import tensorflow as tfprint(tf.__version__)
2.6.0
一、数据集
mnist = tf.keras.datasets.mnist(train_images,train_labels),(test_images,test_labels) = mnist.load_data()
train_images.shape,train_labels.shape
((60000, 28, 28), (60000,))
# 可视化imageimport matplotlib.pyplot as pltdef plot_image(image): plt.imshow(image.reshape(28,28),cmap='binary') plt.show()
plot_image(train_images[0])
# 处理数据集：将数据集拉成二维的x_data = tf.cast(train_images.reshape(-1,28*28),dtype=tf.float32)# 划分数据集valid_percent = 0.2train_num = int(x_data.shape[0] * (1-valid_percent))x_train = x_data[:train_num] # 训练集y_train = train_labels[:train_num]x_valid = x_data[train_num:] # 验证集y_valid = train_labels[train_num:]x_test = tf.cast(test_images.reshape(-1,28*28),dtype=tf.float32) # 测试集y_test = test_labels
x_data.shape[0] * 0.8
48000.0
# 归一化x_train = x_train / 255.0x_valid = x_valid / 255.0x_test = x_test / 255.0
# 对标签进行独热编码y_train = tf.one_hot(y_train,depth=10)y_valid = tf.one_hot(y_valid,depth=10)y_test = tf.one_hot(y_test,depth=10)
二、模型
W = tf.Variable(tf.random.normal(shape=(28*28,10)))B = tf.Variable(tf.zeros(10))
def model(x,w,b): model_ = tf.matmul(x,w) + b return tf.nn.softmax(model_)
def loss(x,y,w,b): pred = model(x,w,b) loss_ = tf.keras.losses.categorical_crossentropy(y_true=y,y_pred=pred) return tf.reduce_mean(loss_)
def grad(x,y,w,b): with tf.GradientTape() as tape: loss_ = loss(x,y,w,b) return tape.gradient(loss_,[w,b])
# 准确率def accuracy(x,y,w,b): pred = model(x,w,b) # print(pred,y) corrent_prediction = tf.equal(tf.argmax(pred,1),tf.argmax(y,1)) # print(corrent_prediction,corrent_prediction.shape) return tf.reduce_mean(tf.cast(corrent_prediction,tf.float32))
三、训练
training_epochs = 20learining_rate = 0.001batch_size = 50
total_step = train_num // batch_sizeoptimizer = tf.keras.optimizers.Adam(learining_rate)train_loss_list = []valid_loss_list = []train_acc_list = []valid_acc_list = []for epoch in range(training_epochs): for step in range(total_step): xs = x_train[step*batch_size:(step+1)*batch_size] ys = y_train[step*batch_size:(step+1)*batch_size]# print(xs.shape,ys.shape) grads = grad(xs,ys,W,B) optimizer.apply_gradients(zip(grads,[W,B])) train_loss = loss(x_train,y_train,W,B).numpy() valid_loss = loss(x_valid,y_valid,W,B).numpy() train_acc = accuracy(x_train,y_train,W,B).numpy() valid_acc = accuracy(x_valid,y_valid,W,B).numpy() train_acc_list.append(train_acc) valid_acc_list.append(valid_acc) train_loss_list.append(train_loss) valid_loss_list.append(valid_loss) print(f"{epoch+1}:train_loss:{train_loss}train_acc:{train_acc}valid_loss:{valid_loss}valid_acc:{valid_acc}") 1:train_loss:1.5718567371368408train_acc:0.706458330154419valid_loss:1.4870891571044922valid_acc:0.71974998712539672:train_loss:0.9688711762428284train_acc:0.799958348274231valid_loss:0.9067797660827637valid_acc:0.81441664695739753:train_loss:0.7616745829582214train_acc:0.8380833268165588valid_loss:0.7226295471191406valid_acc:0.84616667032241824:train_loss:0.6529313921928406train_acc:0.8577708601951599valid_loss:0.6307686567306519valid_acc:0.86374998092651375:train_loss:0.5832834839820862train_acc:0.8693541884422302valid_loss:0.5737946629524231valid_acc:0.87291663885116586:train_loss:0.5349117517471313train_acc:0.8781874775886536valid_loss:0.5341132283210754valid_acc:0.87949997186660777:train_loss:0.49877774715423584train_acc:0.8850208520889282valid_loss:0.504259467124939valid_acc:0.88458335399627698:train_loss:0.4704902470111847train_acc:0.8897500038146973valid_loss:0.4805634021759033valid_acc:0.88824999332427989:train_loss:0.4476514160633087train_acc:0.8929374814033508valid_loss:0.4613181948661804valid_acc:0.8904166817665110:train_loss:0.42855820059776306train_acc:0.8967708349227905valid_loss:0.44507089257240295valid_acc:0.893333315849304211:train_loss:0.41240566968917847train_acc:0.8991041779518127valid_loss:0.4314454197883606valid_acc:0.895416676998138412:train_loss:0.39855533838272095train_acc:0.9014999866485596valid_loss:0.4198707342147827valid_acc:0.897333323955535913:train_loss:0.38649922609329224train_acc:0.9040833115577698valid_loss:0.40982159972190857valid_acc:0.899083316326141414:train_loss:0.3758573532104492train_acc:0.9057083129882812valid_loss:0.40100958943367004valid_acc:0.901333332061767615:train_loss:0.3664068877696991train_acc:0.9076458215713501valid_loss:0.3931334316730499valid_acc:0.902416646480560316:train_loss:0.35793840885162354train_acc:0.9091874957084656valid_loss:0.3861088752746582valid_acc:0.903750002384185817:train_loss:0.3502780497074127train_acc:0.9105416536331177valid_loss:0.37972745299339294valid_acc:0.904999971389770518:train_loss:0.34334996342658997train_acc:0.9118541479110718valid_loss:0.3739684522151947valid_acc:0.90591669082641619:train_loss:0.3370097279548645train_acc:0.9130833148956299valid_loss:0.3686791658401489valid_acc:0.90616667270660420:train_loss:0.33120784163475037train_acc:0.91427081823349valid_loss:0.3638934791088104valid_acc:0.9070833325386047
# 可视化损失import matplotlib.pyplot as pltplt.plot(train_loss_list,'r',label='train')plt.plot(valid_loss_list,'g',label='valid')plt.legend(loc=1)
# 可视化准确率import matplotlib.pyplot as pltplt.plot(train_acc_list,'r',label='trainacc')plt.plot(valid_acc_list,'g',label='validacc')plt.legend(loc=1)
四、预测
# 在测试集上评估模型准确率acc_test = accuracy(x_test,y_test,W,B).numpy()acc_test
0.9087
# 定义预测函数def predict(x,w,b): pred = model(x,w,b) res = tf.argmax(pred,1).numpy() return res
pred_test = predict(x_test,W,B)
# 定义可视化函数import numpy as npimport matplotlib.pyplot as pltdef plot_image_labels_prediction(images, # 图像列表 labels, # 标签列表 preds, # 预测值列表 index=0, # 从第index个开始 num=10 # 缺省一次显示10幅 ): fig = plt.gcf() # 获取当前图表 fig.set_size_inches(10,4) # 1英寸等于2.54cm if num > 10: num = 10 # 最多显示10个子图 for i in range(0,num): ax = plt.subplot(2,5,i+1) ax.imshow(np.reshape(images[index],(28,28)),cmap='binary') # 显示第index个图像 title = 'label='+str(labels[index]) # 构建该图上要显示的title信息 if len(preds) > 0: title += ',predict='+str(preds[index]) ax.set_title(title,fontsize=10) # 显示图上的title信息 ax.set_xticks([]) # 不显示坐标轴 ax.set_yticks([]) index = index + 1 plt.show()
plot_image_labels_prediction(test_images,test_labels,pred_test,9000,20)
