Merge branch 'master' of https://gitee.com/lzhcoooode/machine_learning_projects
This commit is contained in:
[wrh]
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -0,0 +1,863 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "a31d3c40-3593-4ca3-980e-578ee51e171a",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"# 用神经网络进行回归预测"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 1,
|
||||
"id": "738c03b4-3ca0-4a87-9143-53ddea5179be",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"import numpy as np\n",
|
||||
"import pandas as pd\n",
|
||||
"import torch\n",
|
||||
"import torch.nn as nn\n",
|
||||
"import torch.optim as optim\n",
|
||||
"from torch.utils.data import Dataset, DataLoader\n",
|
||||
"from sklearn.preprocessing import StandardScaler\n",
|
||||
"from sklearn.model_selection import train_test_split"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "637063c9-5924-416c-8201-adaae8514ffd",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"设置神经网络超参数，批大小为500，学习率为0.01，一共分别进行100次正向传播和反向传播"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 2,
|
||||
"id": "b909615c-783d-4bfb-acf6-1bc3bacfbb18",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# torch参数\n",
|
||||
"batch_size = 500\n",
|
||||
"lr = 0.01\n",
|
||||
"max_epochs = 100\n",
|
||||
"num_workers = 0\n",
|
||||
"device = torch.device(\"cuda:1\" if torch.cuda.is_available() else \"cpu\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "904cfade-5667-440a-a13c-34fdf57f7b0f",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"定义所需数据集类"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 3,
|
||||
"id": "f6e65156-8535-49c2-8d9d-ecb8639855c9",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"variables = ['number_of_reviews', 'price', 'review_scores_rating']\n",
|
||||
"# 数据集类\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class USDataset(Dataset):\n",
|
||||
" def __init__(self, df):\n",
|
||||
" '''\n",
|
||||
" 初始化\n",
|
||||
" df: 处理后的数据集\n",
|
||||
" '''\n",
|
||||
" self.df = df\n",
|
||||
" self.info = df[['number_of_reviews', 'price']].values\n",
|
||||
" self.target = df['review_scores_rating'].values\n",
|
||||
"\n",
|
||||
" def __getitem__(self, index):\n",
|
||||
" '''\n",
|
||||
" 根据编号返回信息\n",
|
||||
" index: 样本编号\n",
|
||||
" '''\n",
|
||||
" info = self.info[index]\n",
|
||||
" target = self.target[index]\n",
|
||||
" return info, target\n",
|
||||
"\n",
|
||||
" def __len__(self):\n",
|
||||
" '''\n",
|
||||
" 返回数据集样本个数\n",
|
||||
" '''\n",
|
||||
" return len(self.df)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "fc574f2e-6443-48af-9fee-dcf5fb29af58",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"读入数据，由于样本量很大，直接删除有缺失值的样本。"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 4,
|
||||
"id": "17f8aedb-abf1-4622-98ec-d525a779274b",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/html": [
|
||||
"<div>\n",
|
||||
"<style scoped>\n",
|
||||
" .dataframe tbody tr th:only-of-type {\n",
|
||||
" vertical-align: middle;\n",
|
||||
" }\n",
|
||||
"\n",
|
||||
" .dataframe tbody tr th {\n",
|
||||
" vertical-align: top;\n",
|
||||
" }\n",
|
||||
"\n",
|
||||
" .dataframe thead th {\n",
|
||||
" text-align: right;\n",
|
||||
" }\n",
|
||||
"</style>\n",
|
||||
"<table border=\"1\" class=\"dataframe\">\n",
|
||||
" <thead>\n",
|
||||
" <tr style=\"text-align: right;\">\n",
|
||||
" <th></th>\n",
|
||||
" <th>price</th>\n",
|
||||
" <th>number_of_reviews</th>\n",
|
||||
" <th>review_scores_rating</th>\n",
|
||||
" </tr>\n",
|
||||
" </thead>\n",
|
||||
" <tbody>\n",
|
||||
" <tr>\n",
|
||||
" <th>0</th>\n",
|
||||
" <td>120.0</td>\n",
|
||||
" <td>90.0</td>\n",
|
||||
" <td>4.50</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>1</th>\n",
|
||||
" <td>90.0</td>\n",
|
||||
" <td>351.0</td>\n",
|
||||
" <td>4.58</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>2</th>\n",
|
||||
" <td>66.0</td>\n",
|
||||
" <td>67.0</td>\n",
|
||||
" <td>4.52</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>3</th>\n",
|
||||
" <td>33.0</td>\n",
|
||||
" <td>297.0</td>\n",
|
||||
" <td>4.70</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>4</th>\n",
|
||||
" <td>125.0</td>\n",
|
||||
" <td>58.0</td>\n",
|
||||
" <td>4.96</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>...</th>\n",
|
||||
" <td>...</td>\n",
|
||||
" <td>...</td>\n",
|
||||
" <td>...</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>203252</th>\n",
|
||||
" <td>152.0</td>\n",
|
||||
" <td>1.0</td>\n",
|
||||
" <td>4.00</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>203253</th>\n",
|
||||
" <td>45.0</td>\n",
|
||||
" <td>1.0</td>\n",
|
||||
" <td>3.00</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>203254</th>\n",
|
||||
" <td>40.0</td>\n",
|
||||
" <td>1.0</td>\n",
|
||||
" <td>1.00</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>203276</th>\n",
|
||||
" <td>43.0</td>\n",
|
||||
" <td>1.0</td>\n",
|
||||
" <td>5.00</td>\n",
|
||||
" </tr>\n",
|
||||
" <tr>\n",
|
||||
" <th>203308</th>\n",
|
||||
" <td>110.0</td>\n",
|
||||
" <td>1.0</td>\n",
|
||||
" <td>5.00</td>\n",
|
||||
" </tr>\n",
|
||||
" </tbody>\n",
|
||||
"</table>\n",
|
||||
"<p>162429 rows × 3 columns</p>\n",
|
||||
"</div>"
|
||||
],
|
||||
"text/plain": [
|
||||
" price number_of_reviews review_scores_rating\n",
|
||||
"0 120.0 90.0 4.50\n",
|
||||
"1 90.0 351.0 4.58\n",
|
||||
"2 66.0 67.0 4.52\n",
|
||||
"3 33.0 297.0 4.70\n",
|
||||
"4 125.0 58.0 4.96\n",
|
||||
"... ... ... ...\n",
|
||||
"203252 152.0 1.0 4.00\n",
|
||||
"203253 45.0 1.0 3.00\n",
|
||||
"203254 40.0 1.0 1.00\n",
|
||||
"203276 43.0 1.0 5.00\n",
|
||||
"203308 110.0 1.0 5.00\n",
|
||||
"\n",
|
||||
"[162429 rows x 3 columns]"
|
||||
]
|
||||
},
|
||||
"execution_count": 4,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"# 读入数据\n",
|
||||
"df = pd.read_csv('./data/2022-01(US_25).csv', usecols=variables)\n",
|
||||
"df['price'] = df['price'].replace('\\$', '', regex=True)\n",
|
||||
"df['price'] = df['price'].replace('\\,', '', regex=True).astype(float)\n",
|
||||
"df[['number_of_reviews']] = df[['number_of_reviews']].astype(float)\n",
|
||||
"for col in variables:\n",
|
||||
" df[col] = df[col].astype(np.float32)\n",
|
||||
" df = df[np.isnan(df[col]) != 1]\n",
|
||||
"df"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "758d8171-9310-4a51-b9a6-b0fa6274ffe1",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"将数据标准化"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 5,
|
||||
"id": "2cbe7cab-42cd-4b3b-bf11-080cc0a0f4c4",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"df_train:\n",
|
||||
" price number_of_reviews review_scores_rating\n",
|
||||
"0 -0.331270 -0.218031 4.71\n",
|
||||
"1 -0.030076 0.450276 4.99\n",
|
||||
"2 0.425112 -0.603593 5.00\n",
|
||||
"3 -0.324476 -0.590741 2.50\n",
|
||||
"4 0.017481 -0.577889 4.00\n",
|
||||
"... ... ... ...\n",
|
||||
"113695 -0.381091 0.128974 4.98\n",
|
||||
"113696 -0.220303 -0.282292 4.96\n",
|
||||
"113697 -0.165953 -0.603593 3.00\n",
|
||||
"113698 -0.152365 -0.539333 4.00\n",
|
||||
"113699 0.162417 -0.590741 5.00\n",
|
||||
"\n",
|
||||
"[113700 rows x 3 columns]\n",
|
||||
"============================================\n",
|
||||
"df_test:\n",
|
||||
" price number_of_reviews review_scores_rating\n",
|
||||
"0 -0.249743 0.411720 4.84\n",
|
||||
"1 -0.211245 -0.565037 5.00\n",
|
||||
"2 -0.376562 -0.603593 5.00\n",
|
||||
"3 -0.088956 -0.539333 4.33\n",
|
||||
"4 0.517962 0.013306 4.90\n",
|
||||
"... ... ... ...\n",
|
||||
"48724 0.028804 -0.192327 4.91\n",
|
||||
"48725 -0.367503 -0.500777 5.00\n",
|
||||
"48726 -0.360710 -0.436517 4.86\n",
|
||||
"48727 -0.084426 -0.603593 1.00\n",
|
||||
"48728 0.067303 -0.603593 5.00\n",
|
||||
"\n",
|
||||
"[48729 rows x 3 columns]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"# 固定划分测试集和训练集\n",
|
||||
"col_df = df.columns\n",
|
||||
"info = df.iloc[:, :-1].values\n",
|
||||
"target = df.iloc[:, -1].values\n",
|
||||
"# 标准化\n",
|
||||
"stdscaler = StandardScaler()\n",
|
||||
"info_train, info_test, target_train, target_test = train_test_split(\n",
|
||||
" info, target, test_size=0.3, random_state=420)\n",
|
||||
"info_train = stdscaler.fit_transform(info_train)\n",
|
||||
"info_test = stdscaler.transform(info_test)\n",
|
||||
"df_train = np.hstack((info_train, target_train.reshape(-1, 1)))\n",
|
||||
"df_test = np.hstack((info_test, target_test.reshape(-1, 1)))\n",
|
||||
"df_train = pd.DataFrame(df_train, columns=col_df)\n",
|
||||
"df_test = pd.DataFrame(df_test, columns=col_df)\n",
|
||||
"print(\"df_train:\\n\", df_train)\n",
|
||||
"print(\"============================================\")\n",
|
||||
"print(\"df_test:\\n\", df_test)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "1a9661ee-3dc4-4d5c-8a8d-bab3dc299d41",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"设置训练集和验证集"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 6,
|
||||
"id": "a10021bb-4e08-49b2-8269-114b8ce37be6",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# 初始化数据集\n",
|
||||
"train_data = USDataset(df_train)\n",
|
||||
"test_data = USDataset(df_test)\n",
|
||||
"train_loader = DataLoader(train_data, batch_size=batch_size,\n",
|
||||
" num_workers=num_workers, shuffle=True, drop_last=True)\n",
|
||||
"test_loader = DataLoader(test_data, batch_size=batch_size,\n",
|
||||
" num_workers=num_workers, shuffle=False)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "41a15589-9aaf-4541-a603-30a9f433c385",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"设计并初始化网络，其中神经网络是有一个输入层，一个输出层，三个隐藏层的前馈神经网络，激活函数是Sigmoid函数"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 7,
|
||||
"id": "764f28bc-81dd-4924-9039-5eceba54d739",
|
||||
"metadata": {
|
||||
"scrolled": true
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"[Parameter containing:\n",
|
||||
"tensor([[-0.3798, 0.4787],\n",
|
||||
" [-0.6687, 0.3603],\n",
|
||||
" [-0.4871, 0.1701],\n",
|
||||
" [ 0.1127, -0.2635],\n",
|
||||
" [ 0.6846, 0.6578],\n",
|
||||
" [-0.2820, 0.6887],\n",
|
||||
" [ 0.6581, -0.6334],\n",
|
||||
" [ 0.1057, -0.2249],\n",
|
||||
" [ 0.0650, 0.6174],\n",
|
||||
" [ 0.1768, -0.5373],\n",
|
||||
" [-0.5809, 0.6241],\n",
|
||||
" [ 0.1711, 0.3307],\n",
|
||||
" [ 0.0546, 0.6606],\n",
|
||||
" [ 0.0528, 0.1988],\n",
|
||||
" [ 0.6816, -0.2682]], device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([-0.0835, -0.5775, -0.2021, 0.6027, -0.6389, 0.3593, 0.1442, 0.0746,\n",
|
||||
" -0.6356, 0.3029, -0.5204, -0.4724, 0.2451, -0.0103, -0.6133],\n",
|
||||
" device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([[-0.2190, 0.1091, 0.1611, 0.0540, 0.2386, -0.0399, 0.0086, -0.2273,\n",
|
||||
" 0.1421, 0.0286, 0.0507, -0.1721, 0.1504, -0.1714, 0.0113],\n",
|
||||
" [ 0.1312, 0.0581, 0.1805, 0.1385, -0.1394, -0.2539, -0.0047, 0.0753,\n",
|
||||
" 0.2134, 0.1119, 0.0534, -0.1073, 0.2128, 0.1909, 0.2403],\n",
|
||||
" [ 0.0414, 0.1597, -0.2470, -0.0118, 0.1226, -0.1825, 0.2529, 0.2161,\n",
|
||||
" 0.0246, 0.2333, -0.0733, 0.1474, 0.2101, 0.0246, 0.0514],\n",
|
||||
" [ 0.1203, 0.0301, 0.1293, -0.0800, -0.1533, -0.1585, -0.0815, -0.1242,\n",
|
||||
" 0.1415, -0.1520, 0.1287, -0.0701, -0.2560, -0.0094, -0.2361],\n",
|
||||
" [-0.1045, 0.0645, 0.2181, -0.0610, -0.0585, -0.1396, 0.2380, 0.2466,\n",
|
||||
" -0.2495, -0.0690, 0.1078, -0.1838, 0.0231, -0.0264, -0.1335],\n",
|
||||
" [ 0.0328, 0.2338, 0.0761, 0.2368, -0.0891, -0.1581, 0.1605, -0.0699,\n",
|
||||
" 0.1658, -0.0683, -0.2559, -0.0122, 0.1695, 0.2399, -0.1445],\n",
|
||||
" [-0.2489, 0.1559, -0.0083, 0.0980, -0.1653, 0.0678, -0.1833, -0.0217,\n",
|
||||
" -0.0746, 0.0633, -0.0559, 0.0334, 0.1273, -0.1259, -0.1527],\n",
|
||||
" [ 0.2253, -0.0182, 0.1613, -0.1602, -0.2205, -0.2047, 0.2046, -0.0824,\n",
|
||||
" 0.1261, -0.1015, 0.2257, -0.1276, -0.0197, -0.0505, -0.2408],\n",
|
||||
" [ 0.1883, -0.0590, -0.1909, -0.0710, -0.0304, -0.0085, -0.1848, 0.0480,\n",
|
||||
" -0.0447, 0.1001, 0.2330, -0.1368, -0.1478, -0.1676, 0.1942],\n",
|
||||
" [ 0.0554, -0.1686, -0.2217, -0.0343, 0.0218, 0.2300, 0.0303, 0.0711,\n",
|
||||
" -0.0932, -0.1791, 0.0978, -0.1902, -0.2322, -0.1179, -0.1592]],\n",
|
||||
" device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([-0.0498, -0.1153, 0.0559, 0.1391, -0.0231, 0.1752, 0.1824, -0.1185,\n",
|
||||
" -0.1845, 0.1776], device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([[-0.1260, -0.2593, -0.2229, 0.0065, 0.0831, -0.2702, -0.0523, 0.2718,\n",
|
||||
" 0.2547, 0.0824],\n",
|
||||
" [ 0.2193, -0.2204, 0.2573, 0.2804, 0.2895, -0.1308, 0.0558, -0.2521,\n",
|
||||
" -0.2073, -0.2508],\n",
|
||||
" [ 0.2057, -0.2991, 0.0932, -0.1904, 0.2148, -0.1540, 0.0756, -0.1294,\n",
|
||||
" -0.2314, -0.3153],\n",
|
||||
" [ 0.0636, -0.1243, 0.0715, 0.0041, -0.0620, -0.0212, 0.0812, 0.1220,\n",
|
||||
" -0.1704, 0.1767],\n",
|
||||
" [ 0.0371, -0.1627, 0.1664, 0.0266, 0.2154, -0.2726, -0.2169, -0.2568,\n",
|
||||
" 0.0378, -0.1306],\n",
|
||||
" [-0.0030, -0.1174, -0.0169, -0.2378, 0.2668, -0.1097, -0.0550, -0.3043,\n",
|
||||
" 0.1614, 0.2220]], device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([ 0.2263, -0.1245, -0.0600, -0.1303, 0.2717, -0.2886], device='cuda:0',\n",
|
||||
" requires_grad=True), Parameter containing:\n",
|
||||
"tensor([[-0.0118, -0.0705, -0.3342, -0.0979, -0.0048, -0.1048]],\n",
|
||||
" device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([0.3463], device='cuda:0', requires_grad=True)]\n",
|
||||
"==================================================\n",
|
||||
"[Parameter containing:\n",
|
||||
"tensor([[ 0.1109, -0.5283],\n",
|
||||
" [-0.3936, -0.6835],\n",
|
||||
" [ 1.6326, -1.0302],\n",
|
||||
" [ 1.3571, 1.2765],\n",
|
||||
" [-2.0980, 1.2285],\n",
|
||||
" [-0.2051, 0.3710],\n",
|
||||
" [-0.8627, 0.1593],\n",
|
||||
" [ 0.2369, 0.0855],\n",
|
||||
" [ 1.7011, -1.2683],\n",
|
||||
" [ 0.0239, -0.0731],\n",
|
||||
" [ 0.3759, 0.6586],\n",
|
||||
" [ 0.1748, 1.2761],\n",
|
||||
" [ 0.7805, -0.3940],\n",
|
||||
" [-1.7492, -0.3534],\n",
|
||||
" [-0.1294, 2.1254]], device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
|
||||
" device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([[-1.0957, 1.0378, 0.3581, 0.6767, -0.0237, 0.6258, -0.6001, 0.4281,\n",
|
||||
" 0.0584, -0.7526, -0.7916, 0.6348, 0.2372, -0.0493, -1.0156],\n",
|
||||
" [ 0.5839, 0.1032, 1.3735, -0.1767, -1.5280, 0.9311, 1.2752, 0.0721,\n",
|
||||
" -0.0195, -1.0396, -1.2622, -0.2573, 1.1950, 1.2330, -1.1773],\n",
|
||||
" [-1.7392, -0.7176, 0.2932, -1.0405, 1.1690, -0.2543, -0.3711, -0.5608,\n",
|
||||
" 0.2679, -0.1302, 1.1744, 0.6578, 0.2809, -0.4317, 0.3950],\n",
|
||||
" [ 0.4147, -0.5277, -1.3396, 0.6607, 0.3036, 1.3448, -0.6774, 0.2073,\n",
|
||||
" 1.4682, 1.1449, 0.9241, 1.1630, 0.8778, -0.8538, 1.1645],\n",
|
||||
" [-0.2414, -1.1791, -0.9192, -1.3586, 0.7888, -1.2873, -0.7585, 0.1478,\n",
|
||||
" -0.2461, -1.7109, -1.3293, -0.1607, -0.0435, 0.7059, 0.8339],\n",
|
||||
" [ 0.5776, -1.4641, -0.8456, 1.4503, -0.0230, 0.9254, -1.1829, -0.7663,\n",
|
||||
" -0.5102, -1.7416, -0.2776, 1.8553, 0.0287, 1.3071, -0.0577],\n",
|
||||
" [-2.0167, 0.5340, 0.2045, -1.0659, 0.1410, -0.2611, 1.8272, -0.3517,\n",
|
||||
" -1.3150, 0.9499, 1.4403, 2.6355, 1.0081, -0.1915, -1.7888],\n",
|
||||
" [ 0.5569, -0.7237, -0.2397, 1.7227, 1.0570, 0.1441, -0.0228, 1.0863,\n",
|
||||
" 0.5095, -0.1426, -1.6044, -0.4300, 1.5130, 0.7247, -0.1672],\n",
|
||||
" [-0.2202, 0.8956, -0.5899, 0.6426, 1.0715, 2.8716, 0.6675, -0.3307,\n",
|
||||
" 0.4560, -1.6545, 0.6043, 1.4544, 0.3106, 2.0657, 0.4080],\n",
|
||||
" [-1.5387, 0.4301, 1.4037, 1.2928, 0.9105, 0.2546, -1.9276, -0.9019,\n",
|
||||
" -0.0115, 0.4342, -0.1543, 1.0857, 0.7773, 0.0530, 1.0929]],\n",
|
||||
" device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], device='cuda:0',\n",
|
||||
" requires_grad=True), Parameter containing:\n",
|
||||
"tensor([[ 3.2868, 2.1744, -0.4586, 0.3916, 1.0370, 0.2906, -0.6354, -1.7215,\n",
|
||||
" -1.3256, 1.4936],\n",
|
||||
" [ 0.5906, -1.0098, -1.7117, 0.8507, 0.3034, 1.7005, -2.9092, -0.1906,\n",
|
||||
" -1.4068, 2.3292],\n",
|
||||
" [ 0.2435, -0.9280, -0.4463, -0.4837, -2.3409, -0.5112, 0.6242, 0.9315,\n",
|
||||
" 1.6558, -1.3278],\n",
|
||||
" [ 0.4638, 0.5903, -0.7866, 1.3214, 0.6855, 0.7143, 1.2887, -0.2633,\n",
|
||||
" -0.0572, 1.0767],\n",
|
||||
" [ 1.8814, -1.2857, 0.4642, -0.2101, 2.0882, 1.3320, -0.3117, -0.7563,\n",
|
||||
" 0.7099, 0.2467],\n",
|
||||
" [ 0.7781, -1.1896, -0.2684, 1.3948, -0.8702, 0.3317, 0.6835, 0.2728,\n",
|
||||
" 0.3942, 1.5071]], device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([0., 0., 0., 0., 0., 0.], device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([[ 1.5669, 0.3820, 1.0288, 1.5494, -0.9633, -0.8551]],\n",
|
||||
" device='cuda:0', requires_grad=True), Parameter containing:\n",
|
||||
"tensor([0.], device='cuda:0', requires_grad=True)]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"def initialize(self):\n",
|
||||
" '''\n",
|
||||
" 初始化网络参数\n",
|
||||
" '''\n",
|
||||
" for m in self.modules():\n",
|
||||
" if isinstance(m, nn.Linear):\n",
|
||||
" torch.nn.init.normal_(m.weight.data, 0.1)\n",
|
||||
" if m.bias is not None:\n",
|
||||
" torch.nn.init.zeros_(m.bias.data)\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class Net(nn.Module):\n",
|
||||
" '''\n",
|
||||
" 网络结构\n",
|
||||
" '''\n",
|
||||
"\n",
|
||||
" def __init__(self, **kwargs):\n",
|
||||
" super(Net, self).__init__()\n",
|
||||
" self.fc = nn.Sequential(\n",
|
||||
" nn.Linear(2, 15),\n",
|
||||
" nn.Sigmoid(),\n",
|
||||
" nn.Linear(15, 10),\n",
|
||||
" nn.Sigmoid(),\n",
|
||||
" nn.Linear(10, 6),\n",
|
||||
" nn.Sigmoid(),\n",
|
||||
" nn.Linear(6, 1)\n",
|
||||
" )\n",
|
||||
"\n",
|
||||
" def forward(self, x):\n",
|
||||
" x = x.view(-1, 2)\n",
|
||||
" x = self.fc(x)\n",
|
||||
" return x.squeeze(-1)\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"# 初始化网络\n",
|
||||
"model = Net()\n",
|
||||
"model = model.cuda()\n",
|
||||
"print(list(model.parameters()))\n",
|
||||
"initialize(model)\n",
|
||||
"print(\"==================================================\")\n",
|
||||
"print(list(model.parameters()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "20fcccfe-3f82-44c3-bba2-9c66fb15cecd",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"定义损失函数为均方误差损失函数，优化算法为随机梯度下降"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 8,
|
||||
"id": "7b530a85-50b1-4d0f-b47e-14831d0ee85d",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# 损失函数\n",
|
||||
"criterion = nn.MSELoss()\n",
|
||||
"# 优化器\n",
|
||||
"optimizer = optim.SGD(model.parameters(), lr=lr)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 9,
|
||||
"id": "8694737f-afb5-4e58-ad14-3220971b142f",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"def train(epoch):\n",
|
||||
" '''\n",
|
||||
" 训练器\n",
|
||||
"\n",
|
||||
" Parameters\n",
|
||||
" ----------\n",
|
||||
" epoch : int\n",
|
||||
"\n",
|
||||
" Returns\n",
|
||||
" -------\n",
|
||||
" None.\n",
|
||||
"\n",
|
||||
" '''\n",
|
||||
" model.train()\n",
|
||||
" train_loss = 0\n",
|
||||
" for info, target in train_loader:\n",
|
||||
" info = info.cuda()\n",
|
||||
" target = target.cuda()\n",
|
||||
" optimizer.zero_grad()\n",
|
||||
" output = model(info)\n",
|
||||
" loss = criterion(output, target)\n",
|
||||
" loss.backward()\n",
|
||||
" optimizer.step()\n",
|
||||
" train_loss += loss.item()*info.size(0)\n",
|
||||
" train_loss = train_loss/len(train_loader.dataset)\n",
|
||||
" print(\"epoch:%d, train loss:%.4f\" % (epoch, train_loss))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 10,
|
||||
"id": "bb6be6f8-16e6-4c2b-a56c-232b09faa654",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"def validate(epoch):\n",
|
||||
" '''\n",
|
||||
" 验证\n",
|
||||
"\n",
|
||||
" Parameters\n",
|
||||
" ----------\n",
|
||||
" epoch : int\n",
|
||||
"\n",
|
||||
" Returns\n",
|
||||
" -------\n",
|
||||
" None.\n",
|
||||
"\n",
|
||||
" '''\n",
|
||||
" model.eval()\n",
|
||||
" val_loss = 0\n",
|
||||
" with torch.no_grad():\n",
|
||||
" for info, target in test_loader:\n",
|
||||
" info, target = info.cuda(), target.cuda()\n",
|
||||
" output = model(info)\n",
|
||||
" loss = criterion(output, target)\n",
|
||||
" val_loss += loss.item()*info.size(0)\n",
|
||||
" val_loss = val_loss/len(test_loader.dataset)\n",
|
||||
" print(\"epoch:%d, validation loss:%.4f\" %\n",
|
||||
" (epoch, val_loss))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 11,
|
||||
"id": "40532ea8-acfe-469a-aa9f-14d3bec2619d",
|
||||
"metadata": {
|
||||
"scrolled": true
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"epoch:1, train loss:0.5139\n",
|
||||
"epoch:1, validation loss:0.3730\n",
|
||||
"epoch:2, train loss:0.3496\n",
|
||||
"epoch:2, validation loss:0.3698\n",
|
||||
"epoch:3, train loss:0.3471\n",
|
||||
"epoch:3, validation loss:0.3677\n",
|
||||
"epoch:4, train loss:0.3450\n",
|
||||
"epoch:4, validation loss:0.3662\n",
|
||||
"epoch:5, train loss:0.3439\n",
|
||||
"epoch:5, validation loss:0.3652\n",
|
||||
"epoch:6, train loss:0.3429\n",
|
||||
"epoch:6, validation loss:0.3644\n",
|
||||
"epoch:7, train loss:0.3425\n",
|
||||
"epoch:7, validation loss:0.3635\n",
|
||||
"epoch:8, train loss:0.3418\n",
|
||||
"epoch:8, validation loss:0.3631\n",
|
||||
"epoch:9, train loss:0.3417\n",
|
||||
"epoch:9, validation loss:0.3625\n",
|
||||
"epoch:10, train loss:0.3411\n",
|
||||
"epoch:10, validation loss:0.3621\n",
|
||||
"epoch:11, train loss:0.3405\n",
|
||||
"epoch:11, validation loss:0.3620\n",
|
||||
"epoch:12, train loss:0.3406\n",
|
||||
"epoch:12, validation loss:0.3617\n",
|
||||
"epoch:13, train loss:0.3393\n",
|
||||
"epoch:13, validation loss:0.3615\n",
|
||||
"epoch:14, train loss:0.3401\n",
|
||||
"epoch:14, validation loss:0.3616\n",
|
||||
"epoch:15, train loss:0.3396\n",
|
||||
"epoch:15, validation loss:0.3610\n",
|
||||
"epoch:16, train loss:0.3400\n",
|
||||
"epoch:16, validation loss:0.3608\n",
|
||||
"epoch:17, train loss:0.3390\n",
|
||||
"epoch:17, validation loss:0.3607\n",
|
||||
"epoch:18, train loss:0.3392\n",
|
||||
"epoch:18, validation loss:0.3608\n",
|
||||
"epoch:19, train loss:0.3396\n",
|
||||
"epoch:19, validation loss:0.3606\n",
|
||||
"epoch:20, train loss:0.3396\n",
|
||||
"epoch:20, validation loss:0.3604\n",
|
||||
"epoch:21, train loss:0.3396\n",
|
||||
"epoch:21, validation loss:0.3604\n",
|
||||
"epoch:22, train loss:0.3389\n",
|
||||
"epoch:22, validation loss:0.3603\n",
|
||||
"epoch:23, train loss:0.3389\n",
|
||||
"epoch:23, validation loss:0.3601\n",
|
||||
"epoch:24, train loss:0.3387\n",
|
||||
"epoch:24, validation loss:0.3602\n",
|
||||
"epoch:25, train loss:0.3390\n",
|
||||
"epoch:25, validation loss:0.3600\n",
|
||||
"epoch:26, train loss:0.3393\n",
|
||||
"epoch:26, validation loss:0.3600\n",
|
||||
"epoch:27, train loss:0.3385\n",
|
||||
"epoch:27, validation loss:0.3599\n",
|
||||
"epoch:28, train loss:0.3388\n",
|
||||
"epoch:28, validation loss:0.3598\n",
|
||||
"epoch:29, train loss:0.3388\n",
|
||||
"epoch:29, validation loss:0.3598\n",
|
||||
"epoch:30, train loss:0.3388\n",
|
||||
"epoch:30, validation loss:0.3597\n",
|
||||
"epoch:31, train loss:0.3387\n",
|
||||
"epoch:31, validation loss:0.3596\n",
|
||||
"epoch:32, train loss:0.3390\n",
|
||||
"epoch:32, validation loss:0.3596\n",
|
||||
"epoch:33, train loss:0.3384\n",
|
||||
"epoch:33, validation loss:0.3596\n",
|
||||
"epoch:34, train loss:0.3389\n",
|
||||
"epoch:34, validation loss:0.3595\n",
|
||||
"epoch:35, train loss:0.3385\n",
|
||||
"epoch:35, validation loss:0.3594\n",
|
||||
"epoch:36, train loss:0.3381\n",
|
||||
"epoch:36, validation loss:0.3594\n",
|
||||
"epoch:37, train loss:0.3380\n",
|
||||
"epoch:37, validation loss:0.3593\n",
|
||||
"epoch:38, train loss:0.3387\n",
|
||||
"epoch:38, validation loss:0.3593\n",
|
||||
"epoch:39, train loss:0.3388\n",
|
||||
"epoch:39, validation loss:0.3592\n",
|
||||
"epoch:40, train loss:0.3385\n",
|
||||
"epoch:40, validation loss:0.3592\n",
|
||||
"epoch:41, train loss:0.3386\n",
|
||||
"epoch:41, validation loss:0.3593\n",
|
||||
"epoch:42, train loss:0.3384\n",
|
||||
"epoch:42, validation loss:0.3595\n",
|
||||
"epoch:43, train loss:0.3373\n",
|
||||
"epoch:43, validation loss:0.3590\n",
|
||||
"epoch:44, train loss:0.3376\n",
|
||||
"epoch:44, validation loss:0.3591\n",
|
||||
"epoch:45, train loss:0.3385\n",
|
||||
"epoch:45, validation loss:0.3590\n",
|
||||
"epoch:46, train loss:0.3381\n",
|
||||
"epoch:46, validation loss:0.3589\n",
|
||||
"epoch:47, train loss:0.3382\n",
|
||||
"epoch:47, validation loss:0.3591\n",
|
||||
"epoch:48, train loss:0.3381\n",
|
||||
"epoch:48, validation loss:0.3589\n",
|
||||
"epoch:49, train loss:0.3381\n",
|
||||
"epoch:49, validation loss:0.3588\n",
|
||||
"epoch:50, train loss:0.3379\n",
|
||||
"epoch:50, validation loss:0.3587\n",
|
||||
"epoch:51, train loss:0.3383\n",
|
||||
"epoch:51, validation loss:0.3587\n",
|
||||
"epoch:52, train loss:0.3379\n",
|
||||
"epoch:52, validation loss:0.3587\n",
|
||||
"epoch:53, train loss:0.3378\n",
|
||||
"epoch:53, validation loss:0.3587\n",
|
||||
"epoch:54, train loss:0.3379\n",
|
||||
"epoch:54, validation loss:0.3585\n",
|
||||
"epoch:55, train loss:0.3380\n",
|
||||
"epoch:55, validation loss:0.3586\n",
|
||||
"epoch:56, train loss:0.3378\n",
|
||||
"epoch:56, validation loss:0.3585\n",
|
||||
"epoch:57, train loss:0.3378\n",
|
||||
"epoch:57, validation loss:0.3586\n",
|
||||
"epoch:58, train loss:0.3379\n",
|
||||
"epoch:58, validation loss:0.3587\n",
|
||||
"epoch:59, train loss:0.3374\n",
|
||||
"epoch:59, validation loss:0.3583\n",
|
||||
"epoch:60, train loss:0.3380\n",
|
||||
"epoch:60, validation loss:0.3583\n",
|
||||
"epoch:61, train loss:0.3375\n",
|
||||
"epoch:61, validation loss:0.3587\n",
|
||||
"epoch:62, train loss:0.3375\n",
|
||||
"epoch:62, validation loss:0.3583\n",
|
||||
"epoch:63, train loss:0.3377\n",
|
||||
"epoch:63, validation loss:0.3582\n",
|
||||
"epoch:64, train loss:0.3373\n",
|
||||
"epoch:64, validation loss:0.3584\n",
|
||||
"epoch:65, train loss:0.3374\n",
|
||||
"epoch:65, validation loss:0.3581\n",
|
||||
"epoch:66, train loss:0.3373\n",
|
||||
"epoch:66, validation loss:0.3581\n",
|
||||
"epoch:67, train loss:0.3378\n",
|
||||
"epoch:67, validation loss:0.3581\n",
|
||||
"epoch:68, train loss:0.3370\n",
|
||||
"epoch:68, validation loss:0.3581\n",
|
||||
"epoch:69, train loss:0.3370\n",
|
||||
"epoch:69, validation loss:0.3582\n",
|
||||
"epoch:70, train loss:0.3374\n",
|
||||
"epoch:70, validation loss:0.3580\n",
|
||||
"epoch:71, train loss:0.3374\n",
|
||||
"epoch:71, validation loss:0.3580\n",
|
||||
"epoch:72, train loss:0.3375\n",
|
||||
"epoch:72, validation loss:0.3579\n",
|
||||
"epoch:73, train loss:0.3373\n",
|
||||
"epoch:73, validation loss:0.3580\n",
|
||||
"epoch:74, train loss:0.3371\n",
|
||||
"epoch:74, validation loss:0.3583\n",
|
||||
"epoch:75, train loss:0.3376\n",
|
||||
"epoch:75, validation loss:0.3578\n",
|
||||
"epoch:76, train loss:0.3369\n",
|
||||
"epoch:76, validation loss:0.3577\n",
|
||||
"epoch:77, train loss:0.3371\n",
|
||||
"epoch:77, validation loss:0.3577\n",
|
||||
"epoch:78, train loss:0.3373\n",
|
||||
"epoch:78, validation loss:0.3577\n",
|
||||
"epoch:79, train loss:0.3373\n",
|
||||
"epoch:79, validation loss:0.3578\n",
|
||||
"epoch:80, train loss:0.3374\n",
|
||||
"epoch:80, validation loss:0.3577\n",
|
||||
"epoch:81, train loss:0.3368\n",
|
||||
"epoch:81, validation loss:0.3576\n",
|
||||
"epoch:82, train loss:0.3365\n",
|
||||
"epoch:82, validation loss:0.3577\n",
|
||||
"epoch:83, train loss:0.3371\n",
|
||||
"epoch:83, validation loss:0.3576\n",
|
||||
"epoch:84, train loss:0.3371\n",
|
||||
"epoch:84, validation loss:0.3575\n",
|
||||
"epoch:85, train loss:0.3371\n",
|
||||
"epoch:85, validation loss:0.3574\n",
|
||||
"epoch:86, train loss:0.3370\n",
|
||||
"epoch:86, validation loss:0.3574\n",
|
||||
"epoch:87, train loss:0.3371\n",
|
||||
"epoch:87, validation loss:0.3574\n",
|
||||
"epoch:88, train loss:0.3369\n",
|
||||
"epoch:88, validation loss:0.3574\n",
|
||||
"epoch:89, train loss:0.3364\n",
|
||||
"epoch:89, validation loss:0.3573\n",
|
||||
"epoch:90, train loss:0.3367\n",
|
||||
"epoch:90, validation loss:0.3573\n",
|
||||
"epoch:91, train loss:0.3367\n",
|
||||
"epoch:91, validation loss:0.3575\n",
|
||||
"epoch:92, train loss:0.3367\n",
|
||||
"epoch:92, validation loss:0.3573\n",
|
||||
"epoch:93, train loss:0.3372\n",
|
||||
"epoch:93, validation loss:0.3572\n",
|
||||
"epoch:94, train loss:0.3368\n",
|
||||
"epoch:94, validation loss:0.3572\n",
|
||||
"epoch:95, train loss:0.3369\n",
|
||||
"epoch:95, validation loss:0.3571\n",
|
||||
"epoch:96, train loss:0.3366\n",
|
||||
"epoch:96, validation loss:0.3571\n",
|
||||
"epoch:97, train loss:0.3368\n",
|
||||
"epoch:97, validation loss:0.3571\n",
|
||||
"epoch:98, train loss:0.3367\n",
|
||||
"epoch:98, validation loss:0.3571\n",
|
||||
"epoch:99, train loss:0.3367\n",
|
||||
"epoch:99, validation loss:0.3573\n",
|
||||
"epoch:100, train loss:0.3365\n",
|
||||
"epoch:100, validation loss:0.3571\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"# 训练过程\n",
|
||||
"for epoch in range(1, max_epochs+1):\n",
|
||||
" train(epoch)\n",
|
||||
" validate(epoch)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "e3f438d3-813f-4db5-b8e9-da71d54730d4",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": "Python 3 (ipykernel)",
|
||||
"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": "3.10.12"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 5
|
||||
}
|
||||
@@ -0,0 +1,86 @@
|
||||
import tensorflow as tf
|
||||
dataset_path = keras.utils.get_file("housing.data",
|
||||
"https://archive.ics.uci.edu/ml/machine-learning-databases/housing/housing.data")
|
||||
column_names = ['CRIM','ZN','INDUS','CHAS','NOX',
|
||||
'RM', 'AGE', 'DIS','RAD','TAX','PTRATION', 'B', 'LSTAT', 'MEDV']
|
||||
raw_dataset = pd.read_csv(dataset_path, names=column_names,
|
||||
na_values = "?", comment='\t',
|
||||
sep=" ", skipinitialspace=True)
|
||||
dataset = raw_dataset.copy()
|
||||
#下面的函数用以返回最后n行。
|
||||
#dataset.tail(n=10)
|
||||
# 将数据集分为训练集和测试集
|
||||
# p 为训练集所占数据比例
|
||||
p=0.8
|
||||
trainDataset = dataset.sample(frac=p,random_state=0)
|
||||
testDataset = dataset.drop(trainDataset.index)
|
||||
|
||||
import matplotlib.pyplot as plt
|
||||
fig, ax = plt.subplots()
|
||||
x = trainDataset['RM']
|
||||
y = trainDataset['MEDV']
|
||||
ax.scatter(x, y, edgecolors=(0, 0, 0))
|
||||
ax.set_xlabel('RM')
|
||||
ax.set_ylabel('MEDV')
|
||||
plt.show()
|
||||
trainInput = trainDataset['RM']
|
||||
trainTarget = trainDataset['MEDV']
|
||||
testInput = testDataset['RM']
|
||||
testTarget = testDataset['MEDV']
|
||||
|
||||
model = keras.Sequential([
|
||||
layers.Dense(1, use_bias=True, input_shape=(1,))
|
||||
])
|
||||
|
||||
optimizer = tf.keras.optimizers.Adam(
|
||||
learning_rate=0.01, beta_1=0.9, beta_2=0.99, epsilon=1e-05, amsgrad=False,
|
||||
name='Adam')
|
||||
model.compile(loss='mse', optimizer=optimizer, metrics=['mae','mse'])
|
||||
|
||||
|
||||
n_idle_epochs = 100
|
||||
earlyStopping = tf.keras.callbacks.EarlyStopping(monitor='val_loss',
|
||||
patience=n_idle_epochs, min_delta=0.01)
|
||||
class NEPOCHLogger(tf.keras.callbacks.Callback):
|
||||
def __init__(self,per_epoch=100):
|
||||
'''3 实例 4
|
||||
display: Number of batches to wait before outputting loss
|
||||
'''
|
||||
self.seen = 0
|
||||
self.per_epoch = per_epoch
|
||||
def on_epoch_end(self, epoch, logs=None):
|
||||
if epoch % self.per_epoch == 0:
|
||||
print('Epoch {}, loss {:.2f}, val_loss {:.2f},
|
||||
mae {:.2f}, val_mae {:.2f}, mse {:.2f}, val_mse {:.2f}'\
|
||||
.format(epoch, logs['loss'], logs['val_loss'],logs['mae'],
|
||||
logs['val_mae'],logs['mse'], logs['val_mse']))
|
||||
log_display = NEPOCHLogger(per_epoch=100)
|
||||
n_epochs = 2000
|
||||
history = model.fit(
|
||||
trainDataOne, trainLabelOne, batch_size=256,
|
||||
epochs=n_epochs, validation_split = 0.1, verbose=0,
|
||||
callbacks=[earlyStopping,log_display])
|
||||
#打印训练集和验证集MAE
|
||||
import numpy as np
|
||||
import pandas as pd
|
||||
import seaborn as sns
|
||||
mae = np.asarray(history.history['mae'])
|
||||
val_mae = np.asarray(history.history['val_mae'])
|
||||
num_values = (len(mae))
|
||||
values = np.zeros((num_values,2), dtype=float)
|
||||
values[:,0] = mae
|
||||
values[:,1] = val_mae
|
||||
steps = pd.RangeIndex(start=0,stop=num_values)
|
||||
data = pd.DataFrame(values, steps, columns=["mae", "va-mae"])
|
||||
sns.set(style="whitegrid")
|
||||
sns.lineplot(data=data, palette="tab10", linewidth=2.5)
|
||||
|
||||
predictions = model.predict(testInput).flatten()
|
||||
a = plt.axes(aspect='equal')
|
||||
plt.scatter(predictions, testTarget, edgecolors=(0, 0, 0))
|
||||
plt.xlabel('True Values')
|
||||
plt.ylabel('Predictions')
|
||||
lims = [0, 50]
|
||||
plt.xlim(lims)
|
||||
plt.ylim(lims)
|
||||
_ = plt.plot(lims, lims)
|
||||
File diff suppressed because one or more lines are too long
Binary file not shown.
File diff suppressed because it is too large
Load Diff
@@ -1,7 +1,17 @@
|
||||
## kaggle共享单车租用数据，通过历史租车数据并结合天气数据来预测租车需求
|
||||
#通过历史租车数据并结合天气数据来预测租车需求
|
||||
|
||||
通过历史租车数据并结合天气数据来预测租车需求
|
||||
### 字段说明：
|
||||
### 目录结构
|
||||
```
|
||||
|-src
|
||||
| |-main.py
|
||||
|-test.csv
|
||||
|-train.csv
|
||||
|-readme.md
|
||||
|-main.ipynb
|
||||
```
|
||||
src/main.py为前馈型神经网络对结果的预测
|
||||
|
||||
### 数据集字段说明：
|
||||
|
||||
datetime：时间
|
||||
|
||||
@@ -0,0 +1,135 @@
|
||||
import torch
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
import os
|
||||
import time
|
||||
import torch.optim as optim
|
||||
import matplotlib.pyplot as plt
|
||||
|
||||
train_data = pd.read_csv('../train.csv')
|
||||
test_data = pd.read_csv('../test.csv')
|
||||
#print(train_data)
|
||||
datetime=train_data['datetime'].values
|
||||
season=train_data['season'].values
|
||||
holiday=train_data['holiday'].values
|
||||
workingday=train_data['workingday'].values
|
||||
weather=train_data['weather'].values
|
||||
temp=train_data['temp'].values
|
||||
atemp=train_data['atemp'].values
|
||||
humidity=train_data['humidity'].values
|
||||
windspeed=train_data['windspeed'].values
|
||||
casual=train_data['casual'].values
|
||||
registered=train_data['registered'].values
|
||||
|
||||
count=train_data['count'].values
|
||||
|
||||
'''
|
||||
转换datetime,去掉不必要的符号和0
|
||||
'''
|
||||
def datetime_convert(datetime):
|
||||
for i in range(len(datetime)):
|
||||
datetime[i]=datetime[i].replace("-", "").replace(" ", "").replace(":00:00","")
|
||||
datetime[i]=int(datetime[i])
|
||||
return datetime
|
||||
datetime_convert(datetime)
|
||||
train_data['datetime']=datetime
|
||||
datetime_convert(test_data['datetime'].values)
|
||||
test_data['datetime']=test_data['datetime'].values
|
||||
'''
|
||||
标准化数据
|
||||
'''
|
||||
def normalize(ndarry):
|
||||
ndarry=(ndarry-ndarry.mean())/ndarry.std()
|
||||
return ndarry
|
||||
|
||||
for i in range(12):
|
||||
train_data.iloc[:,i]=normalize(train_data.iloc[:,i])
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
X=torch.tensor(train_data.iloc[:,:9].to_numpy().astype(float),dtype=torch.float32)
|
||||
Y=torch.tensor(train_data.iloc[:,9:12].to_numpy().astype(float),dtype=torch.float32)
|
||||
X_test=torch.tensor(test_data.iloc[:,:9].to_numpy().astype(float),dtype=torch.float32)
|
||||
torch_train_dataset = torch.utils.data.TensorDataset(X, Y)
|
||||
batch_size = 10
|
||||
torch.manual_seed(seed=2023)
|
||||
|
||||
training, vertification = torch.utils.data.random_split(torch_train_dataset, [10000, 886], )
|
||||
training_data = torch.utils.data.DataLoader(training,batch_size=batch_size,shuffle=True)
|
||||
vertification_data = torch.utils.data.DataLoader(training,batch_size=batch_size,shuffle=True)
|
||||
|
||||
feature_number = 9 # 设置特征数目
|
||||
out_prediction = 3 # 设置输出数目
|
||||
learning_rate = 0.01 # 设置学习率
|
||||
epochs = 10 # 设置训练代数
|
||||
|
||||
|
||||
class Net(torch.nn.Module):
|
||||
def __init__(self, n_feature, n_output, n_neuron1, n_neuron2,
|
||||
n_layer): # n_feature为特征数目，这个数字不能随便取,n_output为特征对应的输出数目，也不能随便取
|
||||
self.n_feature = n_feature
|
||||
self.n_output = n_output
|
||||
self.n_neuron1 = n_neuron1
|
||||
self.n_neuron2 = n_neuron2
|
||||
self.n_layer = n_layer
|
||||
super(Net, self).__init__()
|
||||
self.input_layer = torch.nn.Linear(self.n_feature, self.n_neuron1) # 输入层
|
||||
self.hidden1 = torch.nn.Linear(self.n_neuron1, self.n_neuron2) # 1类隐藏层
|
||||
self.hidden2 = torch.nn.Linear(self.n_neuron2, self.n_neuron2) # 2类隐藏
|
||||
self.predict = torch.nn.Linear(self.n_neuron2, self.n_output) # 输出层
|
||||
|
||||
def forward(self, x):
|
||||
'''定义前向传递过程'''
|
||||
out = self.input_layer(x)
|
||||
out = torch.relu(out) # 使用relu函数非线性激活
|
||||
out = self.hidden1(out)
|
||||
out = torch.relu(out)
|
||||
for i in range(self.n_layer):
|
||||
out = self.hidden2(out)
|
||||
out = torch.relu(out)
|
||||
out = self.predict( # 回归问题最后一层不需要激活函数
|
||||
out
|
||||
) # 除去feature_number与out_prediction不能随便取，隐藏层数与其他神经元数目均可以适当调整以得到最佳预测效果
|
||||
#print(out.shape)
|
||||
return out
|
||||
|
||||
net = Net(n_feature=feature_number,
|
||||
n_output=out_prediction,
|
||||
n_layer=1,
|
||||
n_neuron1=20,
|
||||
n_neuron2=20) # 这里直接确定了隐藏层数目以及神经元数目，实际操作中需要遍历
|
||||
optimizer = optim.Adam(net.parameters(), learning_rate) # 使用Adam算法更新参数
|
||||
criteon = torch.nn.MSELoss() # 误差计算公式，回归问题采用均方误差
|
||||
average_losses=[]
|
||||
for epoch in range(epochs): # 整个数据集迭代次数
|
||||
net.train() # 启动训练模式
|
||||
for batch_idx, (data, target) in enumerate(training_data):
|
||||
logits = net.forward(data) # 前向计算结果（预测结果）
|
||||
loss = criteon(logits, target) # 计算损失
|
||||
optimizer.zero_grad() # 梯度清零
|
||||
loss.backward() # 后向传递过程
|
||||
optimizer.step() # 优化权重与偏差矩阵
|
||||
#print(logits)
|
||||
|
||||
logit = [] # 这个是验证集，可以根据验证集的结果进行调参，这里根据验证集的结果选取最优的神经网络层数与神经元数目
|
||||
target = []
|
||||
net.eval() # 启动测试模式
|
||||
for data, targets in vertification: # 输出验证集的平均误差
|
||||
logits = net.forward(data).detach().numpy()
|
||||
targets = targets.detach().numpy()
|
||||
target.append(targets[0])
|
||||
logit.append(logits[0])
|
||||
average_loss = criteon(torch.tensor(logit), torch.tensor(target))
|
||||
average_losses.append(average_loss)
|
||||
print("epoch={},the average loss is {}".format(epoch,average_loss))
|
||||
|
||||
Y_test=net.forward(X_test)
|
||||
print(Y_test)
|
||||
xx=range(epochs)
|
||||
yy=average_losses
|
||||
plt.xlim(0,epochs)
|
||||
plt.ylim(0,10)
|
||||
plt.plot(xx,yy)
|
||||
plt.show()
|
||||
|
Can't render this file because it is too large.
|
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
Reference in New Issue
Block a user