{
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Using the SceneWalk Model"
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import shutil\n",
        "import sys\n",
        "from collections import OrderedDict\n",
        "import numpy as np\n",
        "from matplotlib import pyplot as plt\n",
        "from scipy.stats import norm as normal, truncnorm as truncated_normal\n",
        "import seaborn as sns"
      ],
      "outputs": [],
      "execution_count": 1,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:44.688Z",
          "iopub.execute_input": "2020-03-28T10:08:44.692Z",
          "iopub.status.idle": "2020-03-28T10:08:45.692Z",
          "shell.execute_reply": "2020-03-28T10:08:45.707Z"
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "First of all- get data. This will depend on your config file. Mine has the following options."
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        }
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from scenewalk.utils import loadData\n",
        "loadData.get_set_names()"
      ],
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "spst_all\n",
            "spst_training\n",
            "spst_test\n",
            "corpus_test\n",
            "corpus_training\n"
          ]
        }
      ],
      "execution_count": 2,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:45.698Z",
          "iopub.execute_input": "2020-03-28T10:08:45.701Z",
          "iopub.status.idle": "2020-03-28T10:08:45.718Z",
          "shell.execute_reply": "2020-03-28T10:08:45.741Z"
        }
      }
    },
    {
      "cell_type": "code",
      "source": [
        "dataDict = loadData.load_data(\"spst_all\")\n",
        "x_dat, y_dat, dur_dat, im_dat, densities_dat, d_range = loadData.dataDict2vars(dataDict)"
      ],
      "outputs": [],
      "execution_count": 3,
      "metadata": {
        "collapsed": false,
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:45.725Z",
          "iopub.execute_input": "2020-03-28T10:08:45.729Z",
          "iopub.status.idle": "2020-03-28T10:08:45.751Z",
          "shell.execute_reply": "2020-03-28T10:08:45.744Z"
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "This data is in the format: subj[Trials[Fixations[]]]"
      ],
      "metadata": {}
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Scenewalk Model\n",
        "Here you define the setup of the model.The model object can "
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "from scenewalk.scenewalk_model_object import scenewalk as scenewalk_model"
      ],
      "outputs": [],
      "execution_count": 4,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:46.834Z",
          "iopub.execute_input": "2020-03-28T10:08:46.838Z",
          "iopub.status.idle": "2020-03-28T10:08:46.845Z",
          "shell.execute_reply": "2020-03-28T10:08:46.852Z"
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The model takes the settings:\n",
        "\n",
        "- how is the inhibition combined (subtractive or divisive)\n",
        "- how is the attention map initialized (zero or cb)\n",
        "- is there a post-saccadic attention shift? (on or off)\n",
        "- is there a pre-saccadic attention shift? (on or off)\n",
        "- is there an occulomotor potential? (on, off)\n",
        "- is there slower decay on the previous fixation? (on or off)\n",
        "\n",
        "It also takes the range of the data. All data for fixations must be provided in degrees. All duration information in seconds (NOT ms).\n",
        "\n",
        "Lastly it takes a dictionary of other settings for the model, relating to how parameters are passed in and represented in the model. In this case we are estimating log zeta and coupling the two omega parameters by a divisive parameter instead of estimating them separately.\n",
        "\n",
        "The configuration below is the original SceneWalk model without attention shifts"
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "sw = scenewalk_model(\"subtractive\", \"zero\", \"off\", \"off\", \"off\", {'x': d_range[0], 'y': d_range[1]}, {\"logged_z\":True, \"coupled_oms\":True})\n",
        "\n",
        "sw.whoami()"
      ],
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 5,
          "data": {
            "text/plain": [
              "'I am a subtractive scenewalk model, initialized with zero activation, in 2 exponents mode, with om_i as a fraction, with logged z'"
            ]
          },
          "metadata": {}
        }
      ],
      "execution_count": 5,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:47.338Z",
          "iopub.execute_input": "2020-03-28T10:08:47.341Z",
          "iopub.status.idle": "2020-03-28T10:08:47.350Z",
          "shell.execute_reply": "2020-03-28T10:08:47.354Z"
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "It will tell you which parameters it needs and the order if you're passing them in as a list:"
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "sw.get_param_list_order()"
      ],
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 6,
          "data": {
            "text/plain": [
              "['omegaAttention',\n",
              " 'omfrac',\n",
              " 'sigmaAttention',\n",
              " 'sigmaInhib',\n",
              " 'gamma',\n",
              " 'lamb',\n",
              " 'inhibStrength',\n",
              " 'zeta']"
            ]
          },
          "metadata": {}
        }
      ],
      "execution_count": 6,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:47.884Z",
          "iopub.execute_input": "2020-03-28T10:08:47.888Z",
          "iopub.status.idle": "2020-03-28T10:08:47.898Z",
          "shell.execute_reply": "2020-03-28T10:08:47.902Z"
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can pass them in as a list **in that order(!)** or as a dictionary with those keys"
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "sw.update_params([1,2,3,4,5,6,7,1])\n",
        "sw.get_params()"
      ],
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 7,
          "data": {
            "text/plain": [
              "OrderedDict([('omegaAttention', 1),\n",
              "             ('omegaInhib', 0.5),\n",
              "             ('sigmaAttention', 3),\n",
              "             ('sigmaInhib', 4),\n",
              "             ('gamma', 5),\n",
              "             ('lamb', 6),\n",
              "             ('inhibStrength', 7),\n",
              "             ('zeta', 10)])"
            ]
          },
          "metadata": {}
        }
      ],
      "execution_count": 7,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:48.571Z",
          "iopub.execute_input": "2020-03-28T10:08:48.576Z",
          "iopub.status.idle": "2020-03-28T10:08:48.586Z",
          "shell.execute_reply": "2020-03-28T10:08:48.590Z"
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The Model is defined within specific bounds. More on this elsewhere, but for roughly\n",
        "\n",
        "- **omegaAttention** float64.eps : 100000\n",
        "- **omegaInhib** float64.eps : 100000\n",
        "- **sigmaAttention** float64.eps : 100000\n",
        "- **sigmaInhib** float64.eps : 100000\n",
        "- **lamb** float64.eps : 15\n",
        "- **gamma** float64.eps : 15\n",
        "- **inhibStrength** float64.eps : 100000\n",
        "- **zeta** float64.eps : 1\n",
        "\n",
        "etc.\n"
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "# config like what the data suggest\n",
        "param_dict = OrderedDict({\n",
        "    \"omegaAttention\": 18.5,\n",
        "    \"omfrac\": 10,\n",
        "    \"sigmaAttention\": 6,\n",
        "    \"sigmaInhib\": 12,\n",
        "    \"gamma\" : 8,\n",
        "    \"lamb\" : 0.6,\n",
        "    \"inhibStrength\" :0.7,\n",
        "    \"zeta\" : -2,\n",
        "})"
      ],
      "outputs": [],
      "execution_count": 8,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:49.369Z",
          "iopub.execute_input": "2020-03-28T10:08:49.373Z",
          "iopub.status.idle": "2020-03-28T10:08:49.378Z",
          "shell.execute_reply": "2020-03-28T10:08:49.381Z"
        }
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sw.update_params(param_dict)\n",
        "#print(sw.get_params())\n",
        "print(\"have we provided all relevant parameters?:\", sw.check_params_for_config())\n",
        "print(\"are the parameters within the defined bounds?:\", sw.check_params_in_bounds())\n"
      ],
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "have we provided all relevant parameters?: True\n",
            "are the parameters within the defined bounds?: True\n"
          ]
        }
      ],
      "execution_count": 9,
      "metadata": {
        "collapsed": false,
        "outputHidden": false,
        "inputHidden": false,
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:49.745Z",
          "iopub.execute_input": "2020-03-28T10:08:49.748Z",
          "iopub.status.idle": "2020-03-28T10:08:49.756Z",
          "shell.execute_reply": "2020-03-28T10:08:49.759Z"
        }
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now we have set up our model object. It has 2 main interface functions:\n",
        "\n",
        "1. get_scanpath_likelihood: gives the likelihood of a scanpath given an image\n",
        "2. simulate_scanpath: simulates a scanpath on the image"
      ],
      "metadata": {
        "nteract": {
          "transient": {
            "deleting": false
          }
        }
      }
    },
    {
      "cell_type": "code",
      "source": [
        "subj = 1\n",
        "trial = 1\n",
        "image = im_dat[subj][trial][0]\n",
        "ll = sw.get_scanpath_likelihood(x_dat[subj][trial],\n",
        "                                y_dat[subj][trial],\n",
        "                                dur_dat[subj][trial],\n",
        "                                densities_dat[image-1])\n",
        "print(\"this scan paths sum ll, given the model and the parameters is:\", ll)"
      ],
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "this scan paths sum ll, given the model and the parameters is: -222.76976052646520093\n"
          ]
        }
      ],
      "execution_count": 10,
      "metadata": {
        "collapsed": false,
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "execution": {
          "iopub.status.busy": "2020-03-28T10:08:50.538Z",
          "iopub.execute_input": "2020-03-28T10:08:50.541Z",
          "iopub.status.idle": "2020-03-28T10:08:50.619Z",
          "shell.execute_reply": "2020-03-28T10:08:50.627Z"
        }
      }
    },
    {
      "cell_type": "code",
      "source": [
        "subj = 1\n",
        "trial = 1\n",
        "image = im_dat[subj][trial][0]\n",
        "x_path, y_path = sw.simulate_scanpath(dur_dat[subj][trial],\n",
        "                                      densities_dat[image-1],\n",
        "                                      [5,5])\n",
        "\n",
        "from scenewalk.plotting import sw_plot\n",
        "sw_plot.plot_path_on_map(densities_dat[image-1], x_path,y_path, sw)\n"
      ],
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": [
              "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\n"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "execution_count": 12,
      "metadata": {
        "collapsed": false,
        "jupyter": {
          "source_hidden": false,
          "outputs_hidden": false
        },
        "nteract": {
          "transient": {
            "deleting": false
          }
        },
        "execution": {
          "iopub.status.busy": "2020-03-28T10:09:03.285Z",
          "iopub.execute_input": "2020-03-28T10:09:03.290Z",
          "iopub.status.idle": "2020-03-28T10:09:03.564Z",
          "shell.execute_reply": "2020-03-28T10:09:03.573Z"
        }
      }
    }
  ],
  "metadata": {
    "kernel_info": {
      "name": "swtemp"
    },
    "language_info": {
      "name": "python",
      "version": "3.8.0",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "kernelspec": {
      "name": "swtemp",
      "language": "python",
      "display_name": "swtemp"
    },
    "nteract": {
      "version": "0.22.0"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 4
}