[人工智能]利用神经网络实现手写字识别

神经网络介绍

神经网络即多层感知机

如果不知道感知机的可以看博主之前的文章感知机及Python实现

神经网络实现及手写字识别

关于数据集：

• 从http://yann.lecun.com/exdb/mnist/下载，下载后将文件解压
• 将main函数中的path改为下载文件的存储路径即可

如果对数据集有问题，可以私信博主

关于实现：

• 基于pytorch实现，包括神经网络的构建，激活函数的选择
• 归一化使用了像素值/255的方式实现，可以尝试用别的方式进行归一化处理

import numpy as np
import torch
import torch.nn.functional as F
import torch.optim as optim
from torch import nn
class Data:
    '''
    this class is about data module
    '''
    def __init__(self):
        self.start_loc_image=16 # the start location of image data
        self.start_loc_label=8 # the start location of label data
        self.num_pixel=28*28 # the number of pixels
        self.choice={'train-image':'train-images.idx3-ubyte','train-label':'train-labels.idx1-ubyte',
                'test-image':'t10k-images.idx3-ubyte','test-label':'t10k-labels.idx1-ubyte'} # the specific file name
    def get(self,path,train_test='',image_label=''):
        '''
        get the data from the file whose path is "path"
        :param path: the saving path of given files, default is "./file/"
        :param train_test: the data category("train" or "test")
        :param image_label: the data information("image" or "label")
        :return: the data you want
        '''
        if train_test not in ['train','test'] or image_label not in ['image','label']:raise NameError(
            'please check you spelling,"train_test" can be "train/test", "image_label" can be "image/label"')
        ch=train_test+'-'+image_label
        data=[]
        if image_label=='image':
            print('loading images ...')
            with open(path+self.choice[ch],'rb',) as f:
                file=f.read()
                for i in range(self.start_loc_image,file.__len__(),self.num_pixel):
                    item=[]
                    pixel=file[i:i+self.num_pixel].hex()
                    for p in range(0,pixel.__len__(),2):
                        item.append(int(pixel[p:p+2],16)) # decode -> get the pixel information from original file
                    data.append(self.transform2image(item))
            f.close()
        elif image_label=='label':
            print('load labels ...')
            with open(path+self.choice[ch],'rb',) as f:
                file=f.read()
                for i in range(self.start_loc_label,file.__len__()):
                    data.append(file[i]) # decode -> get the label from original file
            f.close()
        return data
    def transform2image(self,data:list):
        '''
        transform pixel point to image
        :param data: the original 1D pixel points
        :return: transformed image(28*28)
        '''
        assert data.__len__()==784
        import numpy as np
        return np.reshape(data,(28,-1))
    def transfer_tensor(self,data):
        '''
        transfer data to tensor format
        :param data: the original input data
        :return: transferred data
        '''
        return torch.tensor(data)
    def normalize(self,data,maximum=255):
        '''
        normalize the data with maximum
        :param data: the input data
        :param maximum: the maximum of pixel(is 255)
        :return: normalized data
        '''
        return torch.div(data,maximum)
class Network(nn.Module):
    '''
    this class is about neural network
    '''
    def __init__(self,in_dim,n_hidden,out_dim):
        '''
        define the network
        :param in_dim: the input dimension
        :param n_hidden: the hidden layer dimension
        :param out_dim: the output dimension
        '''
        ### about network
        super(Network, self).__init__()
        self.layer1 = nn.Sequential(nn.Linear(in_dim,n_hidden))
        self.layer2 = nn.Sequential(nn.Linear(n_hidden,out_dim))
        ### other parameter
        self.size_pixel=28*28 # the size of pixed of each picture
        ### learning rate
        self.learning_rate=0.001
        ### optimizer
        self.optimizer=optim.SGD(self.parameters(),lr=self.learning_rate)
    def forward(self,data):
        '''
        forward the input data
        :param data: the data you want to train
        :return: the output or predicted value
        '''
        hidden=F.relu(self.layer1(data))
        out=F.sigmoid(self.layer2(hidden))
        return out
    def accuracy(self,act,pre):
        '''
        calculate the accuracy
        :param act: actual value
        :param pre: predicted value
        :return: accuracy
        '''
        assert act.__len__()==pre.__len__()
        return round((act==pre).sum().item()/act.__len__(),3)
    def pre_process(self,feature,label):
        '''
        pre processing
        :param feature: feature
        :param label: label
        :return: preprocessed feature and label
        '''
        ### transform to the format of tensor
        feature=dat.transfer_tensor(feature)
        feature=dat.normalize(feature)
        feature=feature.view(-1,self.size_pixel)
        label=dat.transfer_tensor(label)
        return feature,torch.tensor(label,dtype=torch.int64)
    def per_train(self,epoch,feature,label,validation=0.2,batch=50,verbose=True,num_view=50):
        '''
        train neural network
        :param epoch: training times
        :param feature: feature
        :param label: label
        :param validation: for using evaluation
        :param batch: batch size
        :param verbose: whether view the training process or not
        :param num_view: view via training "num_view" times
        :return: none
        '''
        assert feature.__len__()==label.__len__()
        print('training neural network ...')
        fea,lab=self.pre_process(feature,label)
        len_train=int(feature.__len__()*(1-validation))
        data_train,label_train=fea[:len_train+1],lab[:len_train+1]
        data_train=[data_train[i:i+batch]for i in range(0,len_train,batch)]
        label_train=[label_train[i:i+batch]for i in range(0,len_train,batch)]
        data_val,label_val=fea[len_train:],lab[len_train:]
        for e in range(epoch+1):
            self.train()
            loss_tmp=[]
            for img,lab in zip(data_train,label_train):
                pre=self(img)
                loss_train=nn.CrossEntropyLoss()(pre,lab)
                loss_tmp.append(loss_train)
                loss_train.backward()
                self.optimizer.step()
                self.optimizer.zero_grad()
            if verbose and e>0 and e%num_view==0:
                self.eval()
                pre=self(data_val)
                loss_val=nn.CrossEntropyLoss()(pre,label_val)
                _,pre_view=pre.max(1)
                acc=self.accuracy(label_val,pre_view)
                print('epoch: '+str(e)+'/'+str(epoch)+' --> training loss:',loss_train.item(),'validation loss:',
                      loss_val.item(),'validation accuracy:',acc)
def set_seed(seed):
    '''
    set random seed in order that result can be replayed
    :param seed: random seed
    :return: none
    '''
    import random
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)
    torch.backends.cudnn.deterministic = True
    np.random.seed(seed)
    random.seed(seed)
Seed=0
Size_pixel= 28*28
Hidden=200
Output=10
set_seed(Seed)
if __name__ == '__main__':
    ### set some necessary parameters
    path='./file/' # the path of saved file
    ### initialize necessary class
    dat=Data() # for data related
    net=Network(Size_pixel,Hidden,Output) # build network
    ### load image and label of training data and testing data
    # train_image=dat.get(path,'train','image')
    # train_label=dat.get(path,'train','label')
    test_image=dat.get(path,'test','image')
    test_label=dat.get(path,'test','label')
    ### train network
    epoch=1000 # training times
    net.per_train(epoch,test_image,test_label)

结果展示

可以看到随着训练次数的增加，预测手写字的准确率也越来越高
运行结果

loading images ...
load labels ...
training neural network ...
epoch: 50/1000 --> training loss: 2.1074414253234863 validation loss: 2.1338484287261963 validation accuracy: 0.716
epoch: 100/1000 --> training loss: 1.8810529708862305 validation loss: 1.9262853860855103 validation accuracy: 0.814
epoch: 150/1000 --> training loss: 1.7580647468566895 validation loss: 1.811558723449707 validation accuracy: 0.844
epoch: 200/1000 --> training loss: 1.6905272006988525 validation loss: 1.7511157989501953 validation accuracy: 0.858
epoch: 250/1000 --> training loss: 1.6461492776870728 validation loss: 1.7131295204162598 validation accuracy: 0.87
epoch: 300/1000 --> training loss: 1.6150131225585938 validation loss: 1.6869707107543945 validation accuracy: 0.877
epoch: 350/1000 --> training loss: 1.5923327207565308 validation loss: 1.6678192615509033 validation accuracy: 0.886
epoch: 400/1000 --> training loss: 1.5752110481262207 validation loss: 1.653057336807251 validation accuracy: 0.892
epoch: 450/1000 --> training loss: 1.5620046854019165 validation loss: 1.6411659717559814 validation accuracy: 0.896
epoch: 500/1000 --> training loss: 1.5515371561050415 validation loss: 1.6312130689620972 validation accuracy: 0.897
epoch: 550/1000 --> training loss: 1.5430898666381836 validation loss: 1.6226950883865356 validation accuracy: 0.903
epoch: 600/1000 --> training loss: 1.5360981225967407 validation loss: 1.615297794342041 validation accuracy: 0.905
epoch: 650/1000 --> training loss: 1.5302610397338867 validation loss: 1.608830451965332 validation accuracy: 0.907
epoch: 700/1000 --> training loss: 1.5252583026885986 validation loss: 1.603135108947754 validation accuracy: 0.908
epoch: 750/1000 --> training loss: 1.5209182500839233 validation loss: 1.5980955362319946 validation accuracy: 0.91
epoch: 800/1000 --> training loss: 1.5170769691467285 validation loss: 1.5936044454574585 validation accuracy: 0.911
epoch: 850/1000 --> training loss: 1.513597846031189 validation loss: 1.5895581245422363 validation accuracy: 0.915
epoch: 900/1000 --> training loss: 1.5103877782821655 validation loss: 1.5858769416809082 validation accuracy: 0.917
epoch: 950/1000 --> training loss: 1.5073943138122559 validation loss: 1.5824891328811646 validation accuracy: 0.918
epoch: 1000/1000 --> training loss: 1.5045477151870728 validation loss: 1.5793448686599731 validation accuracy: 0.92

效果展示