{"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":"# 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 20GB 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","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-07-21T05:52:36.043826Z","iopub.execute_input":"2023-07-21T05:52:36.044498Z","iopub.status.idle":"2023-07-21T05:52:36.059339Z","shell.execute_reply.started":"2023-07-21T05:52:36.044448Z","shell.execute_reply":"2023-07-21T05:52:36.058300Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# zipファイル解凍","metadata":{}},{"cell_type":"code","source":"!rm -r /kaggle/working/train\n!rm -r /kaggle/working/test\n!unzip /kaggle/input/dogs-vs-cats-redux-kernels-edition/train.zip\n!unzip /kaggle/input/dogs-vs-cats-redux-kernels-edition/test.zip","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:52:36.061485Z","iopub.execute_input":"2023-07-21T05:52:36.061893Z","iopub.status.idle":"2023-07-21T05:52:59.958218Z","shell.execute_reply.started":"2023-07-21T05:52:36.061860Z","shell.execute_reply":"2023-07-21T05:52:59.957036Z"},"collapsed":true,"jupyter":{"outputs_hidden":true},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ライブラリ","metadata":{}},{"cell_type":"code","source":"import scipy.io\nfrom PIL import Image\nimport numpy as np\nfrom torchvision import transforms\nimport torch\nimport torch.utils.data as data\nimport os\nimport sys\nimport torch.optim as optim\nfrom tqdm import tqdm\nimport torch.nn as nn\nfrom glob import glob\nimport re\nimport csv\nfrom torch.nn import functional as F\nimport pandas as pd","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:52:59.960870Z","iopub.execute_input":"2023-07-21T05:52:59.961266Z","iopub.status.idle":"2023-07-21T05:53:03.381278Z","shell.execute_reply.started":"2023-07-21T05:52:59.961215Z","shell.execute_reply":"2023-07-21T05:53:03.380288Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 前処理クラス","metadata":{}},{"cell_type":"code","source":"class ImageTransform:\n    '''\n    画像の前処理を行う\n    画像のリサイズ，色の標準化\n    '''\n\n    def __init__(self, resize, mean, std):\n        self.data_transform = transforms.Compose(\n            [\n                transforms.Resize(resize),  # 短辺をの長さをresizeに\n                transforms.CenterCrop(resize),  # 画面中央を大きさresizeに切り取り\n                transforms.ToTensor(),  # テンソルに変換\n                transforms.Normalize(mean, std),  # 色情報の標準化\n            ]\n        )\n\n    def __call__(self, img):\n        return self.data_transform(img)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.382575Z","iopub.execute_input":"2023-07-21T05:53:03.383167Z","iopub.status.idle":"2023-07-21T05:53:03.390320Z","shell.execute_reply.started":"2023-07-21T05:53:03.383133Z","shell.execute_reply":"2023-07-21T05:53:03.389438Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# リスト作成クラス","metadata":{}},{"cell_type":"code","source":"class AnimalList:\n    '''\n    画像とラベルからリストを作成\n    \"train\"指定 : 前半７割のデータ\n    \"val指定 : 後半３割のデータ\n    '''\n\n    def __init__(self):\n        self.m_path = []  # 画像パス\n        self.m_label = []  # ラベル\n        \n        folder_path = \"/kaggle/working/train/\"\n        \n        files = glob(folder_path + \"*.jpg\")\n\n        for file_name in files:\n            # 画像パス と ラベル\n            self.m_path.append(file_name)\n            if file_name.split('.')[-3].split('/')[-1] == \"cat\":\n                self.m_label.append(0)\n            elif file_name.split('.')[-3].split('/')[-1] == \"dog\":\n                self.m_label.append(1)\n\n        # 学習・評価データの区切り\n        self.partition = int(len(self.m_path) * 0.7)\n\n    def __call__(self, phase):\n        if phase == \"train\":\n            return self.m_path[:self.partition], self.m_label[:self.partition]\n        elif phase == \"val\":\n            return self.m_path[self.partition:], self.m_label[self.partition:]\n        else:\n            print(\"Error: GenderList required 'train' or 'val'\", file=sys.stderr)\n            sys.exit(1)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.393180Z","iopub.execute_input":"2023-07-21T05:53:03.394255Z","iopub.status.idle":"2023-07-21T05:53:03.407035Z","shell.execute_reply.started":"2023-07-21T05:53:03.394222Z","shell.execute_reply":"2023-07-21T05:53:03.406116Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# データセットクラス","metadata":{}},{"cell_type":"code","source":"class AnimalDataset(data.Dataset):\n    '''\n    犬猫のDatasetを生成\n    '''\n\n    def __init__(self, file_list, label_list, transform=None):\n        self.file_list = file_list  # 画像ファイルパス\n        self.label_list = label_list  # ラベル\n        self.transform = transform  # 前処理のインスタンス\n\n    def __len__(self):\n        '''\n        画像枚数を返す\n        '''\n        return len(self.file_list)\n\n    def __getitem__(self, index):\n        '''\n        前処理した画像のテンソル形式のデータを取得\n        '''\n\n        # 画像をロード\n        img_path = self.file_list[index]\n        img = Image.open(img_path)\n\n        # 前処理\n        img_t = self.transform(img)\n\n        return img_t, int(self.label_list[index])","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.408573Z","iopub.execute_input":"2023-07-21T05:53:03.408894Z","iopub.status.idle":"2023-07-21T05:53:03.418797Z","shell.execute_reply.started":"2023-07-21T05:53:03.408864Z","shell.execute_reply":"2023-07-21T05:53:03.417921Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 学習関数","metadata":{}},{"cell_type":"code","source":"# モデルを学習させる関数\ndef train_model(net, dataloaders_dict, criterion, optimizer, num_epochs):\n\n    # 初期設定\n    # GPUが使えるかを確認\n    device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n    print(\"使用デバイス：\", device)\n\n    # ネットワークをGPUへ\n    net = net.to(device)\n\n    # ネットワークがある程度固定であれば、高速化させる\n    torch.backends.cudnn.benchmark = True\n\n    end = False\n\n    # epochのループ\n    for epoch in range(num_epochs):\n        print(\"\")\n        print('Epoch {}/{}'.format(epoch+1, num_epochs))\n        print('-------------')\n\n        # epochごとの訓練と検証のループ\n        for phase in ['train', 'val']:\n            if phase == 'train':\n                net.train()  # モデルを訓練モードに\n            else:\n                net.eval()   # モデルを検証モードに\n\n            epoch_loss = 0.0  # epochの損失和\n            epoch_corrects = 0  # epochの正解数\n\n            # 未学習時の検証性能を確かめるため、epoch=0の訓練は省略\n            if (epoch == 0) and (phase == 'train'):\n                continue\n\n            # データローダーからミニバッチを取り出すループ\n            for inputs, labels in tqdm(dataloaders_dict[phase]):\n\n                # GPUが使えるならGPUにデータを送る\n                inputs = inputs.to(device)\n                labels = labels.to(device)\n\n                # optimizerを初期化\n                optimizer.zero_grad()\n\n                # 順伝搬（forward）計算\n                with torch.set_grad_enabled(phase == 'train'):\n                    outputs = net(inputs)\n                    loss = criterion(outputs, labels)  # 損失を計算\n                    _, preds = torch.max(outputs, 1)  # ラベルを予測\n\n                    # 訓練時はバックプロパゲーション\n                    if phase == 'train':\n                        loss.backward()\n                        optimizer.step()\n\n                    # 結果の計算\n                    epoch_loss += loss.item() * inputs.size(0)  # lossの合計を更新\n                    # 正解数の合計を更新\n                    epoch_corrects += torch.sum(preds == labels.data)\n\n            # epochごとのlossと正解率を表示\n            epoch_loss = epoch_loss / len(dataloaders_dict[phase].dataset)\n            epoch_acc = epoch_corrects.double(\n            ) / len(dataloaders_dict[phase].dataset)\n\n            print('{} Loss: {:.4f} Acc: {:.4f}'.format(\n                phase, epoch_loss, epoch_acc))\n            \n        if end:\n            break\n","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.420134Z","iopub.execute_input":"2023-07-21T05:53:03.420542Z","iopub.status.idle":"2023-07-21T05:53:03.436597Z","shell.execute_reply.started":"2023-07-21T05:53:03.420509Z","shell.execute_reply":"2023-07-21T05:53:03.435513Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## ソート関数","metadata":{}},{"cell_type":"code","source":"def atoi(text):\n    return int(text) if text.isdigit() else text\n\ndef natural_keys(text):\n    return [ atoi(c) for c in re.split(r'(\\d+)', text) ]","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.437924Z","iopub.execute_input":"2023-07-21T05:53:03.438425Z","iopub.status.idle":"2023-07-21T05:53:03.451442Z","shell.execute_reply.started":"2023-07-21T05:53:03.438374Z","shell.execute_reply":"2023-07-21T05:53:03.450493Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# パラメータ","metadata":{}},{"cell_type":"code","source":"# 前処理パラメータ\nsize = 224\nmean = (0.485, 0.456, 0.406)\nstd = (0.229, 0.224, 0.225)\n\n# ミニバッチサイズ\nbatch_size = 64\n\n# エポック数\nnum_epochs = 15","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.452835Z","iopub.execute_input":"2023-07-21T05:53:03.453213Z","iopub.status.idle":"2023-07-21T05:53:03.463414Z","shell.execute_reply.started":"2023-07-21T05:53:03.453182Z","shell.execute_reply":"2023-07-21T05:53:03.462428Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# メイン処理(学習)","metadata":{}},{"cell_type":"markdown","source":"## データリスト作成","metadata":{}},{"cell_type":"code","source":"# データリスト\ng_list = AnimalList()\ntrain_data = g_list(\"train\")\nval_data = g_list(\"val\")","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.464766Z","iopub.execute_input":"2023-07-21T05:53:03.465721Z","iopub.status.idle":"2023-07-21T05:53:03.596777Z","shell.execute_reply.started":"2023-07-21T05:53:03.465685Z","shell.execute_reply":"2023-07-21T05:53:03.595774Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## データセット作成","metadata":{}},{"cell_type":"code","source":"# データセット\ntrain_dataset = AnimalDataset(train_data[0], train_data[1], ImageTransform(size, mean, std))\nval_dataset = AnimalDataset(val_data[0], val_data[1], ImageTransform(size, mean, std))","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.598286Z","iopub.execute_input":"2023-07-21T05:53:03.598653Z","iopub.status.idle":"2023-07-21T05:53:03.604240Z","shell.execute_reply.started":"2023-07-21T05:53:03.598619Z","shell.execute_reply":"2023-07-21T05:53:03.603214Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## データローダ作成","metadata":{}},{"cell_type":"code","source":"# DataLoader\ntrain_dataloader = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size, shuffle=True)\nval_dataloader = torch.utils.data.DataLoader(val_dataset, batch_size=batch_size,shuffle=True)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.609479Z","iopub.execute_input":"2023-07-21T05:53:03.609936Z","iopub.status.idle":"2023-07-21T05:53:03.620350Z","shell.execute_reply.started":"2023-07-21T05:53:03.609903Z","shell.execute_reply":"2023-07-21T05:53:03.619553Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 学習済みモデル読み込み","metadata":{}},{"cell_type":"code","source":"# モデル読み込み\nmodel = torch.hub.load('pytorch/vision:v0.10.0', 'mobilenet_v2', pretrained=True)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:03.621910Z","iopub.execute_input":"2023-07-21T05:53:03.622243Z","iopub.status.idle":"2023-07-21T05:53:05.654326Z","shell.execute_reply.started":"2023-07-21T05:53:03.622213Z","shell.execute_reply":"2023-07-21T05:53:05.653425Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 出力層変更","metadata":{}},{"cell_type":"code","source":"# 最後の層を変更\nmodel.classifier[1] = nn.Linear(in_features=1280, out_features=2)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:05.655713Z","iopub.execute_input":"2023-07-21T05:53:05.656193Z","iopub.status.idle":"2023-07-21T05:53:05.664115Z","shell.execute_reply.started":"2023-07-21T05:53:05.656156Z","shell.execute_reply":"2023-07-21T05:53:05.662962Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(model)\nparams_to_update_1 = []\nparams_to_update_2 = []\n\nupdate_param_names_1 = ['features']\nupdate_param_names_2 = ['classifier.1.weight', 'classifier.1.bias']\n\nfor name, param in model.named_parameters():\n    if update_param_names_1[0] in name:\n        param.requires_grad = True\n        params_to_update_1.append(param)\n        print(\"1\", name)\n    elif name in update_param_names_2:\n        param.requires_grad = True\n        params_to_update_2.append(param)\n        print(\"2\", name)\n    else:\n        param.requires_grad = False","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:05.665596Z","iopub.execute_input":"2023-07-21T05:53:05.665973Z","iopub.status.idle":"2023-07-21T05:53:07.192688Z","shell.execute_reply.started":"2023-07-21T05:53:05.665939Z","shell.execute_reply":"2023-07-21T05:53:07.191772Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 損失関数　最適化手法　設定","metadata":{}},{"cell_type":"code","source":"# 損失関数\ncriterion = torch.nn.CrossEntropyLoss()","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:07.193923Z","iopub.execute_input":"2023-07-21T05:53:07.194810Z","iopub.status.idle":"2023-07-21T05:53:07.208941Z","shell.execute_reply.started":"2023-07-21T05:53:07.194770Z","shell.execute_reply":"2023-07-21T05:53:07.207944Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# 最適化手法の設定\noptimizer = optim.SGD([\n    {'params' : params_to_update_1, 'lr' : 5e-4},\n    {'params' : params_to_update_2, 'lr' : 1e-3},\n], momentum=0.9)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:07.210562Z","iopub.execute_input":"2023-07-21T05:53:07.210902Z","iopub.status.idle":"2023-07-21T05:53:07.220524Z","shell.execute_reply.started":"2023-07-21T05:53:07.210871Z","shell.execute_reply":"2023-07-21T05:53:07.219433Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 学習","metadata":{}},{"cell_type":"code","source":"# 辞書型にする\ndataloaders_dict = {\"train\": train_dataloader, \"val\": val_dataloader}","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:07.222134Z","iopub.execute_input":"2023-07-21T05:53:07.222521Z","iopub.status.idle":"2023-07-21T05:53:07.231637Z","shell.execute_reply.started":"2023-07-21T05:53:07.222490Z","shell.execute_reply":"2023-07-21T05:53:07.230786Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# 学習・検証を実行する\ntrain_model(model, dataloaders_dict, criterion, optimizer, num_epochs=num_epochs)\n\ntorch.save(model, \"/kaggle/working/model_weight.pth\")","metadata":{"execution":{"iopub.status.busy":"2023-07-21T05:53:07.233099Z","iopub.execute_input":"2023-07-21T05:53:07.233728Z","iopub.status.idle":"2023-07-21T06:27:36.236479Z","shell.execute_reply.started":"2023-07-21T05:53:07.233696Z","shell.execute_reply":"2023-07-21T06:27:36.235411Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# メイン処理(推論)","metadata":{}},{"cell_type":"markdown","source":"## テストデータ読み込み","metadata":{}},{"cell_type":"code","source":"folder_path = \"/kaggle/working/test/\"\nfiles = sorted(glob(folder_path + \"*.jpg\"), key=natural_keys)\nlabel = [1] * len(files)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:27:36.238301Z","iopub.execute_input":"2023-07-21T06:27:36.239039Z","iopub.status.idle":"2023-07-21T06:27:36.361124Z","shell.execute_reply.started":"2023-07-21T06:27:36.239001Z","shell.execute_reply":"2023-07-21T06:27:36.360033Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## データセット作成","metadata":{}},{"cell_type":"code","source":"test_dataset = AnimalDataset(files, label, ImageTransform(size, mean, std))","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:27:36.362724Z","iopub.execute_input":"2023-07-21T06:27:36.363124Z","iopub.status.idle":"2023-07-21T06:27:36.369808Z","shell.execute_reply.started":"2023-07-21T06:27:36.363088Z","shell.execute_reply":"2023-07-21T06:27:36.367503Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## データローダ作成","metadata":{}},{"cell_type":"code","source":"test_dataloader = torch.utils.data.DataLoader(test_dataset, shuffle=False)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:27:36.371195Z","iopub.execute_input":"2023-07-21T06:27:36.371781Z","iopub.status.idle":"2023-07-21T06:27:36.381961Z","shell.execute_reply.started":"2023-07-21T06:27:36.371748Z","shell.execute_reply":"2023-07-21T06:27:36.380968Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## GPU確認","metadata":{}},{"cell_type":"code","source":"# GPUが使えるかを確認\ndevice = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\nprint(\"使用デバイス：\", device)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:27:36.383697Z","iopub.execute_input":"2023-07-21T06:27:36.384193Z","iopub.status.idle":"2023-07-21T06:27:36.394762Z","shell.execute_reply.started":"2023-07-21T06:27:36.384162Z","shell.execute_reply":"2023-07-21T06:27:36.393783Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 学習モデルロード","metadata":{}},{"cell_type":"code","source":"model = torch.load(\"/kaggle/working/model_weight.pth\", map_location={'cuda:0': 'cpu'})\n\n# ネットワークをGPUへ\nmodel = model.to(device)","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:27:36.396167Z","iopub.execute_input":"2023-07-21T06:27:36.396957Z","iopub.status.idle":"2023-07-21T06:27:36.449103Z","shell.execute_reply.started":"2023-07-21T06:27:36.396926Z","shell.execute_reply":"2023-07-21T06:27:36.448239Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 推論モード","metadata":{}},{"cell_type":"code","source":"model.eval()","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:27:36.450256Z","iopub.execute_input":"2023-07-21T06:27:36.450659Z","iopub.status.idle":"2023-07-21T06:27:36.464636Z","shell.execute_reply.started":"2023-07-21T06:27:36.450628Z","shell.execute_reply":"2023-07-21T06:27:36.463687Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 推論　保存","metadata":{}},{"cell_type":"code","source":"with open('/kaggle/working/submission.csv', 'w') as f:\n    writer= csv.writer(f, lineterminator='\\n')\n    writer.writerow(['id', 'label'])\n    count = 1\n    for x, y in test_dataloader:\n        with torch.no_grad():\n            x = x.to(device)\n            pred = model(x)\n            predicted = float(F.softmax(pred, dim=1)[:, 1])\n            \n            writer.writerow([count, predicted])\n            count += 1","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:27:36.466061Z","iopub.execute_input":"2023-07-21T06:27:36.466837Z","iopub.status.idle":"2023-07-21T06:29:34.050290Z","shell.execute_reply.started":"2023-07-21T06:27:36.466806Z","shell.execute_reply":"2023-07-21T06:29:34.049200Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"check_data = pd.read_csv('/kaggle/working/submission.csv')\ncheck_data","metadata":{"execution":{"iopub.status.busy":"2023-07-21T06:29:34.051739Z","iopub.execute_input":"2023-07-21T06:29:34.052343Z","iopub.status.idle":"2023-07-21T06:29:34.083879Z","shell.execute_reply.started":"2023-07-21T06:29:34.052309Z","shell.execute_reply":"2023-07-21T06:29:34.082451Z"},"trusted":true},"execution_count":null,"outputs":[]}]}