1068 lines
736 KiB
Plaintext
1068 lines
736 KiB
Plaintext
|
{
|
|||
|
|
"nbformat": 4,
|
|||
|
|
"nbformat_minor": 0,
|
|||
|
|
"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": "3.6.3"
|
|||
|
|
},
|
|||
|
|
"toc": {
|
|||
|
|
"base_numbering": 1,
|
|||
|
|
"nav_menu": {},
|
|||
|
|
"number_sections": true,
|
|||
|
|
"sideBar": true,
|
|||
|
|
"skip_h1_title": false,
|
|||
|
|
"title_cell": "Table of Contents",
|
|||
|
|
"title_sidebar": "Contents",
|
|||
|
|
"toc_cell": false,
|
|||
|
|
"toc_position": {},
|
|||
|
|
"toc_section_display": true,
|
|||
|
|
"toc_window_display": true
|
|||
|
|
},
|
|||
|
|
"varInspector": {
|
|||
|
|
"cols": {
|
|||
|
|
"lenName": 16,
|
|||
|
|
"lenType": 16,
|
|||
|
|
"lenVar": 40
|
|||
|
|
},
|
|||
|
|
"kernels_config": {
|
|||
|
|
"python": {
|
|||
|
|
"delete_cmd_postfix": "",
|
|||
|
|
"delete_cmd_prefix": "del ",
|
|||
|
|
"library": "var_list.py",
|
|||
|
|
"varRefreshCmd": "print(var_dic_list())"
|
|||
|
|
},
|
|||
|
|
"r": {
|
|||
|
|
"delete_cmd_postfix": ") ",
|
|||
|
|
"delete_cmd_prefix": "rm(",
|
|||
|
|
"library": "var_list.r",
|
|||
|
|
"varRefreshCmd": "cat(var_dic_list()) "
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
"types_to_exclude": [
|
|||
|
|
"module",
|
|||
|
|
"function",
|
|||
|
|
"builtin_function_or_method",
|
|||
|
|
"instance",
|
|||
|
|
"_Feature"
|
|||
|
|
],
|
|||
|
|
"window_display": false
|
|||
|
|
},
|
|||
|
|
"colab": {
|
|||
|
|
"name": "Cheng - TrendWeightedFTS.ipynb",
|
|||
|
|
"provenance": [],
|
|||
|
|
"include_colab_link": true
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
"cells": [
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "view-in-github",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"<a href=\"https://colab.research.google.com/github/PYFTS/notebooks/blob/master/Cheng%20-%20TrendWeightedFTS.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "MlD52EhraLen",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"# Trend Weighted Fuzzy Time Series by Cheng, Chen and Wu (2009)\n",
|
|||
|
|
"\n",
|
|||
|
|
"C.-H. Cheng, Y.-S. Chen, and Y.-L. Wu, “Forecasting innovation diffusion of products using trend-weighted fuzzy time-series model,” \n",
|
|||
|
|
"Expert Syst. Appl., vol. 36, no. 2, pp. 1826–1832, 2009."
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "eD3INDn2aLev",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"## Environment Setup"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "yCH4DKhSaLez",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### Library install/update"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "iQG1IaIEaLe3",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 211
|
|||
|
|
},
|
|||
|
|
"outputId": "9495c610-b91a-441b-f262-ef4b22a43808"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"!pip3 install -U git+https://github.com/PYFTS/pyFTS"
|
|||
|
|
],
|
|||
|
|
"execution_count": 1,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"Collecting git+https://github.com/PYFTS/pyFTS\n",
|
|||
|
|
" Cloning https://github.com/PYFTS/pyFTS to /tmp/pip-req-build-gnxa9jyh\n",
|
|||
|
|
" Running command git clone -q https://github.com/PYFTS/pyFTS /tmp/pip-req-build-gnxa9jyh\n",
|
|||
|
|
"Building wheels for collected packages: pyFTS\n",
|
|||
|
|
" Building wheel for pyFTS (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
|
|||
|
|
" Created wheel for pyFTS: filename=pyFTS-1.6-cp36-none-any.whl size=197025 sha256=8b66da66f433712d5c383b263b25ba70a8ed0f21447bbfa4910f327a57311074\n",
|
|||
|
|
" Stored in directory: /tmp/pip-ephem-wheel-cache-6yizw1gx/wheels/e7/32/a9/230470113df5a73242a5a6d05671cb646db97abf14bbce2644\n",
|
|||
|
|
"Successfully built pyFTS\n",
|
|||
|
|
"Installing collected packages: pyFTS\n",
|
|||
|
|
"Successfully installed pyFTS-1.6\n"
|
|||
|
|
],
|
|||
|
|
"name": "stdout"
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "ojhK_4hPaLfE",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### External libraries import"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "XDYBqSKoaLfH",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 34
|
|||
|
|
},
|
|||
|
|
"outputId": "5d8f1683-f3d9-42fa-c42b-3fe4738864b5"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"import warnings\n",
|
|||
|
|
"warnings.filterwarnings('ignore')\n",
|
|||
|
|
"\n",
|
|||
|
|
"import numpy as np\n",
|
|||
|
|
"import pandas as pd\n",
|
|||
|
|
"import matplotlib.pylab as plt\n",
|
|||
|
|
"import seaborn as sns\n",
|
|||
|
|
"\n",
|
|||
|
|
"%pylab inline"
|
|||
|
|
],
|
|||
|
|
"execution_count": 6,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"Populating the interactive namespace from numpy and matplotlib\n"
|
|||
|
|
],
|
|||
|
|
"name": "stdout"
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "LfQjfxheaLfO",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### Common pyFTS imports "
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "HJlbJ-CGaLfP",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.common import Util as cUtil\n",
|
|||
|
|
"from pyFTS.benchmarks import benchmarks as bchmk, Util as bUtil\n",
|
|||
|
|
"from pyFTS.partitioners import Util as pUtil\n",
|
|||
|
|
"\n",
|
|||
|
|
"from pyFTS.models import cheng\n"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "Qim33pXQaLfV",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"## Common data transformations"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "fxA_4i4gaLfX",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.common import Transformations\n",
|
|||
|
|
"\n",
|
|||
|
|
"tdiff = Transformations.Differential(1)\n",
|
|||
|
|
"\n",
|
|||
|
|
"boxcox = Transformations.BoxCox(0)"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "1fzMnkb9aLfe",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"## Datasets"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "hzVKwoAdaLfh",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### Data Loading"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "jSGmlkhDaLfi",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.data import TAIEX, NASDAQ, SP500\n",
|
|||
|
|
"\n",
|
|||
|
|
"dataset_names = [\"TAIEX\", \"SP500\",\"NASDAQ\"]\n",
|
|||
|
|
"\n",
|
|||
|
|
"def get_dataset(name):\n",
|
|||
|
|
" if dataset_name == \"TAIEX\":\n",
|
|||
|
|
" return TAIEX.get_data()\n",
|
|||
|
|
" elif dataset_name == \"SP500\":\n",
|
|||
|
|
" return SP500.get_data()[11500:16000]\n",
|
|||
|
|
" elif dataset_name == \"NASDAQ\":\n",
|
|||
|
|
" return NASDAQ.get_data()\n",
|
|||
|
|
"\n",
|
|||
|
|
"\n",
|
|||
|
|
"train_split = 2000\n",
|
|||
|
|
"test_length = 200"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "nU7HTU-qaLfp",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### Visualization"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "flCVsUxdaLfq",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {},
|
|||
|
|
"outputId": "99d071c7-295d-4267-fd19-73b41cc64078"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"fig, ax = plt.subplots(nrows=2, ncols=3, figsize=[10,5])\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
" dataset_diff = tdiff.apply(dataset)\n",
|
|||
|
|
"\n",
|
|||
|
|
" ax[0][count].plot(dataset)\n",
|
|||
|
|
" ax[1][count].plot(dataset_diff)\n",
|
|||
|
|
" ax[0][count].set_title(dataset_name)"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "display_data",
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAE/CAYAAAD2ee+mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkEP7iTB/N7Eez
|
|||
|
|
"text/plain": [
|
|||
|
|
"<matplotlib.figure.Figure at 0x7fdb3b6d4eb8>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
}
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "ziev89ZoaLfv",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### Statistics"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "Ds8pkjy9aLfw",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {},
|
|||
|
|
"outputId": "4400eda8-cf84-4262-e273-5b742861321a"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from statsmodels.tsa.stattools import adfuller\n",
|
|||
|
|
"\n",
|
|||
|
|
"rows =[]\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" row = [dataset_name]\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
" result = adfuller(dataset)\n",
|
|||
|
|
" row.extend([result[0],result[1]])\n",
|
|||
|
|
" row.extend([value for key, value in result[4].items()])\n",
|
|||
|
|
" rows.append(row)\n",
|
|||
|
|
" \n",
|
|||
|
|
"pd.DataFrame(rows,columns=['Dataset','ADF Statistic','p-value','Cr. Val. 1%','Cr. Val. 5%','Cr. Val. 10%'])\n"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "execute_result",
|
|||
|
|
"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>Dataset</th>\n",
|
|||
|
|
" <th>ADF Statistic</th>\n",
|
|||
|
|
" <th>p-value</th>\n",
|
|||
|
|
" <th>Cr. Val. 1%</th>\n",
|
|||
|
|
" <th>Cr. Val. 5%</th>\n",
|
|||
|
|
" <th>Cr. Val. 10%</th>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" </thead>\n",
|
|||
|
|
" <tbody>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>0</th>\n",
|
|||
|
|
" <td>TAIEX</td>\n",
|
|||
|
|
" <td>-2.656728</td>\n",
|
|||
|
|
" <td>0.081830</td>\n",
|
|||
|
|
" <td>-3.431601</td>\n",
|
|||
|
|
" <td>-2.862093</td>\n",
|
|||
|
|
" <td>-2.567064</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>1</th>\n",
|
|||
|
|
" <td>SP500</td>\n",
|
|||
|
|
" <td>-1.747171</td>\n",
|
|||
|
|
" <td>0.406987</td>\n",
|
|||
|
|
" <td>-3.431811</td>\n",
|
|||
|
|
" <td>-2.862186</td>\n",
|
|||
|
|
" <td>-2.567114</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>2</th>\n",
|
|||
|
|
" <td>NASDAQ</td>\n",
|
|||
|
|
" <td>0.476224</td>\n",
|
|||
|
|
" <td>0.984132</td>\n",
|
|||
|
|
" <td>-3.432022</td>\n",
|
|||
|
|
" <td>-2.862279</td>\n",
|
|||
|
|
" <td>-2.567163</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" </tbody>\n",
|
|||
|
|
"</table>\n",
|
|||
|
|
"</div>"
|
|||
|
|
],
|
|||
|
|
"text/plain": [
|
|||
|
|
" Dataset ADF Statistic p-value Cr. Val. 1% Cr. Val. 5% Cr. Val. 10%\n",
|
|||
|
|
"0 TAIEX -2.656728 0.081830 -3.431601 -2.862093 -2.567064\n",
|
|||
|
|
"1 SP500 -1.747171 0.406987 -3.431811 -2.862186 -2.567114\n",
|
|||
|
|
"2 NASDAQ 0.476224 0.984132 -3.432022 -2.862279 -2.567163"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
},
|
|||
|
|
"execution_count": 6
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "YrAsqYvIaLf2",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"## Partitioning\n",
|
|||
|
|
"\n",
|
|||
|
|
"The best number of partitions of the Universe of Discourse is an optimization problem. The know more about partitioning schemes please look on the Partitioners notebook. To know more about benchmarking look on the Benchmarks notebook."
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "QLhm0lrlaLf4",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.partitioners import Grid, Util as pUtil\n",
|
|||
|
|
"from pyFTS.benchmarks import benchmarks as bchmk\n",
|
|||
|
|
"from pyFTS.models import chen\n",
|
|||
|
|
"\n",
|
|||
|
|
"tag = 'cheng_partitioning'\n",
|
|||
|
|
"_type = 'point'\n",
|
|||
|
|
"\n",
|
|||
|
|
"for dataset_name in dataset_names:\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
"\n",
|
|||
|
|
" bchmk.sliding_window_benchmarks(dataset, 1000, train=0.8, inc=0.2,\n",
|
|||
|
|
" methods=[cheng.TrendWeightedFTS],\n",
|
|||
|
|
" benchmark_models=False,\n",
|
|||
|
|
" transformations=[None],\n",
|
|||
|
|
" partitions=np.arange(10,100,2), \n",
|
|||
|
|
" progress=False, type=_type,\n",
|
|||
|
|
" file=\"benchmarks.db\", dataset=dataset_name, tag=tag)\n",
|
|||
|
|
"\n",
|
|||
|
|
" bchmk.sliding_window_benchmarks(dataset, 1000, train=0.8, inc=0.2,\n",
|
|||
|
|
" methods=[cheng.TrendWeightedFTS],\n",
|
|||
|
|
" benchmark_models=False,\n",
|
|||
|
|
" transformations=[tdiff],\n",
|
|||
|
|
" partitions=np.arange(3,30,1), \n",
|
|||
|
|
" progress=False, type=_type,\n",
|
|||
|
|
" file=\"benchmarks.db\", dataset=dataset_name, tag=tag)\n"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "S5YAHPOQaLf9",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.benchmarks import Util as bUtil\n",
|
|||
|
|
"\n",
|
|||
|
|
"df1 = bUtil.get_dataframe_from_bd(\"benchmarks.db\",\n",
|
|||
|
|
" \"tag = 'cheng_partitioning' and measure = 'rmse'and transformation is null\")\n",
|
|||
|
|
"\n",
|
|||
|
|
"df2 = bUtil.get_dataframe_from_bd(\"benchmarks.db\",\n",
|
|||
|
|
" \"tag = 'cheng_partitioning' and measure = 'rmse' and transformation is not null\")\n",
|
|||
|
|
"\n",
|
|||
|
|
"fig, ax = plt.subplots(nrows=2, ncols=1, figsize=[15,7])\n",
|
|||
|
|
"\n",
|
|||
|
|
"g1 = sns.boxplot(x='Partitions', y='Value', hue='Dataset', data=df1, showfliers=False, ax=ax[0], \n",
|
|||
|
|
" palette=\"Set3\")\n",
|
|||
|
|
"box = g1.get_position()\n",
|
|||
|
|
"g1.set_position([box.x0, box.y0, box.width * 0.85, box.height]) \n",
|
|||
|
|
"g1.legend(loc='right', bbox_to_anchor=(1.15, 0.5), ncol=1)\n",
|
|||
|
|
"ax[0].set_title(\"Original data\")\n",
|
|||
|
|
"ax[0].set_ylabel(\"RMSE\")\n",
|
|||
|
|
"ax[0].set_xlabel(\"\")\n",
|
|||
|
|
"\n",
|
|||
|
|
"g2 = sns.boxplot(x='Partitions', y='Value', hue='Dataset', data=df2, showfliers=False, ax=ax[1], \n",
|
|||
|
|
" palette=\"Set3\")\n",
|
|||
|
|
"box = g2.get_position()\n",
|
|||
|
|
"g2.set_position([box.x0, box.y0, box.width * 0.85, box.height]) \n",
|
|||
|
|
"g2.legend(loc='right', bbox_to_anchor=(1.15, 0.5), ncol=1)\n",
|
|||
|
|
"ax[1].set_title(\"Differentiated data\")\n",
|
|||
|
|
"ax[1].set_ylabel(\"RMSE\")\n",
|
|||
|
|
"ax[1].set_xlabel(\"Number of partitions of the UoD\")"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "aq9WQjbCaLgB",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### Comparing the partitioning schemas"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "Bhc6kikTaLgD",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 367
|
|||
|
|
},
|
|||
|
|
"outputId": "0243d8fc-ef71-484b-b2b8-a11303a76814"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.partitioners import Grid, Util as pUtil\n",
|
|||
|
|
"\n",
|
|||
|
|
"fig, ax = plt.subplots(nrows=2, ncols=3, figsize=[20,5])\n",
|
|||
|
|
"\n",
|
|||
|
|
"\n",
|
|||
|
|
"partitioners = {}\n",
|
|||
|
|
"partitioners_diff = {}\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
"\n",
|
|||
|
|
" partitioner = Grid.GridPartitioner(data=dataset, npart=30)\n",
|
|||
|
|
" partitioners[dataset_name] = partitioner\n",
|
|||
|
|
" partitioner_diff = Grid.GridPartitioner(data=dataset, npart=10, transformation=tdiff)\n",
|
|||
|
|
" partitioners_diff[dataset_name] = partitioner_diff\n",
|
|||
|
|
"\n",
|
|||
|
|
" pUtil.plot_sets(dataset, [partitioner.sets], titles=[dataset_name], axis=ax[0][count])\n",
|
|||
|
|
" pUtil.plot_sets(dataset, [partitioner_diff.sets], titles=[''], axis=ax[1][count])"
|
|||
|
|
],
|
|||
|
|
"execution_count": 8,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "display_data",
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAABIQAAAFKCAYAAABsCIRVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ndbBZlqbW8RBw8
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 1440x360 with 6 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
}
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "6Hnk7-VSaLgI",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"## Fitting models"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "GUWKh7mFaLgK",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### With original data"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "rtCsD04xaLgL",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
"\n",
|
|||
|
|
" model1 = cheng.TrendWeightedFTS(partitioner=partitioners[dataset_name])\n",
|
|||
|
|
" model1.name=dataset_name\n",
|
|||
|
|
" model1.fit(dataset[:train_split], save_model=True, file_path='model1'+dataset_name, order=1)\n"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "RTIB8ZCMaLgP",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"### With transformed data"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "4XleptoCaLgQ",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
"\n",
|
|||
|
|
" model2 = cheng.TrendWeightedFTS(partitioner=partitioners_diff[dataset_name])\n",
|
|||
|
|
" model2.name=dataset_name\n",
|
|||
|
|
" model2.append_transformation(tdiff)\n",
|
|||
|
|
" model2.fit(dataset[:train_split], save_model=True, file_path='model2'+dataset_name, order=1)\n"
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "1hyAaSmraLgW",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"## Predicting with the models"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "M8bWiw5ZaLgX",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 749
|
|||
|
|
},
|
|||
|
|
"outputId": "7b3b4396-56e9-4a29-a022-895246047227"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"fig, ax = plt.subplots(nrows=3, ncols=1, figsize=[20,10])\n",
|
|||
|
|
"\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
" ax[count].plot(dataset[train_split:train_split+200])\n",
|
|||
|
|
"\n",
|
|||
|
|
" model1 = cUtil.load_obj('model1'+dataset_name)\n",
|
|||
|
|
"\n",
|
|||
|
|
" forecasts = model1.predict(dataset[train_split:train_split+200])\n",
|
|||
|
|
" \n",
|
|||
|
|
" ax[count].plot(forecasts)\n",
|
|||
|
|
" \n",
|
|||
|
|
" ax[count].set_title(dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
"plt.tight_layout()"
|
|||
|
|
],
|
|||
|
|
"execution_count": 11,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "display_data",
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAABZgAAALICAYAAADyhJW9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlzcWb2dbXhFfr
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 1440x720 with 3 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
}
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "7aDQjrFOaLgb",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 143
|
|||
|
|
},
|
|||
|
|
"outputId": "3b9aec08-e67c-49c9-fc1e-e4cabbf8d678"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.benchmarks import Measures\n",
|
|||
|
|
"\n",
|
|||
|
|
"rows = []\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" row = [dataset_name]\n",
|
|||
|
|
" \n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
" test = dataset[train_split:train_split+200]\n",
|
|||
|
|
"\n",
|
|||
|
|
" model1 = cUtil.load_obj('model1'+dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
" row.extend(Measures.get_point_statistics(test, model1))\n",
|
|||
|
|
" \n",
|
|||
|
|
" rows.append(row)\n",
|
|||
|
|
" \n",
|
|||
|
|
" \n",
|
|||
|
|
"pd.DataFrame(rows,columns=[\"Dataset\",\"RMSE\",\"SMAPE\",\"Theil's U\"])\n"
|
|||
|
|
],
|
|||
|
|
"execution_count": 12,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "execute_result",
|
|||
|
|
"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>Dataset</th>\n",
|
|||
|
|
" <th>RMSE</th>\n",
|
|||
|
|
" <th>SMAPE</th>\n",
|
|||
|
|
" <th>Theil's U</th>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" </thead>\n",
|
|||
|
|
" <tbody>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>0</th>\n",
|
|||
|
|
" <td>TAIEX</td>\n",
|
|||
|
|
" <td>94.31</td>\n",
|
|||
|
|
" <td>1.62</td>\n",
|
|||
|
|
" <td>1.43</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>1</th>\n",
|
|||
|
|
" <td>SP500</td>\n",
|
|||
|
|
" <td>14.90</td>\n",
|
|||
|
|
" <td>1.17</td>\n",
|
|||
|
|
" <td>2.68</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>2</th>\n",
|
|||
|
|
" <td>NASDAQ</td>\n",
|
|||
|
|
" <td>48.54</td>\n",
|
|||
|
|
" <td>2.37</td>\n",
|
|||
|
|
" <td>2.03</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" </tbody>\n",
|
|||
|
|
"</table>\n",
|
|||
|
|
"</div>"
|
|||
|
|
],
|
|||
|
|
"text/plain": [
|
|||
|
|
" Dataset RMSE SMAPE Theil's U\n",
|
|||
|
|
"0 TAIEX 94.31 1.62 1.43\n",
|
|||
|
|
"1 SP500 14.90 1.17 2.68\n",
|
|||
|
|
"2 NASDAQ 48.54 2.37 2.03"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
},
|
|||
|
|
"execution_count": 12
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "0_eJ9WQAaLgg",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 749
|
|||
|
|
},
|
|||
|
|
"outputId": "bbed10a8-6e18-4f3f-a715-ba761d93a18b"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"fig, ax = plt.subplots(nrows=3, ncols=1, figsize=[20,10])\n",
|
|||
|
|
"\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
" ax[count].plot(dataset[train_split:train_split+200])\n",
|
|||
|
|
"\n",
|
|||
|
|
" model1 = cUtil.load_obj('model2'+dataset_name)\n",
|
|||
|
|
"\n",
|
|||
|
|
" forecasts = model1.predict(dataset[train_split:train_split+200])\n",
|
|||
|
|
" \n",
|
|||
|
|
" ax[count].plot(forecasts)\n",
|
|||
|
|
" \n",
|
|||
|
|
" ax[count].set_title(dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
"plt.tight_layout()"
|
|||
|
|
],
|
|||
|
|
"execution_count": 13,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "display_data",
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAABZgAAALICAYAAADyhJW9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3Xd4lMXexvHvbHognRpaQi8iLRSl\nS5EmCCiCFRs2EMR2fPUU9di72MAOgqA06U2a0hM6SO+hphBCerLz/kHOOaggSdiwCbk/17UXu88z\nz8z9SHHz29kZY61FRERERERERERERCS/HO4OICIiIiIiIiIiIiLFkwrMIiIiIiIiIiIiIlIgKjCL\niIiIiIiIiIiISIGowCwiIiIiIiIiIiIiBaICs4iIiIiIiIiIiIgUiArMIiIiIiIiIiIiIlIgKjCL\niIiIiIiIiIiISIGowCwiIiIi8gfGmLPnPZzGmLTzXt9xXrvBxhhrjLntD9d3MMYcOe/1UmNM+h/6\nnZl77iZjzHFjTOh57fsYY2KNMUFX4n5FRERERApKBWYRERERkT+w1pb+zwM4BNx03rHx5zW9B0gA\n7s5Dt0PP79dae1PuWDOBxcB7AMaYYOBT4BFrbZIr70tERERExNU83R1ARERERKQ4MsZUA9oDtwKT\njDEVrLXHC9jd48B2Y8yNwEBgmbV2houiioiIiIgUGhWYRUREREQK5m4g2lo7xRjzG3AH8E5BOrLW\nxhljhgPjAQvUd11MEREREZHCoyUyREREREQK5m5gQu7zCVx6mYwPjTGnz3u8/Ifzq4EgYIG19pSL\ns4qIiIiIFAoVmEVERERE8skY0xqIBCbmHpoANDTGNP6Lyx631gaf9/j7H86PAcYCPYwx17k+tYiI\niIiI62mJDBERERGR/LsHMMBGY8wfj2/Mb2fGmPuBKkAvIBr4whjTxFqb6YKsIiIiIiKFRjOYRURE\nRETywRjjCwwAhgCNz3sMA243xuRrEocxJhx4C3jQWpsBfAbEA8+7MreIiIiISGFQgVlEREREJH9u\nBtKAsdba4/95AF9x7huC3S5y3UfGmLPnPWJyj38CTLTW/gJgrbXAg8AIY0yDwr0VEREREZHLY869\nfxURERERERERERERyR/NYBYRERERERERERGRAlGBWUREREREREREREQKRAVmERERERERERERESkQ\nFZhFREREREREREREpEA83R3gUsqUKWMjIiLcHUNERERERERERESkxIiJiYmz1pa9VLsiX2COiIgg\nOjra3TFERERERERERERESgxjzMG8tNMSGSIiIiIiIiIiIiJSICowi4iIiIiIiIiIiEiBqMAsIiIi\nIiIiIiIiIgWSpwKzMeaAMWaLMWajMSb6D+eeNMZYY0yZ3NfGGPOhMWaPMWazMabpeW3vMcbszn3c\n49pbEREREREREREREZErKT+b/HW01sadf8AYUwXoChw673B3oFbuoyXwKdDSGBMK/BOIAiwQY4yZ\nYa1NvIz8IiIiIiIiIiIiIuIml7tExnvAM5wrGP9HH2CsPWc1EGyMqQjcCCy01ibkFpUXAt0uc3wR\nERERERERERERcZO8FpgtsMAYE2OMGQJgjOkDxFprN/2hbSXg8Hmvj+Qeu9jxPzHGDDHGRBtjok+d\nOpXHiCIiIiIiIiIiIlJS5TgtP69YzfTpPxCXnO7uOCVGXpfIaGOtjTXGlAMWGmN2AP/HueUxXM5a\nOwYYAxAVFWUv0VxERERERERERERKqOwcJ0tX/IJd/g43ZC3Hw1hiNrzNzPpP07PHTZQL8HV3xKta\nngrM1trY3F9PGmOmAe2BSGCTMQagMrDeGNMCiAWqnHd55dxjsUCHPxxfennxRURERERERERExB0O\nxadyJDGVY4lniUtIIDExgaQziaSkZdCq5fUMbBGJw2EKbfysHCc/L1uK14p3uCF7BZnGm4O176V0\nhRrUWvkOzX57gFnbrmfvtSMZ2LUt5QNVaC4Mxtq/niBsjCkFOKy1ybnPFwIvWWvnndfmABBlrY0z\nxvQEhgI9OLfJ34fW2ha5m/zFAE1zL1sPNLPWJvzV+FFRUTY6OrpgdyciIiIiIiIiIiIu9/Pkz2iw\n5XWCSMHPZP7pfIyzFuPLP83wgb2oFlbKpWM7nZaFSxbhs+JtOjhXk4ofx+veTUTPp3EElD3XKCOZ\nxIVvUyrmU6wzh7G2O3GNHmPETS3w8/ZwaZ6rlTEmxlobdcl2eSgwVwem5b70BCZYa1/5Q5sD/K/A\nbICPOLeBXypwr7U2OrfdfZxbWgPgFWvt15cKqAKziIiIiIiIiIhI0XH64Ba8v+rESe/KeNXsiH9A\nMKUDg/HyCwTvUti0RDIXvozJTOFj25/ATk8yuG1tPFwwm3nb3oMc/PE5uqXNIc3hx8n69xLR8ymM\nf+iFL0iK5ey8F/H/7QcSbWm+q/0hw+/od9k5SgKXFZjdTQVmERERERERERGRIiIrjRPvtMYz7SRn\nBi8hMrLWhdudPUnaTyPx2z2Tbc5qfBn2FI8M7Eut8gEFGjbxbDqLJ75Hh8MfE2xSOBA5iOq3voLx\nD8lbB8c2k/xVX45n+LC/32y6No4sUI6SJK8FZseVCCMiIiIiIiIiIiLFX/JPz1A+fS8zIv9x8eIy\nQOly+N3xHXbAOGr4pfBm4nAWfTSUqWv35ms8p9Myf+E8Dr/dhv5HXie1dATp9/5MjXs+yXtxGaDi\ntfje+jm1HLGcnv4Mx5PS85VDLk4FZhEREREREREREbm0bdMI2DqWL5296NHv7jxdYur3xndENNkN\nBvCIx3RKzR1KelZOnq49k5rGnLfupvOvA6nmOMnRju9S5cnllKrW9NIXX4BX7Rs43fghBrCA8WM/\nxeks2is7FBcqMIuIiIiIiIiIiMhfSzxAzvRhbHTWIOm65ygf6Jv3a/1C8L11NLENhtDFuYpFK1fn\n6bK10z+lV9oM9kfcSuDTmwhvfz84Lq+cGdzrZRID6zI47h3GL1pzWX3JOSowi4iIiIiIiIiIyMVl\nZ2In30d6tpPnPZ7gwY51CtRNeNcROI2DzJWjudS+cGkZ2UTs+orDXpHUHDwG45eP5TD+iqcPwXeN\nJcCRSeSvT7H1SKJr+i3BVGAWERERERERERGRi1v8EiY2hqcyHqB/pzYE+HoVqBsTVInY8BvpnL6A\nNTsO/mXblfO/pyaHyWz5GBhToPEumqNsHbK7vEIbxxaWj32J1Mxsl/Zf0qjALCIiIiIiIiIiIhe2\nawGsHMVs7+5sDe7AHa2qXlZ3Fbo+QaBJY9/CMRdtk5XjJHjjZ8Q7wqjeIW9rPeeX/3UPEF+5M/dn\njOWLH2cUyhglhae7A4iIiIiIiIiIiEgRteifnAmowchTt/HmwDr4eHpcVnc+ES04GtCQ6+Mnc+DU\nc0SUDfhTm1+XL6Kjcyu7rn2WME+fyxrvoowhbNAYkt9vQY+dz/PM9+GEhQQT4u9FsL83If7ehJby\nIrSUD5FlShVOhquECswiIiIiIiIiIiLyZwn74eR2vvG8l9qVynLTteEu6bZUu2GEzx7ChLkTiLj7\nod+ds9ZiVn5ICv7U7P6YS8a7eJAwfG/9nJoT+vLqzh6ctqVItAEkUprTtjS7bQBnfMoz5O8Xn20t\nKjCLiIiIiIiIiIjIheyaB8DklGt5bWBdHA7XrIUc1LQ/ifNfoPresZxJv4/A89Z0XhWznraZv7K3\n5mBq+wW5ZLy/4lX7Brj9BzwPrSYsNYGglHiqnI3HpibgSD9MtuehQs9Q3KnALCIiIiIiIiIiIn9i\nd87lgKlMlRr1aV2zjOs69vAkvcn9tFr3GlMX/0y/Ht3+e+r04g+wxkFkryddN96l1L4Rat+IAbxy\nH/9RSAt0XFW0yZ+IiIiIiIiIiIj8XnoS9sAK5mc1YWDzy9vY70IqdhxCOj54x4whx2kBWL9zHx1S\n5nGgYne8Qqq4f
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 1440x720 with 3 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
}
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "kEojM8cnaLgk",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 143
|
|||
|
|
},
|
|||
|
|
"outputId": "707b3ea5-ba55-4147-a33d-26c8719bc5bd"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.benchmarks import Measures\n",
|
|||
|
|
"\n",
|
|||
|
|
"rows = []\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" row = [dataset_name]\n",
|
|||
|
|
" \n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
" test = dataset[train_split:train_split+200]\n",
|
|||
|
|
"\n",
|
|||
|
|
" model1 = cUtil.load_obj('model2'+dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
" row.extend(Measures.get_point_statistics(test, model1))\n",
|
|||
|
|
" \n",
|
|||
|
|
" rows.append(row)\n",
|
|||
|
|
" \n",
|
|||
|
|
" \n",
|
|||
|
|
"pd.DataFrame(rows,columns=[\"Dataset\",\"RMSE\",\"SMAPE\",\"Theil's U\"])\n",
|
|||
|
|
" \n",
|
|||
|
|
" \n",
|
|||
|
|
" "
|
|||
|
|
],
|
|||
|
|
"execution_count": 14,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "execute_result",
|
|||
|
|
"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>Dataset</th>\n",
|
|||
|
|
" <th>RMSE</th>\n",
|
|||
|
|
" <th>SMAPE</th>\n",
|
|||
|
|
" <th>Theil's U</th>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" </thead>\n",
|
|||
|
|
" <tbody>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>0</th>\n",
|
|||
|
|
" <td>TAIEX</td>\n",
|
|||
|
|
" <td>65.69</td>\n",
|
|||
|
|
" <td>1.13</td>\n",
|
|||
|
|
" <td>1.00</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>1</th>\n",
|
|||
|
|
" <td>SP500</td>\n",
|
|||
|
|
" <td>5.19</td>\n",
|
|||
|
|
" <td>0.37</td>\n",
|
|||
|
|
" <td>0.94</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" <tr>\n",
|
|||
|
|
" <th>2</th>\n",
|
|||
|
|
" <td>NASDAQ</td>\n",
|
|||
|
|
" <td>24.52</td>\n",
|
|||
|
|
" <td>1.17</td>\n",
|
|||
|
|
" <td>1.02</td>\n",
|
|||
|
|
" </tr>\n",
|
|||
|
|
" </tbody>\n",
|
|||
|
|
"</table>\n",
|
|||
|
|
"</div>"
|
|||
|
|
],
|
|||
|
|
"text/plain": [
|
|||
|
|
" Dataset RMSE SMAPE Theil's U\n",
|
|||
|
|
"0 TAIEX 65.69 1.13 1.00\n",
|
|||
|
|
"1 SP500 5.19 0.37 0.94\n",
|
|||
|
|
"2 NASDAQ 24.52 1.17 1.02"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
},
|
|||
|
|
"execution_count": 14
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "RC6t9ZKJaLgq",
|
|||
|
|
"colab_type": "text"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"## Residual Analysis"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "HErrYknJaLgs",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {
|
|||
|
|
"base_uri": "https://localhost:8080/",
|
|||
|
|
"height": 1000
|
|||
|
|
},
|
|||
|
|
"outputId": "e37704bf-aac4-4b97-fbb1-1b7667660939"
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
"from pyFTS.benchmarks import ResidualAnalysis as ra\n",
|
|||
|
|
"\n",
|
|||
|
|
"for count,dataset_name in enumerate(dataset_names):\n",
|
|||
|
|
" dataset = get_dataset(dataset_name)\n",
|
|||
|
|
" \n",
|
|||
|
|
" model1 = cUtil.load_obj('model1'+dataset_name)\n",
|
|||
|
|
" model1 = cUtil.load_obj('model2'+dataset_name)\n",
|
|||
|
|
"\n",
|
|||
|
|
" ra.plot_residuals_by_model(dataset, [model1, model2])"
|
|||
|
|
],
|
|||
|
|
"execution_count": 16,
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"output_type": "display_data",
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAkIAAAI4CAYAAACGHoanAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJ7pOfUZ0Hv3Mz
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 576x576 with 8 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"output_type": "display_data",
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAI4CAYAAACC61aaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrxJdt74gOp9+a
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 576x576 with 8 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"output_type": "display_data",
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAkIAAAI4CAYAAACGHoanAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nHW5m2bdKC7kB6
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 576x576 with 8 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {
|
|||
|
|
"tags": []
|
|||
|
|
}
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "kHyf4scQaLgv",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
""
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"metadata": {
|
|||
|
|
"id": "kUN1rHtJaLg0",
|
|||
|
|
"colab_type": "code",
|
|||
|
|
"colab": {}
|
|||
|
|
},
|
|||
|
|
"source": [
|
|||
|
|
""
|
|||
|
|
],
|
|||
|
|
"execution_count": 0,
|
|||
|
|
"outputs": []
|
|||
|
|
}
|
|||
|
|
]
|
|||
|
|
}
|