examples/FTTransformer多分类.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c6a2bd60",
   "metadata": {},
   "source": [
    "# FTTransformer，一个很能打的模型"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "382eb806",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "source": [
    "FTTransformer是一个可以用于结构化(tabular)数据的分类和回归任务的模型。\n",
    "\n",
    "FT 即 Feature Tokenizer的意思，把结构化数据中的离散特征和连续特征都像单词一样编码成一个向量。\n",
    "\n",
    "从而可以像对text数据那样 应用 Transformer对 Tabular数据进行特征抽取。\n",
    "\n",
    "值得注意的是，它对Transformer作了一些微妙的改动以适应 Tabular数据。\n",
    "\n",
    "例如：去除第一个Transformer输入的LayerNorm层，仿照BERT的设计增加了output token(CLS token) 与features token 一起进行进入Transformer参与注意力计算。\n",
    "\n",
    "参考: https://paperswithcode.com/method/ft-transformer\n",
    "\n",
    "本范例我们使用Covertype数据集，它的目标是预测植被覆盖类型，是一个七分类问题。\n",
    "\n",
    "公众号**算法美食屋**后台回复关键词：torchkeras，获取本文notebook源码和所用Covertype数据集下载链接。\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "79c9440c-4e00-4d27-93d6-3f73c40ae149",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('..')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81c4a29a",
   "metadata": {},
   "source": [
    "## 一，准备数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "18d1cca2",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dfdata.shape =  (581012, 13)\n",
      "target_col =  Cover_Type\n",
      "cat_cols =  ['Wilderness_Area', 'Soil_Type']\n",
      "num_cols =  ['Elevation', 'Aspect', 'Slope', '...']\n",
      "len(dftrain) =  371847\n",
      "len(dfval) =  92962\n",
      "len(dftest) =  116203\n"
     ]
    }
   ],
   "source": [
    "import numpy as np \n",
    "import pandas as pd \n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "file_path = \"covertype.parquet\"\n",
    "dfdata = pd.read_parquet(file_path)\n",
    "\n",
    "cat_cols = ['Wilderness_Area', 'Soil_Type']\n",
    "num_cols = ['Elevation','Aspect','Slope','Horizontal_Distance_To_Hydrology',\n",
    " 'Vertical_Distance_To_Hydrology','Horizontal_Distance_To_Roadways',\n",
    " 'Hillshade_9am','Hillshade_Noon','Hillshade_3pm','Horizontal_Distance_To_Fire_Points']\n",
    "target_col = 'Cover_Type'\n",
    "\n",
    "print(\"dfdata.shape = \",dfdata.shape)\n",
    "print(\"target_col = \", target_col)\n",
    "print('cat_cols = ', cat_cols)  \n",
    "print('num_cols = ', num_cols[:3]+['...'])\n",
    "\n",
    "dftmp, dftest_raw = train_test_split(dfdata, random_state=42, test_size=0.2)\n",
    "dftrain_raw, dfval_raw = train_test_split(dftmp, random_state=42, test_size=0.2)\n",
    "\n",
    "print(\"len(dftrain) = \",len(dftrain_raw))\n",
    "print(\"len(dfval) = \",len(dfval_raw))\n",
    "print(\"len(dftest) = \",len(dftest_raw))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "169703c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ce28f8604a624b25a39923053a7235ae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/2 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from torchkeras.tabular import TabularPreprocessor\n",
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "\n",
    "#特征工程\n",
    "pipe = TabularPreprocessor(cat_features = cat_cols, \n",
    "                           embedding_features=cat_cols)\n",
    "encoder = OrdinalEncoder()\n",
    "\n",
    "dftrain = pipe.fit_transform(dftrain_raw.drop(target_col,axis=1))\n",
    "dftrain[target_col] = encoder.fit_transform(\n",
    "    dftrain_raw[target_col].values.reshape(-1,1)).astype(np.int32)\n",
    "\n",
    "dfval = pipe.transform(dfval_raw.drop(target_col,axis=1))\n",
    "dfval[target_col] = encoder.transform(\n",
    "    dfval_raw[target_col].values.reshape(-1,1)).astype(np.int32)\n",
    "\n",
    "dftest = pipe.transform(dftest_raw.drop(target_col,axis=1))\n",
    "dftest[target_col] = encoder.transform(\n",
    "    dftest_raw[target_col].values.reshape(-1,1)).astype(np.int32)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7768b968",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchkeras.tabular import TabularDataset\n",
    "from torch.utils.data import Dataset,DataLoader \n",
    "\n",
    "def get_dataset(dfdata):\n",
    "    return TabularDataset(\n",
    "                data = dfdata,\n",
    "                task = 'multiclass', #regression, binary, multiclass\n",
    "                target = [target_col],\n",
    "                continuous_cols = pipe.get_numeric_features(),\n",
    "                categorical_cols = pipe.get_embedding_features()\n",
    "        )\n",
    "\n",
    "def get_dataloader(ds,batch_size=1024,num_workers=0,shuffle=False):\n",
    "    dl = DataLoader(\n",
    "            ds,\n",
    "            batch_size=batch_size,\n",
    "            shuffle=shuffle,\n",
    "            num_workers=num_workers,\n",
    "            pin_memory=False,\n",
    "        )\n",
    "    return dl \n",
    "    \n",
    "ds_train = get_dataset(dftrain)\n",
    "ds_val = get_dataset(dfval)\n",
    "ds_test = get_dataset(dftest)\n",
    "\n",
    "dl_train = get_dataloader(ds_train,shuffle=True)\n",
    "dl_val = get_dataloader(ds_val,shuffle=False)\n",
    "dl_test = get_dataloader(ds_test,shuffle=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ec009434",
   "metadata": {},
   "outputs": [],
   "source": [
    "for batch in dl_train:\n",
    "    break"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb2f0170",
   "metadata": {},
   "source": [
    "## 二，定义模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "767468ab",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32\n",
      "7\n"
     ]
    }
   ],
   "source": [
    "from torchkeras.tabular.models import FTTransformerConfig,FTTransformerModel\n",
    "\n",
    "model_config = FTTransformerConfig(\n",
    "    task=\"multiclass\",  #regression, binary, multiclass\n",
    "    num_attn_blocks=3\n",
    ")\n",
    "\n",
    "config = model_config.merge_dataset_config(ds_train)\n",
    "net = FTTransformerModel(config = config)\n",
    "\n",
    "#初始化参数\n",
    "net.reset_weights()\n",
    "net.data_aware_initialization(dl_train)\n",
    "\n",
    "print(net.backbone.output_dim)\n",
    "print(net.hparams.output_dim)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7e312c76",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor(2.1680, grad_fn=<AddBackward0>)\n"
     ]
    }
   ],
   "source": [
    "output = net.forward(batch)\n",
    "loss = net.compute_loss(output,batch['target'])\n",
    "print(loss)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22001eaf",
   "metadata": {},
   "source": [
    "## 三，训练模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "14bb01e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchkeras import KerasModel \n",
    "from torchkeras.tabular import StepRunner \n",
    "KerasModel.StepRunner = StepRunner \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "780167c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch \n",
    "from torch import nn \n",
    "class Accuracy(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "\n",
    "        self.correct = nn.Parameter(torch.tensor(0.0),requires_grad=False)\n",
    "        self.total = nn.Parameter(torch.tensor(0.0),requires_grad=False)\n",
    "\n",
    "    def forward(self, preds: torch.Tensor, targets: torch.Tensor):\n",
    "        preds = preds.argmax(dim=-1)\n",
    "        targets = targets.reshape(-1)\n",
    "        m = (preds == targets).sum()\n",
    "        n = targets.shape[0] \n",
    "        self.correct += m \n",
    "        self.total += n\n",
    "        \n",
    "        return m/n\n",
    "\n",
    "    def compute(self):\n",
    "        return self.correct.float() / self.total \n",
    "    \n",
    "    def reset(self):\n",
    "        self.correct -= self.correct\n",
    "        self.total -= self.total\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d3c69f48",
   "metadata": {},
   "outputs": [],
   "source": [
    "keras_model = KerasModel(net,\n",
    "                   loss_fn=None,\n",
    "                   optimizer = torch.optim.AdamW(net.parameters(),lr = 1e-3),\n",
    "                   metrics_dict = {\"acc\":Accuracy()}\n",
    "                   )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "08f4ad31",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[0;31m<<<<<< 🚀 mps is used >>>>>>\u001b[0m\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* background: */\n",
       "    progress::-webkit-progress-bar {background-color: #CDCDCD; width: 100%;}\n",
       "    progress {background-color: #CDCDCD;}\n",
       "\n",
       "    /* value: */\n",
       "    progress::-webkit-progress-value {background-color: #00BFFF  !important;}\n",
       "    progress::-moz-progress-bar {background-color: #00BFFF  !important;}\n",
       "    progress {color: #00BFFF ;}\n",
       "\n",
       "    /* optional */\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #000000;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      <progress value='20' class='' max='20' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      100% [20/20] [11:43]\n",
       "      <br>\n",
       "      ████████████████████100.00% [91/91] [val_loss=0.2238, val_acc=0.9087]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>train_acc</th>\n",
       "      <th>lr</th>\n",
       "      <th>val_loss</th>\n",
       "      <th>val_acc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0.666728</td>\n",
       "      <td>0.724241</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.542972</td>\n",
       "      <td>0.764990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0.522805</td>\n",
       "      <td>0.774832</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.442829</td>\n",
       "      <td>0.811557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0.458113</td>\n",
       "      <td>0.805092</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.383394</td>\n",
       "      <td>0.839461</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0.417264</td>\n",
       "      <td>0.824320</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.351998</td>\n",
       "      <td>0.855382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0.392233</td>\n",
       "      <td>0.834725</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.322582</td>\n",
       "      <td>0.868064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0.373089</td>\n",
       "      <td>0.844092</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.306578</td>\n",
       "      <td>0.874906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0.358037</td>\n",
       "      <td>0.850912</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.295564</td>\n",
       "      <td>0.879639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0.346993</td>\n",
       "      <td>0.855629</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.293008</td>\n",
       "      <td>0.880392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>0.338669</td>\n",
       "      <td>0.859682</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.280957</td>\n",
       "      <td>0.885674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>0.329203</td>\n",
       "      <td>0.863264</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.267493</td>\n",
       "      <td>0.891633</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11</td>\n",
       "      <td>0.320611</td>\n",
       "      <td>0.867548</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.267678</td>\n",
       "      <td>0.891579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>0.314633</td>\n",
       "      <td>0.869277</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.256675</td>\n",
       "      <td>0.895237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>13</td>\n",
       "      <td>0.308088</td>\n",
       "      <td>0.872458</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.245271</td>\n",
       "      <td>0.899185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>14</td>\n",
       "      <td>0.302497</td>\n",
       "      <td>0.874586</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.247016</td>\n",
       "      <td>0.899733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>0.298406</td>\n",
       "      <td>0.877143</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.236148</td>\n",
       "      <td>0.903864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>16</td>\n",
       "      <td>0.293300</td>\n",
       "      <td>0.879114</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.230079</td>\n",
       "      <td>0.907188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>17</td>\n",
       "      <td>0.289149</td>\n",
       "      <td>0.881306</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.230419</td>\n",
       "      <td>0.906456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>18</td>\n",
       "      <td>0.286373</td>\n",
       "      <td>0.881908</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.232185</td>\n",
       "      <td>0.905800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>19</td>\n",
       "      <td>0.282401</td>\n",
       "      <td>0.883976</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.221686</td>\n",
       "      <td>0.910974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>20</td>\n",
       "      <td>0.278942</td>\n",
       "      <td>0.885466</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.223757</td>\n",
       "      <td>0.908662</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    epoch  train_loss  train_acc     lr  val_loss   val_acc\n",
       "0       1    0.666728   0.724241  0.001  0.542972  0.764990\n",
       "1       2    0.522805   0.774832  0.001  0.442829  0.811557\n",
       "2       3    0.458113   0.805092  0.001  0.383394  0.839461\n",
       "3       4    0.417264   0.824320  0.001  0.351998  0.855382\n",
       "4       5    0.392233   0.834725  0.001  0.322582  0.868064\n",
       "5       6    0.373089   0.844092  0.001  0.306578  0.874906\n",
       "6       7    0.358037   0.850912  0.001  0.295564  0.879639\n",
       "7       8    0.346993   0.855629  0.001  0.293008  0.880392\n",
       "8       9    0.338669   0.859682  0.001  0.280957  0.885674\n",
       "9      10    0.329203   0.863264  0.001  0.267493  0.891633\n",
       "10     11    0.320611   0.867548  0.001  0.267678  0.891579\n",
       "11     12    0.314633   0.869277  0.001  0.256675  0.895237\n",
       "12     13    0.308088   0.872458  0.001  0.245271  0.899185\n",
       "13     14    0.302497   0.874586  0.001  0.247016  0.899733\n",
       "14     15    0.298406   0.877143  0.001  0.236148  0.903864\n",
       "15     16    0.293300   0.879114  0.001  0.230079  0.907188\n",
       "16     17    0.289149   0.881306  0.001  0.230419  0.906456\n",
       "17     18    0.286373   0.881908  0.001  0.232185  0.905800\n",
       "18     19    0.282401   0.883976  0.001  0.221686  0.910974\n",
       "19     20    0.278942   0.885466  0.001  0.223757  0.908662"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras_model.fit(\n",
    "    train_data = dl_train,\n",
    "    val_data= dl_val,\n",
    "    ckpt_path='checkpoint',\n",
    "    epochs=20,\n",
    "    patience=10,\n",
    "    monitor=\"val_acc\", \n",
    "    mode=\"max\",\n",
    "    plot = True,\n",
    "    wandb = False\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "092528f3",
   "metadata": {},
   "source": [
    "## 四，评估模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "15693dba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████| 364/364 [00:11<00:00, 31.54it/s, val_acc=0.916, val_loss=0.206]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'val_loss': 0.20587581506633496, 'val_acc': 0.9163365364074707}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras_model.evaluate(dl_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "4f52f49a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████| 91/91 [00:03<00:00, 29.18it/s, val_acc=0.911, val_loss=0.222]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'val_loss': 0.22168648832446927, 'val_acc': 0.910974383354187}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras_model.evaluate(dl_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6050f49a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████| 114/114 [00:04<00:00, 27.62it/s, val_acc=0.909, val_loss=0.223]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'val_loss': 0.2229656313072171, 'val_acc': 0.9093396663665771}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras_model.evaluate(dl_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ced7f651",
   "metadata": {},
   "source": [
    "## 五，使用模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "0c3f0dd5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 114/114 [00:15<00:00,  7.47it/s]\n"
     ]
    }
   ],
   "source": [
    "from tqdm import tqdm \n",
    "net = net.cpu()\n",
    "net.eval()\n",
    "preds = []\n",
    "with torch.no_grad():\n",
    "    for batch in tqdm(dl_test):\n",
    "        preds.append(net.predict(batch))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e330d9f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "yhat_list = [yd.argmax(dim=-1).tolist() for yd in preds]\n",
    "yhat = []\n",
    "for yd in yhat_list:\n",
    "    yhat.extend(yd)\n",
    "yhat = encoder.inverse_transform(np.array(yhat).reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "ac0b9cad",
   "metadata": {},
   "outputs": [],
   "source": [
    "dftest_raw = dftest_raw.rename(columns = {target_col: 'y'})\n",
    "dftest_raw['yhat'] = yhat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7cd623c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.92      0.89      0.91     42557\n",
      "           2       0.91      0.94      0.92     56500\n",
      "           3       0.92      0.89      0.91      7121\n",
      "           4       0.82      0.81      0.81       526\n",
      "           5       0.81      0.73      0.77      1995\n",
      "           6       0.82      0.82      0.82      3489\n",
      "           7       0.91      0.90      0.91      4015\n",
      "\n",
      "    accuracy                           0.91    116203\n",
      "   macro avg       0.87      0.85      0.86    116203\n",
      "weighted avg       0.91      0.91      0.91    116203\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_true = dftest_raw['y'],y_pred = dftest_raw['yhat']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "75b5f2ee",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "\n",
    "# 计算混淆矩阵\n",
    "cm = confusion_matrix(dftest_raw['y'], dftest_raw['yhat'])\n",
    "\n",
    "# 将混淆矩阵转换为DataFrame\n",
    "df_cm = pd.DataFrame(cm, index=['Actual {}'.format(i) for i in range(cm.shape[0])],\n",
    "                     columns=['Predicted {}'.format(i) for i in range(cm.shape[1])])\n",
    "\n",
    "# 使用seaborn绘制混淆矩阵\n",
    "plt.figure(figsize=(10,7))\n",
    "sns.heatmap(df_cm, annot=True, fmt='d', cmap='Blues', cbar=False)\n",
    "plt.title('Confusion Matrix')\n",
    "plt.xlabel('Predicted labels')\n",
    "plt.ylabel('True labels')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bd458bc",
   "metadata": {},
   "source": [
    "## 六，保存模型"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84fd68e5",
   "metadata": {},
   "source": [
    "最佳模型权重已经保存在ckpt_path = 'checkpoint'位置了。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "039d1fd4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<All keys matched successfully>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "net.load_state_dict(torch.load('checkpoint',weights_only=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6af119bc",
   "metadata": {},
   "source": [
    "## 七，与LightGBM对比"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "5a217aa4",
   "metadata": {
    "editable": true,
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/liangyun2/miniforge3/lib/python3.10/site-packages/lightgbm/engine.py:204: UserWarning: Found `n_estimators` in params. Will use it instead of argument\n",
      "  _log_warning(f\"Found `{alias}` in params. Will use it instead of argument\")\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[LightGBM] [Info] Auto-choosing row-wise multi-threading, the overhead of testing was 0.001100 seconds.\n",
      "You can set `force_row_wise=true` to remove the overhead.\n",
      "And if memory is not enough, you can set `force_col_wise=true`.\n",
      "[LightGBM] [Info] Total Bins 2219\n",
      "[LightGBM] [Info] Number of data points in the train set: 371847, number of used features: 12\n",
      "[LightGBM] [Info] Start training from score -1.009334\n",
      "[LightGBM] [Info] Start training from score -0.717530\n",
      "[LightGBM] [Info] Start training from score -2.789050\n",
      "[LightGBM] [Info] Start training from score -5.354306\n",
      "[LightGBM] [Info] Start training from score -4.127223\n",
      "[LightGBM] [Info] Start training from score -3.510637\n",
      "[LightGBM] [Info] Start training from score -3.341909\n",
      "Training until validation scores don't improve for 50 rounds\n",
      "Did not meet early stopping. Best iteration is:\n",
      "[500]\tvalidate's multi_logloss: 0.408343\n",
      "val_score =  0.8321464684494739\n",
      "test_score =  0.8329389086340284\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd \n",
    "import lightgbm as lgb\n",
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "from sklearn.metrics import accuracy_score \n",
    "\n",
    "dftmp, dftest_raw = train_test_split(dfdata, random_state=42, test_size=0.2)\n",
    "dftrain_raw, dfval_raw = train_test_split(dftmp, random_state=42, test_size=0.2)\n",
    "\n",
    "dftrain = dftrain_raw.copy()\n",
    "dfval = dfval_raw.copy()\n",
    "dftest = dftest_raw.copy()\n",
    "\n",
    "target_col = 'Cover_Type'\n",
    "cat_cols = ['Wilderness_Area', 'Soil_Type']\n",
    "\n",
    "encoder = OrdinalEncoder()\n",
    "\n",
    "dftrain[target_col] = encoder.fit_transform(dftrain[target_col].values.reshape(-1,1)) \n",
    "dfval[target_col] = encoder.transform(dfval[target_col].values.reshape(-1,1))\n",
    "dftest[target_col] = encoder.transform(dftest[target_col].values.reshape(-1,1))\n",
    "\n",
    "for col in cat_cols:\n",
    "    dftrain[col] = dftrain[col].astype(int)\n",
    "    dfval[col] = dfval[col].astype(int)\n",
    "    dftest[col] = dftest[col].astype(int)\n",
    "\n",
    "ds_train = lgb.Dataset(dftrain.drop(columns=[target_col]), label=dftrain[target_col],categorical_feature=cat_cols)\n",
    "ds_val = lgb.Dataset(dfval.drop(columns=[target_col]), label=dfval[target_col],categorical_feature=cat_cols)\n",
    "ds_test = lgb.Dataset(dftest.drop(columns=[target_col]), label=dftest[target_col],categorical_feature=cat_cols)\n",
    "\n",
    "\n",
    "import lightgbm as lgb\n",
    "\n",
    "params = {\n",
    "    'n_estimators':500,\n",
    "    'boosting_type': 'gbdt',\n",
    "    'objective':'multiclass',\n",
    "    'num_class': 7,  # 类别数量\n",
    "    'metric': 'multi_logloss', \n",
    "    'learning_rate': 0.01,\n",
    "    'verbose': 1,\n",
    "    'early_stopping_round':50\n",
    "}\n",
    "model = lgb.train(params, ds_train, \n",
    "        valid_sets=[ds_val], \n",
    "        valid_names=['validate']\n",
    "        )\n",
    "\n",
    "y_pred_val = model.predict(dfval.drop(target_col,axis = 1), num_iteration=model.best_iteration)\n",
    "y_pred_val = np.argmax(y_pred_val, axis=1)\n",
    "\n",
    "y_pred_test = model.predict(dftest.drop(target_col,axis = 1), num_iteration=model.best_iteration)\n",
    "y_pred_test = np.argmax(y_pred_test, axis=1)\n",
    "\n",
    "val_score = accuracy_score(dfval[target_col], y_pred_val)\n",
    "test_score = accuracy_score(dftest[target_col], y_pred_test) \n",
    "\n",
    "print('val_score = ',val_score)\n",
    "print('test_score = ' , test_score)"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "gpu",
   "dataSources": [],
   "dockerImageVersionId": 30747,
   "isGpuEnabled": true,
   "isInternetEnabled": true,
   "language": "python",
   "sourceType": "notebook"
  },
  "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": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}