{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "execution": {},
    "id": "view-in-github"
   },
   "source": [
    "<a href=\"https://colab.research.google.com/github/NeuromatchAcademy/course-content/blob/main/projects/modelingsteps/TrainIllusionDataProject.ipynb\" target=\"_blank\"><img alt=\"Open In Colab\" src=\"https://colab.research.google.com/assets/colab-badge.svg\"/></a>   <a href=\"https://kaggle.com/kernels/welcome?src=https://raw.githubusercontent.com/NeuromatchAcademy/course-content/main/projects/modelingsteps/TrainIllusionDataProject.ipynb\" target=\"_blank\"><img alt=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"/></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "# Example Data Project: the Train Illusion\n",
    "\n",
    "Marius 't Hart, Megan Peters, Paul Schrater, Jean Laurens, Gunnar Blohm\n",
    "\n",
    "**Disclaimer**:  this is a \"toy\" data neuroscience project used to demonstrate the [10 step procedure of how-to-model](https://doi.org/10.1523/ENEURO.0352-19.2019). It is not meant to be state of the art research."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "---\n",
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Install dependencies\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "form",
    "execution": {},
    "tags": [
     "hide-input"
    ]
   },
   "outputs": [],
   "source": [
    "# @title Install dependencies\n",
    "!pip install tqdm --quiet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [],
   "source": [
    "# Imports\n",
    "\n",
    "# for matrices and plotting\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# for random distributions\n",
    "from scipy.stats import norm, poisson\n",
    "\n",
    "# for logistic regression\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import cross_val_score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Helper functions\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " Generate the Data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "form",
    "execution": {},
    "tags": [
     "hide-input"
    ]
   },
   "outputs": [],
   "source": [
    "# @title Helper functions\n",
    "# @markdown Generate the Data\n",
    "\n",
    "\n",
    "def generateSpikeTrains():\n",
    "\n",
    "  gain = 2\n",
    "  neurons = 50\n",
    "  movements = [0, 1, 2]\n",
    "  repetitions = 800\n",
    "\n",
    "  np.random.seed(37)\n",
    "\n",
    "  # set up the basic parameters:\n",
    "  dt = 1/100\n",
    "  start, stop = -1.5, 1.5\n",
    "  t = np.arange(start, stop + dt, dt)  # a time interval\n",
    "  Velocity_sigma = 0.5  # std dev of the velocity profile\n",
    "  Velocity_Profile = norm.pdf(t, 0, Velocity_sigma)/norm.pdf(0, 0, Velocity_sigma)  # The Gaussian velocity profile, normalized to a peak of 1\n",
    "\n",
    "  # set up the neuron properties:\n",
    "  Gains = np.random.rand(neurons) * gain  # random sensitivity between 0 and `gain`\n",
    "  FRs = (np.random.rand(neurons) * 60 ) - 10  # random base firing rate between -10 and 50\n",
    "\n",
    "  # output matrix will have this shape:\n",
    "  target_shape = [len(movements), repetitions, neurons, len(Velocity_Profile)]\n",
    "\n",
    "  # build matrix for spikes, first, they depend on the velocity profile:\n",
    "  Spikes = np.repeat(Velocity_Profile.reshape([1, 1, 1, len(Velocity_Profile)]),\n",
    "                     len(movements)*repetitions*neurons, axis=2).reshape(target_shape)\n",
    "\n",
    "  # multiplied by gains:\n",
    "  S_gains = np.repeat(np.repeat(Gains.reshape([1, 1, neurons]),\n",
    "                                len(movements)*repetitions, axis=1).reshape(target_shape[:3]),\n",
    "                      len(Velocity_Profile)).reshape(target_shape)\n",
    "  Spikes = Spikes * S_gains\n",
    "\n",
    "  # and multiplied by the movement:\n",
    "  S_moves = np.repeat( np.array(movements).reshape([len(movements), 1, 1, 1]),\n",
    "                      repetitions*neurons*len(Velocity_Profile), axis=3 ).reshape(target_shape)\n",
    "  Spikes = Spikes * S_moves\n",
    "\n",
    "  # on top of a baseline firing rate:\n",
    "  S_FR = np.repeat(np.repeat(FRs.reshape([1, 1, neurons]), len(movements)*repetitions, axis=1).reshape(target_shape[:3]),\n",
    "                   len(Velocity_Profile)).reshape(target_shape)\n",
    "  Spikes = Spikes + S_FR\n",
    "\n",
    "  # can not run the poisson random number generator on input lower than 0:\n",
    "  Spikes = np.where(Spikes < 0, 0, Spikes)\n",
    "\n",
    "  # so far, these were expected firing rates per second, correct for dt:\n",
    "  Spikes = poisson.rvs(Spikes * dt)\n",
    "\n",
    "  return Spikes\n",
    "\n",
    "\n",
    "def subsetPerception(spikes):\n",
    "\n",
    "  movements = [0, 1, 2]\n",
    "  split = 400\n",
    "  subset = 40\n",
    "  hwin = 3\n",
    "\n",
    "  [num_movements, repetitions, neurons, timepoints] = np.shape(spikes)\n",
    "\n",
    "  decision = np.zeros([num_movements, repetitions])\n",
    "\n",
    "  # ground truth for logistic regression:\n",
    "  y_train = np.repeat([0, 1, 1], split)\n",
    "  y_test = np.repeat([0, 1, 1], repetitions - split)\n",
    "\n",
    "  m_train = np.repeat(movements, split)\n",
    "  m_test = np.repeat(movements, split)\n",
    "\n",
    "  # reproduce the time points:\n",
    "  dt = 1/100\n",
    "  start, stop = -1.5, 1.5\n",
    "  t = np.arange(start, stop + dt, dt)\n",
    "\n",
    "  w_idx = list((abs(t) < (hwin*dt)).nonzero()[0])\n",
    "  w_0 = min(w_idx)\n",
    "  w_1 = max(w_idx) + 1  # python...\n",
    "\n",
    "  # get the total spike counts from stationary and movement trials:\n",
    "  spikes_stat = np.sum( spikes[0, :, :, :], axis=2)\n",
    "  spikes_move = np.sum( spikes[1:, :, :, :], axis=3)\n",
    "\n",
    "  train_spikes_stat = spikes_stat[:split, :]\n",
    "  train_spikes_move = spikes_move[:, :split, :].reshape([-1, neurons])\n",
    "\n",
    "  test_spikes_stat = spikes_stat[split:, :]\n",
    "  test_spikes_move = spikes_move[:, split:, :].reshape([-1, neurons])\n",
    "\n",
    "  # data to use to predict y:\n",
    "  x_train = np.concatenate((train_spikes_stat, train_spikes_move))\n",
    "  x_test = np.concatenate((test_spikes_stat, test_spikes_move))\n",
    "\n",
    "  # this line creates a logistics regression model object, and immediately fits it:\n",
    "  population_model = LogisticRegression(solver='liblinear',\n",
    "                                        random_state=0).fit(x_train, y_train)\n",
    "\n",
    "  # solver, one of: 'liblinear', 'newton-cg', 'lbfgs', 'sag', and 'saga'\n",
    "  # some of those require certain other options\n",
    "  #print(population_model.coef_)       # slope\n",
    "  #print(population_model.intercept_)  # intercept\n",
    "\n",
    "  ground_truth = np.array(population_model.predict(x_test))\n",
    "  ground_truth = ground_truth.reshape([3, -1])\n",
    "\n",
    "  output = {}\n",
    "  output['perception'] = ground_truth\n",
    "  output['spikes'] = spikes[:, split:, :subset, :]\n",
    "\n",
    "  return output\n",
    "\n",
    "\n",
    "def getData():\n",
    "\n",
    "  spikes = generateSpikeTrains()\n",
    "\n",
    "  dataset = subsetPerception(spikes=spikes)\n",
    "\n",
    "  return dataset\n",
    "\n",
    "\n",
    "dataset = getData()\n",
    "perception = dataset['perception']\n",
    "spikes = dataset['spikes']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Plot Functions\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "form",
    "execution": {},
    "tags": [
     "hide-input"
    ]
   },
   "outputs": [],
   "source": [
    "# @title Plot Functions\n",
    "\n",
    "def rasterplot(spikes,movement,trial):\n",
    "\n",
    "  [movements, trials, neurons, timepoints] = np.shape(spikes)\n",
    "  trial_spikes = spikes[movement,trial, :, :]\n",
    "  trial_events = [((trial_spikes[x, :] > 0).nonzero()[0] - 150)/100 for x in range(neurons)]\n",
    "  dt = 1/100\n",
    "\n",
    "  plt.figure()\n",
    "  plt.eventplot(trial_events, linewidths=1);\n",
    "  plt.title('movement: %d - trial: %d'%(movement, trial))\n",
    "  plt.ylabel('neuron')\n",
    "  plt.xlabel('time [s]')\n",
    "\n",
    "\n",
    "def plotCrossValAccuracies(accuracies):\n",
    "  f, ax = plt.subplots(figsize=(8, 3))\n",
    "  ax.boxplot(accuracies, vert=False, widths=.7)\n",
    "  ax.scatter(accuracies, np.ones(8))\n",
    "  ax.set(\n",
    "    xlabel=\"Accuracy\",\n",
    "    yticks=[],\n",
    "    title=f\"Average test accuracy: {accuracies.mean():.2%}\"\n",
    "  )\n",
    "  ax.spines[\"left\"].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Phenomenon\n",
    "*Part of Steps 1-2*\n",
    "\n",
    "The train illusion occurs when sitting on a train and viewing another train outside the window. Suddenly, the other train *seems* to move, i.e. you experience visual motion of the other train relative to your train. But which train is actually moving?\n",
    "\n",
    "Often people mix this up. In particular, they think their own train might be moving when it's the other train that moves; or vice versa. The illusion is usually resolved once you gain vision of the surroundings that lets you disambiguate the relative motion; or if you experience strong vibrations indicating that it is indeed your own train that is in motion."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Question\n",
    "\n",
    "*Part of Step 1*\n",
    "\n",
    "We assume that we have build the train illusion model (see the other example project colab). That model predicts that accumulated sensory evidence from vestibular signals determines the decision of whether self-motion is experienced or not. We now have vestibular neuron data (simulated in our case, but let's pretend) and would like to see if that prediction holds true.\n",
    "\n",
    "The data contains $N$ neurons and $M$ trials for each of 3 motion conditions: no self-motion, slowly accelerating self-motion and faster accelerating self-motion.\n",
    "\n",
    "\\begin{align}\n",
    "N &= 40\\\\\n",
    "M &= 400\\\\\n",
    "\\end{align}\n",
    "\n",
    "**So we can ask the following question**: \"Does accumulated vestibular neuron activity correlate with self-motion judgements?\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "# Background\n",
    "\n",
    "_Part of step 2_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "While it seems a well-known fact that vestibular signals are noisy, we should check if we can also find this in the literature."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "Let's also see what's in our data, there should be a 4d array called `spikes` that has spike counts (positive integers), a 2d array called `perception` with self-motion judgements (0=no motion or 1=motion). Let's see what this data looks like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3, 400, 40, 301)\n",
      "(3, 400)\n",
      "[0.355  0.7575 0.975 ]\n"
     ]
    }
   ],
   "source": [
    "print(np.shape(spikes))\n",
    "print(np.shape(perception))\n",
    "print(np.mean(perception, axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "In the `spikes` array, we see our 3 acceleration conditions (first dimension), with 400 trials each (second dimensions) and simultaneous recordings from 40 neurons (third dimension), across 3 seconds in 10 ms bins (fourth dimension). The first two dimensions are also there in the `perception` array.\n",
    "\n",
    "Perfect perception would have looked like [0, 1, 1]. The average judgements are far from correct (lots of self-motion illusions) but they do make some sense: it's closer to 0 in the no-motion condition and closer to 1 in both of the real-motion conditions.\n",
    "\n",
    "The idea of our project is that the vestibular signals are noisy so that they might be mis-interpreted by the brain. Let's see if we can reproduce the stimuli from the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for move_no in range(3):\n",
    "  plt.plot(np.arange(-1.5, 1.5 + (1/100),\n",
    "                     (1/100)),\n",
    "           np.mean(np.mean(spikes[move_no, :, :, :], axis=0), axis=0),\n",
    "           label=['no motion', '$1 m/s^2$', '$2 m/s^2$'][move_no])\n",
    "plt.xlabel('time [s]');\n",
    "plt.ylabel('averaged spike counts');\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "Blue is the no-motion condition, and produces flat average spike counts across the 3 s time interval. The orange and green line do show a bell-shaped curve that corresponds to the acceleration profile. But there also seems to be considerable noise: exactly what we need. Let's see what the spike trains for a single trial look like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for move in range(3):\n",
    "  rasterplot(spikes=spikes, movement=move, trial=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "You can change the trial number in the bit of code above to compare what the rasterplots look like in different trials. You'll notice that they all look kind of the same: the 3 conditions are very hard (impossible?) to distinguish by eye-balling.\n",
    "\n",
    "Now that we have seen the data, let's see if we can extract self-motion judgements from the spike counts."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Ingredients\n",
    "\n",
    "*Part of step 3*\n",
    "\n",
    "In order to address our question we need to design an appropriate computational data analysis pipeline. We did some brainstorming and think that we need to somehow extract the self-motion judgements from the spike counts of our neurons. Based on that, our algorithm needs to make a decision: was there self motion or not? This is a classical 2-choice classification problem. We will have to transform the raw spike data into the right input for the algorithm (spike pre-processing).\n",
    "\n",
    "So we determined that we probably needed the following ingredients:\n",
    "\n",
    "* spike trains $S$ of 3-second trials (10ms spike bins)\n",
    "* ground truth movement $m_r$ (real) and perceived movement $m_p$\n",
    "* some form of classifier $C$ giving us a classification $c$\n",
    "* spike pre-processing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Hypotheses\n",
    "\n",
    "*Part of step 4*\n",
    "\n",
    "We think that noise in the signal drives whether or not people perceive self motion. Maybe the brain uses the strongest signal at peak acceleration to decide on self motion, but we actually think it is better to accumulate evidence over some period of time. We want to test this. The noise idea also means that when the signal-to-noise ratio is higher, the brain does better, and this would be in the faster acceleration condition. We want to test this too.\n",
    "\n",
    "We came up with the following hypotheses focussing on specific details of our overall research question:\n",
    "\n",
    "* Hyp 1: Accumulated vestibular spike rates explain self-motion judgements better than average spike rates around peak acceleration.\n",
    "* Hyp 2: Classification performance should be better for faster vs slower self-motion.\n",
    "\n",
    "> There are many other hypotheses you could come up with, but for simplicity, let's go with those.\n",
    "\n",
    "Mathematically, we can write our hypotheses as follows (using our above ingredients):\n",
    "* Hyp 1: $\\mathbb{E}[c_{accum}]>\\mathbb{E}[c_{win}]$\n",
    "* Hyp 2: $\\mathbb{E}[c_{fast}]>\\mathbb{E}[c_{slow}]$\n",
    "\n",
    "Where $\\mathbb{E}[\\cdot]$ denotes the expected value (in this case the mean) of its argument, i.e., classification outcome in a given trial type."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Selected toolkit\n",
    "\n",
    "*Part of step 5*\n",
    "\n",
    "We want to build some form of classification pipeline. There are many options here, but a simple decoder would be a good start. Since we will learn all about Generalized Linear Models soon, we decided to use that! From W1D4: Generalized Linear Models there were two models that looked relevant:\n",
    "\n",
    "* Linear-Nonlinear-Poisson GLM: predict spike counts from stimulus\n",
    "* Logistic Regression: predict stimulus from spike counts\n",
    "\n",
    "Here we want to predict self-motion from spike counts, so we pick Logistic Regression."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Model draft\n",
    "\n",
    "*Part of step 6*\n",
    "\n",
    "Brainstorming on the whiteboard, we came up with this set of steps:\n",
    "\n",
    "1. convert spike count data into design matrix: trials x neurons (with accumulated or windowed data)\n",
    "2. get the classes each trial belongs to: stimulus (stationary / moving) or self-motion judgements (moving or not?) as a vector array\n",
    "3. split data into train and test sets (200 trials each for the 3 conditions?) with an equal number of each class\n",
    "4. fit logistic regression model on training set, and get predictions for test set\n",
    "\n",
    "<p align='center'><img src='https://github.com/NeuromatchAcademy/course-content/blob/main/tutorials/static/data_analysis_step6.jpeg?raw=True'/></p>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Model implementation\n",
    "\n",
    "*Part of step 7*\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "Below is a function that gets us the design matrix of observations X features (in this case: trials X neurons). It sums all spike counts, either across the whole time interval, or a window, specified in seconds before and after the peak acceleration at time=0 (it is the half-width of the window: `halfwin`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(800, 40)\n",
      "[76.33475 77.53275 78.61975]\n",
      "76.33475\n",
      "77.53275\n"
     ]
    }
   ],
   "source": [
    "m_r = np.repeat([0, 1, 1], 400)\n",
    "m_p = perception.reshape([-1])\n",
    "a_r = np.repeat([0, 1, 2], 400)  # for grouping in cross validation? maybe not\n",
    "\n",
    "\n",
    "def getDesignMatrix(spikes, motions, classes, halfwin=None):\n",
    "  \"\"\"\n",
    "    Get the design matrix and vector of classes for a logistic regression.\n",
    "\n",
    "    Args:\n",
    "        spikes (numpy.ndarray): a 4d matrix with spike counts for 1) three motion\n",
    "        conditions, 2) 400 trials, 3) 40 neurons, 4) 301 time bins\n",
    "        motions (list): motion conditions to include, any subset of [0,1,2]\n",
    "        classes (list): list of length 3 * 400 with classes to use for the\n",
    "        logistic regression\n",
    "        halfwin (None or float): None to use the whole 3 s time interval, or a\n",
    "        value between 0 and 1.5 seconds as the extent of the window from 0 (when\n",
    "        acceleration peaks) in both directions.\n",
    "\n",
    "    Returns:\n",
    "        (list of numpy.ndarray): first element of the list is the design matrix\n",
    "        of shape [trials] X [neurons], second elements is a 1d matrix to use as\n",
    "        a vector of classes to predict from the design matrix\n",
    "\n",
    "    This function can be used to get the right design matrix and set of class\n",
    "    labels for a variety of logistic regressions we might want to do.\n",
    "  \"\"\"\n",
    "\n",
    "  # select motion conditions:\n",
    "  spikes = spikes[motions, :, :, :]\n",
    "  y = (np.array(classes).reshape([3, -1])[motions, :]).reshape([-1])\n",
    "\n",
    "  [movstims, trials, neurons, timepoints] = np.shape(spikes)\n",
    "\n",
    "  # first we get the window (if any):\n",
    "  if halfwin is None:\n",
    "    return [np.sum(spikes, axis=3).reshape([movstims*trials, neurons]), y]\n",
    "  else:\n",
    "    dt = 1/100\n",
    "    t = np.arange(-1.5,1.5+dt,dt)\n",
    "    win_idx = (abs(t) < halfwin).nonzero()[0]\n",
    "    w_0, w_1 = min(win_idx), max(win_idx)+1\n",
    "    return [np.sum(spikes[:, :, :, w_0:w_1], axis=3).reshape([movstims*trials, neurons]), y]\n",
    "\n",
    "\n",
    "# test the function:\n",
    "[desmat, y] = getDesignMatrix(spikes, motions=[0,1], classes=m_r)\n",
    "# let's check the shape:\n",
    "print(np.shape(desmat))\n",
    "# and let's check if the mean sum of spikes checks out:\n",
    "print(np.mean(np.sum(spikes, axis=3),axis=(1, 2)))\n",
    "for mov in range(2):\n",
    "  print(np.mean(desmat[(mov*400):((mov+1)*400), :]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "First, the design matrix: `desmat` has the right shape: 800 trials (for two conditions), by 40 neurons.\n",
    "\n",
    "We also get the same average spike counts from our function (76.3 and 77.5) as when calculating it 'by hand' ([76.33475 77.53275 78.61975]).\n",
    "\n",
    "This means our function `getDesignMatrix()` works correctly (unit test)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "It also gets us the classes for each trial. We can use both judgements, $m_p$, and stimuli, $m_r$, as one vector, and `getDesignMatrix()` returns the correct subset (depending on which motion conditions are used) as the second variable."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "We wanted to split the data into a training and test set, but we can have scikit-learn do this for us with the `cross_val_score()` function, as we saw in the GLM day:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "accuracies = cross_val_score(LogisticRegression(solver='liblinear'), X=desmat, y=y, cv=8)\n",
    "plotCrossValAccuracies(accuracies) # this plotting function is copied from W1D4: Generalized Linear Models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "We asked for 8 cross validations, which show up as the blue dots in the graph (two have the same accuracy). Prediction accuracy ranges from 56% to 72%, with the average at 65%, and the orange line is the median. Given the noisy data, that is not too bad actually.\n",
    "\n",
    "But maybe it's better to split it according to the conditions where there is motion but of different magnitude. It should work better to classify higher acceleration motion from no motion as compared to classifying the lower acceleration motion.\n",
    "\n",
    "Maybe it also works better to ignore some of the noise at the beginning and end of each trial by focussing on the spikes around the maximum acceleration, using our window option. The average spike count plot above, seems to best discriminate between the three motion conditions around peak acceleration (at 0 s).\n",
    "\n",
    "We also want to test if it's possible to predict the self-motion judgements rather than the actual motion.\n",
    "\n",
    "So we write a function that brings it all together:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "m_r = np.repeat([0, 1, 1], 400)\n",
    "m_p = perception.reshape([-1])\n",
    "\n",
    "\n",
    "def classifyMotionFromSpikes(spikes, classes,\n",
    "                             halfwin=None,\n",
    "                             motions=[0, 1, 2],\n",
    "                             cv=8):\n",
    "  \"\"\"\n",
    "    Runs one logistic regression using the specified parameters.\n",
    "\n",
    "    Args:\n",
    "        spikes (numpy.ndarray): a 4d matrix with spike counts for 1) three motion\n",
    "        conditions, 2) 400 trials, 3) 40 neurons, 4) 301 time bins\n",
    "        classes (list): list of length 3 * 400 with classes to use for the\n",
    "        logistic regression\n",
    "        halfwin: None to use the whole 3 s time interval, or a value between 0\n",
    "        and 0.750 seconds as the extent of the window from 0 (when acceleration\n",
    "        peaks)\n",
    "        motions (list): motion conditions to include, any subset of [0,1,2]\n",
    "        cv (int): number of cross validations to do in logistic regression\n",
    "\n",
    "    Returns:\n",
    "        (numpy.ndarray): 1d array of shape (cv,) with prediction accuracies for\n",
    "        each cross validation\n",
    "\n",
    "    This function uses our getDesignMatrix() function according to specification\n",
    "    and then uses cross-validated logistic regression and returns the accuracy\n",
    "    for each run of the model.\n",
    "  \"\"\"\n",
    "\n",
    "  # get the right design matrix:\n",
    "  X, y = getDesignMatrix(spikes, motions, classes=classes, halfwin=halfwin)\n",
    "\n",
    "  # right now, we are not using regularization:\n",
    "  return cross_val_score(LogisticRegression(solver='liblinear'), X=X, y=y, cv=cv)\n",
    "\n",
    "\n",
    "accuracies = classifyMotionFromSpikes(spikes, m_r, motions=[0, 1])\n",
    "plotCrossValAccuracies(accuracies)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "This is the exact same figure as before, so our function `classifyMotionFromSpikes()` also works as intended."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Model completion\n",
    "\n",
    "*Part of step 8*\n",
    "\n",
    "**Can we answer our question?** Question: \"Does accumulated vestibular neuron activity correlate with self-motion judgements?\" Yes, we can get an answer to the question by applying our analysis to predict self-motion judgements.\n",
    "\n",
    "**Can we speak to our hypothesis?** We had two hypotheses: 1) prediction is better with total accumulated spike counts than from a window around peak acceleration, and 2) prediction is better with higher acceleration as the signal-to-noise ratio increases.\n",
    "\n",
    "**Does the model reach our goals?** We want to understand if motion judgement and not true motion correlates better with vestibular neural activity.\n",
    "\n",
    "Note: We have somewhat refined our goal here by explicitly contrasting true motion with motion judgements."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Model evaluation & testing\n",
    "\n",
    "*Part of step 9*\n",
    "\n",
    "We can now evaluate the performance of our analysis pipeline.\n",
    "\n",
    "To do so, we will run the logistic regression quite a few times: on all data or split by the velocities, and then we use the data to predict the presence of real motion, or the judgements of self-motion. We repeat this for total accumulated spikes and for a window around the peak acceleration. And then we'll plot the average classification performance in those 12 cases."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [],
   "source": [
    "def runAnalysis(spikes):\n",
    "  \"\"\"\n",
    "    Runs one logistic regression using the specified parameters.\n",
    "\n",
    "    Args:\n",
    "        spikes (numpy.ndarray): a 4d matrix with spike counts for 1) three motion\n",
    "        conditions, 2) 400 trials, 3) 40 neurons, 4) 301 time bins\n",
    "\n",
    "    Returns:\n",
    "        (numpy.ndarray): array of shape (2, 2, 3, 12), where the first dimension\n",
    "        denotes full interval or 100 ms window, the second dimension denotes\n",
    "        real motion or self-motion judgements, the third dimension denotes slow,\n",
    "        fast or all data, and for each there 12 prediction accuracies for each\n",
    "        of the 12 cross validations\n",
    "\n",
    "    This function uses our classifyMotionFromSpikes()) function according to\n",
    "    specification for all the different ways we want to classify the data and\n",
    "    gives us the classification performance.\n",
    "  \"\"\"\n",
    "\n",
    "  # variables we use for testing our analysis pipeline:\n",
    "  m_r = np.repeat([0, 1, 1], 400)\n",
    "  m_p = perception.reshape([-1])\n",
    "  halfwins = [None, 0.050]\n",
    "  motions = [[0, 1], [0, 2], [0, 1, 2]]\n",
    "  class_sets = [m_r, m_p]\n",
    "  cv = 12\n",
    "\n",
    "  # empty array to collect classification performance:\n",
    "  accuracies = np.zeros([len(halfwins), len(class_sets), len(motions), cv])\n",
    "\n",
    "  for halfwin_no in range(len(halfwins)):\n",
    "\n",
    "    halfwin = halfwins[halfwin_no]\n",
    "    lty = ['-', '--'][halfwin_no]\n",
    "    leg_hw = ['accumulated', '100 ms window'][halfwin_no]\n",
    "\n",
    "    for classes_no in range(len(class_sets)):\n",
    "\n",
    "      classes = class_sets[classes_no]\n",
    "      leg_class = ['real', 'judgements'][classes_no]\n",
    "      color = ['orange','purple'][classes_no]\n",
    "\n",
    "      for motions_no in range(len(motions)):\n",
    "\n",
    "        motion = motions[motions_no]\n",
    "        cond_acc = classifyMotionFromSpikes(spikes = spikes,\n",
    "                                            classes = classes,\n",
    "                                            halfwin = halfwin,\n",
    "                                            motions = motion,\n",
    "                                            cv=12)\n",
    "        accuracies[halfwin_no, classes_no, motions_no, :] = cond_acc\n",
    "\n",
    "      m_acc = np.mean(accuracies[halfwin_no, classes_no, :, :], axis=1)\n",
    "\n",
    "  return accuracies\n",
    "\n",
    "\n",
    "# here we run the above function and store the output:\n",
    "accuracies = runAnalysis(spikes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "We will need to plot this output, and to simplify the code, we do that in a separate function, which we write here:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plotAccuracies(accuracies):\n",
    "  \"\"\"\n",
    "    Plot the accuracies from our main fitting function.\n",
    "\n",
    "    Args:\n",
    "        accuarcies (numpy.ndarray): a 4d matrix with performance accuracy for\n",
    "        1) full interval or 100 ms window, 2) real motion or self-motion\n",
    "        judgements, 3) slow, fast or all conditions, and 4) the 12 cross\n",
    "        validations\n",
    "\n",
    "    Returns:\n",
    "\n",
    "    This function plots the output of our data analysis pipeline and does not\n",
    "    return anything.\n",
    "  \"\"\"\n",
    "  ax = plt.figure(figsize=(6, 4))\n",
    "  plt.title('classification accuracy')\n",
    "  plt.xlim([0.5, 3.5])\n",
    "  plt.ylim([0.4, 1.0])\n",
    "  plt.ylabel('proportion correct')\n",
    "  plt.xlabel('velocity condition vs. stationary')\n",
    "\n",
    "  for halfwin_no in range(2):\n",
    "\n",
    "    lty = ['-', '--'][halfwin_no]\n",
    "    leg_hw = ['accumulated', '100 ms window'][halfwin_no]\n",
    "\n",
    "    for classes_no in range(2):\n",
    "\n",
    "      leg_class = ['real', 'judgements'][classes_no]\n",
    "      color = ['orange','purple'][classes_no]\n",
    "\n",
    "      m_acc = np.mean(accuracies[halfwin_no,classes_no, :, :], axis=1)\n",
    "      plt.plot([1, 2, 3], m_acc, lty, color=color,\n",
    "               label=f\"{leg_class} {leg_hw}\")\n",
    "\n",
    "  plt.xticks(ticks=[1, 2, 3],\n",
    "             labels=['slow ($1 m/s^2$)','fast ($2 m/s^2$)','both'])\n",
    "  plt.legend(loc='lower right')\n",
    "  plt.show()\n",
    "\n",
    "\n",
    "plotAccuracies(accuracies)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "Well, that's interesting! The logistic regression doesn't do a perfect job, but there is information in these results.\n",
    "\n",
    "1. The dashed lines reflect predictions based on a small window of data, and they do worse than the full data set: this could mean that the brain also integrates signals across longer time frames for perception.\n",
    "\n",
    "2. In the predictions based on accumulated spike counts, the slow movements are harder to separate from no movements than the faster movements. This is clearer when predicting real motion than when predicting self-motion judgements.\n",
    "\n",
    "Those were our two hypotheses, but we notice something else:\n",
    "\n",
    "3. Self-motion judgments display higher decoding accuracy than the actual motion.\n",
    "\n",
    "Actually, if self-motion judgements and our logistic regression use input from the same noisy sensors, it kind of makes sense that they would both give similar output. This is in line the notion that self-motion judgements can be wrong because the underlying sensory signals are noisy. Of course, this only works if we record activity from neuronal populations that contribute to self-motion judgements. On the other hand, we would also see this result if the sensory signal was not noisy and we recorded from one of several populations that contribute to self-motion judgements in a noisy way. So we need to do more research here.\n",
    "\n",
    "Either way, we learned something today!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Summary\n",
    "*Part of Step 10*\n",
    "\n",
    "Let's write a simple abstract following the guidelines.\n",
    "\n",
    "**A. What is the phenomena?** Here summarize the part of the phenomena which your model addresses.\n",
    "\n",
    "_When sitting in a stationary train and seeing the train next to them move, people may experience the \"train illusion\": they feel like they are moving when they are not, or vice versa. Vestibular information can disambiguate self motion from motion of the adjacent train._\n",
    "\n",
    "**B. What is the key scientific question?** Clearly articulate the question which your model tries to answer.\n",
    "\n",
    "_However, it is unknown if accumulated, noisy vestibular neuron activity correlates with self-motion judgement._\n",
    "\n",
    "**C. What was our hypothesis?** Explain the key relationships which we relied on to simulate the phenomena.\n",
    "\n",
    "_Based on previous modeling efforts, we hypothesized that cumulative spike counts (not instantaneous) recorded from vestibular neurons can better discriminate experienced self motion judgements as opposed to true self motion._\n",
    "\n",
    "**D. How did your model work?** Give an overview of the model, it's main components, and how the model works. \"Here we ...\"\n",
    "\n",
    "_Here we use logistic regression to classify both true self motion and experienced self motion judgements from vestibular neuron activity. This was done both with the vestibular signal accumulated across the full trial and with a 100 ms window ('instantaneous') around peak acceleration, and with 2 different speeds compared to a no-motion condition._\n",
    "\n",
    "**E. What did we find?** Did the model work? Explain the key outcomes of your model evaluation.\n",
    "\n",
    "_Decoding performance for accumulated vestibular signals was higher than for instantaneous signals. Further, experienced self motion judgements could be classified better than true self motion._\n",
    "\n",
    "**F. What can we conclude?** Conclude as much as you can with reference to the hypothesis, within the limits of the model.\n",
    "\n",
    "_We conclude that accumulated, noisy vestibular signals during slowly-accelerating self motion may drive the train illusion._\n",
    "\n",
    "**G. What are the limitations and future directions?** What is left to be learned? Briefly argue the plausibility of the approach and/or what you think is essential that may have been left out.\n",
    "\n",
    "_Future research should examine how visual signals may combine with vestibular signals in real-world scenarios to causally link these sensory signals to self-motion experience and illusions._\n",
    "\n",
    "\n",
    "If we put this all in one paragraph, we have our final complete abstract. But, first, do not include the letters in _your_ abstract, and second, we did paraphrase the answers a little so they fit together.\n",
    "\n",
    "<br>\n",
    "**Abstract**\n",
    "\n",
    "(A) When sitting in a stationary train and seeing the train next to them move, people may experience the \"train illusion\": they feel like they are moving when they are not, or vice versa. Vestibular information can disambiguate self motion from motion of the adjacent train. (B) However, it is unknown if accumulated, noisy vestibular neuron activity correlates with self-motion judgement. (C) Based on previous modeling efforts, we hypothesized that cumulative spike counts (not instantaneous) recorded from vestibular neurons can better discriminate experienced self motion judgements as opposed to true self motion. (D) Here we use logistic regression to classify both true self motion and experienced self motion judgements from vestibular neuron activity. This was done both with the vestibular signal accumulated across the full trial and with a 100 ms window (instantaneous) around peak acceleration, and with 2 different speeds compared to a no-motion condition. (E) Decoding performance for accumulated vestibular signals was higher than for instantaneous signals. Further, experienced self motion judgements could be classified better than true self motion. (F) We conclude that accumulated, noisy vestibular signals during slowly-accelerating self motion may drive the train illusion. (G) Future research should model possible mechanisms by which visual signals may combine with vestibular signals to causally link these sensory signals to self-motion experience and illusions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {}
   },
   "source": [
    "----\n",
    "# Final thoughts\n",
    "\n",
    "Note that the analysis pipeline we built here was extremely simple and we used artificial data on purpose. It allowed us to go through all the steps of building a data neuroscience project, and hopefully you noticed that it is not always a linear process, you will go back to different steps if you hit a roadblock somewhere.\n",
    "\n",
    "There are many issues that we did not address, such as:\n",
    "* we could have tried different decoders\n",
    "* or find the optimal window\n",
    "* what if we had more data...\n",
    "* how do different neurons contribute to the decision? And why?\n",
    "  * this is something that you want to explore as a neuroscientist\n",
    "* we could have run a GLM on the neurons*time array and then analyze the weightings to see if all acceleration steps are weighted equally (perfect integration)... or not?\n",
    "* what is the role of visual motion in this phenomenon?\n",
    "\n",
    "However, this project is not meant to be complete, and yours doesn't have to be either. The goal of the projects is to go through _the process_ of a modeling or data science project and put into practice one or more of the toolkits you learn at NMA with your group.\n"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "include_colab_link": true,
   "name": "TrainIllusionDataProject",
   "provenance": [],
   "toc_visible": true
  },
  "kernel": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}