AttMoE-NASA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Reference:**\n",
    "\n",
    "Chen, D., & Zhou, X. (2024). AttMoE: Attention with Mixture of Experts for remaining useful life prediction of lithium-ion batteries. Journal of Energy Storage, 84, 110780."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda3\\envs\\torch_11\\lib\\site-packages\\numpy\\_distributor_init.py:32: UserWarning: loaded more than 1 DLL from .libs:\n",
      "D:\\Anaconda3\\envs\\torch_11\\lib\\site-packages\\numpy\\.libs\\libopenblas.WCDJNK7YVMPZQ2ME2ZZHJJRJ3JIKNDB7.gfortran-win_amd64.dll\n",
      "D:\\Anaconda3\\envs\\torch_11\\lib\\site-packages\\numpy\\.libs\\libopenblas.XWYDX2IKJW2NMTWSFYNGFUWKQU3LYTCZ.gfortran-win_amd64.dll\n",
      "  stacklevel=1)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import random\n",
    "import math\n",
    "import os\n",
    "import scipy.io\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torchvision\n",
    "import transformers\n",
    "%matplotlib inline\n",
    "\n",
    "from math import sqrt\n",
    "from datetime import datetime\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 1. get device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:0\n"
     ]
    }
   ],
   "source": [
    "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
    "print(device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 2. load data from mat data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert str to datatime\n",
    "def convert_to_time(hmm):\n",
    "    year, month, day, hour, minute, second = int(hmm[0]), int(hmm[1]), int(hmm[2]), int(hmm[3]), int(hmm[4]), int(hmm[5])\n",
    "    return datetime(year=year, month=month, day=day, hour=hour, minute=minute, second=second)\n",
    "\n",
    "\n",
    "# load .mat file\n",
    "def loadMat(matfile):\n",
    "    data = scipy.io.loadmat(matfile)\n",
    "    filename = matfile.split(\".\")[0]\n",
    "    col = data[filename]\n",
    "    col = col[0][0][0][0]\n",
    "    size = col.shape[0]\n",
    "\n",
    "    data = []\n",
    "    for i in range(size):\n",
    "        k = list(col[i][3][0].dtype.fields.keys())\n",
    "        d1, d2 = {}, {}\n",
    "        if str(col[i][0][0]) != 'impedance':\n",
    "            for j in range(len(k)):\n",
    "                t = col[i][3][0][0][j][0];\n",
    "                l = [t[m] for m in range(len(t))]\n",
    "                d2[k[j]] = l\n",
    "        d1['type'], d1['temp'], d1['time'], d1['data'] = str(col[i][0][0]), int(col[i][1][0]), str(convert_to_time(col[i][2][0])), d2\n",
    "        data.append(d1)\n",
    "\n",
    "    return data\n",
    "\n",
    "\n",
    "# get capacity data of batteries\n",
    "def getBatteryCapacity(Battery):\n",
    "    cycle, capacity = [], []\n",
    "    i = 1\n",
    "    for Bat in Battery:\n",
    "        if Bat['type'] == 'discharge':\n",
    "            capacity.append(Bat['data']['Capacity'][0])\n",
    "            cycle.append(i)\n",
    "            i += 1\n",
    "    return [cycle, capacity]\n",
    "\n",
    "\n",
    "# get battery data\n",
    "def getBatteryValues(Battery, Type='charge'):\n",
    "    data=[]\n",
    "    for Bat in Battery:\n",
    "        if Bat['type'] == Type:\n",
    "            data.append(Bat['data'])\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load Dataset B0005.mat ...\n",
      "Load Dataset B0006.mat ...\n",
      "Load Dataset B0007.mat ...\n",
      "Load Dataset B0018.mat ...\n"
     ]
    }
   ],
   "source": [
    "Battery_list = ['B0005', 'B0006', 'B0007', 'B0018']\n",
    "\n",
    "Battery = {}\n",
    "for name in Battery_list:\n",
    "    print('Load Dataset ' + name + '.mat ...')\n",
    "    path = name + '.mat'\n",
    "    data = loadMat(path)\n",
    "    Battery[name] = getBatteryCapacity(data)              # 放电时的容量数据"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# or step 2. load from .npy file\n",
    "\n",
    "**If the above read fails, please load the .npy file.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Battery_list = ['B0005', 'B0006', 'B0007', 'B0018']\n",
    "Battery = np.load('datasets/NASA/NASA.npy', allow_pickle=True)\n",
    "Battery = Battery.item()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 3. capacity figure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1df2d4817b8>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, figsize=(12, 8))\n",
    "color_list = ['b:', 'g--', 'r-.', 'c.']\n",
    "c = 0\n",
    "for name,color in zip(Battery_list, color_list):\n",
    "    df_result = Battery[name]\n",
    "    ax.plot(df_result[0], df_result[1], color, label=name)\n",
    "ax.set(xlabel='Discharge cycles', ylabel='Capacity (Ah)', title='Capacity degradation at ambient temperature of 24°C')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 4. define functions for data processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_sequences(text, window_size):\n",
    "    #text:list of capacity\n",
    "    x, y = [],[]\n",
    "    for i in range(len(text) - window_size):\n",
    "        sequence = text[i:i+window_size]\n",
    "        target = text[i+1:i+1+window_size]\n",
    "\n",
    "        x.append(sequence)\n",
    "        y.append(target)\n",
    "        \n",
    "    return np.array(x), np.array(y)\n",
    "\n",
    "\n",
    "def split_dataset(data_sequence, train_ratio=0.0, capacity_threshold=0.0):\n",
    "    if capacity_threshold > 0:\n",
    "        max_capacity = max(data_sequence)\n",
    "        capacity = max_capacity * capacity_threshold\n",
    "        point = [i for i in range(len(data_sequence)) if data_sequence[i] < capacity]\n",
    "    else:\n",
    "        point = int(train_ratio + 1)\n",
    "        if 0 < train_ratio <= 1:\n",
    "            point = int(len(data_sequence) * train_ratio)\n",
    "    train_data, test_data = data_sequence[:point], data_sequence[point:]\n",
    "    return train_data, test_data\n",
    "\n",
    "\n",
    "# 留一评估：一组数据为测试集，其他所有数据全部拿来训练\n",
    "def get_train_test(data_dict, name, window_size=8):\n",
    "    data_sequence=data_dict[name][1]\n",
    "    train_data, test_data = data_sequence[:window_size+1], data_sequence[window_size+1:]\n",
    "    train_x, train_y = build_sequences(text=train_data, window_size=window_size)\n",
    "    for k, v in data_dict.items():\n",
    "        if k != name:\n",
    "            data_x, data_y = build_sequences(text=v[1], window_size=window_size)\n",
    "            train_x, train_y = np.r_[train_x, data_x], np.r_[train_y, data_y]\n",
    "            \n",
    "    return train_x, train_y, list(train_data), list(test_data)\n",
    "\n",
    "\n",
    "def relative_error(y_test, y_predict, threshold):\n",
    "    true_re, pred_re = len(y_test), 0\n",
    "    for i in range(len(y_test)-1):\n",
    "        if y_test[i] <= threshold >= y_test[i+1]:\n",
    "            true_re = i - 1\n",
    "            break\n",
    "    for i in range(len(y_predict)-1):\n",
    "        if y_predict[i] <= threshold:\n",
    "            pred_re = i - 1\n",
    "            break\n",
    "    return abs(true_re - pred_re)/true_re\n",
    "\n",
    "\n",
    "def evaluation(y_test, y_predict):\n",
    "    mae = mean_absolute_error(y_test, y_predict)\n",
    "    mse = mean_squared_error(y_test, y_predict)\n",
    "    rmse = sqrt(mean_squared_error(y_test, y_predict))\n",
    "    return mae, rmse\n",
    "    \n",
    "    \n",
    "def setup_seed(seed):\n",
    "    np.random.seed(seed)\n",
    "    random.seed(seed) \n",
    "    os.environ['PYTHONHASHSEED'] = str(seed)\n",
    "    torch.manual_seed(seed) \n",
    "    if torch.cuda.is_available():\n",
    "        torch.cuda.manual_seed(seed)\n",
    "        torch.cuda.manual_seed_all(seed)\n",
    "        torch.backends.cudnn.benchmark = False\n",
    "        torch.backends.cudnn.deterministic = True"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 5. build net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mixture_of_experts import MoE\n",
    "\n",
    "class Attention(nn.Module):\n",
    "    def __init__(self, feature_size, hidden_dim, nhead=4, dropout=0.0):\n",
    "        super(Attention, self).__init__()\n",
    "        self.query = nn.Linear(feature_size, hidden_dim)\n",
    "        self.key = nn.Linear(feature_size, hidden_dim)\n",
    "        self.value = nn.Linear(feature_size, hidden_dim)\n",
    "        self.attn = nn.MultiheadAttention(embed_dim=hidden_dim, num_heads=nhead, dropout=dropout, batch_first=True)\n",
    "        \n",
    "    def forward(self, x):\n",
    "        query, key, value = self.query(x), self.key(x), self.value(x)\n",
    "        out, _ = self.attn(query, key, value)\n",
    "        return out\n",
    "\n",
    "\n",
    "class AttMoE(nn.Module):\n",
    "    def __init__(self, feature_size=16, hidden_dim=8, num_layers=1, nhead=4, dropout=0., dropout_rate=0.2, \n",
    "                 num_experts=8, device='cpu'):\n",
    "        super(AttMoE, self).__init__()\n",
    "        self.feature_size, self.hidden_dim = feature_size, hidden_dim\n",
    "        self.dropout = nn.Dropout(dropout_rate)\n",
    "        self.cell = Attention(feature_size=feature_size, hidden_dim=hidden_dim, nhead=nhead, dropout=dropout)\n",
    "        self.linear = nn.Linear(hidden_dim, 1)\n",
    "        \n",
    "        experts = nn.Linear(hidden_dim, hidden_dim)\n",
    "        # create moe layers based on the number of experts\n",
    "        self.moe = MoE(dim=hidden_dim, num_experts=num_experts, experts=experts)\n",
    "        self.moe = self.moe.to(device)\n",
    " \n",
    "    def forward(self, x): \n",
    "        out = self.dropout(x)\n",
    "        out = self.cell(x)   # cell 输出 shape (batch_size, seq_len=1, feature_size)\n",
    "        out,_ = self.moe(out)\n",
    "        out = out.reshape(-1, self.hidden_dim) # (batch_size, hidden_dim)\n",
    "        out = self.linear(out)  # shape: (batch_size, 1)\n",
    "        \n",
    "        return out"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 6. define train function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train(lr=0.01, feature_size=8, hidden_dim=32, num_layers=1, nhead=8, weight_decay=0.0, EPOCH=1000, \n",
    "          seed=0, dropout=0.0, metric='re', num_experts=8, device='cpu'):\n",
    "    score_list, result_list = [], []\n",
    "    for i in range(4):\n",
    "        name = Battery_list[i]\n",
    "        window_size = feature_size\n",
    "        train_x, train_y, train_data, test_data = get_train_test(Battery, name, window_size)\n",
    "        train_size = len(train_x)\n",
    "        #print('sample size: {}'.format(train_size))\n",
    "\n",
    "        setup_seed(seed)\n",
    "        model = AttMoE(feature_size=feature_size, hidden_dim=hidden_dim, num_layers=num_layers, \n",
    "                        nhead=nhead, dropout=dropout, num_experts=num_experts, device=device)\n",
    "        model = model.to(device)\n",
    "        optimizer = torch.optim.Adam(model.parameters(), lr=lr, weight_decay=weight_decay)\n",
    "        criterion = nn.MSELoss()\n",
    "\n",
    "        test_x = train_data.copy()\n",
    "        loss_list, y_ = [0], []\n",
    "        mae, rmse, re = 1, 1, 1\n",
    "        score_, score = [1],[1]\n",
    "        for epoch in range(EPOCH):\n",
    "            X = np.reshape(train_x/Rated_Capacity,(-1, 1, feature_size)).astype(np.float32)#(batch_size, seq_len, input_size)\n",
    "            y = np.reshape(train_y[:,-1]/Rated_Capacity,(-1,1)).astype(np.float32)# shape 为 (batch_size, 1)\n",
    "\n",
    "            X, y = torch.from_numpy(X), torch.from_numpy(y)\n",
    "            X, y = X.to(device), y.to(device)\n",
    "            output = model(X)\n",
    "            output = output.reshape(-1, 1)\n",
    "            loss = criterion(output, y)\n",
    "            optimizer.zero_grad()              # clear gradients for this training step\n",
    "            loss.backward()                    # backpropagation, compute gradients\n",
    "            optimizer.step()                   # apply gradients\n",
    "\n",
    "            if (epoch + 1)%100 == 0:\n",
    "                test_x = train_data.copy() #每100次重新预测一次\n",
    "                point_list = []\n",
    "                while (len(test_x) - len(train_data)) < len(test_data):\n",
    "                    x = np.reshape(np.array(test_x[-feature_size:])/Rated_Capacity,(-1, 1, feature_size)).astype(np.float32)\n",
    "                    x = torch.from_numpy(x) #(batch_size,feature_size=1,input_size)\n",
    "                    x = x.to(device)\n",
    "                    pred = model(x) # 测试集 模型预测#pred shape为(batch_size=1, feature_size=1)\n",
    "                    next_point = pred.cpu().data.numpy()[0,0] * Rated_Capacity\n",
    "                    test_x.append(next_point)#测试值加入原来序列用来继续预测下一个点\n",
    "                    point_list.append(next_point)#保存输出序列最后一个点的预测值\n",
    "                y_.append(point_list)#保存本次预测所有的预测值\n",
    "                loss_list.append(loss)\n",
    "                mae, rmse = evaluation(y_test=test_data, y_predict=y_[-1])\n",
    "                re = relative_error(y_test=test_data, y_predict=y_[-1], threshold=Rated_Capacity*0.7)\n",
    "                #print('epoch:{:<2d} | loss:{:<6.4f} | MAE:{:<6.4f} | RMSE:{:<6.4f} | RE:{:<6.4f}'.format(epoch, loss, mae, rmse, re))\n",
    "            if metric == 're':\n",
    "                score = [re]\n",
    "            elif metric == 'mae':\n",
    "                score = [mae]\n",
    "            elif metric == 'rmse':\n",
    "                score = [rmse]\n",
    "            else:\n",
    "                score = [re, mae, rmse]\n",
    "            if (loss < 1e-3) and (score_[0] < score[0]):\n",
    "                break\n",
    "            score_ = score.copy()\n",
    "            \n",
    "        score_list.append(score_)\n",
    "        result_list.append(y_[-1])\n",
    "    return score_list, result_list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 7. grid-search to find key parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lr:0.0001, hidden_dim:32, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.5295\n",
      "mae mean: 0.1714\n",
      "rmse mean: 0.1822\n",
      "===================================================================\n",
      "lr:0.0001, hidden_dim:32, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.4866\n",
      "mae mean: 0.1552\n",
      "rmse mean: 0.1668\n",
      "===================================================================\n",
      "lr:0.0001, hidden_dim:64, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2286\n",
      "mae mean: 0.0809\n",
      "rmse mean: 0.0921\n",
      "===================================================================\n",
      "lr:0.0001, hidden_dim:64, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2000\n",
      "mae mean: 0.0760\n",
      "rmse mean: 0.0872\n",
      "===================================================================\n",
      "lr:0.0001, hidden_dim:128, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2743\n",
      "mae mean: 0.0810\n",
      "rmse mean: 0.0912\n",
      "===================================================================\n",
      "lr:0.0001, hidden_dim:128, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2572\n",
      "mae mean: 0.0810\n",
      "rmse mean: 0.0912\n",
      "===================================================================\n",
      "lr:0.0001, hidden_dim:256, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2744\n",
      "mae mean: 0.0785\n",
      "rmse mean: 0.0886\n",
      "===================================================================\n",
      "lr:0.0001, hidden_dim:256, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2680\n",
      "mae mean: 0.0767\n",
      "rmse mean: 0.0871\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:32, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2514\n",
      "mae mean: 0.0794\n",
      "rmse mean: 0.0899\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:32, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2712\n",
      "mae mean: 0.0779\n",
      "rmse mean: 0.0875\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:64, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2781\n",
      "mae mean: 0.0838\n",
      "rmse mean: 0.0944\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:64, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2830\n",
      "mae mean: 0.0833\n",
      "rmse mean: 0.0937\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:128, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2833\n",
      "mae mean: 0.0809\n",
      "rmse mean: 0.0905\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:128, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2955\n",
      "mae mean: 0.0813\n",
      "rmse mean: 0.0905\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:256, num_experts:4\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2573\n",
      "mae mean: 0.0747\n",
      "rmse mean: 0.0833\n",
      "===================================================================\n",
      "lr:0.001, hidden_dim:256, num_experts:8\n",
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2407\n",
      "mae mean: 0.0693\n",
      "rmse mean: 0.0779\n",
      "===================================================================\n"
     ]
    }
   ],
   "source": [
    "Rated_Capacity = 2.0\n",
    "feature_size = 16\n",
    "EPOCH = 2000\n",
    "num_layers = 1\n",
    "nhead = 4\n",
    "weight_decay = 0.0\n",
    "dropout = 0.0\n",
    "metric = 'all'\n",
    "\n",
    "for lr in [1e-4, 1e-3]:\n",
    "    for hidden_dim in [32, 64, 128, 256]:\n",
    "        for num_experts in [4, 8]:\n",
    "            print('lr:{}, hidden_dim:{}, num_experts:{}'.format(lr, hidden_dim, num_experts))\n",
    "\n",
    "            SCORE = []\n",
    "            for seed in range(5):\n",
    "                print('seed:{}'.format(seed))\n",
    "                score_list, _ = train(lr=lr, feature_size=feature_size, hidden_dim=hidden_dim, num_layers=num_layers, \n",
    "                                      nhead=nhead, weight_decay=weight_decay, EPOCH=EPOCH, seed=seed, dropout=dropout, \n",
    "                                      metric=metric, num_experts=num_experts, device=device)\n",
    "                print('------------------------------------------------------------------')\n",
    "                for s in score_list:\n",
    "                    SCORE.append(s)\n",
    "\n",
    "            if metric != 'all':\n",
    "                print(metric + ' mean: {:<6.4f}'.format(np.mean(np.array(SCORE))))\n",
    "            else:\n",
    "                mlist = ['re', 'mae', 'rmse']\n",
    "                for i in range(3):\n",
    "                    s = [line[i] for line in SCORE]\n",
    "                    print(mlist[i] + ' mean: {:<6.4f}'.format(np.mean(np.array(s))))\n",
    "            print('===================================================================')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# step 8. set the obtained best parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "seed:0\n",
      "------------------------------------------------------------------\n",
      "seed:1\n",
      "------------------------------------------------------------------\n",
      "seed:2\n",
      "------------------------------------------------------------------\n",
      "seed:3\n",
      "------------------------------------------------------------------\n",
      "seed:4\n",
      "------------------------------------------------------------------\n",
      "re mean: 0.2000\n",
      "mae mean: 0.0760\n",
      "rmse mean: 0.0872\n",
      "===================================================================\n"
     ]
    }
   ],
   "source": [
    "lr = 0.0001    # learning rate\n",
    "hidden_dim = 64\n",
    "num_experts = 8\n",
    "\n",
    "Rated_Capacity = 2.0\n",
    "feature_size = 16\n",
    "EPOCH = 2000\n",
    "num_layers = 1\n",
    "nhead = 4\n",
    "weight_decay = 0.0\n",
    "dropout = 0.0\n",
    "metric = 'all'\n",
    "\n",
    "\n",
    "SCORE = []\n",
    "for seed in range(5):\n",
    "    print('seed:{}'.format(seed))\n",
    "    score_list, _ = train(lr=lr, feature_size=feature_size, hidden_dim=hidden_dim, num_layers=num_layers, \n",
    "                          nhead=nhead, weight_decay=weight_decay, EPOCH=EPOCH, seed=seed, dropout=dropout, \n",
    "                          metric=metric, num_experts=num_experts, device=device)\n",
    "    print('------------------------------------------------------------------')\n",
    "    for s in score_list:\n",
    "        SCORE.append(s)\n",
    "\n",
    "if metric != 'all':\n",
    "    print(metric + ' mean: {:<6.4f}'.format(np.mean(np.array(SCORE))))\n",
    "else:\n",
    "    mlist = ['re', 'mae', 'rmse']\n",
    "    for i in range(3):\n",
    "        s = [line[i] for line in SCORE]\n",
    "        print(mlist[i] + ' mean: {:<6.4f}'.format(np.mean(np.array(s))))\n",
    "print('===================================================================')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "torch_11",
   "language": "python",
   "name": "torch_11"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}