{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 5GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"train = pd.read_csv(\"/kaggle/input/digit-recognizer/train.csv\")\ntrain","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y = train[\"label\"]\ny[:10]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train2 = train.drop(\"label\",axis=1)\ntrain2","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"効率的なやり方\n# Process 1 : Simpleにしてできるところからやる。例えば1つだけやる\n# Process 2 : いったんまとめてみる\n# Process 3 : for文にして回してみる\n# Process 4 : 汎用性を持たせる (数字 → 文字化)\n# Process 5 : さらに汎用性を持たせる (関数化)\n# Process 6 : class"},{"metadata":{"trusted":true},"cell_type":"code","source":"X = train2.iloc[0].values\nX","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"\nX2 = X.reshape(28,28)\nimport matplotlib.pyplot as plt\nplt.imshow(X2)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 1次元データで処理していくのか、\n# RGB3次元データで処理していくのか。"},{"metadata":{"trusted":true},"cell_type":"code","source":"import cv2","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cv2.imwrite(\"tmp.bmp\",X2)\ntmp = cv2.imread(\"tmp.bmp\")\ntmp.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.imshow(tmp)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"X = train2.iloc[0].values\nX2 = X.reshape(28,28)\ncv2.imwrite(\"tmp.bmp\",X2)\ntmp = cv2.imread(\"tmp.bmp\")\nplt.imshow(tmp)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"for i in range(10):\n    X = train2.iloc[i].values\n    X2 = X.reshape(28,28)\n    cv2.imwrite(\"tmp.bmp\",X2)\n    tmp = cv2.imread(\"tmp.bmp\")\n    plt.figure()\n    plt.imshow(tmp)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def makeimage(num):\n    for i in range(num):\n        X = train2.iloc[i].values\n        X2 = X.reshape(28,28)\n        cv2.imwrite(\"tmp.bmp\",X2)\n        tmp = cv2.imread(\"tmp.bmp\")\n        plt.figure()\n        plt.imshow(tmp)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"makeimage(15)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Pytorch"},{"metadata":{"trusted":true},"cell_type":"code","source":"import torch\nfrom torch.utils.data import Dataset, DataLoader\nfrom torch import nn, optim\nfrom torch.nn import functional as F\nfrom torchvision.models import resnet18\nfrom albumentations import Normalize, Compose\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport cv2\nfrom tqdm.notebook import tqdm\nfrom sklearn.model_selection import train_test_split\nimport os\nimport glob\nimport multiprocessing as mp\n\n\n\nif torch.cuda.is_available():\n    device = 'cuda:0'\n    torch.set_default_tensor_type('torch.cuda.FloatTensor')\nelse:\n    device = 'cpu'\nprint(f'Running on device: {device}')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 1. transformの定義"},{"metadata":{"trusted":true},"cell_type":"code","source":"preprocess = Compose([\n    Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225], p=1)\n])\n\n# resnextなどのpre-trainモデルは全て、同じ方法で正規化された入力画像を使用しなければならない。それの変換をこの関数で行う。値はdefault。\n# Composeは今回あまり、意味をなさない\n# https://betashort-lab.com/%E3%83%87%E3%83%BC%E3%82%BF%E3%82%B5%E3%82%A4%E3%82%A8%E3%83%B3%E3%82%B9/albumentations%E3%81%AE%E3%81%BE%E3%81%A8%E3%82%81/ に詳細は書いてある","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"preprocess2 = Compose([\n    Normalize(mean=[0.5,], std=[0.5, ])\n])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 2.Dataset"},{"metadata":{"trusted":true},"cell_type":"code","source":"# 画像をどれだけ小さくするかの処理\nROWS = 32\nCOLS = 32","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"class GLDataset(Dataset):\n    \n    def __init__(self,df,labels,preprocess=None):\n        self.df = df\n        self.labels = labels\n        self.preprocess = preprocess\n        \n    def __len__(self):\n        return len(self.df)\n    \n    def __getitem__(self,idx):\n        \n        # ここからdatasetに食わせる前の前処理の記述。\n        \n        \n        \n        label = self.labels[idx]\n        \n        #img_pass = self.img_pass[idx]\n        \n        X = self.df.iloc[idx].values\n        X2 = X.reshape(28,28)\n        cv2.imwrite(\"tmp.bmp\",X2)\n        land = cv2.imread(\"tmp.bmp\") \n        \n        land = cv2.resize(land,(ROWS,COLS),interpolation = cv2.INTER_CUBIC)\n       \n        land = cv2.cvtColor(land,cv2.COLOR_BGR2RGB) # augmentを使うときにBGRからRGBにする必要があるのかもしれない。\n        \n        if self.preprocess is not None: # ここで、前処理を入れてnormalizationしている。\n                augmented = self.preprocess(image=land) # preprocessのimageをfaceで読む\n                land = augmented['image'] # https://betashort-lab.com/%E3%83%87%E3%83%BC%E3%82%BF%E3%82%B5%E3%82%A4%E3%82%A8%E3%83%B3%E3%82%B9/albumentations%E3%81%AE%E3%81%BE%E3%81%A8%E3%82%81/　に書いてある\n                \n        return {'landmarks': land.transpose(2, 0, 1), 'label': np.array(label, dtype=int)}  # pytorchはchannnl, x, yの形。これは辞書型で返している。(扱いやすいというだけかも。)\n        \n        \n        \n        \n        \n        \n        \n        ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train2","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"traindf = train2.iloc[:30000,:]\nvaldf = train2.iloc[30000:,:]\n\ntrainlabel = y[:30000]\nvallabel = y[30000:]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# instance化\ntrain_dataset = GLDataset(\n    df=traindf,\n    labels=trainlabel.to_numpy(),\n    preprocess=preprocess\n)\n\nval_dataset = GLDataset(\n    df=valdf,\n    labels=vallabel.to_numpy(),\n    preprocess=preprocess\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 3.Dataloader"},{"metadata":{"trusted":true},"cell_type":"code","source":"BATCH_SIZE = 1028\n\n#NUM_WORKERS = mp.cpu_count()\nNUM_WORKERS = 0 # ここを0にしないと動かない。cpuの仕様個数。←実は動くことが判明。classの中身次第！","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"## DataLoaderはimport torch.utils.data.Datasetでimport済みのもの\ntrain_dataloader = DataLoader(\n    train_dataset,\n    batch_size=BATCH_SIZE,\n    shuffle=True, #https://schemer1341.hatenablog.com/entry/2019/01/06/024605 を参考. idがわからなくなる\n    num_workers=NUM_WORKERS\n)\nval_dataloader = DataLoader(\n    val_dataset,\n    batch_size=BATCH_SIZE,\n    shuffle=False,\n    num_workers=NUM_WORKERS\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Deep learning"},{"metadata":{"trusted":true},"cell_type":"code","source":"encoder = resnet18(pretrained=True) #  pretrained = Trueはimagenetからpre-trainモデルを使う","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"class LandmarkClassifier(nn.Module): # nn.Moduleが入っているのが、モデルの定義っぽい\n    \n    def __init__(self, encoder, in_channels=3, num_classes=10): \n        \n        super(LandmarkClassifier, self).__init__() # nn.Moduleの__init__を継承。https://blog.codecamp.jp/python-class-2\n        \n        self.encoder = encoder\n        \n        # Modify input layer. # 入口のチャンネル数を合わせる。defaultのResnetのclassは64 channelになっているので、それをin_channelsにする。\n        # ここの記述がencoderごとに代わるので、efficientnet使うときは変えなければいけない\n        \n        self.encoder.conv1 = nn.Conv2d(\n            in_channels,\n            64,\n            kernel_size=7,\n            stride=2,\n            padding=3,\n            bias=False\n        )\n        \n        # Modify output layer.# 出口の個数も合わせる。defaultのResnetの出口は、1000になっているので、num_classesに変更\n        # ここの記述がencoderごとに代わるので、efficientnet使うときは変えなければいけない\n        \n        self.encoder.fc = nn.Linear(512 * 1, num_classes)\n\n    def forward(self, x): # 呼び出されたときに、xの引数があると、sigmoidで返す。ex) 後ほどのclassifier(sample_batched[\"landmark\"]) みたいなところ。\n        \n        # sigmoidで返すときは以下の感じ　https://aidiary.hatenablog.com/entry/20180203/1517629555\n        # return torch.sigmoid(self.encoder(x))\n        \n        # 多値問題でも、softmaxはここでは使わない。nn.CrossEntropyに既に入っているため\n        return self.encoder(x)\n    \n    \n    \n    \n    \n    ### 以下、どこのパラメーターを最適化するか。簡易テスト用は真ん中。フルオプションは下。ここはご参考。###\n    \n    def freeze_all_layers(self):# 中間層のパラメーター(重みづけ)を全部変えない。\n        for param in self.encoder.parameters():\n            param.requires_grad = False\n\n    def freeze_middle_layers(self):\n        self.freeze_all_layers()\n        \n        for param in self.encoder.conv1.parameters():# 最初のパラメーター(重みづけ)を変える。\n            param.requires_grad = True\n            \n        for param in self.encoder.fc.parameters():# 最後の(重みづけ)を変える。\n            param.requires_grad = True\n\n    def unfreeze_all_layers(self):# 中間層のパラメーター(重みづけ)を全部変える。\n        for param in self.encoder.parameters():\n            param.requires_grad = True","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"\nclassifier = LandmarkClassifier(encoder=encoder, in_channels=3, num_classes=10) # classifierはDeepfakeClassifierのインスタンス化\n\nclassifier = classifier.to(device) # ここがKerasとは違うところ。GPUに送りますよーという意味。今回はcpuなので、pass.\n\nclassifier.train() # ここがKerasとは違うところ。 訓練モードの場合 classifier.train() , 推論の場合 classifier.eval() と書く。Normalizeのプロセスなどが違うらしい。","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# 今回は全部最適化\nclassifier.unfreeze_all_layers()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"criterion = nn.CrossEntropyLoss()\n# 多値分類はcrossentropy","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#optimizer = optim.Adam(filter(lambda p: p.requires_grad, classifier.parameters()),lr=1e-5)\noptimizer = optim.Adam(classifier.parameters(),lr=0.01)\n#optimizer = optim.SGD(classifier.parameters(),lr=1e-5)\n\n# conv層(model.features)のパラメータを固定し、全結合層(model.classifier)のみをfinetuneする設定です。\n# 全層のうち、requires_gradがTrueの層のみをfinetuneするという意味です。\n# lr : 学習率","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def trainmodel(train_dataloader):\n    classifier.train()\n    \n    for a in train_dataloader:\n        \n        input1 = a[\"landmarks\"].to(device)\n        \n        y_pred = classifier(input1) # onenoteでいうoutput = model(train_x)\n        label = a[\"label\"].to(device)\n\n        loss = criterion(y_pred, label)\n\n        # Zero gradients, perform a backward pass, and update the weights.\n        optimizer.zero_grad() # おきまり。初期化。\n        loss.backward() # 後方にバック\n        optimizer.step() # 重みづけの最適化\n        \n    return(loss.item())\n    \n    ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def valmodel(val_dataloader):\n    \n    classifier.eval()\n    \n    for a in val_dataloader:\n        \n        input1 = a[\"landmarks\"].to(device)\n\n        y_pred = classifier(input1) # onenoteでいうoutput = model(train_x)\n        label = a[\"label\"].to(device)\n\n        loss = criterion(y_pred, label)\n        \n    return(loss.item())\n    \n    ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"epochs = 3\nsavename = \"resnet18.pth\"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def savemodel(bestloss,valloss):\n    \n    #######modelをsave#######\n    \n    if bestloss is None:\n        bestloss = valloss[-1]\n        state = {\n                'state_dict': classifier.state_dict(),\n                'optimizer_dict': optimizer.state_dict(),\n                \"bestloss\":bestloss,\n            }\n\n        torch.save(state, savename)\n        \n        print(\"save the first model\")\n    \n    elif valloss[-1] < bestloss:\n        \n        bestloss = valloss[-1]\n        state = {\n                'state_dict': classifier.state_dict(),\n                'optimizer_dict': optimizer.state_dict(),\n                \"bestloss\":bestloss,\n            }\n\n        torch.save(state, savename)\n        \n        print(\"found a better point\")\n    \n    else:\n        pass\n    \n    return bestloss\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"trainloss = []\nvalloss = []\n\nbestloss = None\n\nfor epoch in range(epochs):\n    \n    trainloss.append(trainmodel(train_dataloader))\n    valloss.append(valmodel(val_dataloader))\n    \n    print(str(epoch) + \"_end\")\n    \n    bestloss= savemodel(bestloss,valloss)\n    \n    print(bestloss)\n    \n    \n    ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 推論"},{"metadata":{"trusted":true},"cell_type":"code","source":"x = np.arange(epochs)\nplt.scatter(x,trainloss)\nplt.scatter(x,valloss)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"testdf = pd.read_csv(\"/kaggle/input/digit-recognizer/test.csv\")\ntestdf","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"class GLDataset_inf(Dataset):\n    \n    def __init__(self,df,testid,preprocess=None):\n        self.df = df\n        self.testid = testid\n        self.preprocess = preprocess\n        \n    def __len__(self):\n        return len(self.df)\n    \n    def __getitem__(self,idx):\n        \n        # ここからdatasetに食わせる前の前処理の記述。\n        \n        \n        \n        testid = self.testid[idx]\n        \n        #img_pass = self.img_pass[idx]\n        \n        X = self.df.iloc[idx].values\n        X2 = X.reshape(28,28)\n        cv2.imwrite(\"tmp.bmp\",X2)\n        land = cv2.imread(\"tmp.bmp\") \n        \n        land = cv2.resize(land,(ROWS,COLS),interpolation = cv2.INTER_CUBIC)\n       \n        land = cv2.cvtColor(land,cv2.COLOR_BGR2RGB) # augmentを使うときにBGRからRGBにする必要があるのかもしれない。\n        \n        if self.preprocess is not None: # ここで、前処理を入れてnormalizationしている。\n                augmented = self.preprocess(image=land) # preprocessのimageをfaceで読む\n                land = augmented['image'] # https://betashort-lab.com/%E3%83%87%E3%83%BC%E3%82%BF%E3%82%B5%E3%82%A4%E3%82%A8%E3%83%B3%E3%82%B9/albumentations%E3%81%AE%E3%81%BE%E3%81%A8%E3%82%81/　に書いてある\n                \n        return {'landmarks': land.transpose(2, 0, 1), 'testid': testid}  # pytorchはchannnl, x, yの形。これは辞書型で返している。(扱いやすいというだけかも。)\n        \n        \n        \n        \n        \n        \n        \n        ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"testid2 = testdf.index.values + 1","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# instance化\ntest_dataset = GLDataset_inf(\n    df=testdf,\n    testid=testid2,\n    preprocess=preprocess\n)\n\n## DataLoaderはimport torch.utils.data.Datasetでimport済みのもの\ntest_dataloader = DataLoader(\n    test_dataset,\n    batch_size=BATCH_SIZE,\n    shuffle=False, #https://schemer1341.hatenablog.com/entry/2019/01/06/024605 を参考. idがわからなくなる\n    num_workers=NUM_WORKERS\n)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"classifier.eval()\npresub = []\n\n    \nfor a in test_dataloader:\n\n    input1 = a[\"landmarks\"].to(device)\n\n    y_pred = classifier(input1) # onenoteでいうoutput = model(train_x)\n    \n    soft = F.softmax(y_pred).cpu().detach().numpy()\n    \n    y_pred = y_pred.cpu().detach().numpy()\n    \n    for b in range(len(y_pred)):\n        \n        predid = np.argmax(soft[b])\n        conf = np.max(soft[b])\n        \n        testid = a[\"testid\"][b]\n        \n        presub.append([testid,predid,conf])\n    \n    \n    \n\n    ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission = pd.DataFrame(presub,columns = [\"ImageId\",\"labels\",\"conf\"])\nsubmission","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample = pd.read_csv(\"/kaggle/input/digit-recognizer/sample_submission.csv\")\nsample","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample[\"Label\"]=submission[\"labels\"]\nsample","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample.to_csv(\"submission.csv\",index=False)","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}