{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0",
   "metadata": {},
   "source": [
    "# Preprocessing Raw Gaze Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1",
   "metadata": {},
   "source": [
    "## What you will learn in this tutorial:\n",
    "\n",
    "* how to transform pixel coordinates into degrees of visual angle\n",
    "* how to transform positional data into velocity data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2",
   "metadata": {},
   "source": [
    "## Preparations"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3",
   "metadata": {},
   "source": [
    "We import `pymovements` as the alias `pm` for convenience."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pymovements as pm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5",
   "metadata": {},
   "source": [
    "Let's start by downloading our `ToyDataset` and loading in its data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = pm.Dataset('ToyDataset', path='data/ToyDataset')\n",
    "dataset.download()\n",
    "dataset.load()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7",
   "metadata": {},
   "source": [
    "We can verify that all files have been loaded in by checking the `fileinfo` attribute:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.fileinfo"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9",
   "metadata": {},
   "source": [
    "Now let's inpect our gaze dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "10",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.gaze[0].frame.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11",
   "metadata": {},
   "source": [
    "Apart from some trial identifier columns we see the columns `time` and `pixel`."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12",
   "metadata": {},
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13",
   "metadata": {},
   "source": [
    "We now want to transform these pixel position coordinates into coordinates in degrees of visual angle.\n",
    "This is simply done by:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "14",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.pix2deg()\n",
    "\n",
    "dataset.gaze[0].frame"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15",
   "metadata": {},
   "source": [
    "The processed result has been added as a new column named `position` to our gaze dataframe."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16",
   "metadata": {},
   "source": [
    "Additionally we would like to have velocity data available too.\n",
    "We have four different methods available:\n",
    "\n",
    "* `preceding`: this will just take the single preceding sample in account for velocity calculation. Most noisy variant.\n",
    "* `neighbors`: this will take the neighboring samples in account for velocity calculation. A bit less noisy.\n",
    "* `smooth`: this will increase the neighboring samples to two on each side. You can get a smooth conversion this way.\n",
    "* `savitzky_golay`: this is using the *Savitzky-Golay* differentiation filter for conversion. You can specify additional parameters like `window_length` and `degree`. Depending on your parameters this will lead to the best results.\n",
    "\n",
    "Let's use the `fivepoint` method first:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "17",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.pos2vel(method='fivepoint')\n",
    "\n",
    "dataset.gaze[0].frame"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "18",
   "metadata": {},
   "source": [
    "The processed result has been added as a new column named `velocity` to our gaze dataframe."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19",
   "metadata": {},
   "source": [
    "We can also use the *Savitzky-Golay* differentiation filter with some additional parameters like this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "20",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.pos2vel(method='savitzky_golay', degree=2, window_length=7)\n",
    "\n",
    "dataset.gaze[0].frame"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21",
   "metadata": {},
   "source": [
    "This has overwritten our velocity columns.\n",
    "As we see, the values in the velocity columns are slightly different."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22",
   "metadata": {},
   "source": [
    " "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23",
   "metadata": {},
   "source": [
    "## What you have learned in this tutorial:\n",
    "\n",
    "* transforming pixel coordinates into degrees of visual angle by using `Dataset.pix2deg()`\n",
    "* transforming positional data into velocity data by using `Dataset.pos2vel()`\n",
    "* passing additional keyword arguments when using the *Savitzky-Golay* differentiation filter"
   ]
  }
 ],
 "metadata": {},
 "nbformat": 4,
 "nbformat_minor": 5
}