model/BITCOIN BiLSTM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from datetime import datetime, timedelta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def rmse(predictions, targets):\n",
    "    return np.sqrt(((predictions - targets) ** 2).mean())\n",
    "\n",
    "def mape(predictions, targets):\n",
    "    return np.mean(np.abs((targets - predictions) / targets)) * 100\n",
    "\n",
    "def weighted_mape(predictions, targets, weights):\n",
    "    errors = np.abs(targets - predictions)\n",
    "    weighted_errors = errors * weights\n",
    "    weighted_mape = np.sum(weighted_errors) / np.sum(np.abs(targets) * weights) * 100\n",
    "    return weighted_mape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "class BiLSTM(nn.Module):\n",
    "    def __init__(self, input_size, hidden_size, num_layers, output_size):\n",
    "        super(BiLSTM, self).__init__()\n",
    "        self.hidden_size = hidden_size\n",
    "        self.num_layers = num_layers\n",
    "        self.lstm = nn.LSTM(input_size, hidden_size, num_layers, batch_first=True, bidirectional=True)\n",
    "        self.fc = nn.Linear(hidden_size*2, output_size)  # Умножаем на 2 из-за двунаправленности\n",
    "\n",
    "    def forward(self, x):\n",
    "        h0 = torch.zeros(self.num_layers*2, x.size(0), self.hidden_size).to(x.device)  # 2 для bidirectional\n",
    "        c0 = torch.zeros(self.num_layers*2, x.size(0), self.hidden_size).to(x.device)\n",
    "\n",
    "        out, _ = self.lstm(x, (h0, c0))\n",
    "        out = self.fc(out[:, -1, :])\n",
    "        return out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('BTC-USD_data.csv')\n",
    "df['Date'] = pd.to_datetime(df['Date'])\n",
    "df.sort_values('Date', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "scaled_data = scaler.fit_transform(df['Adj Close'].values.reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_dataset(data, time_step=1):\n",
    "    X, Y = [], []\n",
    "    for i in range(len(data)-time_step-1):\n",
    "        a = data[i:(i+time_step), 0]\n",
    "        X.append(a)\n",
    "        Y.append(data[i + time_step, 0])\n",
    "    return np.array(X), np.array(Y)\n",
    "\n",
    "time_step = 14"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, y = create_dataset(scaled_data, time_step)\n",
    "\n",
    "# Разделяем данные на обучающую и тестовую выборки\n",
    "test_size = 14  # 14 дней для теста\n",
    "train_size = len(X) - test_size\n",
    "X_train, X_test = X[:train_size], X[train_size:]\n",
    "y_train, y_test = y[:train_size], y[train_size:]\n",
    "\n",
    "X_train = torch.Tensor(X_train).unsqueeze(-1)  # Добавляем размерность\n",
    "X_test = torch.Tensor(X_test).unsqueeze(-1)    # Для совместимости с LSTM\n",
    "y_train = torch.Tensor(y_train)\n",
    "y_test = torch.Tensor(y_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_size = 1  # Входной размер равен 1, так как мы рассматриваем одну фичу - цену\n",
    "hidden_size = 128\n",
    "num_layers = 3\n",
    "output_size = 1\n",
    "\n",
    "model = BiLSTM(input_size, hidden_size, num_layers, output_size)\n",
    "criterion = nn.MSELoss()\n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Задаем путь для сохранения весов\n",
    "weights_path = 'model_weights3.pth'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BiLSTM(\n",
       "  (lstm): LSTM(1, 128, num_layers=3, batch_first=True, bidirectional=True)\n",
       "  (fc): Linear(in_features=256, out_features=1, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Загружаем веса модели\n",
    "model = BiLSTM(input_size, hidden_size, num_layers, output_size)\n",
    "model.load_state_dict(torch.load(weights_path))\n",
    "model.eval()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Loss: 0.0008\n",
      "Test RMSE: 1441.8396\n",
      "Test MAPE: 2.1854%\n",
      "Test Weighted MAPE: 2.1708%\n"
     ]
    }
   ],
   "source": [
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    test_predictions = model(X_test)\n",
    "    test_loss = criterion(test_predictions, y_test.view(-1, 1))\n",
    "\n",
    "    test_predictions = scaler.inverse_transform(test_predictions.cpu().numpy())\n",
    "    y_test = scaler.inverse_transform(y_test.view(-1, 1).cpu().numpy())\n",
    "    \n",
    "    test_rmse = rmse(test_predictions, y_test)\n",
    "    test_mape = mape(test_predictions, y_test)\n",
    "    weights = np.array([1.0, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4])\n",
    "    test_weighted_mape = weighted_mape(test_predictions, y_test, weights)\n",
    "\n",
    "    print(f'Test Loss: {test_loss.item():.4f}')\n",
    "    print(f'Test RMSE: {test_rmse:.4f}')\n",
    "    print(f'Test MAPE: {test_mape:.4f}%')\n",
    "    print(f'Test Weighted MAPE: {test_weighted_mape:.4f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Получите предсказания с использованием вашей модели на тестовых данных или другим способом\n",
    "# Пример:\n",
    "with torch.no_grad():\n",
    "    test_predictions = model(X_test)\n",
    "    adjusted_future_predictions = scaler.inverse_transform(test_predictions.cpu().numpy())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Пример загрузки и предварительной обработки данных\n",
    "import pandas as pd\n",
    "\n",
    "# Загрузка и предобработка данных, например:\n",
    "df = pd.read_csv('BTC-USD_data.csv')\n",
    "# Выполните предварительную обработку данных, например, сортировку и масштабирование\n",
    "\n",
    "# После предварительной обработки сохраните DataFrame в переменной filtered_df\n",
    "filtered_df = df\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Прогнозы с учетом трендов\n",
    "trend_first_day = adjusted_future_predictions[0] - filtered_df['Adj Close'].iloc[-1]\n",
    "trend_14_days = adjusted_future_predictions[-1] - adjusted_future_predictions[0]\n",
    "adjusted_future_predictions = [filtered_df['Adj Close'].iloc[-1] + trend_first_day + (trend_14_days / 14) * i for i in range(14)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Определение переменной end_date\n",
    "end_date = filtered_df['Date'].max()  # Находим максимальную дату в filtered_df\n",
    "\n",
    "# Преобразование 'end_date' в объект datetime\n",
    "end_date = pd.to_datetime(end_date)\n",
    "\n",
    "# Вычисляем дату, которая находится на 3 месяца назад\n",
    "start_date = end_date - pd.DateOffset(months=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Вычисление даты, которая находится на 3 месяца назад\n",
    "start_date = end_date - timedelta(days=90)\n",
    "\n",
    "# Преобразование строк столбца 'Date' в объекты datetime\n",
    "df['Date'] = pd.to_datetime(df['Date'])\n",
    "\n",
    "# Фильтрация данных за последние 3 месяца\n",
    "filtered_df = df[(df['Date'] >= start_date) & (df['Date'] <= end_date)]\n",
    "\n",
    "# Генерация дат для будущих предсказаний\n",
    "future_dates = [filtered_df['Date'].values[-1] + np.timedelta64(i+1, 'D') for i in range(14)]\n",
    "\n",
    "# Расчет тренда на первый день\n",
    "trend_first_day = adjusted_future_predictions[0] - filtered_df['Adj Close'].iloc[-1]\n",
    "\n",
    "# Расчет тренда за 14 дней (как разницу между первым и последним днем)\n",
    "trend_14_days = adjusted_future_predictions[-1] - adjusted_future_predictions[0]\n",
    "\n",
    "# Прогнозы с учетом трендов\n",
    "adjusted_future_predictions = [filtered_df['Adj Close'].iloc[-1] + trend_first_day + (trend_14_days / 14) * i for i in range(14)]\n",
    "\n",
    "# Создание графика\n",
    "plt.figure(figsize=(12,6))\n",
    "plt.plot(filtered_df['Date'], filtered_df['Adj Close'], label='Actual', linewidth=2)\n",
    "plt.plot(future_dates, adjusted_future_predictions, label='Future Predictions (with Trends)', linestyle='--', linewidth=2)\n",
    "\n",
    "plt.title('Bitcoin Price Prediction (Last 3 Months)')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price')\n",
    "plt.xticks(rotation=45)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Прогноз на 14 дней вперед:\n",
      "День 1 (2024-01-20T00:00:00.000000000): Прогноз: 44059.55\n",
      "День 2 (2024-01-21T00:00:00.000000000): Прогноз: 43979.44\n",
      "День 3 (2024-01-22T00:00:00.000000000): Прогноз: 43899.33\n",
      "День 4 (2024-01-23T00:00:00.000000000): Прогноз: 43819.22\n",
      "День 5 (2024-01-24T00:00:00.000000000): Прогноз: 43739.11\n",
      "День 6 (2024-01-25T00:00:00.000000000): Прогноз: 43659.00\n",
      "День 7 (2024-01-26T00:00:00.000000000): Прогноз: 43578.89\n",
      "День 8 (2024-01-27T00:00:00.000000000): Прогноз: 43498.77\n",
      "День 9 (2024-01-28T00:00:00.000000000): Прогноз: 43418.66\n",
      "День 10 (2024-01-29T00:00:00.000000000): Прогноз: 43338.55\n",
      "День 11 (2024-01-30T00:00:00.000000000): Прогноз: 43258.44\n",
      "День 12 (2024-01-31T00:00:00.000000000): Прогноз: 43178.33\n",
      "День 13 (2024-02-01T00:00:00.000000000): Прогноз: 43098.21\n",
      "День 14 (2024-02-02T00:00:00.000000000): Прогноз: 43018.11\n",
      "Тренд на первый день: Рост\n",
      "Тренд на последний день: Падение\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\SoloYolo\\AppData\\Local\\Temp\\ipykernel_7288\\648475885.py:13: DeprecationWarning: Conversion of an array with ndim > 0 to a scalar is deprecated, and will error in future. Ensure you extract a single element from your array before performing this operation. (Deprecated NumPy 1.25.)\n",
      "  print(f\"День {i+1} ({date}): Прогноз: {float(pred):.2f}\")\n"
     ]
    }
   ],
   "source": [
    "# Генерация дат для будущих предсказаний\n",
    "future_dates = [df['Date'].values[-1] + np.timedelta64(i+1, 'D') for i in range(14)]\n",
    "\n",
    "# Расчет тренда на первый день\n",
    "trend_first_day = adjusted_future_predictions[0] - filtered_df['Adj Close'].iloc[-1]\n",
    "\n",
    "# Расчет тренда на последний день\n",
    "trend_last_day = adjusted_future_predictions[-1] - adjusted_future_predictions[0]\n",
    "\n",
    "# Вывод прогнозов\n",
    "print(\"Прогноз на 14 дней вперед:\")\n",
    "for i, (date, pred) in enumerate(zip(future_dates, adjusted_future_predictions)):\n",
    "    print(f\"День {i+1} ({date}): Прогноз: {float(pred):.2f}\")\n",
    "\n",
    "# Вывод тренда на первый и последний день\n",
    "trend_first_day_text = \"Рост\" if trend_first_day > 0 else \"Падение\" if trend_first_day < 0 else \"Нет тренда\"\n",
    "trend_last_day_text = \"Рост\" if trend_last_day > 0 else \"Падение\" if trend_last_day < 0 else \"Нет тренда\"\n",
    "\n",
    "print(f\"Тренд на первый день: {trend_first_day_text}\")\n",
    "print(f\"Тренд на последний день: {trend_last_day_text}\")\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "af",
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}