{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import xarray as xr\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (8,5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# notice that the time grid is 15th of each month!\n",
    "da_orig = xr.DataArray(np.linspace(0,11,num=12),\\\n",
    "                  coords=[pd.date_range('15/12/1999',\\\n",
    "                  periods=12, freq=pd.DateOffset(months=1))],\\\n",
    "                  dims='time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Ingrid type time grid:\n",
    "year=2000\n",
    "ensostart = (year-1960)*12\n",
    "da_ensotime=xr.DataArray(np.linspace(0,119,num=120),\\\n",
    "            coords=[np.linspace(ensostart+0.5,ensostart+119.5,num=120)],\\\n",
    "            dims='time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "da_dtI = xr.DataArray(np.linspace(0,119,num=120),\\\n",
    "                  coords=[pd.date_range('15/12/1999',\\\n",
    "                  periods=120, freq=pd.DateOffset(months=1))],\\\n",
    "                  dims='time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dag = da_dtI.groupby('time.month')\n",
    "dag.groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "da_clim = da_dtI.groupby('time.month').mean(dim='time')\n",
    "da_anom = da_dtI.groupby('time.month') - da_clim\n",
    "da_clim.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "da_clim.rolling(month=3).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "da_clim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# want to do [month] 3 boxAverage\n",
    "print(da_clim.groupby_bins('month', 4).groups,'\\n')\n",
    "num_of_bins = da_clim.month.shape[0]/3\n",
    "da_3boxAverage = da_clim.groupby_bins('month', num_of_bins).mean()\n",
    "da_3boxAverage.groupby('month_bins').mean()\n",
    "# nope this is getting too messy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nmonths = 3\n",
    "da3 = da_clim.rolling(month=nmonths,center=True).mean()\n",
    "da3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "da3f = da3[nmonths-1::nmonths]\n",
    "da3f['month'] = da3.month[nmonths-1::nmonths] -(nmonths-1)/2\n",
    "da3f"
   ]
  }
 ],
 "metadata": {
  "gist_info": {
   "gist_id": null,
   "gist_url": null
  },
  "kernelspec": {
   "display_name": "Py3 pangeo",
   "language": "python",
   "name": "pangeo"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}