Software for Scientists - October 2015

genetic data.ipynb

python_project_demo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "!curl http://synesthesiam.com/assets/weather.zip > weather.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "!unzip weather.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y1 = ['1', '2', '3']\n",
    "y2 = ['a', 'b', 'c']\n",
    "x = zip (y1,y2)\n",
    "\n",
    "print x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as pyplot\n",
    "from datetime import datetime\n",
    "import os, calendar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "event_types = ['Rain', 'Thunderstorm', 'Snow', 'Fog']\n",
    "num_events = len(event_types)\n",
    "\n",
    "def event2int(event):\n",
    "    return event_types.index(event)\n",
    "\n",
    "def date2int(date_str):\n",
    "    date = datetime.strptime(date_str, '%Y-%m-%d')\n",
    "    return date.toordinal()\n",
    "\n",
    "def r_squared(actual, ideal):\n",
    "    actual_mean = np.mean(actual)\n",
    "    ideal_dev = np.sum([(val - actual_mean)**2 for val in ideal])\n",
    "    actual_dev = np.sum([(val - actual_mean)**2 for val in actual])\n",
    "\n",
    "    return ideal_dev / actual_dev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def read_weather(file_name):\n",
    "    dtypes = np.dtype({ 'names' : ('timestamp', 'max temp', 'mean temp', 'min temp', 'events'),\n",
    "                        'formats' : [np.int, np.float, np.float, np.float, 'S100'] })\n",
    "\n",
    "    data = np.loadtxt(file_name, delimiter=',', skiprows=1, \n",
    "            converters = { 0 : date2int },\n",
    "            usecols=(0,1,2,3,21), dtype=dtypes)\n",
    "\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def temp_plot(dates, mean_temps, min_temps = None, max_temps = None):\n",
    "\n",
    "    year_start = datetime(2012, 1, 1)\n",
    "    days = np.array([(d - year_start).days + 1 for d in dates])\n",
    "\n",
    "    fig = pyplot.figure()\n",
    "    pyplot.title('Temperatures in Bloomington 2012')\n",
    "    pyplot.ylabel('Mean Temperature (F)')\n",
    "    pyplot.xlabel('Day of Year')\n",
    "\n",
    "    if (max_temps is None or min_temps is None):\n",
    "        pyplot.plot(days, mean_temps, marker='o')\n",
    "    else:\n",
    "        temp_err = np.vstack((mean_temps - min_temps,\n",
    "                              max_temps - mean_temps))\n",
    "\n",
    "        pyplot.errorbar(days, mean_temps, marker='o', yerr=temp_err)\n",
    "        pyplot.title('Temperatures in Bloomington 2012 (max/min)')\n",
    "\n",
    "    slope, intercept = np.polyfit(days, mean_temps, 1)\n",
    "    ideal_temps = intercept + (slope * days)\n",
    "    r_sq = r_squared(mean_temps, ideal_temps)\n",
    "\n",
    "    fit_label = 'Linear fit ({0:.3f})'.format(slope)\n",
    "    pyplot.plot(days, ideal_temps, color='red', linestyle='--', label=fit_label)\n",
    "    pyplot.annotate('r^2 = {0:.3f}'.format(r_sq), (0.05, 0.9), xycoords='axes fraction')\n",
    "    pyplot.legend(loc='lower right')\n",
    "\n",
    "    return fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def hist_events(dates, events):\n",
    "    event_months = []\n",
    "\n",
    "    for i in range(num_events):\n",
    "        event_months.append([])\n",
    "\n",
    "    # Build up lists of months where events occurred\n",
    "    min_month = 11\n",
    "    max_month = 0\n",
    "\n",
    "    for date, event_str in zip(dates, events):\n",
    "        if len(event_str) == 0:\n",
    "            continue  # Skip blank events\n",
    "\n",
    "        month = date.month\n",
    "        min_month = min(month, min_month)\n",
    "        max_month = max(month, max_month)\n",
    "\n",
    "        for event in event_str.split('-'):\n",
    "            event_code = event2int(event)\n",
    "            event_months[event_code].append(month)\n",
    "\n",
    "    # Plot histogram\n",
    "    fig = pyplot.figure()\n",
    "    pyplot.title('Weather Events in Bloomington 2012')\n",
    "    pyplot.xlabel('Month')\n",
    "    pyplot.ylabel('Event Count')\n",
    "    pyplot.axes().yaxis.grid()\n",
    "\n",
    "    num_months = max_month - min_month + 1;\n",
    "    bins = np.arange(1, num_months + 2)  # Need extra bin\n",
    "    pyplot.hist(event_months, bins=bins, label=event_types)\n",
    "\n",
    "    # Align month labels to bin centers\n",
    "    month_names = calendar.month_name[min_month:max_month+1]\n",
    "    pyplot.xticks(bins + 0.5, month_names)\n",
    "\n",
    "    pyplot.legend()\n",
    "    \n",
    "    return fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "# Read data and extract dates, temperatures, and events\n",
    "data = read_weather(os.path.join('data', 'weather.csv'))\n",
    "\n",
    "min_temps = data['min temp']\n",
    "mean_temps = data['mean temp']\n",
    "max_temps = data['max temp']\n",
    "dates = [datetime.fromordinal(d) for d in data['timestamp']]\n",
    "events = data['events']\n",
    "\n",
    "# Do plotting\n",
    "if not os.path.exists('plots'):\n",
    "    os.mkdir('plots')\n",
    "\n",
    "temp_plot(dates, mean_temps)\n",
    "fig.show()\n",
    "\n",
    "temp_plot(dates, mean_temps, min_temps, max_temps)\n",
    "fig.show()\n",
    "\n",
    "hist_events(dates, events)\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}

weather_example.ipynb