{"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":6799,"databundleVersionId":4225553,"sourceType":"competition"}],"dockerImageVersionId":30918,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import torch\nimport torch.nn as nn\nimport torch.optim as optim\nimport numpy as np\nimport pandas as pd\nimport os\nimport torchvision ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:17:09.041074Z","iopub.execute_input":"2025-04-05T10:17:09.041629Z","iopub.status.idle":"2025-04-05T10:17:16.770249Z","shell.execute_reply.started":"2025-04-05T10:17:09.041583Z","shell.execute_reply":"2025-04-05T10:17:16.769188Z"}},"outputs":[],"execution_count":1},{"cell_type":"code","source":"from torch.utils.data import Dataset \nimport cv2\nimport matplotlib.pyplot as plt\nfrom xml.dom import minidom\nfrom os.path import basename\nimport glob","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:17:16.771527Z","iopub.execute_input":"2025-04-05T10:17:16.772101Z","iopub.status.idle":"2025-04-05T10:17:17.119266Z","shell.execute_reply.started":"2025-04-05T10:17:16.772067Z","shell.execute_reply":"2025-04-05T10:17:17.118375Z"}},"outputs":[],"execution_count":2},{"cell_type":"code","source":"def unpickle(file):\n   import pickle\n   with open(file, 'rb') as fo:\n      dict = pickle.load(fo, encoding='bytes')\n   return dict","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-04T20:57:47.773371Z","iopub.execute_input":"2025-04-04T20:57:47.773745Z","iopub.status.idle":"2025-04-04T20:57:47.778768Z","shell.execute_reply.started":"2025-04-04T20:57:47.773696Z","shell.execute_reply":"2025-04-04T20:57:47.77767Z"}},"outputs":[],"execution_count":3},{"cell_type":"code","source":"training=torchvision.datasets.ImageNet(root='/kaggle/input/imagenet-object-localization-challenge/ILSVRC', train=True, download=True)","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def generate_noise(path,t,ct):\n    img=cv2.imread(path)\n    img=np.array(img)\n    img=img/255.0\n    \n    epsilon=np.random.randn(*img.shape).astype(np.float32)\n    noise = np.exp(-t / ct)\n    noise=np.clip(noise,0,1)\n    new_img=np.sqrt(noise)*img + np.sqrt(1-noise)*epsilon\n    new_img=np.clip(new_img,0,1)\n    \n    return noise","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-04T20:57:50.270635Z","iopub.execute_input":"2025-04-04T20:57:50.271102Z","iopub.status.idle":"2025-04-04T20:57:50.27727Z","shell.execute_reply.started":"2025-04-04T20:57:50.271072Z","shell.execute_reply":"2025-04-04T20:57:50.275941Z"}},"outputs":[],"execution_count":4},{"cell_type":"code","source":"xml_folder=r'/kaggle/input/imagenet-object-localization-challenge/ILSVRC/Annotations/CLS-LOC/train'\ntrain_folder=r'/kaggle/input/imagenet-object-localization-challenge/ILSVRC/Data/CLS-LOC/train/*'","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T09:39:48.230839Z","iopub.execute_input":"2025-04-05T09:39:48.231252Z","iopub.status.idle":"2025-04-05T09:39:48.235291Z","shell.execute_reply.started":"2025-04-05T09:39:48.231222Z","shell.execute_reply":"2025-04-05T09:39:48.234083Z"}},"outputs":[],"execution_count":3},{"cell_type":"code","source":"train_folders = glob.glob('/kaggle/input/imagenet-object-localization-challenge/ILSVRC/Data/CLS-LOC/train/*')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:18:20.994745Z","iopub.execute_input":"2025-04-05T10:18:20.995235Z","iopub.status.idle":"2025-04-05T10:18:21.020325Z","shell.execute_reply.started":"2025-04-05T10:18:20.995197Z","shell.execute_reply":"2025-04-05T10:18:21.019308Z"}},"outputs":[],"execution_count":3},{"cell_type":"code","source":"folder=(train_folders[0]+'/*')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:18:22.401256Z","iopub.execute_input":"2025-04-05T10:18:22.401664Z","iopub.status.idle":"2025-04-05T10:18:22.406776Z","shell.execute_reply.started":"2025-04-05T10:18:22.401632Z","shell.execute_reply":"2025-04-05T10:18:22.405674Z"}},"outputs":[],"execution_count":4},{"cell_type":"code","source":"folder_name=glob.glob(folder)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:18:23.164238Z","iopub.execute_input":"2025-04-05T10:18:23.16461Z","iopub.status.idle":"2025-04-05T10:18:24.636476Z","shell.execute_reply.started":"2025-04-05T10:18:23.164584Z","shell.execute_reply":"2025-04-05T10:18:24.635255Z"}},"outputs":[],"execution_count":5},{"cell_type":"code","source":"(folder_name)","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"for i in range(1,101):\n    generate_noise(folder_name[0],0.0001,0.02,i,100)","metadata":{}},{"cell_type":"code","source":"class Diffusion(nn.Module):\n    def __init__(self,batch) -> None:\n        super().__init__()\n        self.batch=batch\n        self.time_embed = nn.Linear(1, 32).float()\n        self.conv1 = nn.Conv2d(3, 32, kernel_size=(3,3), stride=1, padding=1)\n        self.act1 = nn.ReLU()\n        self.drop1 = nn.Dropout(0.3)\n \n        self.conv2 = nn.Conv2d(32, 32, kernel_size=(3,3), stride=1, padding=1)\n        self.act2 = nn.ReLU()\n        self.conv3 = nn.Conv2d(32, 3, kernel_size=(3,3), stride=1, padding=1)\n\n    def forward(self,x,t):\n\n        t=t.view(self.batch,1).float()\n        x = self.act1(self.conv1(x))\n        x = self.drop1(x)\n        t_emb = self.time_embed(t).unsqueeze(-1).unsqueeze(-1)\n        x= x + t_emb\n        x = self.act2(self.conv2(x))\n        x = x+ t_emb\n        x = self.conv3(x)\n        return x\n\nmodel = Diffusion(1)\nloss_fn = nn.MSELoss()\noptimizer = optim.SGD(model.parameters(), lr=0.0001, momentum=0.1)\ndef generate(path,t,ct):\n    img=cv2.imread(path)\n    img=np.array(img)\n    img=img/255.0\n    \n    epsilon=np.random.randn(*img.shape).astype(np.float32)\n    noise = np.exp(-t / ct)\n    noise=np.clip(noise,0,1)\n    epsilon=np.clip(epsilon,0,1)\n    new_img=np.sqrt(noise)*img + np.sqrt(1-noise)*epsilon\n    new_img=np.clip(new_img,0,1)\n    return torch.tensor(new_img,dtype=torch.float32).permute(2,0,1).unsqueeze(0), torch.tensor(epsilon,dtype=torch.float32).permute(2,0,1).unsqueeze(0)\n\ndef noising(path,t,epoch):\n    new_img,error=generate(path,t,epoch)\n    timestep = torch.tensor([t/epoch],dtype=torch.float32) \n    pred_error=model.forward(new_img,timestep)\n    loss = loss_fn(pred_error, error)\n    print(loss)\n    optimizer.zero_grad()\n    loss.backward()\n    optimizer.step()\n    \ndef train(folder,epoch,T):\n        for i in range(epoch):\n            path=folder[np.random.randint(len(folder))]\n            print(f\"path : {path}, epoch: {i}\")\n            t=np.random.randint(T)\n            noising(path,t,T)\n        ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:42:11.421429Z","iopub.execute_input":"2025-04-05T10:42:11.421879Z","iopub.status.idle":"2025-04-05T10:42:11.439421Z","shell.execute_reply.started":"2025-04-05T10:42:11.421844Z","shell.execute_reply":"2025-04-05T10:42:11.438182Z"}},"outputs":[],"execution_count":50},{"cell_type":"code","source":"train(folder_name,1000,100)","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"img,_=generate(folder_name[0],100,100)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:48:13.557476Z","iopub.execute_input":"2025-04-05T10:48:13.557969Z","iopub.status.idle":"2025-04-05T10:48:13.601036Z","shell.execute_reply.started":"2025-04-05T10:48:13.557925Z","shell.execute_reply":"2025-04-05T10:48:13.59984Z"}},"outputs":[],"execution_count":52},{"cell_type":"code","source":"test=img","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:48:23.982472Z","iopub.execute_input":"2025-04-05T10:48:23.98286Z","iopub.status.idle":"2025-04-05T10:48:23.987286Z","shell.execute_reply.started":"2025-04-05T10:48:23.982831Z","shell.execute_reply":"2025-04-05T10:48:23.986094Z"}},"outputs":[],"execution_count":53},{"cell_type":"code","source":"def inference(img,ct):\n    x=img\n    for t in range(ct-1,-1,-1):\n        t = torch.tensor([t/ct],dtype=torch.float32)\n        noise = torch.exp(-t / ct)\n        noise=torch.clip(noise,0,1)\n        noise= noise.view(1, 1, 1, 1)\n        eps=model(x,t)\n        new_x= (x- torch.sqrt(1-noise)*eps)/(torch.sqrt(noise))\n        x=new_x\n    return new_x","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:48:27.298862Z","iopub.execute_input":"2025-04-05T10:48:27.299285Z","iopub.status.idle":"2025-04-05T10:48:27.305322Z","shell.execute_reply.started":"2025-04-05T10:48:27.299252Z","shell.execute_reply":"2025-04-05T10:48:27.304233Z"}},"outputs":[],"execution_count":54},{"cell_type":"code","source":"fin_img=inference(test,100)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:48:31.513621Z","iopub.execute_input":"2025-04-05T10:48:31.514139Z","execution_failed":"2025-04-05T10:48:53.725Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"fin_img","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import torchvision.transforms.functional as TF\nfrom PIL import Image","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:39:27.658865Z","iopub.execute_input":"2025-04-05T10:39:27.659342Z","iopub.status.idle":"2025-04-05T10:39:27.663953Z","shell.execute_reply.started":"2025-04-05T10:39:27.659305Z","shell.execute_reply":"2025-04-05T10:39:27.662436Z"}},"outputs":[],"execution_count":39},{"cell_type":"code","source":"image_tensor = torch.clamp(fin_img, 0, 1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:39:32.051084Z","iopub.execute_input":"2025-04-05T10:39:32.05151Z","iopub.status.idle":"2025-04-05T10:39:32.057714Z","shell.execute_reply.started":"2025-04-05T10:39:32.05147Z","shell.execute_reply":"2025-04-05T10:39:32.056382Z"}},"outputs":[],"execution_count":40},{"cell_type":"code","source":"img1=cv2.imread(folder_name[0])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:40:54.709722Z","iopub.execute_input":"2025-04-05T10:40:54.710132Z","iopub.status.idle":"2025-04-05T10:40:54.722478Z","shell.execute_reply.started":"2025-04-05T10:40:54.710097Z","shell.execute_reply":"2025-04-05T10:40:54.72145Z"}},"outputs":[],"execution_count":46},{"cell_type":"code","source":"img1","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:41:37.885021Z","iopub.execute_input":"2025-04-05T10:41:37.88546Z","iopub.status.idle":"2025-04-05T10:41:37.89309Z","shell.execute_reply.started":"2025-04-05T10:41:37.885424Z","shell.execute_reply":"2025-04-05T10:41:37.891892Z"}},"outputs":[{"execution_count":49,"output_type":"execute_result","data":{"text/plain":"array([[[ 45, 186, 123],\n        [ 53, 190, 128],\n        [ 58, 188, 127],\n        ...,\n        [ 78, 167, 117],\n        [ 80, 172, 121],\n        [ 89, 181, 130]],\n\n       [[ 56, 183, 121],\n        [ 47, 176, 115],\n        [ 47, 176, 119],\n        ...,\n        [ 83, 178, 128],\n        [ 77, 180, 129],\n        [ 82, 187, 138]],\n\n       [[ 50, 175, 106],\n        [ 44, 170, 105],\n        [ 47, 175, 116],\n        ...,\n        [ 89, 189, 129],\n        [ 75, 183, 124],\n        [ 73, 186, 128]],\n\n       ...,\n\n       [[ 78, 188, 140],\n        [ 77, 186, 138],\n        [ 77, 185, 139],\n        ...,\n        [ 87, 182, 148],\n        [ 88, 184, 148],\n        [ 86, 184, 148]],\n\n       [[ 89, 191, 136],\n        [ 78, 182, 129],\n        [ 73, 183, 131],\n        ...,\n        [ 92, 185, 148],\n        [ 89, 186, 146],\n        [ 87, 187, 145]],\n\n       [[ 92, 191, 129],\n        [ 81, 186, 123],\n        [ 87, 202, 139],\n        ...,\n        [ 85, 181, 134],\n        [ 79, 182, 131],\n        [ 77, 183, 130]]], dtype=uint8)"},"metadata":{}}],"execution_count":49},{"cell_type":"code","source":"img_tensor","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:40:08.358561Z","iopub.execute_input":"2025-04-05T10:40:08.358967Z","iopub.status.idle":"2025-04-05T10:40:08.368776Z","shell.execute_reply.started":"2025-04-05T10:40:08.358937Z","shell.execute_reply":"2025-04-05T10:40:08.367816Z"}},"outputs":[{"execution_count":44,"output_type":"execute_result","data":{"text/plain":"tensor([[[0.6688, 0.0000, 0.0000,  ..., 0.1804, 0.0000, 0.3042],\n         [1.0000, 0.0000, 0.0754,  ..., 0.0000, 0.0242, 0.2287],\n         [0.8401, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.5889],\n         ...,\n         [1.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.5467],\n         [0.7735, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.6393],\n         [0.8327, 1.0000, 0.3841,  ..., 1.0000, 0.3811, 0.7275]],\n\n        [[0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0257, 0.6401],\n         [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 1.0000],\n         [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 1.0000],\n         ...,\n         [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 1.0000],\n         [0.0000, 0.0000, 0.0590,  ..., 0.0000, 0.0504, 1.0000],\n         [0.0000, 0.9274, 0.7631,  ..., 0.6340, 1.0000, 1.0000]],\n\n        [[0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.2404],\n         [0.0790, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.1478],\n         [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0262],\n         ...,\n         [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.2197],\n         [0.1955, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.2516],\n         [0.6128, 0.0602, 0.4235,  ..., 0.4805, 0.1753, 0.8881]]],\n       grad_fn=<SqueezeBackward1>)"},"metadata":{}}],"execution_count":44},{"cell_type":"code","source":"img_tensor\nimg_tensor = img_tensor.squeeze(0).cpu()  # remove batch dimension\nimg_pil = TF.to_pil_image(img_tensor) ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:39:32.832255Z","iopub.execute_input":"2025-04-05T10:39:32.832626Z","iopub.status.idle":"2025-04-05T10:39:32.840251Z","shell.execute_reply.started":"2025-04-05T10:39:32.832598Z","shell.execute_reply":"2025-04-05T10:39:32.839143Z"}},"outputs":[],"execution_count":41},{"cell_type":"code","source":"img_pil.save(\"output.png\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:39:42.609321Z","iopub.execute_input":"2025-04-05T10:39:42.609798Z","iopub.status.idle":"2025-04-05T10:39:42.630261Z","shell.execute_reply.started":"2025-04-05T10:39:42.609763Z","shell.execute_reply":"2025-04-05T10:39:42.628863Z"}},"outputs":[],"execution_count":42},{"cell_type":"code","source":"img_pil","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-05T10:39:48.040295Z","iopub.execute_input":"2025-04-05T10:39:48.04076Z","iopub.status.idle":"2025-04-05T10:39:48.059674Z","shell.execute_reply.started":"2025-04-05T10:39:48.040723Z","shell.execute_reply":"2025-04-05T10:39:48.058495Z"}},"outputs":[{"execution_count":43,"output_type":"execute_result","data":{"text/plain":"<PIL.Image.Image image mode=RGB size=500x375>","image/png":"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\n","image/jpeg":"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\n"},"metadata":{}}],"execution_count":43},{"cell_type":"code","source":"import tensorflow as tf\nfrom tensorflow.keras import layers\n\nmodel = tf.keras.Sequential()\nmodel.add(layers.Dense(16, activation='relu', input_shape=(1,)))\nmodel.add(layers.Dense(16, activation='relu'))\nmodel.add(layers.Dense(16, activation='relu'))\nmodel.add(layers.Dense(1))\nmodel.compile(optimizer='rmsprop', loss='mse', metrics=['mae'])\nmodel.summary()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-11T16:13:32.853344Z","iopub.execute_input":"2025-04-11T16:13:32.853691Z","iopub.status.idle":"2025-04-11T16:13:51.553179Z","shell.execute_reply.started":"2025-04-11T16:13:32.853661Z","shell.execute_reply":"2025-04-11T16:13:51.551866Z"}},"outputs":[{"name":"stderr","text":"/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"\u001b[1mModel: \"sequential\"\u001b[0m\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n│ dense (\u001b[38;5;33mDense\u001b[0m)                        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)                  │              \u001b[38;5;34m32\u001b[0m │\n├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)                  │             \u001b[38;5;34m272\u001b[0m │\n├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)                  │             \u001b[38;5;34m272\u001b[0m │\n├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n│ dense_3 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)                   │              \u001b[38;5;34m17\u001b[0m │\n└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)                  │              <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span> │\n├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)                  │             <span style=\"color: #00af00; text-decoration-color: #00af00\">272</span> │\n├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)                  │             <span style=\"color: #00af00; text-decoration-color: #00af00\">272</span> │\n├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)                   │              <span style=\"color: #00af00; text-decoration-color: #00af00\">17</span> │\n└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m593\u001b[0m (2.32 KB)\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">593</span> (2.32 KB)\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m593\u001b[0m (2.32 KB)\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">593</span> (2.32 KB)\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n</pre>\n"},"metadata":{}}],"execution_count":1},{"cell_type":"code","source":"converter = tf.lite.TFLiteConverter.from_keras_model(model)\ntflite_model = converter.convert()\nopen(\"sine_model.tflite\", \"wb\").write(tflite_model)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-11T16:14:33.027552Z","iopub.execute_input":"2025-04-11T16:14:33.027981Z","iopub.status.idle":"2025-04-11T16:14:34.166564Z","shell.execute_reply.started":"2025-04-11T16:14:33.027943Z","shell.execute_reply":"2025-04-11T16:14:34.165493Z"}},"outputs":[{"name":"stdout","text":"Saved artifact at '/tmp/tmpsjkqbnw_'. The following endpoints are available:\n\n* Endpoint 'serve'\n  args_0 (POSITIONAL_ONLY): TensorSpec(shape=(None, 1), dtype=tf.float32, name='keras_tensor')\nOutput Type:\n  TensorSpec(shape=(None, 1), dtype=tf.float32, name=None)\nCaptures:\n  134700732169584: TensorSpec(shape=(), dtype=tf.resource, name=None)\n  134700732207472: TensorSpec(shape=(), dtype=tf.resource, name=None)\n  134700732212224: TensorSpec(shape=(), dtype=tf.resource, name=None)\n  134700732215744: TensorSpec(shape=(), dtype=tf.resource, name=None)\n  134700732217856: TensorSpec(shape=(), dtype=tf.resource, name=None)\n  134700732762240: TensorSpec(shape=(), dtype=tf.resource, name=None)\n  134700732767344: TensorSpec(shape=(), dtype=tf.resource, name=None)\n  134700732768224: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","output_type":"stream"},{"execution_count":2,"output_type":"execute_result","data":{"text/plain":"4060"},"metadata":{}}],"execution_count":2},{"cell_type":"code","source":"!apt-get update\n!apt-get -qq install xxd\n!xxd -i /kaggle/working/sine_model.tflite > sine_model.h","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-11T16:32:06.616447Z","iopub.execute_input":"2025-04-11T16:32:06.616935Z","iopub.status.idle":"2025-04-11T16:32:21.74241Z","shell.execute_reply.started":"2025-04-11T16:32:06.616879Z","shell.execute_reply":"2025-04-11T16:32:21.740828Z"}},"outputs":[{"name":"stdout","text":"Get:1 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  InRelease [1,581 B]\nGet:2 https://cloud.r-project.org/bin/linux/ubuntu jammy-cran40/ InRelease [3,632 B]                \nGet:3 http://security.ubuntu.com/ubuntu jammy-security InRelease [129 kB]                           \nGet:4 https://r2u.stat.illinois.edu/ubuntu jammy InRelease [6,555 B]                                \nHit:5 http://archive.ubuntu.com/ubuntu jammy InRelease                                              \nGet:6 http://archive.ubuntu.com/ubuntu jammy-updates InRelease [128 kB]                             \nGet:7 https://cloud.r-project.org/bin/linux/ubuntu jammy-cran40/ Packages [70.9 kB]       \nGet:8 http://archive.ubuntu.com/ubuntu jammy-backports InRelease [127 kB]                           \nGet:9 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages [1,383 kB]\nGet:10 https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy InRelease [18.1 kB]  \nGet:11 https://ppa.launchpadcontent.net/graphics-drivers/ppa/ubuntu jammy InRelease [24.3 kB]\nGet:12 https://r2u.stat.illinois.edu/ubuntu jammy/main amd64 Packages [2,690 kB]\nHit:13 https://ppa.launchpadcontent.net/ubuntugis/ppa/ubuntu jammy InRelease\nGet:14 https://r2u.stat.illinois.edu/ubuntu jammy/main all Packages [8,833 kB]\nGet:15 http://archive.ubuntu.com/ubuntu jammy-updates/restricted amd64 Packages [4,154 kB]\nGet:16 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 Packages [1,542 kB]        \nGet:17 http://archive.ubuntu.com/ubuntu jammy-updates/multiverse amd64 Packages [55.7 kB]\nGet:18 http://archive.ubuntu.com/ubuntu jammy-updates/main amd64 Packages [3,099 kB]\nGet:19 http://archive.ubuntu.com/ubuntu jammy-backports/main amd64 Packages [82.7 kB]\nGet:20 http://archive.ubuntu.com/ubuntu jammy-backports/universe amd64 Packages [35.2 kB]\nGet:21 http://security.ubuntu.com/ubuntu jammy-security/main amd64 Packages [2,788 kB]\nGet:22 https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy/main amd64 Packages [34.3 kB]\nGet:23 https://ppa.launchpadcontent.net/graphics-drivers/ppa/ubuntu jammy/main amd64 Packages [46.8 kB]\nGet:24 http://security.ubuntu.com/ubuntu jammy-security/multiverse amd64 Packages [47.7 kB]\nGet:25 http://security.ubuntu.com/ubuntu jammy-security/universe amd64 Packages [1,243 kB]\nGet:26 http://security.ubuntu.com/ubuntu jammy-security/restricted amd64 Packages [4,000 kB]\nFetched 30.5 MB in 3s (9,687 kB/s)                             \nReading package lists... Done\nW: Skipping acquire of configured file 'main/source/Sources' as repository 'https://r2u.stat.illinois.edu/ubuntu jammy InRelease' does not seem to provide it (sources.list entry misspelt?)\n(Reading database ... 127400 files and directories currently installed.)\nPreparing to unpack .../xxd_2%3a8.2.3995-1ubuntu2.24_amd64.deb ...\nUnpacking xxd (2:8.2.3995-1ubuntu2.24) over (2:8.2.3995-1ubuntu2.21) ...\nSetting up xxd (2:8.2.3995-1ubuntu2.24) ...\nProcessing triggers for man-db (2.10.2-1) ...\n","output_type":"stream"}],"execution_count":7},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}