Masked_Arrays_Gradient_MetPy.ipynb

Masked_Arrays_Gradient_MetPy.ipynb

{
  "cells": [
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import metpy.calc as metcalc\nimport numpy as np",
      "execution_count": 1,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from collections import namedtuple",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "def deriv_2d_data():\n    \"\"\"Return 2-dimensional data for analytic function for testing derivative functions.\"\"\"\n    ret = namedtuple('D_2D_Test_Data', 'x y x0 y0 a b f')(\n        np.array([0., 2., 7.]), np.array([1., 5., 11., 13.]), 3, 1.5, 0.5, 0.25, 0)\n\n    # Makes a value array with y changing along rows (axis 0) and x along columns (axis 1)\n    return ret._replace(f=ret.a * (ret.x - ret.x0)**2 + ret.b * (ret.y[:, None] - ret.y0)**2)",
      "execution_count": 3,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "res = metcalc.kinematics.gradient(deriv_2d_data().f, x=(deriv_2d_data().y, deriv_2d_data().x))",
      "execution_count": 4,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "res",
      "execution_count": 5,
      "outputs": [
        {
          "data": {
            "text/plain": "(<Quantity([[-0.25 -0.25 -0.25]\n  [ 1.75  1.75  1.75]\n  [ 4.75  4.75  4.75]\n  [ 5.75  5.75  5.75]], 'dimensionless')>, <Quantity([[-3. -1.  4.]\n  [-3. -1.  4.]\n  [-3. -1.  4.]\n  [-3. -1.  4.]], 'dimensionless')>)"
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "def deriv_2d_data_masked():\n    \"\"\"Return 2-dimensional data for analytic function for testing derivative functions.\"\"\"\n    ret = namedtuple('D_2D_Test_Data', 'x y x0 y0 a b f')(\n        np.array([0., 2., 7.]), np.array([1., 5., 11., 13.]), 3, 1.5, 0.5, 0.25, 0)\n\n    # Makes a value array with y changing along rows (axis 0) and x along columns (axis 1)\n    return ret._replace(f=np.ma.masked_outside(ret.a * (ret.x - ret.x0)**2 + ret.b * (ret.y[:, None] - ret.y0)**2, 5, 30))",
      "execution_count": 6,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "res_masked = metcalc.kinematics.gradient(deriv_2d_data_masked().f, x=(deriv_2d_data().y, deriv_2d_data().x))",
      "execution_count": 7,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "deriv_2d_data_masked().f",
      "execution_count": 8,
      "outputs": [
        {
          "data": {
            "text/plain": "masked_array(\n  data=[[--, --, 8.0625],\n        [7.5625, --, 11.0625],\n        [27.0625, 23.0625, --],\n        [--, --, --]],\n  mask=[[ True,  True, False],\n        [False,  True, False],\n        [False, False,  True],\n        [ True,  True,  True]],\n  fill_value=1e+20)"
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "res_masked",
      "execution_count": 9,
      "outputs": [
        {
          "data": {
            "text/plain": "(<Quantity([[ -0.35        -0.35         1.7875    ]\n  [ -0.15        -0.15        -0.2875    ]\n  [ -9.3359375   -0.04166667  -0.4609375 ]\n  [-17.7265625    0.04166667   0.4609375 ]], 'dimensionless')>,\n <Quantity([[-0.64285714 -0.35714286  0.35714286]\n  [-4.86160714 -2.70089286  2.70089286]\n  [-1.25357143 -2.74642857 -6.47857143]\n  [-0.64285714 -0.35714286  0.35714286]], 'dimensionless')>)"
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "res_masked[0].m",
      "execution_count": 10,
      "outputs": [
        {
          "data": {
            "text/plain": "masked_array(\n  data=[[ -0.35      ,  -0.35      ,   1.7875    ],\n        [ -0.15      ,  -0.15      ,  -0.2875    ],\n        [ -9.3359375 ,  -0.04166667,  -0.4609375 ],\n        [-17.7265625 ,   0.04166667,   0.4609375 ]],\n  mask=False,\n  fill_value=1e+20)"
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "So instead of having half of the values masked, the result has erroneous values like -0.35..."
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "conda-env-phd-py",
      "display_name": "Python [conda env:phd]",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.6.5",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "gist": {
      "id": "",
      "data": {
        "description": "Masked_Arrays_Gradient_MetPy.ipynb",
        "public": true
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}