{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"#Pseudo-label 伪标签\nimport pandas as pd#导入csv文件的库\nimport math\nimport numpy as np#进行矩阵运算的库\nimport matplotlib.pyplot as plt#作图的库\nimport torch#一个深度学习的库Pytorch\nimport torch.nn as nn#neural network,神经网络\nimport torch.optim as optim#一个实现了各种优化算法的库\nimport torch.nn.functional as F#神经网络函数库\nimport warnings#避免一些可以忽略的报错\nwarnings.filterwarnings('ignore')","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-07-10T07:52:45.760532Z","iopub.execute_input":"2023-07-10T07:52:45.761162Z","iopub.status.idle":"2023-07-10T07:52:50.279860Z","shell.execute_reply.started":"2023-07-10T07:52:45.761136Z","shell.execute_reply":"2023-07-10T07:52:50.278848Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#设置随机种子\nimport random\ntorch.backends.cudnn.deterministic = True#将cudnn框架中的随机数生成器设为确定性模式\ntorch.backends.cudnn.benchmark = False#关闭CuDNN框架的自动寻找最优卷积算法的功能，以避免不同的算法对结果产生影响\ntorch.manual_seed(42)\nnp.random.seed(42)\nrandom.seed(42)","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:50.281775Z","iopub.execute_input":"2023-07-10T07:52:50.282387Z","iopub.status.idle":"2023-07-10T07:52:50.292441Z","shell.execute_reply.started":"2023-07-10T07:52:50.282354Z","shell.execute_reply":"2023-07-10T07:52:50.291269Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train=pd.read_csv(\"/kaggle/input/ece597697-sp2023/train.csv/train.csv\")\ntrain.head()","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:50.294276Z","iopub.execute_input":"2023-07-10T07:52:50.294678Z","iopub.status.idle":"2023-07-10T07:52:56.859092Z","shell.execute_reply.started":"2023-07-10T07:52:50.294649Z","shell.execute_reply":"2023-07-10T07:52:56.857977Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"datay=train['label'].values\ntrain=train.drop(['label','id'],axis=1)\ndataX=train.values","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:56.861917Z","iopub.execute_input":"2023-07-10T07:52:56.862613Z","iopub.status.idle":"2023-07-10T07:52:57.005615Z","shell.execute_reply.started":"2023-07-10T07:52:56.862580Z","shell.execute_reply":"2023-07-10T07:52:57.004448Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#划分训练集和测试集的函数\ndef train_test_split(dataX,datay,shuffle=True,percentage=0.8):\n    \"\"\"\n    将训练数据X和标签y以numpy.array数组的形式传入\n    划分的比例定为训练集:测试集=8:2\n    \"\"\"\n    if shuffle:\n        random_num=[index for index in range(len(dataX))]\n        np.random.shuffle(random_num)\n        dataX=dataX[random_num]\n        datay=datay[random_num]\n    split_num=int(len(dataX)*percentage)\n    train_X=dataX[:split_num]\n    train_y=datay[:split_num]\n    test_X=dataX[split_num:]\n    test_y=datay[split_num:]\n    return train_X,train_y,test_X,test_y","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.007488Z","iopub.execute_input":"2023-07-10T07:52:57.008247Z","iopub.status.idle":"2023-07-10T07:52:57.017709Z","shell.execute_reply.started":"2023-07-10T07:52:57.008211Z","shell.execute_reply":"2023-07-10T07:52:57.016597Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_X,train_y,valid_X,valid_y=train_test_split(dataX,datay,percentage=0.9)\nlabel_count=np.max(train_y)+1\nfor index in range(label_count):\n    print(f\"np.sum(train_y==index)={np.sum(train_y==index)}\")\nprint(f\"train_X.shape:{train_X.shape},valid_X.shape:{valid_X.shape}\")","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.019453Z","iopub.execute_input":"2023-07-10T07:52:57.020438Z","iopub.status.idle":"2023-07-10T07:52:57.698701Z","shell.execute_reply.started":"2023-07-10T07:52:57.020403Z","shell.execute_reply":"2023-07-10T07:52:57.697673Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#多头注意力机制\nclass MultiHeadSelfAttention(nn.Module):\n    #定义初始化函数\n    def __init__(self,dim_in,d_model,num_heads=4):\n\n        super(MultiHeadSelfAttention,self).__init__()\n        \n        self.dim_in=dim_in\n        \n        self.d_model=d_model\n        \n        self.num_heads=num_heads\n        \n        #向量的维度必须被头的个数整除,否则会抛出异常.\n        assert d_model %num_heads==0,\"d_model must be multiple of num_heads\"\n        \n        #定义线性变换矩阵\n        self.linear_q=nn.Linear(dim_in,d_model)\n        self.linear_k=nn.Linear(dim_in,d_model)\n        self.linear_v=nn.Linear(dim_in,d_model)\n        self.scale=1/math.sqrt(d_model//num_heads)\n        \n        #最后的线性层\n        self.fc=nn.Linear(d_model,d_model)\n        \n    def forward(self,x):\n        batch,n,dim_in=x.shape\n        \n        assert dim_in==self.dim_in\n        \n        nh=self.num_heads\n        \n        dk=self.d_model//nh\n        \n        q=self.linear_q(x).reshape(batch,n,nh,dk).transpose(1,2)\n        k=self.linear_k(x).reshape(batch,n,nh,dk).transpose(1,2)\n        v=self.linear_v(x).reshape(batch,n,nh,dk).transpose(1,2)\n        \n        \n        dist=torch.matmul(q,k.transpose(2,3))*self.scale\n        dist=torch.softmax(dist,dim=-1)\n        \n        att=torch.matmul(dist,v)\n        \n        att=att.transpose(1,2).reshape(batch,n,self.d_model)\n        \n        output=self.fc(att)\n        \n        return output ","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.699975Z","iopub.execute_input":"2023-07-10T07:52:57.700577Z","iopub.status.idle":"2023-07-10T07:52:57.724726Z","shell.execute_reply.started":"2023-07-10T07:52:57.700543Z","shell.execute_reply":"2023-07-10T07:52:57.723642Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class SELayer(nn.Module):\n    def __init__(self,channel=256,reduction=16):\n        super(SELayer,self).__init__()\n        self.avg_pool=nn.AdaptiveAvgPool2d(1)#输出为1*1\n        self.fc=nn.Sequential(\n            nn.Linear(channel,channel//reduction,bias=True),\n            nn.BatchNorm1d(channel//reduction),\n            nn.GELU(),\n            nn.Linear(channel//reduction,channel,bias=True),\n            nn.Sigmoid(),\n        )\n    def forward(self,x):\n        #batch_size,channels,\n        b,c,_,_=x.shape\n        \n        y=self.avg_pool(x).view(b,c)\n        y=self.fc(y).view(b,c,1,1)\n        return x*y.expand_as(x)#将y扩展到x同样的shape","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.726405Z","iopub.execute_input":"2023-07-10T07:52:57.726735Z","iopub.status.idle":"2023-07-10T07:52:57.741722Z","shell.execute_reply.started":"2023-07-10T07:52:57.726710Z","shell.execute_reply":"2023-07-10T07:52:57.740789Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#对特征图的每个位置也提取一次权重\nclass SEConv(nn.Module):\n    def __init__(self,area=36,reduction=4):\n        super(SEConv,self).__init__()\n        self.fc=nn.Sequential(\n            nn.Linear(area,area//reduction,bias=True),\n            nn.BatchNorm1d(area//reduction),\n            nn.GELU(),\n            nn.Linear(area//reduction,area,bias=True),\n            nn.Sigmoid(),\n        )\n    def forward(self,x):\n        #batch_size,channels,(height*width)\n        b,c,area=x.shape\n        y=torch.mean(x,dim=1)\n        y=self.fc(y)\n        y=y.view(-1,1,area)\n        return x*y.expand_as(x)#将y扩展到x同样的shape","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.743138Z","iopub.execute_input":"2023-07-10T07:52:57.743499Z","iopub.status.idle":"2023-07-10T07:52:57.756594Z","shell.execute_reply.started":"2023-07-10T07:52:57.743468Z","shell.execute_reply":"2023-07-10T07:52:57.755635Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class Active_Rotating_Filters(nn.Module):\n    #定义初始化函数\n    def __init__(self,in_channels=3,out_channels=64,kernel_size=3,stride=1,padding=0):\n        super(Active_Rotating_Filters,self).__init__()\n        self.in_channels=in_channels\n        self.out_channels=out_channels\n        self.kernel_size=kernel_size\n        self.stride=stride\n        self.padding=padding\n        \n        self.conv=nn.Conv2d(in_channels,out_channels,kernel_size,stride,padding)\n        self.rot_conv=nn.Conv2d(in_channels,out_channels,kernel_size,stride,padding)#用于存放旋转后卷积的参数\n    \n    #对卷积核的权重进行旋转\n    def rotate_weight(self,index,weight):#out_channel*in_channel*kernel_size*kernel_size\n        #逆时针旋转90度旋转一次\n        rot_weight=torch.rot90(weight,index,(2,3))\n        return rot_weight\n        \n    def forward(self,x):\n        rot_x=self.conv(x)\n        for index in range(1,4):\n            #旋转后的参数\n            self.rot_conv.weight=nn.Parameter(self.rotate_weight(index,self.conv.weight))\n            rot_x=torch.cat((rot_x,self.rot_conv(x)),dim=1)\n        b,c,h,w=rot_x.shape\n        rot_x=rot_x.reshape(b,c//4,4,h,w)\n        rot_x,_=torch.max(rot_x,dim=2)#,_忽略索引\n        rot_x=rot_x.view(b,-1,h,w)\n        return rot_x","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.761408Z","iopub.execute_input":"2023-07-10T07:52:57.762480Z","iopub.status.idle":"2023-07-10T07:52:57.773490Z","shell.execute_reply.started":"2023-07-10T07:52:57.762450Z","shell.execute_reply":"2023-07-10T07:52:57.772388Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#score:0.92976\nclass CNN(nn.Module):\n\n    def __init__(self):\n        \n        super(CNN,self).__init__()\n        \n        self.flag=0\n        self.center_array=[]\n        \n        #创建自注意机制的类\n        self.Self_Attention=MultiHeadSelfAttention(dim_in=36,d_model=36,num_heads=3)\n        \n        self.selayer=SELayer()\n        \n        self.seconv=SEConv()\n        \n        self.conv=nn.Sequential(\n        nn.Conv2d(1,16,3,1,0),#30*30->28*28\n        nn.BatchNorm2d(16),\n        nn.SiLU(),\n        nn.Conv2d(16,32,3,1,0),#28*28->26*26\n        nn.BatchNorm2d(32),\n        nn.ReLU(),\n        nn.Conv2d(32,72,5,1,0),#26*26->22*22\n        nn.BatchNorm2d(72),\n        nn.Tanh(),\n        nn.Conv2d(72,128,5,1,1),#20*20\n        nn.GELU(),\n        nn.MaxPool2d(2,2,0),#10*10\n        nn.BatchNorm2d(128),\n        nn.Conv2d(128,256,5,1,0),#6*6\n        nn.SiLU(),\n        nn.BatchNorm2d(256),\n        self.selayer,\n        )\n        \n        self.fc=nn.Sequential( \n        nn.Linear(256*6*6,128),\n        nn.BatchNorm1d(128),\n        nn.Dropout(p=0.2),\n        nn.GELU(),\n        nn.Linear(128,10),\n        )\n    def circle_padding(sellf,image, pad=1):\n        \"\"\"\n        input: torch.Tensor image (batch_size, channels, height, width)\n        output: pad_image (batch_size, channels, height+2*pad, width+2*pad)\n        \"\"\"\n        b, c, h, w = image.shape\n        pad_image = torch.zeros((b, c, h+2*pad, w+2*pad), device=image.device)\n\n        pad_image[:, :, pad:h+pad, pad:w+pad] = image\n        pad_image[:, :, 0:pad, pad:w+pad] = image[:, :, -pad:, :]\n        pad_image[:, :, -pad:, pad:w+pad] = image[:, :, 0:pad, :]\n        pad_image[:, :, pad:h+pad, 0:pad] = image[:, :, :, -pad:]\n        pad_image[:, :, pad:h+pad, -pad:] = image[:, :, :, 0:pad]\n        pad_image[:, :, 0:pad, 0:pad] = image[:, :, -pad:, -pad:]\n        pad_image[:, :, -pad:, -pad:] = image[:, :, 0:pad, 0:pad]\n        pad_image[:, :, 0:pad, -pad:] = image[:, :, -pad:, 0:pad]\n        pad_image[:, :, -pad:, 0:pad] = image[:, :, 0:pad, -pad:]\n\n        return pad_image\n    \n    def get_centering_array(self,image):\n        if self.flag==0:\n            #传入的是np.array的数组\n            batch_size,channels,row,col=image.shape\n            center_x=row/2\n            center_y=col/2\n            array=torch.randn(row,col).to(device)\n            for i in range(row):\n                for j in range(col):\n                    array[i][j]=math.sqrt(1- ( (i-center_x)**2+(j-center_y)**2)/( center_x**2+center_y**2 )/10)\n            self.flag=1\n            self.center_array=array\n        return image*self.center_array\n    #噪音注入\n    def noise_injection(self,features,noise_factor=0.002):\n        noise = torch.randn_like(features) * noise_factor\n        return features + noise\n    \n    def forward(self,input): \n        output=self.circle_padding(input)\n        output=self.get_centering_array(output)\n        output=self.noise_injection(output)\n        output=self.conv(output)\n        output=output.view(-1,256,6*6)#reshape\n        output=self.seconv(output)\n        output=output+self.Self_Attention(output)#跳连接\n        output=output.view(-1,256*6*6)#reshape\n        output=self.fc(output)\n        return F.log_softmax(output/5,dim=1)","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.776405Z","iopub.execute_input":"2023-07-10T07:52:57.776660Z","iopub.status.idle":"2023-07-10T07:52:57.800408Z","shell.execute_reply.started":"2023-07-10T07:52:57.776638Z","shell.execute_reply":"2023-07-10T07:52:57.799552Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n\ndevice='cuda' if torch.cuda.is_available() else\"cpu\"\n\n\ntrain_X=np.array(train_X).reshape(-1,1,28,28)/255.0-0.5\ntrain_y=np.array(train_y)\nvalid_X=np.array(valid_X).reshape(-1,1,28,28)/255.0-0.5\n\nvalid_X1=torch.Tensor(valid_X).to(device)\nnetC=CNN()\ntrain_accs=[]#存储训练集的准确率\nvalid_accs=[]#存储测试集的准确率\n#训练周期为1000次\nnum_epochs=1000\nbatch_size=200#一次训练的数据量\nweight_decay=0.0001\n\n#优化器\noptimizer=optim.Adam(netC.parameters(),lr=0.0005,betas=(0.5,0.999))\n#损失函数\ncriterion=nn.CrossEntropyLoss()\nnetC=netC.to(device)\nfor epoch in range(num_epochs):\n    \n    random_num=[index for index in range(len(train_X))]\n    np.random.shuffle(random_num)\n    \n    train_X=train_X[random_num]\n    \n    train_y=train_y[random_num]\n    \n    train_X1=torch.Tensor(train_X[:batch_size]).to(device)\n    train_y1=torch.Tensor(train_y[:batch_size]).long().to(device)\n    \n    #训练\n    netC.train()\n    #将数据放进去训练\n    output=netC(train_X1).to(device)\n    #计算每次的损失函数\n    loss=criterion(output,F.one_hot(train_y1,10).type(torch.FloatTensor).to(device)).to(device)\n    #L1正则化\n    l1_regularization = torch.tensor(0.).to(device)\n    for param in netC.parameters():\n        l1_regularization += torch.norm(param, 1)\n    loss += weight_decay * l1_regularization\n    #反向传播\n    loss.backward()\n    #优化器进行优化(梯度下降,降低误差)\n    optimizer.step()\n    #将梯度清空\n    optimizer.zero_grad()\n    if epoch%50==0:\n        netC.eval()\n        with torch.no_grad():\n            output=netC(train_X1).to(device)\n            output=output.detach().cpu().numpy()\n            train_pred=np.argmax(output,axis=1)\n\n            output=netC(valid_X1).to(device)\n            output=output.detach().cpu().numpy()\n            valid_pred=np.argmax(output,axis=1)\n        \n        train_acc=np.sum(train_pred==train_y[:batch_size])/len(train_y[:batch_size])\n        valid_acc=np.sum(valid_pred==valid_y)/len(valid_y)\n\n        train_accs.append(train_acc)\n        valid_accs.append(valid_acc)\n        print(f\"epoch:{epoch},loss:{loss}\\n train_acc:{train_acc},valid_acc:{valid_acc}\")","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:52:57.801987Z","iopub.execute_input":"2023-07-10T07:52:57.802328Z","iopub.status.idle":"2023-07-10T07:55:43.705037Z","shell.execute_reply.started":"2023-07-10T07:52:57.802298Z","shell.execute_reply":"2023-07-10T07:55:43.704063Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"percentage=0.25#剪枝的比例\nparameter=[]#每层的参数\nparameter_sorted=np.array([])\nfor name,module in netC.named_modules():\n    if isinstance(module,nn.BatchNorm2d): \n        parameter.append(module.weight.detach().cpu().numpy())\n        parameter_sorted=np.concatenate((parameter_sorted,module.weight.detach().cpu().numpy()),axis=0)\nparameter_sorted,index_sorted=torch.sort(torch.abs(torch.Tensor(parameter_sorted)))\nsplit_weight=parameter_sorted[int(percentage*len(parameter_sorted))]#小于这个数字的特征图不要\nfor i in range(len(parameter)):\n    for j in range(len(parameter[i])):\n        if abs(parameter[i][j])<split_weight:\n            parameter[i][j]=0\nfor index in range(len(parameter)):\n    print(f\"np.sum(parameter[index]!=0)={np.sum(parameter[index]!=0)}\")","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:55:43.706483Z","iopub.execute_input":"2023-07-10T07:55:43.706819Z","iopub.status.idle":"2023-07-10T07:55:43.740720Z","shell.execute_reply.started":"2023-07-10T07:55:43.706788Z","shell.execute_reply":"2023-07-10T07:55:43.739695Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#存储卷积层和batchNorm层剪枝剩下的参数\nlayer=0#记录一下这是第几层\ntotal_parameter=[]\nfor name,module in netC.named_modules():\n    if isinstance(module,nn.Conv2d):\n        conv_parameter=[]\n        conv_weight=module.weight.detach().cpu().numpy()\n        for index in range(len(conv_weight)):\n            if parameter[layer][index]:#如果这个特征图重要\n                #conv_weight[index]\n                weight=conv_weight[index]\n                remain_weight=[]\n                if layer==0:\n                    remain_weight=weight\n                for j in range(len(weight)):\n                    if layer and parameter[layer-1][j]:#上一层的这个位置重要吗?\n                        remain_weight.append(weight[j])\n                conv_parameter.append(np.array(remain_weight))\n        #print(f\"conv_weight[index].shape:{conv_weight[index].shape}\")\n        total_parameter.append(conv_parameter)\n    if isinstance(module,nn.BatchNorm2d): \n        bn_parameter=[]\n        bn_weight=module.weight.detach().cpu().numpy()\n        for index in range(len(bn_weight)):\n            if parameter[layer][index]:#如果这个特征图重要\n                bn_parameter.append(bn_weight[index])\n        total_parameter.append(bn_parameter)\n        layer+=1\nprint(f\"len(total_parameter):{len(total_parameter)}\")","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:55:43.742166Z","iopub.execute_input":"2023-07-10T07:55:43.742497Z","iopub.status.idle":"2023-07-10T07:55:43.798797Z","shell.execute_reply.started":"2023-07-10T07:55:43.742466Z","shell.execute_reply":"2023-07-10T07:55:43.797852Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#score:0.92976\nclass prune_CNN(nn.Module):\n\n    def __init__(self):\n        \n        super(prune_CNN,self).__init__()\n        \n        self.flag=0\n        self.center_array=[]\n        \n        #创建自注意机制的类\n        self.Self_Attention=MultiHeadSelfAttention(dim_in=36,d_model=36,num_heads=3)\n        \n        self.seconv=SEConv()\n        \n        self.conv=nn.Sequential(\n        nn.Conv2d(1,16,3,1,0),#30*30->28*28\n        nn.BatchNorm2d(16),\n        nn.SiLU(),\n        nn.Conv2d(16,32,3,1,0),#28*28->26*26\n        nn.BatchNorm2d(32),\n        nn.ReLU(),\n        nn.Conv2d(32,72,5,1,0),#26*26->22*22\n        nn.BatchNorm2d(72),\n        nn.Tanh(),\n        nn.Conv2d(72,115,5,1,1),#20*20\n        nn.GELU(),\n        nn.MaxPool2d(2,2,0),#10*10\n        nn.BatchNorm2d(115),\n        nn.Conv2d(115,143,5,1,0),#6*6\n        nn.SiLU(),\n        nn.BatchNorm2d(143),\n        )\n        \n        self.fc=nn.Sequential( \n        nn.Linear(143*6*6,128),\n        nn.BatchNorm1d(128),\n        nn.Dropout(p=0.2),\n        nn.GELU(),\n        nn.Linear(128,10),\n        )\n    def circle_padding(sellf,image, pad=1):\n        \"\"\"\n        input: torch.Tensor image (batch_size, channels, height, width)\n        output: pad_image (batch_size, channels, height+2*pad, width+2*pad)\n        \"\"\"\n        b, c, h, w = image.shape\n        pad_image = torch.zeros((b, c, h+2*pad, w+2*pad), device=image.device)\n\n        pad_image[:, :, pad:h+pad, pad:w+pad] = image\n        pad_image[:, :, 0:pad, pad:w+pad] = image[:, :, -pad:, :]\n        pad_image[:, :, -pad:, pad:w+pad] = image[:, :, 0:pad, :]\n        pad_image[:, :, pad:h+pad, 0:pad] = image[:, :, :, -pad:]\n        pad_image[:, :, pad:h+pad, -pad:] = image[:, :, :, 0:pad]\n        pad_image[:, :, 0:pad, 0:pad] = image[:, :, -pad:, -pad:]\n        pad_image[:, :, -pad:, -pad:] = image[:, :, 0:pad, 0:pad]\n        pad_image[:, :, 0:pad, -pad:] = image[:, :, -pad:, 0:pad]\n        pad_image[:, :, -pad:, 0:pad] = image[:, :, 0:pad, -pad:]\n\n        return pad_image\n    \n    def get_centering_array(self,image):\n        if self.flag==0:\n            #传入的是np.array的数组\n            batch_size,channels,row,col=image.shape\n            center_x=row/2\n            center_y=col/2\n            array=torch.randn(row,col).to(device)\n            for i in range(row):\n                for j in range(col):\n                    array[i][j]=math.sqrt(1- ( (i-center_x)**2+(j-center_y)**2)/( center_x**2+center_y**2 )/10)\n            self.flag=1\n            self.center_array=array\n        return image*self.center_array\n    #噪音注入\n    def noise_injection(self,features,noise_factor=0.002):\n        noise = torch.randn_like(features) * noise_factor\n        return features + noise\n    \n    def forward(self,input): \n        output=self.circle_padding(input)\n        output=self.get_centering_array(output)\n        output=self.noise_injection(output)\n        output=self.conv(output)\n        output=output.view(-1,143,6*6)#reshape\n        output=self.seconv(output)\n        output=output+self.Self_Attention(output)#跳连接\n        output=output.view(-1,143*6*6)#reshape\n        output=self.fc(output)\n        return F.log_softmax(output/5,dim=1)","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:55:43.800272Z","iopub.execute_input":"2023-07-10T07:55:43.800598Z","iopub.status.idle":"2023-07-10T07:55:43.824430Z","shell.execute_reply.started":"2023-07-10T07:55:43.800568Z","shell.execute_reply":"2023-07-10T07:55:43.823456Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"netC1=prune_CNN()\nlayer=0\nfor name,module in netC1.named_modules():\n    if isinstance(module,nn.Conv2d) or isinstance(module,nn.BatchNorm2d):  \n        module.weight.data=torch.Tensor(total_parameter[layer]).clone()\n        print(f\"module.weight.shape:{module.weight.shape}\\n total_parameter[layer].shape:{np.array(total_parameter[layer]).shape:}\")\n        print(f\"len(total_parameter[layer]):{len(total_parameter[layer])}\")\n        layer+=1","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:55:43.825576Z","iopub.execute_input":"2023-07-10T07:55:43.826593Z","iopub.status.idle":"2023-07-10T07:55:44.454620Z","shell.execute_reply.started":"2023-07-10T07:55:43.826559Z","shell.execute_reply":"2023-07-10T07:55:44.453667Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n\ntrain_accs=[]#存储训练集的准确率\nvalid_accs=[]#存储测试集的准确率\n#训练周期为1000次\nnum_epochs=1000\nbatch_size=200#一次训练的数据量\n\n#优化器\noptimizer=optim.Adam(netC1.parameters(),lr=0.00025,betas=(0.5,0.999))\n#交叉熵损失函数\ncriterion=nn.CrossEntropyLoss()\n\nnetC1.to(device)\n\nfor epoch in range(num_epochs):\n    \n    random_num=[index for index in range(len(train_X))]\n    np.random.shuffle(random_num)\n    \n    train_X=train_X[random_num]\n    \n    train_y=train_y[random_num]\n    \n    train_X1=torch.Tensor(train_X[:batch_size]).to(device)\n    train_y1=torch.Tensor(train_y[:batch_size]).long().to(device)\n    \n    #训练\n    netC1.train()\n    #将数据放进去训练\n    output=netC1(train_X1).to(device)\n    #计算每次的损失函数\n    loss=criterion(output,F.one_hot(train_y1,10).type(torch.FloatTensor).to(device)).to(device)\n    #反向传播\n    loss.backward()\n    #优化器进行优化(梯度下降,降低误差)\n    optimizer.step()\n    #将梯度清空\n    optimizer.zero_grad()\n    if epoch%50==0:\n        netC.eval()\n        with torch.no_grad():\n            output=netC1(train_X1).to(device)\n            output=output.detach().cpu().numpy()\n            train_pred=np.argmax(output,axis=1)\n\n            output=netC1(valid_X1).to(device)\n            output=output.detach().cpu().numpy()\n            valid_pred=np.argmax(output,axis=1)\n        \n        train_acc=np.sum(train_pred==train_y[:batch_size])/len(train_y[:batch_size])\n        valid_acc=np.sum(valid_pred==valid_y)/len(valid_y)\n\n        train_accs.append(train_acc)\n        valid_accs.append(valid_acc)\n        print(f\"epoch:{epoch},loss:{loss}\\n train_acc:{train_acc},valid_acc:{valid_acc}\")","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:55:44.456112Z","iopub.execute_input":"2023-07-10T07:55:44.456848Z","iopub.status.idle":"2023-07-10T07:58:13.272125Z","shell.execute_reply.started":"2023-07-10T07:55:44.456815Z","shell.execute_reply":"2023-07-10T07:58:13.271058Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test=pd.read_csv(\"/kaggle/input/ece597697-sp2023/test.csv/test.csv\")\ntest=test.drop(['id'],axis=1)\ntest_X=(test.values/255.0-0.5).reshape(-1,1,28,28)\ntest_y=[]\nfor i in range(len(test_X)):\n    netC1.eval()\n    with torch.no_grad():\n        pred=netC1(torch.Tensor([test_X[i]]).to(device))[0]\n        pred=np.argmax(pred.detach().cpu().numpy())\n        test_y.append(pred)\ntest_y=np.array(test_y)\nprint(f\"test_X.shape:{test_X.shape},test_y.shape:{test_y.shape}\")","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:58:13.273570Z","iopub.execute_input":"2023-07-10T07:58:13.273923Z","iopub.status.idle":"2023-07-10T07:58:41.385636Z","shell.execute_reply.started":"2023-07-10T07:58:13.273876Z","shell.execute_reply":"2023-07-10T07:58:41.384669Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"total_X=np.concatenate((train_X,test_X),axis=0)\ntotal_y=np.concatenate((train_y,test_y),axis=0)","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:58:41.390079Z","iopub.execute_input":"2023-07-10T07:58:41.392754Z","iopub.status.idle":"2023-07-10T07:58:41.570404Z","shell.execute_reply.started":"2023-07-10T07:58:41.392718Z","shell.execute_reply":"2023-07-10T07:58:41.569114Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\ntrain_accs=[]#存储训练集的准确率\nvalid_accs=[]#存储测试集的准确率\n#训练周期为100次\nnum_epochs=100\nbatch_size=200#一次训练的数据量\nweight_decay=0.001\n\n#优化器\noptimizer=optim.Adam(netC1.parameters(),lr=0.0002,betas=(0.5,0.999))\n#损失函数\ncriterion=nn.CrossEntropyLoss()\n\nfor epoch in range(num_epochs):\n    \n    random_num=[index for index in range(len(total_X))]\n    np.random.shuffle(random_num)\n    \n    total_X=total_X[random_num]\n    \n    total_y=total_y[random_num]\n    \n    train_X1=torch.Tensor(total_X[:batch_size]).to(device)\n    train_y1=torch.Tensor(total_y[:batch_size]).long().to(device)\n    \n    #训练\n    netC1.train()\n    #将数据放进去训练\n    output=netC1(train_X1).to(device)\n    #计算每次的损失函数\n    loss=criterion(output,F.one_hot(train_y1,10).type(torch.FloatTensor).to(device)).to(device)\n    #L2正则化\n    l2_regularization = torch.tensor(0.).to(device)\n    for param in netC1.parameters():\n        l2_regularization += torch.norm(param, 2)\n    loss += weight_decay * l2_regularization\n    #反向传播\n    loss.backward()\n    #优化器进行优化(梯度下降,降低误差)\n    optimizer.step()\n    #将梯度清空\n    optimizer.zero_grad()\n    if epoch%50==0:\n        netC1.eval()\n        with torch.no_grad():\n            output=netC1(train_X1).to(device)\n            output=output.detach().cpu().numpy()\n            train_pred=np.argmax(output,axis=1)\n\n            output=netC1(valid_X1).to(device)\n            output=output.detach().cpu().numpy()\n            valid_pred=np.argmax(output,axis=1)\n        \n        train_acc=np.sum(train_pred==total_y[:batch_size])/len(total_y[:batch_size])\n        valid_acc=np.sum(valid_pred==valid_y)/len(valid_y)\n\n        train_accs.append(train_acc)\n        valid_accs.append(valid_acc)\n        print(f\"epoch:{epoch},loss:{loss}\\n train_acc:{train_acc},valid_acc:{valid_acc}\")","metadata":{"execution":{"iopub.status.busy":"2023-07-10T07:58:41.575348Z","iopub.execute_input":"2023-07-10T07:58:41.575712Z","iopub.status.idle":"2023-07-10T08:01:39.403178Z","shell.execute_reply.started":"2023-07-10T07:58:41.575681Z","shell.execute_reply":"2023-07-10T08:01:39.402036Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test=pd.read_csv(\"/kaggle/input/ece597697-sp2023/test.csv/test.csv\")\ntest=test.drop(['id'],axis=1)\ntest_X=test.values/255.0-0.5\ny_pred=[]\nfor i in range(len(test_X)):\n    netC1.eval()\n    with torch.no_grad():\n        image=test_X[i].reshape((1,1,28,28))\n        pred=netC1(torch.Tensor(image).to(device))[0]\n        pred=np.argmax(pred.detach().cpu().numpy())\n        y_pred.append(pred)","metadata":{"execution":{"iopub.status.busy":"2023-07-10T08:01:41.811764Z","iopub.execute_input":"2023-07-10T08:01:41.812465Z","iopub.status.idle":"2023-07-10T08:02:05.491081Z","shell.execute_reply.started":"2023-07-10T08:01:41.812429Z","shell.execute_reply":"2023-07-10T08:02:05.490078Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#写入文件\nimport csv\nhead=[\"id\",\"label\"]\ndata=[]\nfor index in range(len(y_pred)):\n    data.append([index,y_pred[index]])\nwith open(\"fashion_mnsit_submission.csv\",'w',newline='') as f:\n    writer=csv.writer(f)\n    writer.writerow(head)\n    writer.writerows(data)","metadata":{"execution":{"iopub.status.busy":"2023-07-10T08:02:05.494151Z","iopub.execute_input":"2023-07-10T08:02:05.494549Z","iopub.status.idle":"2023-07-10T08:02:05.524654Z","shell.execute_reply.started":"2023-07-10T08:02:05.494518Z","shell.execute_reply":"2023-07-10T08:02:05.523768Z"},"trusted":true},"execution_count":null,"outputs":[]}]}