{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "Sequential_method_to_build_a_neural_network.ipynb",
"provenance": [],
"include_colab_link": true
},
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.6"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"![\"Open](\"https://colab.research.google.com/assets/colab-badge.svg\")
"
]
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:17.305235Z",
"start_time": "2020-09-25T19:49:17.302498Z"
},
"id": "D5_lUQ_JzxNQ"
},
"source": [
"x = [[1,2],[3,4],[5,6],[7,8]]\n",
"y = [[3],[7],[11],[15]]"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:17.611616Z",
"start_time": "2020-09-25T19:49:17.306365Z"
},
"id": "TG0fNwONz6yn"
},
"source": [
"import torch\n",
"import torch.nn as nn\n",
"import numpy as np\n",
"from torch.utils.data import Dataset, DataLoader\n",
"device = 'cuda' if torch.cuda.is_available() else 'cpu'"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:17.616192Z",
"start_time": "2020-09-25T19:49:17.613054Z"
},
"id": "f4-xTYoCz8U9"
},
"source": [
"class MyDataset(Dataset):\n",
" def __init__(self, x, y):\n",
" self.x = torch.tensor(x).float().to(device)\n",
" self.y = torch.tensor(y).float().to(device)\n",
" def __getitem__(self, ix):\n",
" return self.x[ix], self.y[ix]\n",
" def __len__(self): \n",
" return len(self.x)"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:19.209881Z",
"start_time": "2020-09-25T19:49:17.617151Z"
},
"id": "WeBe83XQz9we"
},
"source": [
"ds = MyDataset(x, y)\n",
"dl = DataLoader(ds, batch_size=2, shuffle=True)"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:19.213866Z",
"start_time": "2020-09-25T19:49:19.210841Z"
},
"id": "Vcg57P86z_oF"
},
"source": [
"model = nn.Sequential(\n",
" nn.Linear(2, 8),\n",
" nn.ReLU(),\n",
" nn.Linear(8, 1)\n",
").to(device)"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:21.005946Z",
"start_time": "2020-09-25T19:49:19.215403Z"
},
"id": "7FGa-UWK0BIX",
"outputId": "570c4f77-ef48-46c7-85b9-49b41eec4088",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 85
}
},
"source": [
"!pip install torch_summary\n",
"from torchsummary import summary"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"text": [
"Requirement already satisfied: torch_summary in /home/yyr/anaconda3/lib/python3.7/site-packages (1.4.1)\n",
"\u001b[33mWARNING: You are using pip version 20.2.2; however, version 20.2.3 is available.\n",
"You should consider upgrading via the '/home/yyr/anaconda3/bin/python -m pip install --upgrade pip' command.\u001b[0m\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:21.040105Z",
"start_time": "2020-09-25T19:49:21.011241Z"
},
"id": "UVZlHyXh0Fyd",
"outputId": "1b7c50ea-f954-4a56-8eb0-8095891c943c",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 595
}
},
"source": [
"summary(model, torch.zeros(1,2));"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"text": [
"==========================================================================================\n",
"Layer (type:depth-idx) Output Shape Param #\n",
"==========================================================================================\n",
"├─Linear: 1-1 [-1, 8] 24\n",
"├─ReLU: 1-2 [-1, 8] --\n",
"├─Linear: 1-3 [-1, 1] 9\n",
"==========================================================================================\n",
"Total params: 33\n",
"Trainable params: 33\n",
"Non-trainable params: 0\n",
"Total mult-adds (M): 0.00\n",
"==========================================================================================\n",
"Input size (MB): 0.00\n",
"Forward/backward pass size (MB): 0.00\n",
"Params size (MB): 0.00\n",
"Estimated Total Size (MB): 0.00\n",
"==========================================================================================\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:21.127594Z",
"start_time": "2020-09-25T19:49:21.044743Z"
},
"id": "NDHfUDbW0Lh_",
"outputId": "c0f4620b-4479-4ecc-d3e2-77e7d067b5d8",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
}
},
"source": [
"loss_func = nn.MSELoss()\n",
"from torch.optim import SGD\n",
"opt = SGD(model.parameters(), lr = 0.001)\n",
"import time\n",
"loss_history = []\n",
"start = time.time()\n",
"for _ in range(50):\n",
" for ix, iy in dl:\n",
" opt.zero_grad()\n",
" loss_value = loss_func(model(ix),iy)\n",
" loss_value.backward()\n",
" opt.step()\n",
" loss_history.append(loss_value)\n",
"end = time.time()\n",
"print(end - start)"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"text": [
"0.0754392147064209\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:21.130860Z",
"start_time": "2020-09-25T19:49:21.128656Z"
},
"id": "-Y-j0JeW0WKz"
},
"source": [
"val = [[8,9],[10,11],[1.5,2.5]]\n",
"val = torch.tensor(val).float()"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:21.141201Z",
"start_time": "2020-09-25T19:49:21.132039Z"
},
"id": "KdNMIy4u0Xkt",
"outputId": "1fc3883d-0692-409d-ecb8-d5dd98583285",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 68
}
},
"source": [
"model(val.to(device))"
],
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"tensor([[16.7953],\n",
" [20.6512],\n",
" [ 4.2647]], device='cuda:0', grad_fn=)"
]
},
"metadata": {
"tags": []
},
"execution_count": 10
}
]
},
{
"cell_type": "code",
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-25T19:49:21.145210Z",
"start_time": "2020-09-25T19:49:21.142226Z"
},
"id": "0fgQGXEX0YK_",
"outputId": "a8c760d3-bae4-4954-89cf-8e87d222fc5b"
},
"source": [
"val.sum(-1)"
],
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"tensor([17., 21., 4.])"
]
},
"metadata": {
"tags": []
},
"execution_count": 11
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "5HF-DDp50YLB"
},
"source": [
""
],
"execution_count": null,
"outputs": []
}
]
}