pytorch-ts/examples/m5.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import json\n",
    "from functools import partial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from gluonts.dataset.repository.datasets import get_dataset\n",
    "from gluonts.dataset.util import to_pandas\n",
    "from gluonts.evaluation import Evaluator\n",
    "from gluonts.evaluation.backtest import make_evaluation_predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pts.model.deepar import DeepAREstimator\n",
    "from pts.modules import ZeroInflatedNegativeBinomialOutput\n",
    "from pts import Trainer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "saving time-series into /Users/krasul/.mxnet/gluon-ts/datasets/pts_m5/train/data.json\n",
      "saving time-series into /Users/krasul/.mxnet/gluon-ts/datasets/pts_m5/test/data.json\n"
     ]
    }
   ],
   "source": [
    "dataset = get_dataset(\"pts_m5\", regenerate=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "entry = next(iter(dataset.train))\n",
    "train_series = to_pandas(entry)\n",
    "train_series.plot()\n",
    "plt.grid(which=\"both\")\n",
    "plt.legend([\"train series\"], loc=\"upper left\")\n",
    "plt.title(entry['item_id'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "entry = next(iter(dataset.test))\n",
    "test_series = to_pandas(entry)\n",
    "test_series.plot()\n",
    "plt.axvline(train_series.index[-1], color='r') # end of train dataset\n",
    "plt.grid(which=\"both\")\n",
    "plt.legend([\"test series\", \"end of train series\"], loc=\"upper left\")\n",
    "plt.title(entry['item_id'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recommended prediction horizon: 28\n",
      "Frequency of the time series: D\n"
     ]
    }
   ],
   "source": [
    "print(f\"Recommended prediction horizon: {dataset.metadata.prediction_length}\")\n",
    "print(f\"Frequency of the time series: {dataset.metadata.freq}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "estimator = DeepAREstimator(\n",
    "    distr_output=ZeroInflatedNegativeBinomialOutput(),\n",
    "    cell_type='GRU',\n",
    "    input_size=72,\n",
    "    num_cells=64,\n",
    "    num_layers=3,\n",
    "    dropout_rate=0.2,\n",
    "    use_feat_dynamic_real=True,\n",
    "    use_feat_static_cat=True,\n",
    "    cardinality=[int(cat_feat_info.cardinality) for cat_feat_info in dataset.metadata.feat_static_cat],\n",
    "    embedding_dimension = [4, 4, 4, 4, 16],\n",
    "    prediction_length=dataset.metadata.prediction_length,\n",
    "    context_length=dataset.metadata.prediction_length*2,\n",
    "    freq=dataset.metadata.freq,\n",
    "    scaling=True,\n",
    "    trainer=Trainer(device=device,\n",
    "                    epochs=50,\n",
    "                    learning_rate=1e-3,\n",
    "                    num_batches_per_epoch=120,\n",
    "                    batch_size=256,\n",
    "                    num_workers=8,\n",
    "                   )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "119it [02:44,  1.39s/it, avg_epoch_loss=1.18, epoch=0]\n"
     ]
    }
   ],
   "source": [
    "predictor = estimator.train(dataset.train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "forecast_it, ts_it = make_evaluation_predictions(\n",
    "    dataset=dataset.test,  # test dataset\n",
    "    predictor=predictor,  # predictor\n",
    "    num_samples=100,  # number of sample paths we want for evaluation\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "forecasts = list(forecast_it)\n",
    "tss = list(ts_it)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Running evaluation: 100%|██████████| 30490/30490 [03:42<00:00, 137.06it/s]\n"
     ]
    }
   ],
   "source": [
    "evaluator = Evaluator(num_workers=0)\n",
    "agg_metrics, item_metrics = evaluator(iter(tss), iter(forecasts), num_series=len(dataset.test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\n",
      "    \"MSE\": 6.058231599133125,\n",
      "    \"abs_error\": 898994.0,\n",
      "    \"abs_target_sum\": 1231764.0,\n",
      "    \"abs_target_mean\": 1.4428196598416343,\n",
      "    \"seasonal_error\": 1.1272178349378457,\n",
      "    \"MASE\": 0.9187582800838673,\n",
      "    \"MAPE\": 0.3138891162975238,\n",
      "    \"sMAPE\": 0.721700212685564,\n",
      "    \"OWA\": NaN,\n",
      "    \"MSIS\": 8.565326751268298,\n",
      "    \"QuantileLoss[0.1]\": 241247.00000000003,\n",
      "    \"Coverage[0.1]\": 0.002498477252494963,\n",
      "    \"QuantileLoss[0.2]\": 459438.39999999997,\n",
      "    \"Coverage[0.2]\": 0.011113714098299208,\n",
      "    \"QuantileLoss[0.3]\": 646149.0,\n",
      "    \"Coverage[0.3]\": 0.029292976619969074,\n",
      "    \"QuantileLoss[0.4]\": 795096.4,\n",
      "    \"Coverage[0.4]\": 0.06370121351262709,\n",
      "    \"QuantileLoss[0.5]\": 898994.0,\n",
      "    \"Coverage[0.5]\": 0.11985428477721033,\n",
      "    \"QuantileLoss[0.6]\": 954627.2,\n",
      "    \"Coverage[0.6]\": 0.19450873822799047,\n",
      "    \"QuantileLoss[0.7]\": 939651.9999999999,\n",
      "    \"Coverage[0.7]\": 0.3110961439347796,\n",
      "    \"QuantileLoss[0.8]\": 840218.8,\n",
      "    \"Coverage[0.8]\": 0.4717834887316684,\n",
      "    \"QuantileLoss[0.9]\": 618277.3999999999,\n",
      "    \"Coverage[0.9]\": 0.6762791079042308,\n",
      "    \"RMSE\": 2.4613475169372414,\n",
      "    \"NRMSE\": 1.7059287348547787,\n",
      "    \"ND\": 0.7298427296137897,\n",
      "    \"wQuantileLoss[0.1]\": 0.19585488778694624,\n",
      "    \"wQuantileLoss[0.2]\": 0.37299222903088575,\n",
      "    \"wQuantileLoss[0.3]\": 0.5245720771186688,\n",
      "    \"wQuantileLoss[0.4]\": 0.6454941043901267,\n",
      "    \"wQuantileLoss[0.5]\": 0.7298427296137897,\n",
      "    \"wQuantileLoss[0.6]\": 0.775008199622655,\n",
      "    \"wQuantileLoss[0.7]\": 0.7628506759411705,\n",
      "    \"wQuantileLoss[0.8]\": 0.6821264462997783,\n",
      "    \"wQuantileLoss[0.9]\": 0.5019446907037386,\n",
      "    \"mean_absolute_QuantileLoss\": 710411.1333333333,\n",
      "    \"mean_wQuantileLoss\": 0.576742893389751,\n",
      "    \"MAE_Coverage\": 0.29109687277119217\n",
      "}\n"
     ]
    }
   ],
   "source": [
    "print(json.dumps(agg_metrics, indent=4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "item_metrics.plot(x='MSIS', y='MASE', kind='scatter')\n",
    "plt.grid(which=\"both\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}