energy_price_model_prediction/Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6487331d-ee31-46b1-bc35-a9c215e75de4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting xgboost\n",
      "  Downloading xgboost-3.1.1-py3-none-win_amd64.whl.metadata (2.1 kB)\n",
      "Requirement already satisfied: matplotlib in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (3.10.3)\n",
      "Requirement already satisfied: numpy in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from xgboost) (2.2.4)\n",
      "Requirement already satisfied: scipy in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from xgboost) (1.15.2)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (1.3.2)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (0.12.1)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (4.58.5)\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (1.4.8)\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (25.0)\n",
      "Requirement already satisfied: pillow>=8 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (11.2.1)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (3.2.3)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from matplotlib) (2.9.0.post0)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from python-dateutil>=2.7->matplotlib) (1.17.0)\n",
      "Downloading xgboost-3.1.1-py3-none-win_amd64.whl (72.0 MB)\n",
      "   ---------------------------------------- 0.0/72.0 MB ? eta -:--:--\n",
      "   ---------------------------------------- 0.8/72.0 MB 13.0 MB/s eta 0:00:06\n",
      "   --- ------------------------------------ 6.8/72.0 MB 25.0 MB/s eta 0:00:03\n",
      "   --------- ------------------------------ 16.3/72.0 MB 33.4 MB/s eta 0:00:02\n",
      "   -------------- ------------------------- 26.2/72.0 MB 38.0 MB/s eta 0:00:02\n",
      "   ------------------- -------------------- 35.9/72.0 MB 39.9 MB/s eta 0:00:01\n",
      "   ------------------------- -------------- 45.1/72.0 MB 40.1 MB/s eta 0:00:01\n",
      "   ----------------------------- ---------- 53.5/72.0 MB 40.4 MB/s eta 0:00:01\n",
      "   ---------------------------------- ----- 62.7/72.0 MB 40.6 MB/s eta 0:00:01\n",
      "   ---------------------------------------  71.0/72.0 MB 40.8 MB/s eta 0:00:01\n",
      "   ---------------------------------------- 72.0/72.0 MB 38.5 MB/s  0:00:02\n",
      "Installing collected packages: xgboost\n",
      "Successfully installed xgboost-3.1.1\n"
     ]
    }
   ],
   "source": [
    "!pip install xgboost matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9cfedaeb-32a4-4eec-9e6a-22716054c27b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Start verwerking voor training_set_2024_2025.csv...\n",
      "Stap 1: Feature Engineering...\n",
      "Data opgeschoond. 11554 rijen overgebleven.\n",
      "Stap 2: Definiëren Features (X) en Target (y)...\n",
      "Stap 3: Data splitsen op 2025-01-01 00:00:00...\n",
      "Trainingset: 4001 rijen\n",
      "Testset: 7553 rijen\n",
      "Stap 4: XGBoost Model trainen...\n",
      "Model training voltooid.\n",
      "Stap 5: Model evalueren...\n",
      "\n",
      "--- RESULTAAT ---\n",
      "Gemiddelde Fout (MAE): 0.0122\n",
      "Dit betekent dat het model er gemiddeld 1.22 cent naast zat.\n",
      "Stap 6: Plot genereren...\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stap 7: Belangrijkste Features tonen...\n",
      "\n",
      "Belangrijkste features volgens het model:\n",
      "prijs_1u_geleden     0.771534\n",
      "prijs_24u_geleden    0.037891\n",
      "uur_van_de_dag       0.027163\n",
      "temp_avg_3u          0.021784\n",
      "prijs_avg_6u         0.020315\n",
      "dag_van_de_week      0.016608\n",
      "wind_snelheid        0.016497\n",
      "luchtdruk            0.014006\n",
      "maand                0.013830\n",
      "dag_van_het_jaar     0.012987\n",
      "dtype: float32\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Analyse voltooid ---\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import xgboost as xgb\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# --- CONFIGURATIE ---\n",
    "bestandsnaam = 'training_set_2024_2025.csv'\n",
    "split_datum = '2025-01-01 00:00:00'\n",
    "TARGET = 'gemiddelde_prijs'\n",
    "\n",
    "print(f\"Start verwerking voor {bestandsnaam}...\")\n",
    "\n",
    "# --- STAP 1: DATA LADEN & FEATURE ENGINEERING ---\n",
    "try:\n",
    "    df = pd.read_csv(bestandsnaam)\n",
    "except FileNotFoundError:\n",
    "    print(f\"FOUT: Bestand '{bestandsnaam}' niet gevonden.\")\n",
    "    # Stop de rest van het script als het bestand mist\n",
    "    raise\n",
    "\n",
    "print(\"Stap 1: Feature Engineering...\")\n",
    "df['datum_tijd'] = pd.to_datetime(df['datum_tijd'])\n",
    "df = df.set_index('datum_tijd').sort_index() # Sorteer op datum en maak het de index\n",
    "\n",
    "# Tijd-features\n",
    "df['uur_van_de_dag'] = df.index.hour\n",
    "df['dag_van_de_week'] = df.index.dayofweek\n",
    "df['maand'] = df.index.month\n",
    "df['dag_van_het_jaar'] = df.index.dayofyear\n",
    "\n",
    "# Lag-features (historie)\n",
    "df['prijs_1u_geleden'] = df[TARGET].shift(1)\n",
    "df['prijs_24u_geleden'] = df[TARGET].shift(24)\n",
    "\n",
    "# Rolling-features (trends)\n",
    "df['temp_avg_3u'] = df['temperatuur'].rolling(window=3).mean()\n",
    "df['prijs_avg_6u'] = df[TARGET].rolling(window=6).mean()\n",
    "\n",
    "# Verwijder NaNs die door .shift() en .rolling() zijn ontstaan\n",
    "df_clean = df.dropna()\n",
    "print(f\"Data opgeschoond. {len(df_clean)} rijen overgebleven.\")\n",
    "\n",
    "# --- STAP 2: DEFINIEER FEATURES (X) en TARGET (y) ---\n",
    "print(\"Stap 2: Definiëren Features (X) en Target (y)...\")\n",
    "FEATURES = [col for col in df_clean.columns if col not in [TARGET]]\n",
    "\n",
    "X = df_clean[FEATURES]\n",
    "y = df_clean[TARGET]\n",
    "\n",
    "# --- STAP 3: CHRONOLOGISCHE TRAIN-TEST SPLIT ---\n",
    "print(f\"Stap 3: Data splitsen op {split_datum}...\")\n",
    "train_mask = X.index < split_datum\n",
    "test_mask = X.index >= split_datum\n",
    "\n",
    "X_train, y_train = X[train_mask], y[train_mask]\n",
    "X_test, y_test = X[test_mask], y[test_mask]\n",
    "\n",
    "print(f\"Trainingset: {X_train.shape[0]} rijen\")\n",
    "print(f\"Testset: {X_test.shape[0]} rijen\")\n",
    "\n",
    "# --- STAP 4: MODEL TRAINEN (XGBoost) ---\n",
    "print(\"Stap 4: XGBoost Model trainen...\")\n",
    "# Dit is het model dat we wilden gebruiken\n",
    "xgb_model = xgb.XGBRegressor(\n",
    "    n_estimators=1000,       # Max 1000 \"bomen\"\n",
    "    learning_rate=0.01,      # Leer langzaam\n",
    "    early_stopping_rounds=50 # Stop als de score 50 rondes niet verbetert\n",
    ")\n",
    "\n",
    "# Train het model en gebruik de testset om 'early stopping' toe te passen\n",
    "xgb_model.fit(\n",
    "    X_train, y_train,\n",
    "    eval_set=[(X_test, y_test)],\n",
    "    verbose=False # Zet op True (of 100) als je de voortgang wilt zien\n",
    ")\n",
    "print(\"Model training voltooid.\")\n",
    "\n",
    "# --- STAP 5: EVALUATIE ---\n",
    "print(\"Stap 5: Model evalueren...\")\n",
    "# Genereer voorspellingen op de testset\n",
    "voorspellingen = xgb_model.predict(X_test)\n",
    "\n",
    "# Bereken de fout\n",
    "mae = mean_absolute_error(y_test, voorspellingen)\n",
    "print(f\"\\n--- RESULTAAT ---\")\n",
    "print(f\"Gemiddelde Fout (MAE): {mae:.4f}\")\n",
    "print(f\"Dit betekent dat het model er gemiddeld {mae*100:.2f} cent naast zat.\")\n",
    "\n",
    "# Maak een dataframe met de resultaten voor de plot\n",
    "resultaten = pd.DataFrame({'Echte_Prijs': y_test, 'Voorspelde_Prijs': voorspellingen})\n",
    "\n",
    "# --- STAP 6: VISUALISATIE (PLOT) ---\n",
    "print(\"Stap 6: Plot genereren...\")\n",
    "# Plot de eerste week van de voorspellingen\n",
    "resultaten.head(168).plot(figsize=(15, 6)) # 7 dagen * 24 uur = 168\n",
    "plt.title('Voorspelling vs. Echte Prijzen (Eerste Week van 2025)')\n",
    "plt.ylabel('Gemiddelde Prijs')\n",
    "plt.grid(True)\n",
    "plt.show() # Dit toont de grafiek direct in je notebook\n",
    "\n",
    "# --- STAP 7: FEATURE IMPORTANCE ---\n",
    "print(\"Stap 7: Belangrijkste Features tonen...\")\n",
    "# Wat vond het model het belangrijkst?\n",
    "feature_importance = pd.Series(xgb_model.feature_importances_, index=FEATURES).sort_values(ascending=False)\n",
    "\n",
    "print(\"\\nBelangrijkste features volgens het model:\")\n",
    "print(feature_importance.head(10)) # Top 10\n",
    "\n",
    "# Plot de feature importance\n",
    "feature_importance.head(10).sort_values(ascending=True).plot(\n",
    "    kind='barh', \n",
    "    figsize=(10, 6), \n",
    "    title='Top 10 Belangrijkste Features'\n",
    ")\n",
    "plt.xlabel('Belangrijkheid (Importance)')\n",
    "plt.show() # Toont de grafiek\n",
    "\n",
    "print(\"\\n--- Analyse voltooid ---\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b7fa76a8-9cea-4209-9f4d-80dada3e812b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model succesvol opgeslagen als: price_forecast_model.json\n",
      "Je kunt nu het 'voorspel_prijzen.py' script uitvoeren.\n"
     ]
    }
   ],
   "source": [
    "# Definieer de bestandsnaam\n",
    "model_bestandsnaam = 'price_forecast_model.json'\n",
    "\n",
    "# Sla het model op\n",
    "# (De variabele 'xgb_model' bestaat nog van de vorige cel die je hebt gerund)\n",
    "xgb_model.save_model(model_bestandsnaam)\n",
    "\n",
    "print(f\"Model succesvol opgeslagen als: {model_bestandsnaam}\")\n",
    "print(\"Je kunt nu het 'voorspel_prijzen.py' script uitvoeren.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fe34a15b-0df5-43df-91a6-c66864992f93",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7221a2bd-fcbb-433f-b5aa-c76b010ab581",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Laden van model: price_forecast_model.json\n",
      "Model succesvol geladen.\n",
      "Simuleren van recente historische data...\n",
      "Simuleren van toekomstige weersvoorspelling (volgende 24u)...\n",
      "Feature Engineering toepassen op nieuwe data...\n",
      "\n",
      "--- Voorspelling wordt gemaakt ---\n",
      "                            Voorspelde_Prijs\n",
      "2025-11-12 23:31:02.684646          0.289561\n",
      "2025-11-13 00:31:02.684646          0.304078\n",
      "2025-11-13 01:31:02.684646          0.299526\n",
      "2025-11-13 02:31:02.684646          0.311754\n",
      "2025-11-13 03:31:02.684646          0.303218\n",
      "2025-11-13 04:31:02.684646          0.305205\n",
      "2025-11-13 05:31:02.684646          0.327496\n",
      "2025-11-13 06:31:02.684646          0.317999\n",
      "2025-11-13 07:31:02.684646          0.315852\n",
      "2025-11-13 08:31:02.684646          0.311190\n",
      "2025-11-13 09:31:02.684646          0.303526\n",
      "2025-11-13 10:31:02.684646          0.303281\n",
      "2025-11-13 11:31:02.684646          0.300676\n",
      "2025-11-13 12:31:02.684646          0.300875\n",
      "2025-11-13 13:31:02.684646          0.309934\n",
      "2025-11-13 14:31:02.684646          0.316070\n",
      "2025-11-13 15:31:02.684646          0.311042\n",
      "2025-11-13 16:31:02.684646          0.327661\n",
      "2025-11-13 17:31:02.684646          0.311183\n",
      "2025-11-13 18:31:02.684646          0.309919\n",
      "2025-11-13 19:31:02.684646          0.314924\n",
      "2025-11-13 20:31:02.684646          0.301626\n",
      "2025-11-13 21:31:02.684646          0.289536\n",
      "2025-11-13 22:31:02.684646          0.297345\n",
      "\n",
      "Klaar. Je kunt dit script elke keer runnen als je een nieuwe weersvoorspelling hebt.\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import xgboost as xgb\n",
    "\n",
    "# --- CONFIGURATIE ---\n",
    "MODEL_FILE = 'price_forecast_model.json'\n",
    "TARGET = 'gemiddelde_prijs'\n",
    "\n",
    "# Dit zijn de features die het model MOET hebben (kopieer van je trainingsscript)\n",
    "FEATURES = [\n",
    "    'temperatuur', 'gevoelstemperatuur', 'neerslag', 'wind_richting', \n",
    "    'wind_snelheid', 'bewolking', 'luchtdruk', 'luchtvochtigheid', \n",
    "    'uur_van_de_dag', 'dag_van_de_week', 'maand', 'dag_van_het_jaar', \n",
    "    'prijs_1u_geleden', 'prijs_24u_geleden', 'temp_avg_3u', 'prijs_avg_6u'\n",
    "]\n",
    "\n",
    "print(f\"Laden van model: {MODEL_FILE}\")\n",
    "# --- 1. Laad het getrainde model ---\n",
    "model = xgb.XGBRegressor()\n",
    "model.load_model(MODEL_FILE)\n",
    "print(\"Model succesvol geladen.\")\n",
    "\n",
    "\n",
    "# --- 2. Verzamel de benodigde data ---\n",
    "# Dit is de data die je \"live\" moet ophalen.\n",
    "# We simuleren het nu, maar in de praktijk haal je dit uit je database of een API.\n",
    "\n",
    "# Je hebt de data van de afgelopen 24 uur nodig voor de Lag/Rolling features\n",
    "# (HIER GEBRUIKEN WE NEP-DATA TER ILLUSTRATIE)\n",
    "print(\"Simuleren van recente historische data...\")\n",
    "historische_data = {\n",
    "    'datum_tijd': pd.to_datetime(pd.date_range(end=pd.Timestamp.now(), periods=24, freq='h')),\n",
    "    TARGET: np.random.uniform(0.15, 0.35, 24),\n",
    "    'temperatuur': np.random.uniform(5, 15, 24)\n",
    "}\n",
    "hist_df = pd.DataFrame(historische_data).set_index('datum_tijd')\n",
    "\n",
    "\n",
    "# ==============================================================================\n",
    "# === HIER VUL JE DE ECHTE WEERSVOORSPELLING IN ===\n",
    "# ==============================================================================\n",
    "# Dit is de *toekomstige* weersvoorspelling die je van een API (KNMI, OpenWeather) haalt.\n",
    "# We simuleren nu een voorspelling voor de komende 24 uur.\n",
    "print(\"Simuleren van toekomstige weersvoorspelling (volgende 24u)...\")\n",
    "\n",
    "toekomstige_datums = pd.date_range(start=pd.Timestamp.now() + pd.Timedelta(hours=1), periods=24, freq='h')\n",
    "\n",
    "simulated_forecast_df = pd.DataFrame({\n",
    "    'datum_tijd': toekomstige_datums,\n",
    "    'temperatuur': [10.1, 10.0, 9.8, 9.5, 9.2, 9.0, 8.8, 9.1, 10.2, 11.5, 12.0, 12.5, 13.0, 12.8, 12.0, 11.5, 11.0, 10.5, 10.2, 10.0, 9.8, 9.6, 9.4, 9.2],\n",
    "    'gevoelstemperatuur': [7.1, 7.0, 6.8, 6.5, 6.2, 6.0, 5.8, 6.1, 7.2, 8.5, 9.0, 9.5, 10.0, 9.8, 9.0, 8.5, 8.0, 7.5, 7.2, 7.0, 6.8, 6.6, 6.4, 6.2],\n",
    "    'neerslag': [0.0]*24,\n",
    "    'wind_richting': [210]*24,\n",
    "    'wind_snelheid': [18.0]*24,\n",
    "    'bewolking': [80]*24,\n",
    "    'luchtdruk': [1015]*24,\n",
    "    'luchtvochtigheid': [90]*24,\n",
    "}).set_index('datum_tijd')\n",
    "\n",
    "\n",
    "# --- 3. Combineer historie en toekomst voor Feature Engineering ---\n",
    "# We hebben de staart van de historie nodig om de 'lag' en 'rolling' features \n",
    "# voor de *eerste* toekomstige uren te berekenen.\n",
    "combined_df = pd.concat([hist_df, simulated_forecast_df])\n",
    "\n",
    "# --- 4. Pas EXACT dezelfde Feature Engineering toe ---\n",
    "print(\"Feature Engineering toepassen op nieuwe data...\")\n",
    "# Tijd-features\n",
    "combined_df['uur_van_de_dag'] = combined_df.index.hour\n",
    "combined_df['dag_van_de_week'] = combined_df.index.dayofweek\n",
    "combined_df['maand'] = combined_df.index.month\n",
    "combined_df['dag_van_het_jaar'] = combined_df.index.dayofyear\n",
    "\n",
    "# Lag-features (historie)\n",
    "combined_df['prijs_1u_geleden'] = combined_df[TARGET].shift(1)\n",
    "combined_df['prijs_24u_geleden'] = combined_df[TARGET].shift(24)\n",
    "\n",
    "# Rolling-features (trends)\n",
    "combined_df['temp_avg_3u'] = combined_df['temperatuur'].rolling(window=3).mean()\n",
    "combined_df['prijs_avg_6u'] = combined_df[TARGET].rolling(window=6).mean()\n",
    "\n",
    "# --- 5. Selecteer de data die we willen voorspellen ---\n",
    "# We willen alleen de toekomstige rijen voorspellen.\n",
    "# We pakken alleen de rijen waarvoor we een voorspelling willen doen (de toekomst)\n",
    "X_toekomst = combined_df.loc[toekomstige_datums]\n",
    "\n",
    "# Check of alle benodigde kolommen aanwezig zijn\n",
    "X_voorspelling_input = X_toekomst[FEATURES]\n",
    "\n",
    "# BELANGRIJK: De eerste paar rijen kunnen NaNs hebben (van de rolling features)\n",
    "# We vullen die hier op een simpele manier (forward fill)\n",
    "X_voorspelling_input = X_voorspelling_input.ffill()\n",
    "\n",
    "# --- 6. Maak de Voorspelling! ---\n",
    "print(\"\\n--- Voorspelling wordt gemaakt ---\")\n",
    "voorspelde_prijzen = model.predict(X_voorspelling_input)\n",
    "\n",
    "# --- 7. Toon de resultaten ---\n",
    "resultaat = pd.DataFrame({\n",
    "    'Voorspelde_Prijs': voorspelde_prijzen\n",
    "}, index=toekomstige_datums)\n",
    "\n",
    "print(resultaat)\n",
    "\n",
    "print(\"\\nKlaar. Je kunt dit script elke keer runnen als je een nieuwe weersvoorspelling hebt.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fb9a1193-745d-4cd7-afa6-6909a2136b4c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting holidays\n",
      "  Downloading holidays-0.84-py3-none-any.whl.metadata (50 kB)\n",
      "Requirement already satisfied: python-dateutil in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from holidays) (2.9.0.post0)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\markk\\appdata\\roaming\\python\\python313\\site-packages (from python-dateutil->holidays) (1.17.0)\n",
      "Downloading holidays-0.84-py3-none-any.whl (1.3 MB)\n",
      "   ---------------------------------------- 0.0/1.3 MB ? eta -:--:--\n",
      "   ---------------------------------------- 1.3/1.3 MB 12.5 MB/s  0:00:00\n",
      "Installing collected packages: holidays\n",
      "Successfully installed holidays-0.84\n"
     ]
    }
   ],
   "source": [
    "!pip install holidays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f15f6814-aa0e-402d-afd4-c52c8a8d16ae",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Start training Model 1.5...\n",
      "Data geladen: 11578 rijen.\n",
      "Feature Engineering (v1.5) gestart...\n",
      "Feestdagen feature toegevoegd.\n",
      "One-Hot Encoding toepassen...\n",
      "Rijen vóór opschonen: 11578\n",
      "Rijen ná opschonen: 11554\n",
      "\n",
      "Voorbeeld van de nieuwe 'dag' features:\n",
      "                     dag_van_het_jaar  dag_0  dag_1  dag_2  dag_3  dag_4  \\\n",
      "datum_tijd                                                                 \n",
      "2024-07-18 00:00:00               200  False  False  False   True  False   \n",
      "2024-07-18 01:00:00               200  False  False  False   True  False   \n",
      "2024-07-18 02:00:00               200  False  False  False   True  False   \n",
      "2024-07-18 03:00:00               200  False  False  False   True  False   \n",
      "2024-07-18 04:00:00               200  False  False  False   True  False   \n",
      "\n",
      "                     dag_5  dag_6  \n",
      "datum_tijd                         \n",
      "2024-07-18 00:00:00  False  False  \n",
      "2024-07-18 01:00:00  False  False  \n",
      "2024-07-18 02:00:00  False  False  \n",
      "2024-07-18 03:00:00  False  False  \n",
      "2024-07-18 04:00:00  False  False  \n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import holidays\n",
    "import xgboost as xgb\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "\n",
    "print(\"Start training Model 1.5...\")\n",
    "\n",
    "# --- 1. DATA LADEN ---\n",
    "# We gebruiken de CSV die we al hadden\n",
    "try:\n",
    "    df = pd.read_csv(\"training_set_2024_2025.csv\")\n",
    "    df['datum_tijd'] = pd.to_datetime(df['datum_tijd'])\n",
    "    df = df.set_index('datum_tijd').sort_index()\n",
    "    print(f\"Data geladen: {len(df)} rijen.\")\n",
    "except FileNotFoundError:\n",
    "    print(\"Zorg dat 'training_set_2024_2025.csv' in de map staat.\")\n",
    "    raise\n",
    "\n",
    "# --- 2. FEATURE ENGINEERING (v1.5) ---\n",
    "print(\"Feature Engineering (v1.5) gestart...\")\n",
    "\n",
    "# --- 2a. FEESTDAGEN FEATURE ---\n",
    "# Maak een lijst van Nederlandse feestdagen voor de relevante jaren\n",
    "nl_holidays = holidays.Netherlands(years=[2024, 2025])\n",
    "df['is_feestdag'] = df.index.to_series().apply(lambda x: 1 if x in nl_holidays else 0)\n",
    "print(\"Feestdagen feature toegevoegd.\")\n",
    "\n",
    "# --- 2b. TIJD-FEATURES (Basis) ---\n",
    "df['uur_van_de_dag'] = df.index.hour\n",
    "df['dag_van_de_week'] = df.index.dayofweek # 0=Maandag, 6=Zondag\n",
    "df['maand'] = df.index.month\n",
    "df['dag_van_het_jaar'] = df.index.dayofyear\n",
    "\n",
    "# --- 2c. ONE-HOT ENCODING (DE BELANGRIJKSTE FIX) ---\n",
    "# Converteer 'dag_van_de_week' en 'uur_van_de_dag' naar losse kolommen\n",
    "print(\"One-Hot Encoding toepassen...\")\n",
    "df = pd.get_dummies(df, columns=['dag_van_de_week', 'uur_van_de_dag'], \n",
    "                    prefix=['dag', 'uur'])\n",
    "\n",
    "# --- 2d. Basis Lag/Rolling Features (deze hadden we al) ---\n",
    "df['prijs_1u_geleden'] = df['gemiddelde_prijs'].shift(1)\n",
    "df['prijs_24u_geleden'] = df['gemiddelde_prijs'].shift(24)\n",
    "df['temp_avg_3u'] = df['temperatuur'].rolling(window=3).mean()\n",
    "df['prijs_avg_6u'] = df['gemiddelde_prijs'].rolling(window=6).mean()\n",
    "\n",
    "# --- 2e. Opschonen ---\n",
    "# We verliezen nu maar 24 rijen (van de 'prijs_24u_geleden' lag)\n",
    "print(f\"Rijen vóór opschonen: {len(df)}\")\n",
    "df_clean = df.dropna()\n",
    "print(f\"Rijen ná opschonen: {len(df_clean)}\")\n",
    "\n",
    "# Toon ons de nieuwe features\n",
    "print(\"\\nVoorbeeld van de nieuwe 'dag' features:\")\n",
    "print(df_clean.filter(like='dag_').head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "00c9d26c-4078-4577-9575-d263ca22bf60",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Model wordt getraind met 46 features.\n",
      "Trainingset: 4001 rijen\n",
      "Testset: 7553 rijen\n",
      "\n",
      "Model v1.5 aan het trainen...\n",
      "Training voltooid.\n",
      "\n",
      "Nieuwe Model (v1.5) MAE: 0.0122\n",
      "\n",
      "Model 1.5 opgeslagen als: price_forecast_model_v1_5.json\n"
     ]
    }
   ],
   "source": [
    "# --- 3. DEFINIEER FEATURES (X) en TARGET (y) ---\n",
    "TARGET = 'gemiddelde_prijs'\n",
    "# Automatisch alle kolommen als feature gebruiken\n",
    "FEATURES = [col for col in df_clean.columns if col not in [TARGET]]\n",
    "\n",
    "print(f\"\\nModel wordt getraind met {len(FEATURES)} features.\")\n",
    "\n",
    "X = df_clean[FEATURES]\n",
    "y = df_clean[TARGET]\n",
    "\n",
    "# --- 4. CHRONOLOGISCHE SPLIT ---\n",
    "# We splitsen op 1 jan 2025, net als de vorige keer\n",
    "SPLIT_DATE = '2025-01-01 00:00:00'\n",
    "train_mask = X.index < SPLIT_DATE\n",
    "test_mask = X.index >= SPLIT_DATE\n",
    "\n",
    "X_train, y_train = X[train_mask], y[train_mask]\n",
    "X_test, y_test = X[test_mask], y[test_mask]\n",
    "\n",
    "print(f\"Trainingset: {len(X_train)} rijen\")\n",
    "print(f\"Testset: {len(X_test)} rijen\")\n",
    "\n",
    "# --- 5. MODEL TRAINEN ---\n",
    "xgb_model_v1_5 = xgb.XGBRegressor(\n",
    "    n_estimators=1000,\n",
    "    learning_rate=0.01,\n",
    "    early_stopping_rounds=50\n",
    ")\n",
    "\n",
    "print(\"\\nModel v1.5 aan het trainen...\")\n",
    "xgb_model_v1_5.fit(\n",
    "    X_train, y_train,\n",
    "    eval_set=[(X_test, y_test)],\n",
    "    verbose=False # Zet op 100 om voortgang te zien\n",
    ")\n",
    "print(\"Training voltooid.\")\n",
    "\n",
    "# --- 6. EVALUATIE ---\n",
    "voorspellingen_v1_5 = xgb_model_v1_5.predict(X_test)\n",
    "mae_v1_5 = mean_absolute_error(y_test, voorspellingen_v1_5)\n",
    "print(f\"\\nNieuwe Model (v1.5) MAE: {mae_v1_5:.4f}\")\n",
    "\n",
    "# --- 7. MODEL OPSLAAN ---\n",
    "MODEL_FILE = 'price_forecast_model_v1_5.json'\n",
    "xgb_model_v1_5.save_model(MODEL_FILE)\n",
    "print(f\"\\nModel 1.5 opgeslagen als: {MODEL_FILE}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eb4e1adc-007a-4329-906b-9925a7f11ddf",
   "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": "3.13.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}