaler.fit_transform(X_data) scaler.scale_ scaler.min_ print("查看归一化后的X数据: ",X_data) #转换为图片数据的维度 X = X_data.reshape(-1,8,8,1) #对Y进行 one-hot处理 独热编码 from sklearn.preprocessing import OneHotEncoder Y_data = digits.target.reshape(-1,1) Y = OneHotEncoder().fit_transform(Y_data).todense() print("One-hot处理: ",Y) #划分训练集和测试集 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("查看划分的数据集的维度: ",X_train.shape,X_test.shape,Y_train.shape,Y_test.shape)
查看归一化,独热编码和划分测试集的维度:
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。
- 根据经典模型绘制
from tensorflow.keras.models import Sequential from tensorflow.keras.layers import Dense,Dropout,Flatten,Conv2D,MaxPool2D #设计卷积神经网络结构 model = Sequential() ks = (3, 3) # 卷积核的大小 input_shape = X_train.shape[1:] # 第一层卷积,指定input_shape,其他层的数据的input_shape框架会自动推导,padding指定扫描方式 model.add(Conv2D(filters=16, kernel_size=ks, padding='same', input_shape=input_shape, activation='relu')) # 池化层 model.add(MaxPool2D(pool_size=(2, 2))) # 随机丢链接,防止过拟合 model.add(Dropout(0.25)) # 二层卷积 model.add(Conv2D(filters=32, kernel_size=ks, padding='same', activation='relu')) # 池化层 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')) model.summary()
模型结果:
4.模型训练
#对模型进行训练 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) score = model.evaluate(X_test,Y_test) print('查看损失率和精确度:', score)
可视化结果:
5.模型评价
- model.evaluate()
- 交叉表与交叉矩阵
- pandas.crosstab
- seaborn.heatmap
# 模型评价 import seaborn as sns import pandas as pd score = model.evaluate(X_test, Y_test) # 预测值 y_pred = model.predict_classes(X_test) print('预测值与真实值对比:', y_pred[:10]," ",Y_test[:10]) # 交叉表与交叉矩阵 y_test1 = np.argmax(Y_test, axis=1).reshape(-1) y_true = np.array(y_test1)[0] # 交叉表查看预测数据与原数据对比 pd.crosstab(y_true, y_pred, rownames=['true'], colnames=['predict']) # 交叉矩阵 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="YlGn", linewidths=0.2, linecolor='G') plt.show()
可视化结果: