{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Using TensorFlow backend.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Couldn't import dot_parser, loading of dot files will not be possible.\n" ] } ], "source": [ "%matplotlib inline\n", "from matplotlib import pylab as plt\n", "import matplotlib.dates as mdates\n", "plt.rcParams['figure.figsize'] = (15.0, 8.0)\n", "import pandas as pd\n", "import seaborn as sns\n", "\n", "from IPython.display import SVG\n", "from keras.utils.vis_utils import model_to_dot" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": true }, "outputs": [], "source": [ "from pandas.tseries.holiday import USFederalHolidayCalendar as calendar" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " Loading BokehJS ...\n", "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/javascript": [ "\n", "(function(global) {\n", " function now() {\n", " return new Date();\n", " }\n", "\n", " var force = \"1\";\n", "\n", " if (typeof (window._bokeh_onload_callbacks) === \"undefined\" || force !== \"\") {\n", " window._bokeh_onload_callbacks = [];\n", " window._bokeh_is_loading = undefined;\n", " }\n", "\n", "\n", " \n", " if (typeof (window._bokeh_timeout) === \"undefined\" || force !== \"\") {\n", " window._bokeh_timeout = Date.now() + 5000;\n", " window._bokeh_failed_load = false;\n", " }\n", "\n", " var NB_LOAD_WARNING = {'data': {'text/html':\n", " \"
\\n\"+\n", " \"

\\n\"+\n", " \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n", " \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n", " \"

\\n\"+\n", " \"\\n\"+\n", " \"\\n\"+\n", " \"from bokeh.resources import INLINE\\n\"+\n", " \"output_notebook(resources=INLINE)\\n\"+\n", " \"\\n\"+\n", " \"
\"}};\n", "\n", " function display_loaded() {\n", " if (window.Bokeh !== undefined) {\n", " Bokeh.$(\"#428726d7-a0ae-4a09-873b-cfb976255192\").text(\"BokehJS successfully loaded.\");\n", " } else if (Date.now() < window._bokeh_timeout) {\n", " setTimeout(display_loaded, 100)\n", " }\n", " }\n", "\n", " function run_callbacks() {\n", " window._bokeh_onload_callbacks.forEach(function(callback) { callback() });\n", " delete window._bokeh_onload_callbacks\n", " console.info(\"Bokeh: all callbacks have finished\");\n", " }\n", "\n", " function load_libs(js_urls, callback) {\n", " window._bokeh_onload_callbacks.push(callback);\n", " if (window._bokeh_is_loading > 0) {\n", " console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n", " return null;\n", " }\n", " if (js_urls == null || js_urls.length === 0) {\n", " run_callbacks();\n", " return null;\n", " }\n", " console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n", " window._bokeh_is_loading = js_urls.length;\n", " for (var i = 0; i < js_urls.length; i++) {\n", " var url = js_urls[i];\n", " var s = document.createElement('script');\n", " s.src = url;\n", " s.async = false;\n", " s.onreadystatechange = s.onload = function() {\n", " window._bokeh_is_loading--;\n", " if (window._bokeh_is_loading === 0) {\n", " console.log(\"Bokeh: all BokehJS libraries loaded\");\n", " run_callbacks()\n", " }\n", " };\n", " s.onerror = function() {\n", " console.warn(\"failed to load library \" + url);\n", " };\n", " console.log(\"Bokeh: injecting script tag for BokehJS library: \", url);\n", " document.getElementsByTagName(\"head\")[0].appendChild(s);\n", " }\n", " };var element = document.getElementById(\"428726d7-a0ae-4a09-873b-cfb976255192\");\n", " if (element == null) {\n", " console.log(\"Bokeh: ERROR: autoload.js configured with elementid '428726d7-a0ae-4a09-873b-cfb976255192' but no matching script tag was found. \")\n", " return false;\n", " }\n", "\n", " var js_urls = ['https://cdn.pydata.org/bokeh/release/bokeh-0.12.2.min.js', 'https://cdn.pydata.org/bokeh/release/bokeh-widgets-0.12.2.min.js', 'https://cdn.pydata.org/bokeh/release/bokeh-compiler-0.12.2.min.js'];\n", "\n", " var inline_js = [\n", " function(Bokeh) {\n", " Bokeh.set_log_level(\"info\");\n", " },\n", " \n", " function(Bokeh) {\n", " \n", " Bokeh.$(\"#428726d7-a0ae-4a09-873b-cfb976255192\").text(\"BokehJS is loading...\");\n", " },\n", " function(Bokeh) {\n", " console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-0.12.2.min.css\");\n", " Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-0.12.2.min.css\");\n", " console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-widgets-0.12.2.min.css\");\n", " Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-widgets-0.12.2.min.css\");\n", " }\n", " ];\n", "\n", " function run_inline_js() {\n", " \n", " if ((window.Bokeh !== undefined) || (force === \"1\")) {\n", " for (var i = 0; i < inline_js.length; i++) {\n", " inline_js[i](window.Bokeh);\n", " }if (force === \"1\") {\n", " display_loaded();\n", " }} else if (Date.now() < window._bokeh_timeout) {\n", " setTimeout(run_inline_js, 100);\n", " } else if (!window._bokeh_failed_load) {\n", " console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n", " window._bokeh_failed_load = true;\n", " } else if (!force) {\n", " var cell = $(\"#428726d7-a0ae-4a09-873b-cfb976255192\").parents('.cell').data().cell;\n", " cell.output_area.append_execute_result(NB_LOAD_WARNING)\n", " }\n", "\n", " }\n", "\n", " if (window._bokeh_is_loading === 0) {\n", " console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n", " run_inline_js();\n", " } else {\n", " load_libs(js_urls, function() {\n", " console.log(\"Bokeh: BokehJS plotting callback run at\", now());\n", " run_inline_js();\n", " });\n", " }\n", "}(this));" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from bokeh.charts import TimeSeries, output_file, show\n", "from bokeh.io import output_notebook\n", "output_notebook()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Data Preparation\n", "Eleven months data from one bed one bath apartment unit in San Jose, CA region was picked for this experiment. The electricity consumption is recorded in 15 minutes interval by the energy supply company. The raw data contains information such as type, date, start time, end time, usage, units, cost and notes fields. The start time and end time is the measurement interval. In this data, the interval is 15 minutes. The usage in 15 minutes interval is provided in kWh unit, and the cost for the consumption is presented in the dollar. Before we deep dive into the data, some quick feature engineering steps are done to enrich the data with more features. " ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TYPEDATESTART TIMEEND TIMEUSAGEUNITSCOSTNOTES
0Electric usage11/1/160:000:140.07kWh$0.01NaN
1Electric usage11/1/160:150:290.05kWh$0.01NaN
\n", "
" ], "text/plain": [ " TYPE DATE START TIME END TIME USAGE UNITS COST NOTES\n", "0 Electric usage 11/1/16 0:00 0:14 0.07 kWh $0.01 NaN\n", "1 Electric usage 11/1/16 0:15 0:29 0.05 kWh $0.01 NaN" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = pd.read_csv(\"D202.csv\")\n", "data.head(2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Creating Date and Time Filed" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": true }, "outputs": [], "source": [ "data[\"DATE_TIME\"] = pd.to_datetime(data.DATE + \" \" + data[\"END TIME\"])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Working Day or Not" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": true }, "outputs": [], "source": [ "data[\"DAY_TYPE\"] = data.DATE_TIME.apply(lambda x: 1 if x.dayofweek > 5 else 0 )" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Finding Fedaral Holidays" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "collapsed": true }, "outputs": [], "source": [ "cal = calendar()\n", "holidays = cal.holidays(start = data.DATE_TIME.min(), end = data.DATE_TIME.max())\n", "data[\"IS_HOLIDAY\"] = data.DATE_TIME.isin(holidays)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TYPEDATESTART TIMEEND TIMEUSAGEUNITSCOSTNOTESDATE_TIMEDAY_TYPEIS_HOLIDAY
0Electric usage11/1/160:000:140.07kWh$0.01NaN2016-11-01 00:14:000False
1Electric usage11/1/160:150:290.05kWh$0.01NaN2016-11-01 00:29:000False
2Electric usage11/1/160:300:440.06kWh$0.01NaN2016-11-01 00:44:000False
\n", "
" ], "text/plain": [ " TYPE DATE START TIME END TIME USAGE UNITS COST NOTES \\\n", "0 Electric usage 11/1/16 0:00 0:14 0.07 kWh $0.01 NaN \n", "1 Electric usage 11/1/16 0:15 0:29 0.05 kWh $0.01 NaN \n", "2 Electric usage 11/1/16 0:30 0:44 0.06 kWh $0.01 NaN \n", "\n", " DATE_TIME DAY_TYPE IS_HOLIDAY \n", "0 2016-11-01 00:14:00 0 False \n", "1 2016-11-01 00:29:00 0 False \n", "2 2016-11-01 00:44:00 0 False " ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.head(3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Previous Five Observations" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "collapsed": false }, "outputs": [], "source": [ "for obs in range(1,6):\n", " data[\"T_\" + str(obs)] = data.USAGE.shift(obs)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TYPEDATESTART TIMEEND TIMEUSAGEUNITSCOSTNOTESDATE_TIMEDAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5
0Electric usage11/1/160:000:140.07kWh$0.0102016-11-01 00:14:000False0.000.000.000.000.00
1Electric usage11/1/160:150:290.05kWh$0.0102016-11-01 00:29:000False0.070.000.000.000.00
2Electric usage11/1/160:300:440.06kWh$0.0102016-11-01 00:44:000False0.050.070.000.000.00
3Electric usage11/1/160:450:590.05kWh$0.0102016-11-01 00:59:000False0.060.050.070.000.00
4Electric usage11/1/161:001:140.03kWh$0.0102016-11-01 01:14:000False0.050.060.050.070.00
5Electric usage11/1/161:151:290.03kWh$0.0102016-11-01 01:29:000False0.030.050.060.050.07
6Electric usage11/1/161:301:440.03kWh$0.0102016-11-01 01:44:000False0.030.030.050.060.05
7Electric usage11/1/161:451:590.06kWh$0.0102016-11-01 01:59:000False0.030.030.030.050.06
8Electric usage11/1/162:002:140.04kWh$0.0102016-11-01 02:14:000False0.060.030.030.030.05
9Electric usage11/1/162:152:290.03kWh$0.0102016-11-01 02:29:000False0.040.060.030.030.03
\n", "
" ], "text/plain": [ " TYPE DATE START TIME END TIME USAGE UNITS COST NOTES \\\n", "0 Electric usage 11/1/16 0:00 0:14 0.07 kWh $0.01 0 \n", "1 Electric usage 11/1/16 0:15 0:29 0.05 kWh $0.01 0 \n", "2 Electric usage 11/1/16 0:30 0:44 0.06 kWh $0.01 0 \n", "3 Electric usage 11/1/16 0:45 0:59 0.05 kWh $0.01 0 \n", "4 Electric usage 11/1/16 1:00 1:14 0.03 kWh $0.01 0 \n", "5 Electric usage 11/1/16 1:15 1:29 0.03 kWh $0.01 0 \n", "6 Electric usage 11/1/16 1:30 1:44 0.03 kWh $0.01 0 \n", "7 Electric usage 11/1/16 1:45 1:59 0.06 kWh $0.01 0 \n", "8 Electric usage 11/1/16 2:00 2:14 0.04 kWh $0.01 0 \n", "9 Electric usage 11/1/16 2:15 2:29 0.03 kWh $0.01 0 \n", "\n", " DATE_TIME DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 \n", "0 2016-11-01 00:14:00 0 False 0.00 0.00 0.00 0.00 0.00 \n", "1 2016-11-01 00:29:00 0 False 0.07 0.00 0.00 0.00 0.00 \n", "2 2016-11-01 00:44:00 0 False 0.05 0.07 0.00 0.00 0.00 \n", "3 2016-11-01 00:59:00 0 False 0.06 0.05 0.07 0.00 0.00 \n", "4 2016-11-01 01:14:00 0 False 0.05 0.06 0.05 0.07 0.00 \n", "5 2016-11-01 01:29:00 0 False 0.03 0.05 0.06 0.05 0.07 \n", "6 2016-11-01 01:44:00 0 False 0.03 0.03 0.05 0.06 0.05 \n", "7 2016-11-01 01:59:00 0 False 0.03 0.03 0.03 0.05 0.06 \n", "8 2016-11-01 02:14:00 0 False 0.06 0.03 0.03 0.03 0.05 \n", "9 2016-11-01 02:29:00 0 False 0.04 0.06 0.03 0.03 0.03 " ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.fillna(0.00,inplace=True)\n", "data.head(10)" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "collapsed": true }, "outputs": [], "source": [ "data.IS_HOLIDAY = data.IS_HOLIDAY.astype(\"int\")" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TYPEDATESTART TIMEEND TIMEUSAGEUNITSCOSTNOTESDATE_TIMEDAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5
0Electric usage11/1/160:000:140.07kWh$0.0102016-11-01 00:14:00000.000.00.00.00.0
1Electric usage11/1/160:150:290.05kWh$0.0102016-11-01 00:29:00000.070.00.00.00.0
\n", "
" ], "text/plain": [ " TYPE DATE START TIME END TIME USAGE UNITS COST NOTES \\\n", "0 Electric usage 11/1/16 0:00 0:14 0.07 kWh $0.01 0 \n", "1 Electric usage 11/1/16 0:15 0:29 0.05 kWh $0.01 0 \n", "\n", " DATE_TIME DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 \n", "0 2016-11-01 00:14:00 0 0 0.00 0.0 0.0 0.0 0.0 \n", "1 2016-11-01 00:29:00 0 0 0.07 0.0 0.0 0.0 0.0 " ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.head(2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Clean Data" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "collapsed": false }, "outputs": [], "source": [ "clean_data = data[['DAY_TYPE', 'IS_HOLIDAY', 'T_1','T_2', 'T_3', 'T_4', 'T_5','USAGE']]" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5USAGE
0000.000.00.00.00.00.07
1000.070.00.00.00.00.05
\n", "
" ], "text/plain": [ " DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 USAGE\n", "0 0 0 0.00 0.0 0.0 0.0 0.0 0.07\n", "1 0 0 0.07 0.0 0.0 0.0 0.0 0.05" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "clean_data.head(2)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Let's Explore" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/Users/jagan/anaconda2/lib/python2.7/site-packages/bokeh/core/json_encoder.py:33: FutureWarning: pandas.tslib is deprecated and will be removed in a future version.\n", "You can access Timestamp as pandas.Timestamp\n", " if pd and isinstance(obj, pd.tslib.Timestamp):\n" ] }, { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "
\n", "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "all_show = TimeSeries(data,x=\"DATE_TIME\",y=[\"USAGE\"],legend=True,plot_width=900, plot_height=350)\n", "show(all_show)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### A Week ! Yes X'Mas Week" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "collapsed": false }, "outputs": [], "source": [ "xmask = (data.DATE_TIME >= pd.to_datetime(\"12/20/2016\")) & (data.DATE_TIME <= pd.to_datetime(\"12/27/2016\"))" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "collapsed": true }, "outputs": [], "source": [ "xmas_week = data.loc[xmask]" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "
\n", "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "xmas_show = TimeSeries(xmas_week,x=\"DATE_TIME\",y=[\"USAGE\"],legend=True,plot_width=900, plot_height=350)\n", "show(xmas_show)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### A Day ! New Year 2017" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "collapsed": true }, "outputs": [], "source": [ "dmask = (data.DATE_TIME >= pd.to_datetime(\"01/01/2017\")) & (data.DATE_TIME < pd.to_datetime(\"01/02/2017\"))\n", "nyd = data.loc[dmask]" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "
\n", "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "nyd_show = TimeSeries(nyd,x=\"DATE_TIME\",y=[\"USAGE\"],legend=True,plot_width=900, plot_height=350)\n", "show(nyd_show)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Train and Test Data" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "collapsed": true }, "outputs": [], "source": [ "training_data = data[data.DATE_TIME < pd.to_datetime(\"08/01/2017\")]" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "collapsed": true }, "outputs": [], "source": [ "val_mask = (data.DATE_TIME >= pd.to_datetime(\"08/01/2017\")) & (data.DATE_TIME < pd.to_datetime(\"09/01/2017\"))\n", "val_data = data.loc[val_mask]" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "collapsed": true }, "outputs": [], "source": [ "test_data = data[data.DATE_TIME >= pd.to_datetime(\"09/01/2017\")]" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TYPEDATESTART TIMEEND TIMEUSAGEUNITSCOSTNOTESDATE_TIMEDAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5
26006Electric usage7/31/1723:1523:290.12kWh$0.0202017-07-31 23:29:00000.150.100.270.440.12
26007Electric usage7/31/1723:3023:440.11kWh$0.0202017-07-31 23:44:00000.120.150.100.270.44
26008Electric usage7/31/1723:4523:590.10kWh$0.0202017-07-31 23:59:00000.110.120.150.100.27
\n", "
" ], "text/plain": [ " TYPE DATE START TIME END TIME USAGE UNITS COST NOTES \\\n", "26006 Electric usage 7/31/17 23:15 23:29 0.12 kWh $0.02 0 \n", "26007 Electric usage 7/31/17 23:30 23:44 0.11 kWh $0.02 0 \n", "26008 Electric usage 7/31/17 23:45 23:59 0.10 kWh $0.02 0 \n", "\n", " DATE_TIME DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 \n", "26006 2017-07-31 23:29:00 0 0 0.15 0.10 0.27 0.44 0.12 \n", "26007 2017-07-31 23:44:00 0 0 0.12 0.15 0.10 0.27 0.44 \n", "26008 2017-07-31 23:59:00 0 0 0.11 0.12 0.15 0.10 0.27 " ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "training_data.tail(3)" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TYPEDATESTART TIMEEND TIMEUSAGEUNITSCOSTNOTESDATE_TIMEDAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5
28984Electric usage9/1/170:000:140.03kWh$0.0102017-09-01 00:14:00000.030.030.040.040.05
28985Electric usage9/1/170:150:290.04kWh$0.0102017-09-01 00:29:00000.030.030.030.040.04
\n", "
" ], "text/plain": [ " TYPE DATE START TIME END TIME USAGE UNITS COST NOTES \\\n", "28984 Electric usage 9/1/17 0:00 0:14 0.03 kWh $0.01 0 \n", "28985 Electric usage 9/1/17 0:15 0:29 0.04 kWh $0.01 0 \n", "\n", " DATE_TIME DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 \n", "28984 2017-09-01 00:14:00 0 0 0.03 0.03 0.04 0.04 0.05 \n", "28985 2017-09-01 00:29:00 0 0 0.03 0.03 0.03 0.04 0.04 " ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "test_data.head(2)" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "collapsed": true }, "outputs": [], "source": [ "clean_train = training_data[['DAY_TYPE', 'IS_HOLIDAY', 'T_1','T_2', 'T_3', 'T_4', 'T_5','USAGE']]\n", "clean_test = test_data[['DAY_TYPE', 'IS_HOLIDAY', 'T_1','T_2', 'T_3', 'T_4', 'T_5','USAGE']]\n", "clean_val = val_data[['DAY_TYPE', 'IS_HOLIDAY', 'T_1','T_2', 'T_3', 'T_4', 'T_5','USAGE']]" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5USAGE
0000.000.00.00.00.00.07
1000.070.00.00.00.00.05
\n", "
" ], "text/plain": [ " DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 USAGE\n", "0 0 0 0.00 0.0 0.0 0.0 0.0 0.07\n", "1 0 0 0.07 0.0 0.0 0.0 0.0 0.05" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "clean_train.head(2)" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5USAGE
28984000.030.030.040.040.050.03
28985000.030.030.030.040.040.04
\n", "
" ], "text/plain": [ " DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 USAGE\n", "28984 0 0 0.03 0.03 0.04 0.04 0.05 0.03\n", "28985 0 0 0.03 0.03 0.03 0.04 0.04 0.04" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "clean_test.head(2)" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5USAGE
26009000.10.110.120.150.100.10
26010000.10.100.110.120.150.10
26011000.10.100.100.110.120.08
\n", "
" ], "text/plain": [ " DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 USAGE\n", "26009 0 0 0.1 0.11 0.12 0.15 0.10 0.10\n", "26010 0 0 0.1 0.10 0.11 0.12 0.15 0.10\n", "26011 0 0 0.1 0.10 0.10 0.11 0.12 0.08" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "clean_val.head(3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Let's Model and Predict" ] }, { "cell_type": "code", "execution_count": 30, "metadata": { "collapsed": true }, "outputs": [], "source": [ "from sklearn.preprocessing import StandardScaler,MinMaxScaler\n", "from sklearn.ensemble import RandomForestRegressor\n", "from sklearn.metrics import r2_score,mean_squared_error" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "collapsed": true }, "outputs": [], "source": [ "X_train,y_train = clean_train.drop([\"USAGE\"],axis=1),clean_train.USAGE\n", "X_test,y_test = clean_test.drop([\"USAGE\"],axis=1),clean_test.USAGE\n", "X_val,y_val = clean_val.drop([\"USAGE\"],axis=1),clean_val.USAGE" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "collapsed": true }, "outputs": [], "source": [ "scaler = StandardScaler()\n", "#scaler = MinMaxScaler(feature_range=(-1, 1))\n", "rfr = RandomForestRegressor(random_state=2017,verbose=2,n_jobs=5)" ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "collapsed": false }, "outputs": [], "source": [ "X_train_scaled = scaler.fit_transform(X_train)\n", "X_test_scaled = scaler.fit_transform(X_test)\n", "X_valid_scaled = scaler.fit_transform(X_val)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 34, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "building tree 1 of 10\n", "building tree 2 of 10\n", "building tree 3 of 10\n", "building tree 4 of 10\n", "building tree 5 of 10\n", "building tree 6 of 10\n", "building tree 7 of 10\n", "building tree 8 of 10\n", "building tree 9 of 10\n", "building tree 10 of 10\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "[Parallel(n_jobs=5)]: Done 7 out of 10 | elapsed: 0.1s remaining: 0.1s\n", "[Parallel(n_jobs=5)]: Done 10 out of 10 | elapsed: 0.1s finished\n" ] }, { "data": { "text/plain": [ "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n", " max_features='auto', max_leaf_nodes=None,\n", " min_impurity_decrease=0.0, min_impurity_split=None,\n", " min_samples_leaf=1, min_samples_split=2,\n", " min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=5,\n", " oob_score=False, random_state=2017, verbose=2, warm_start=False)" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "rfr.fit(X_train_scaled,y_train)" ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "[Parallel(n_jobs=5)]: Done 7 out of 10 | elapsed: 0.0s remaining: 0.0s\n", "[Parallel(n_jobs=5)]: Done 10 out of 10 | elapsed: 0.0s finished\n" ] }, { "data": { "text/plain": [ "0.26352935135691857" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "rfr.score(X_val,y_val)" ] }, { "cell_type": "code", "execution_count": 36, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "[Parallel(n_jobs=5)]: Done 7 out of 10 | elapsed: 0.0s remaining: 0.0s\n", "[Parallel(n_jobs=5)]: Done 10 out of 10 | elapsed: 0.0s finished\n" ] }, { "data": { "text/plain": [ "0.32707009777378615" ] }, "execution_count": 36, "metadata": {}, "output_type": "execute_result" } ], "source": [ "rfr.score(X_test,y_test)" ] }, { "cell_type": "code", "execution_count": 37, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "[Parallel(n_jobs=5)]: Done 7 out of 10 | elapsed: 0.0s remaining: 0.0s\n", "[Parallel(n_jobs=5)]: Done 10 out of 10 | elapsed: 0.0s finished\n", "/Users/jagan/anaconda2/lib/python2.7/site-packages/ipykernel/__main__.py:1: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n", " if __name__ == '__main__':\n" ] } ], "source": [ "test_data[\"RF_PREDICTED\"] = rfr.predict(X_test_scaled)" ] }, { "cell_type": "code", "execution_count": 38, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TYPEDATESTART TIMEEND TIMEUSAGEUNITSCOSTNOTESDATE_TIMEDAY_TYPEIS_HOLIDAYT_1T_2T_3T_4T_5RF_PREDICTED
28984Electric usage9/1/170:000:140.03kWh$0.0102017-09-01 00:14:00000.030.030.040.040.050.021306
28985Electric usage9/1/170:150:290.04kWh$0.0102017-09-01 00:29:00000.030.030.030.040.040.060113
28986Electric usage9/1/170:300:440.06kWh$0.0102017-09-01 00:44:00000.040.030.030.030.040.048081
28987Electric usage9/1/170:450:590.03kWh$0.0102017-09-01 00:59:00000.060.040.030.030.030.061000
28988Electric usage9/1/171:001:140.03kWh$0.0102017-09-01 01:14:00000.030.060.040.030.030.028246
\n", "
" ], "text/plain": [ " TYPE DATE START TIME END TIME USAGE UNITS COST NOTES \\\n", "28984 Electric usage 9/1/17 0:00 0:14 0.03 kWh $0.01 0 \n", "28985 Electric usage 9/1/17 0:15 0:29 0.04 kWh $0.01 0 \n", "28986 Electric usage 9/1/17 0:30 0:44 0.06 kWh $0.01 0 \n", "28987 Electric usage 9/1/17 0:45 0:59 0.03 kWh $0.01 0 \n", "28988 Electric usage 9/1/17 1:00 1:14 0.03 kWh $0.01 0 \n", "\n", " DATE_TIME DAY_TYPE IS_HOLIDAY T_1 T_2 T_3 T_4 T_5 \\\n", "28984 2017-09-01 00:14:00 0 0 0.03 0.03 0.04 0.04 0.05 \n", "28985 2017-09-01 00:29:00 0 0 0.03 0.03 0.03 0.04 0.04 \n", "28986 2017-09-01 00:44:00 0 0 0.04 0.03 0.03 0.03 0.04 \n", "28987 2017-09-01 00:59:00 0 0 0.06 0.04 0.03 0.03 0.03 \n", "28988 2017-09-01 01:14:00 0 0 0.03 0.06 0.04 0.03 0.03 \n", "\n", " RF_PREDICTED \n", "28984 0.021306 \n", "28985 0.060113 \n", "28986 0.048081 \n", "28987 0.061000 \n", "28988 0.028246 " ] }, "execution_count": 38, "metadata": {}, "output_type": "execute_result" } ], "source": [ "test_data.head(5)" ] }, { "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "### Prediction with Random Forest Model in Test Data" ] }, { "cell_type": "code", "execution_count": 39, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "
\n", "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "pred_show = TimeSeries(test_data,x=\"DATE_TIME\",y=[\"USAGE\",\"RF_PREDICTED\"],legend=True,plot_width=800, plot_height=350)\n", "show(pred_show)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Prediction Single Day in Test Data" ] }, { "cell_type": "code", "execution_count": 40, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "
\n", "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sep_30m = test_data[test_data.DATE_TIME >= pd.to_datetime(\"09/30/2017\")]\n", "\n", "sep_30rf = TimeSeries(sep_30m,x=\"DATE_TIME\",y=[\"USAGE\",\"RF_PREDICTED\"],legend=True,plot_width=900, plot_height=350)\n", "show(sep_30rf)" ] }, { "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "### LSTM Modelling" ] }, { "cell_type": "code", "execution_count": 41, "metadata": { "collapsed": false }, "outputs": [], "source": [ "from keras.models import Sequential\n", "from keras.layers import Dense\n", "from keras.layers import LSTM" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### LSTM Model" ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "collapsed": false }, "outputs": [], "source": [ "model_k = Sequential()\n", "model_k.add(LSTM(1, input_shape=(1,7)))\n", "model_k.add(Dense(1))\n", "model_k.compile(loss='mean_squared_error', optimizer='adam')" ] }, { "cell_type": "code", "execution_count": 43, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/svg+xml": [ "\n", "\n", "G\n", "\n", "\n", "120866850192\n", "\n", "lstm_1_input: InputLayer\n", "\n", "\n", "120866849360\n", "\n", "lstm_1: LSTM\n", "\n", "\n", "120866850192->120866849360\n", "\n", "\n", "\n", "\n", "120866849680\n", "\n", "dense_1: Dense\n", "\n", "\n", "120866849360->120866849680\n", "\n", "\n", "\n", "\n", "" ], "text/plain": [ "" ] }, "execution_count": 43, "metadata": {}, "output_type": "execute_result" } ], "source": [ "SVG(model_to_dot(model_k).create(prog='dot', format='svg'))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Reshape the data to 3D" ] }, { "cell_type": "code", "execution_count": 44, "metadata": { "collapsed": false }, "outputs": [], "source": [ "X_t_reshaped = X_train_scaled.reshape((X_train_scaled.shape[0], 1, X_train_scaled.shape[1]))" ] }, { "cell_type": "code", "execution_count": 45, "metadata": { "collapsed": true }, "outputs": [], "source": [ "X_val_resaped = X_valid_scaled.reshape((X_valid_scaled.shape[0], 1, X_valid_scaled.shape[1]))" ] }, { "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "##### Fit the Model" ] }, { "cell_type": "code", "execution_count": 46, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train on 26009 samples, validate on 2975 samples\n", "Epoch 1/10\n", "2s - loss: 0.0757 - val_loss: 0.0165\n", "Epoch 2/10\n", "1s - loss: 0.0605 - val_loss: 0.0191\n", "Epoch 3/10\n", "1s - loss: 0.0506 - val_loss: 0.0166\n", "Epoch 4/10\n", "1s - loss: 0.0421 - val_loss: 0.0207\n", "Epoch 5/10\n", "1s - loss: 0.0370 - val_loss: 0.0267\n", "Epoch 6/10\n", "1s - loss: 0.0339 - val_loss: 0.0307\n", "Epoch 7/10\n", "1s - loss: 0.0321 - val_loss: 0.0357\n", "Epoch 8/10\n", "1s - loss: 0.0309 - val_loss: 0.0394\n", "Epoch 9/10\n", "1s - loss: 0.0303 - val_loss: 0.0396\n", "Epoch 10/10\n", "1s - loss: 0.0298 - val_loss: 0.0413\n" ] } ], "source": [ "history = model_k.fit(X_t_reshaped, y_train, validation_data=(X_val_resaped, y_val),\\\n", "epochs=10, batch_size=96, verbose=2)" ] }, { "cell_type": "code", "execution_count": 47, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 47, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAFJCAYAAAChG+XKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n27/gDAAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.plot(history.history['loss'], label='train')\n", "plt.plot(history.history['val_loss'], label='test')\n", "plt.legend()" ] }, { "cell_type": "code", "execution_count": 48, "metadata": { "collapsed": false }, "outputs": [], "source": [ "X_te_reshaped = X_test_scaled.reshape((X_test_scaled.shape[0], 1, X_test_scaled.shape[1]))" ] }, { "cell_type": "code", "execution_count": 49, "metadata": { "collapsed": true }, "outputs": [], "source": [ "res = model_k.predict(X_te_reshaped)" ] }, { "cell_type": "code", "execution_count": 50, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/Users/jagan/anaconda2/lib/python2.7/site-packages/ipykernel/__main__.py:1: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n", " if __name__ == '__main__':\n" ] } ], "source": [ "test_data[\"DL_PRED\"] = res" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### LSTM Prediction on Test Data" ] }, { "cell_type": "code", "execution_count": 51, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "
\n", "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "keras_show = TimeSeries(test_data,x=\"DATE_TIME\",y=[\"USAGE\",\"RF_PREDICTED\",\"DL_PRED\"],legend=True,plot_width=900, plot_height=350)\n", "show(keras_show)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### A Day on LSTM Predcted Result" ] }, { "cell_type": "code", "execution_count": 52, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "
\n", "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sep_30m = test_data[test_data.DATE_TIME >= pd.to_datetime(\"09/30/2017\")]\n", "\n", "sep_30 = TimeSeries(sep_30m,x=\"DATE_TIME\",y=[\"USAGE\",\"RF_PREDICTED\",\"DL_PRED\"],legend=True,plot_width=900, plot_height=350)\n", "show(sep_30)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### RMSE Value of Random Forest and LSTM" ] }, { "cell_type": "code", "execution_count": 53, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "0.16743906314666301" ] }, "execution_count": 53, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from numpy import sqrt\n", "sqrt(mean_squared_error(test_data.USAGE,test_data.DL_PRED))" ] }, { "cell_type": "code", "execution_count": 54, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "0.18412616700088405" ] }, "execution_count": 54, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sqrt(mean_squared_error(test_data.USAGE,test_data.RF_PREDICTED))" ] } ], "metadata": { "anaconda-cloud": { "attach-environment": true, "summary": "Time Series Forecasting with LSTM Deep Learning", "thumbnail": "data:image/png;base64,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" }, "kernelspec": { "display_name": "Python [conda root]", "language": "python", "name": "conda-root-py" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 2 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython2", "version": "2.7.13" } }, "nbformat": 4, "nbformat_minor": 1 }