{ "cells": [ { "cell_type": "markdown", "id": "de6415f5-67e5-4335-83a2-b51106db7e85", "metadata": {}, "source": [ "# MNE-RSA: representational similarity analysis on sensor-level MEG data\n", "\n", "Ok, let's go!\n", "\n", "The dataset we will be working with today is the [Wakeman & Nelson (2015) \"faces\" dataset](https://www.nature.com/articles/sdata20151). During this experiment, participants were presented with a series of images, containing:\n", " - Faces of famous people that the participants likely knew\n", " - Faces of people that the participants likely did not know\n", " - Scrambled faces: the images were cut-up and randomly put together again\n", "\n", "In this tutorial, we are going to use this dataset to explore the neural representational code within the visual cortex.\n", "From time to time, there will be green blocks indicating it's up to you to do something, like this one:\n", "\n", "
\n", "EXERCISE:\n", " \n", "In the cell below, update the `data_path` variable to point to where you have extracted the [`rsa-data.zip`](https://github.com/wmvanvliet/neuroscience_tutorials/releases/download/2/rsa-data.zip) file to.\n", "\n", "(If you are running this on MyBinder then the data is located in the `data` folder).\n", "
" ] }, { "cell_type": "code", "execution_count": 1, "id": "99e72799-bae9-4e4c-b335-98f653f7b110", "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", "import matplotlib as mpl\n", "mpl.rcParams['figure.dpi'] = 90 # Tune this to make figures bigger/smaller\n", "\n", "# Set this to where you've extracted `data.zip` to\n", "data_path = \"data\"" ] }, { "cell_type": "markdown", "id": "2561afd3-1314-4e13-aecc-c613af6608e3", "metadata": {}, "source": [ "## A representational code for the stimuli\n", "\n", "Let's start by taking a look at the stimuli that were presented during the experiment.\n", "I've put them in the `stimuli` folder for you as `.bmp` image files.\n", "The Python Imaging Library (PIL) can open them and display them in this notebook.\n", "We can use the notebook's native [`IPython.display.display`](https://ipython.readthedocs.io/en/stable/api/generated/IPython.display.html#IPython.display.display) function if we want to display more than one image at once." ] }, { "cell_type": "code", "execution_count": 2, "id": "83490c2f-bdeb-41de-aa00-ecf478f71440", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Famous face: 128 x 162 pixels\n" ] }, { "data": { "image/jpeg": "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", "image/png": "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", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Scrambled face: 128 x 162 pixels\n" ] }, { "data": { "image/jpeg": "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", "image/png": "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", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from PIL import Image\n", "from IPython.display import display\n", "\n", "# Show the first \"famous\" face and the first \"scrambled\" face\n", "img_famous = Image.open(f\"{data_path}/stimuli/f001.bmp\")\n", "img_scrambled = Image.open(f\"{data_path}/stimuli/s001.bmp\")\n", "\n", "print(f\"Famous face: {img_famous.width} x {img_famous.height} pixels\")\n", "display(img_famous)\n", "print(f\"Scrambled face: {img_scrambled.width} x {img_scrambled.height} pixels\")\n", "display(img_scrambled)" ] }, { "cell_type": "markdown", "id": "e972de76-be98-440b-8e6a-0bf019639f4d", "metadata": {}, "source": [ "Loaded like this, the stimuli are in a representational space defined by their pixels.\n", "Each image is represented by 128 x 162 = 20736 values between 0 (black) and 255 (white).\n", "Let's create a Representational Dissimilarity Matrix (RDM) where images are compared based on the difference between their pixels.\n", "To get the pixels of an image, you can convert it to a NumPy array like this:" ] }, { "cell_type": "code", "execution_count": 3, "id": "791d2d5b-cd93-4865-978b-4241f25d75ba", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Shape of the pixel array for the famous face: (162, 128)\n", "Shape of the pixel array for the scrambled face: (162, 128)\n" ] } ], "source": [ "import numpy as np\n", "pixels_famous = np.array(img_famous)\n", "pixels_scrambled = np.array(img_scrambled)\n", "\n", "print(\"Shape of the pixel array for the famous face:\", pixels_famous.shape)\n", "print(\"Shape of the pixel array for the scrambled face:\", pixels_scrambled.shape)" ] }, { "cell_type": "markdown", "id": "a8d39fd1-aa8e-47cc-a366-80903fe9d08b", "metadata": {}, "source": [ "We can now compute the \"dissimilarity\" between the two images, based on their pixels.\n", "For this, we need to decide on a metric to use.\n", "The default metric used in the original publication ([Kiegeskorte et al. 2008](https://www.frontiersin.org/articles/10.3389/neuro.06.004.2008/full)) was Pearson Correlation, so let's use that.\n", "Of course, correlation is a metric of similarity and we want a metric of *dis*similarity.\n", "Let's make it easy on ourselves and just do $1 - r$." ] }, { "cell_type": "code", "execution_count": 4, "id": "b7cf207a-5534-4813-98b5-1013011722f2", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The dissimilarity between the pixels of the famous and scrambled faces is: 0.418\n" ] } ], "source": [ "from scipy.stats import pearsonr\n", "similarity, _ = pearsonr(pixels_famous.flatten(), pixels_scrambled.flatten())\n", "dissimilarity = 1 - similarity\n", "print(f\"The dissimilarity between the pixels of the famous and scrambled faces is: {dissimilarity:.3f}\")" ] }, { "cell_type": "markdown", "id": "e1cd613f-ee10-48d0-9a63-1da6dbfd7abd", "metadata": {}, "source": [ "To construct the full RDM, we need to do this for all pairs of images.\n", "I'll talk you through the process, but I will let you do the coding for this.\n", "Ready? Let's go!\n", "\n", "
\n", "EXERCISE:\n", " \n", "In the cell below, I've already constructed a list of all image files for you.\n", "For first task is to load all of them (there are 450), convert them to NumPy arrays and concatenate them all together in a single big array called `pixels` of shape `n_images x width x height`.\n", "
" ] }, { "cell_type": "code", "execution_count": 5, "id": "bbae434c-e6b2-49ad-bccd-9a1fea089f4b", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "There are 450 images to read.\n" ] } ], "source": [ "from glob import glob\n", "files = sorted(glob(f\"{data_path}/stimuli/*.bmp\"))\n", "print(f\"There are {len(files)} images to read.\")\n", "\n", "pixels = np.array([np.array(Image.open(f)) for f in files])# write your code here" ] }, { "cell_type": "markdown", "id": "f70c00db-335f-4d63-94c4-0778dcec7a8b", "metadata": {}, "source": [ "If you did it correctly, then executing the cell below should tell us the shape of your big array, and verify its dimensions." ] }, { "cell_type": "code", "execution_count": 6, "id": "43b8ecbd-a359-4d95-a764-01bd88507ec3", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The dimensions of the `pixel` array are: (450, 162, 128)\n", "These dimensions are correct! 😊\n" ] } ], "source": [ "print(\"The dimensions of the `pixel` array are:\", pixels.shape) \n", "if pixels.shape == (450, 162, 128):\n", " print(\"These dimensions are correct! 😊\")\n", "else:\n", " print(\"These dimensions are not correct. 🤔\")" ] }, { "cell_type": "markdown", "id": "a1549800-4549-40d3-b3da-1040022db567", "metadata": {}, "source": [ "## Your first RDM\n", "\n", "Now that you have all the images loaded in, computing the pairwise dissimilarities is a matter of looping over them and computing correlations.\n", "We could do this manually, but we can make our life a lot easier by using MNE-RSA's [`compute_rdm`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.compute_rdm.html) function.\n", "It wants the big matrix as input and also takes a `metric` parameter to select which dissimilarity metric to use.\n", "Setting it to `metric=\"correlation\"`, which is also the default by the way, will make it use (1 - Pearson correlation) as a metric like we did manually above.\n", "\n", "
\n", "EXERCISE:\n", " \n", "In the cell below, I've imported the function for you.\n", "I'll leave it up to you to call it properly (check [its documentation](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.compute_rdm.html) if you're unsure).\n", "
" ] }, { "cell_type": "code", "execution_count": 8, "id": "f728e936-ad7d-402d-aead-897a7bd3542f", "metadata": {}, "outputs": [], "source": [ "from mne_rsa import compute_rdm\n", "pixel_rdm = compute_rdm(pixels) # write the call to compute_dsm() here" ] }, { "cell_type": "markdown", "id": "a912118e-f2e4-4158-a7fe-6d921a8d2912", "metadata": {}, "source": [ "If you did it correctly, executing the cell below will plot your RDM:" ] }, { "cell_type": "code", "execution_count": 9, "id": "717fcba6-8c25-4da0-bf1b-1140f1492921", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAMwAAACkCAYAAADbhcHFAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAN1wAADdcBQiibeAAAZvZJREFUeJzt/XusbNlV349+xphzrVVVe+/z6G633XbjxgHM4/7s2MpVArLCQ3HIRVeRAroxESH8FMXEucaNRMLvFxJiY8AxN4l0/YdpLhLBylMmoEgECYkYK4qEUX4mImn4ifxkjAg22O53n3P2o2qtNecY948xq3Z3fLr3Pu7XOWQPqbQfVbVqrVVzzPH6ju8Qd3cu5EIu5Fyir/QJXMiF3ElyoTAXciG3IBcKcyEXcgtyoTAXciG3IBcKcyEXcgtyoTAXciG3IBcKcyEXcgtyoTAX8orKl39Zh4jc9PHlX/7lr/TpfZHIReHyQl5JERE2n3/DTZ9bvPa/c7stz/xKn8CFXEihvtKncG65UJgLecVldnulT+HccqEwF/KKy8yFwlzIhZxb6m0WpzyfXCjMhbziMnOhMBdyIeeW+c7RlwuFuZBXXmaXV/oUzi0XCnMhr7hMd1D9/EJhLuQVl9nvHIV5Wc90nmfe8573cPXqVe666y4efPBBSikv5ylcyG0oFb3p43aUl/WsPvCBD/CJT3yC//bf/hu/8zu/w6/92q/xwQ9+8OU8hQu5DWV2venjdpSX9aw+8pGP8A/+wT/gvvvu47777uOHf/iH+dmf/dmX8xQu5DaU2fNNH7ejvGwK8/TTT/NHf/RHvOUtb9n97y1veQuf/exnuX79+he9/v3vf/9zolgvHnfe4/lk8nTTx+0oLxta+Q//8A95/etfz+OPP84999wDwOOPP869997LH/7hH3L//fc//4mK8H//6/8fbB6Z71rSrQWpYAsQB0kGa6UOQipGmpy6J8gIdSHoCSBgHaQNWAYUXEGKAA4Ocu/MqBmZBFchH4LiyFeuqWOCpzNyLWGdMF9x6sLx3mFWSO0Ys4CAA6IOCCIGLrCq+GHGFXQSbOHIslJ6kI2iG0EWlalTdIZuDY5QO0jFqVeM1FXKnLjkE8dLpZSEVsGXldUNY/V/9FgneII6xH1KG8dWIGO7X8XxJPE6BTGQ6mgB60FmsF7idQsor6mkRxPWx33UDbjE+fdPxX3cvM7J14XuOkxXBHGYLsF/+4kffE7UsYjwy7//f7vpc//PP/E7//Oilff39wG4fv36TmG2luXg4OBcx5guJdyXyPqEslzGt7R0pDqugjrYwmEGT+13bQrT7rstHATKQgAhzY51EAqjpM6xzuNYvVMdZIL91YbDsmReQO6gLB0bDOmBwTAEyYabIOq4x4JVDHMlqeEupGVhnhRcMMCy02UjKXgGz4p0Tupq7M4Vqink2BTyUEjZqCKIGd3gpGxM646uq6TOKStwdTwL1jlSAXF8ALRtGL0jONYrroDFa7QpimbiPszxPh0qdal4huROFcWSI9mZDwQXYGmUqmCKdaDi0J/9vd6u7tfN5GU706tXr3L//ffz8MMP8xVf8RUAPPzww3zZl30Zly9fPtcx+utgnVCWS7pHN8iwoBYFBMmOnkCtIAXyDEUEHR0MusPY4G0WdHLSSSyq2kMusVtKddI+6EZABN1AvgEiznQ4oEVIm9itpQpTL4gaNQsUQYriyUlTrD8zARUEyMWpruQToU4SC34WUnJydSxJ7O4Wr7UkSAEZlQSYgMxKGmPBlmYNmIVaEiJx3cNckUPHurCa856QN94UL46p1ZES9zKtmwWsjnh73SBhkXpBDGyG+UpC14J3jqmiU/jz1kN/DXBnPBDy0053rbK5K4EIdGd/r7er+3UzeVlV+6//9b/OP/yH/5C3ve1tAHzwgx/kne9857nfr1NYElyRYYHeWGP9Au8UKbGIxUDNyLPhCG6Kjh7vTaEIUpsLYvE/HR3EkeJ0tSLWoep45+gUymgJemaoKSxRe49rWAnvHBcnieFVcQmFFDUQwQ0qkHujrsNdQYDkqBqaBe8EZtBseEpoZ3gWvCqoUZJCErLG4s4e75UsWBVMhOQGU4175FBXgs5x7xgkLGBxdDIsJQQHD8sChhRBs+3utZgjKnQ6I1WoA4AjJrgIWj3uhTuijlZIa0eq4ik2i7PEbtOM2M3kZVWY9773vTz55JN87dd+LQDf/d3fzd//+3//3O+vCw2XaunUoli/QI5GWC3xIb5AV8dSYtwTzAVfQu21uSgRr0iFNIer5iniGs8Rs0w6xMJwkLVSBxCc8XjAE0hO6GXQCWRSioBXkEnC/UuJtKG5UOCz4Emok0MSppMFWgTrQml9VKZ5YB5BR0UqzKWHHqwKchwK66LoJMwnHbVTTIVDDd2o646UK8Uy16fE8kqKxe6OJWW6x5BNuF7JYD4Id1SKIji1E9Q8rKSBd4Ilx7Ky9WXruieJUhA8OVkiSPOaqMvYPLghVIH5UtzP2gP1bJW5sDDPIV3X8dBDD/HQQw99Se/35BGw1tievVNYLUmHaywPiMdC0GKhHFWQ5FjysCzisaO3GKf2xO8Z0NhRvUYQrEVwD5cjj0SCYIpFl82xHAtG51j4OkdIhcbvO5emAgJisdN7J8gE1QgXrEjELAiyaQG5gmi8Po3gRaCGkvogiCmCQNJmIQXmBAnMI3YBMA0rakoocI3ja424TbAI+AlrqICkcMssAckjNtFQJohroxLJDWv3sYvrwgTrA67vGp+V5rO/1zsphrlzbCEw6IQki90zO4rDAHZ1gPUJngt0hi0jBvDe265qqFZcnS4VvHPyYmYpU7hDYqRkaDL6YaYf5lBOcdQNSc7Qz0hy1B2z2JmlM8iOrAxbOHXpWB9Zs7oXQbMP8X9fxGtsZdTB8fDs8IUhg5GWBV8anh1ZVaQ3fHBsOD0mfRxH+nhdkjhn2nloX8ldiXikxubiXct2OKE0WNwbJKytt+daBhEDS45Ki2m8JS6S47mddwdJDdxJ1VA3dDZQQ62ipYYrquCLs9uPZ083fdyK3AqK5HOf+xx/6S/9Je6++27uuece3vGOd/D444+f63PuKIXhiQxrRU8EPRL0WNETRTcJ6Zb0fzSSrkE6FLobTjqCdARyqOQnhe66YIeZ/ilHHs/Y0z16KKRDQa4r3ZPCdNgxHffI1ILyQ0VOhJP1gJqRCugYKWA2kRGqVXGTyIxJBPwuzXKViBvM4jVWI0PmkavARSgosylu8XVIi0fcJVwabcer0o4byuEZNFlLX1tYhI3SX3f66xGM50Olf1LorznpENKJ0B06/fWIPXQUdIpUcHfD6Q4hnwjphpCPJI71mKAnSj5q1vMorK2YIBshHwppHeeajpV8rKRDoX8a0o2zF/6LoTC3giL5vu/7PgA+85nP8N//+39ns9nw/d///ef6nDvHFgLT5UQdIjVcW+Durc5Bl5hevcLrmrq3RC1qDEhkfXABDcuDa6tBEO+lJQAk6iSmDs01q4OQi7Najcym1Dmhk1N6wZYgfYUFmCqydWGkpbU13MCtlasq9H1hTF2zMAK9k7sCA+BQi6DZ0WXBVfFZ41gtlsrLQu4qsymdz9RO0c4i4O4qi5Ux76+wvI3PIpbRKa7Xeo0PwiO13eowdRGZr0Iouk5QFqAVXBUuV+qs1EXcdywSAt4LdYhYxy8Zc1ZqL9SFgIeFPEteDJfsIx/5CB/60Ie47777APjhH/5hfvAHf5D3ve99X/Ta3//93+eHfuiHdqWO7/zO7+QnfuInzvU5d5SF0VpJxdA5MlpaagTwo4ULAtjekvzkCTqWyNzMhlRD5xruVXtobb9XJ03NlagVnR3dtKySQ5rjf6MlbNadu9NNho6x07JuBce1ouvYddORko7b7yeCbxTWQpkyaR07ct6AHgu2Sdhxho2SNoqNCVtn5CTBrFHQXGskGdaZacyUdUcdM9O6w2elHnVMY8e6ZsQKWgo6F/JYn3FtcS/SWNGposXQYqRi5Clen+aZNMbPPFbSZIhVzJqbVyJNH8dyqEZ3MpM3hWogtZKnGanW9qKzg/6K3PQBPAst8P73v/+m779VFMnf/tt/m1/4hV/g+vXrXLt2jY9+9KP8xb/4F8+1Bu8oC5Nno0rsuHmOv8c9iYKfRS1BTWDV0z19AnmFTEKqSndc8VmRVQTkvoA8OUUVrY6WSI2aW6AARsWBbjQUYehG5pphyrBOaHbmSwYLYHBMU1gSNbzE+VgH2iyXuOMm5NXMXPuIXxKwMIbFzDQItk74pAzDxLxUSAolYQq2dFAhDZVhMbEpHct5pg7GlBNlTnSLwv7RTHejx/oEYswHSn/oyMaZ9xNYFC7T7JRFpH1dIE/gJVzOshK6dRQ2ZYzkhGtFHbQqao6UiIO8g25yvDqbLr6Tfu2UwZEJpsXZ3+vzuV/nqfQfHR0BcOXKld3/tr8fHh5+UZ3vbW97Gz/zMz/D1atXAfiGb/gG/t7f+3tnnyh3mIUpB0rdE2wFZU+Yr0i4HhKVcOvDxSh7mflVK3yaKPtCuQLT1cx8WfEF1JXiPdhevKesBNuDclmYc4Ie6qBRO9kTyqAcW89UOspSmPeFegABGZBQxCJIjSybtZhGKnhRvAh1Vrwo5aSLuMQkahmmTCVTp4RvElKgzgmZwh2jNoxNjSC9FmWa4hgn2lFMsSmRDMqsHHvPeDUz7yvzlUQdhPmqMN6dKMuADo1XE5tXZcpKqMt4zJeVsp+YLyXqSpmuJuaD9nNPKd5hvWKdUgfFhu29F+YDpVxWfE5YnyiX4vmyius8S2bLN32cV56JItnKc6FIzIw//+f/PG9729s4Ojri6OiIt73tbXzrt37ruT7rjlIYmWLHYwIdHV9HgXCL4dpme6Q6Mgm2vyA9uUGvOXpisCHcickxJBabE9XyWbFRyVKRZIi1gx0paTIOdELd8FHJN8CONNLOHUhnLZtlYTW6yGpZg814T2S4BsNXFkXO7GHJkqPJkGzY0sIqDRVrEJ14xOt8MCQ7qa94drJWUjbSUPHkdEOl10J33chHhh4BM3BDydectIlAv79uDI8XuhuGrj0eNxwdnXwUlkGOQY+hv1ZJa0dzxQ10MijxWjGPlPOo+EmCzjELF9OBfBIZvrPkhQb9z0SRbOW5UCRPPfUUn/nMZ/j+7/9+VqsVq9WKBx98kE9+8pM88cQTZ37WnaUwFsXGulDKQrEk1F5jJ+sDDlKH2DF9UHxPsEsLZL3BiQDYpVmP3JIGubl0nQbYsivMfRQsSZFarb1QVGDh0HlgqlIAD62PNLCtDNs3ZK9EWnvpoSxLo+4ZdenUlcffi5Yu7hzfq7BX0b0KS8N7R5fxO6t4nw8enzUEUDMPha6vdBppZO8MWVQ0G3mLDshxPdazSzDUQeJaVok6JGyIgL0OGsXKTpoFkVbIBUuxRDw5dQjr7ss4VlnoDg1AK4BuU9GuUbD1/NIrDJyiSB555BEeeeSR50SR3HPPPXzlV34lDz30EJvNhs1mw0MPPcT999+/wzg+n9xRMYwegabY1PIhpMkjSyaCKuQbkfnR2SNmSTkskS7oHl8jtqIule7QsHXgvBh8V3TU2Sn7HUxKDnRJFA4RpikH/KZ4WDrzCMaT4ykho0KrrWgJY0eGZNuaRxwHj1SuJ9BZYGzVTnHYKFKIdLUKMknDbzVs5EbwVXvOCAiOQa0NBmOBzs7HRh3i8qxPdMeOjtYKqVHYzWujLBNpjMxePnEwQ4uQk9Adh4Kk0bEJxo2QjxvieXZkjqKwVw9L0sW9ShsnHzt5Fdg4XZ/tkpUXodL/fCiSv/W3/hYAP/3TPw3Av/t3/44f+IEf4HWvex1mxlvf+lZ+6Zd+6Vyfc0cpTNnT06SL0HawAEdWFXQRqWN6wv/PNL9bEFvBeoOvFtRF7KR4c4s8FrKrkHpDNVw864V5T1F1FsNMRZmlQwzKQpFkqFrs8EV21fRkThUJZECKlG/ujIqQFoWyaTWMClTotDCjpNQgLO0ykhosor6hakinpKGAwjQnDrxAUZI7PiXSUOgWBRuWeKdYDgtjSaALK4LEZ9rWsibiWruIBb3E/60TyhAAzjqALOJz3QlEQWsJqIMw78UG4INTVgldh+U2jSLnWTLbC3d0ng9FslWUrXzd130d//7f//sv6XPuKIWBVktYODY3IGURSE4vFZMcVWwXZEXENsumUEvFVwvkZENdLiLOsYCaQFSwa4IqihSNWLVGJdyB4+MhYCMCdRXvZQrXp64VZgFzHMHGQPUKzQokodYWIIsgc8NoueNV2BwP4TrNGkjnlJGiFBd0lkAXTAkdYT7u8E5R4MQ6ZnN8VlSMsslcHzPLle6cbctC2RPSaNS+IZ1XIItWPCU2ntqFLpm2Ppfc2iBWrbg6ZzQ18KVHps1bO4AWxw3kJGA/kXFsWLJzBP0X4MuXSHR20iZ2SZ0C8yQ1qs+1S/TH1lC6EeDWoujk1CHcsLoQ6nJB/8iaemVJXSj5xBoGLSA3tlZ8YaSTttNPDYKDU44SMiv9cSCUdSaCauKn51CwwJaFcucpFl5q+CsMfNOatEwigIZY4GMsWqlAimvNh1CXgkrDq02CqGCuiAQimypUVzRV0gb6o4C/WA/dUbN661ZXmgDzwNupUvYCsKrFkdlJM5Rl3GtXIa/jWOVEdm0N0WgW34P14U1ac18DAR6f2a8buvwMuV37928md5jCwLwXzV/pJBZmakDKLhUgB6S8Cr4AyU7NEXjaOtwwqU69siQ/uUYvL5kPNOoySaPSrgKbRNmLBQBCTcJm7OkWM2kUyl7s9i5Qk+C5dVMSGTMfBctRma+ZWFgW9SNqxDqWJfpJMpgEmlpcUI1M0xY+E66SRz+MtD4aF1KqzBbPlZroh5lSlUmEskyti7L11YjDMirwaABHpTRXNfrmot1BoS4IF8xCIcpeQmandpBVqCvAhbQ5tVBSnVSjaU4nSBtD9lpa+Rz4/hcjhnm55I5SGITmasSuKOZhNXroxFuWJ6AdeXJsaAVNIyJg95ZpU/TyEjla4/vLOJ5EMS+nitVEckN6BxTrha4vrBYj9fpil8Hx3qN3ZfAolyRIu2yBRAejhtImjYq2Lir1OO8WkiYnDYXaCWbACNpXWCrSNTQzjiYgQe4rfV9wgeWJsQHKooYVU2M1FPBlHNwb7uwkdnzvHKst+66EafC4RrGWIq9Epqvdt0h9w9DNkDK+iAY86xUn3EWsFTkXhkjAaXB2rRNnyWwXCvOSyLyMXS3N0SnpaZttgllzc51idy6q0VTl4SoxRIDvg5BPogLu+0v0aENdLSALRcE9em7SOmKOuog2ZtQYDweqZ5JB3WbXjiM+yWPDrnnA8iNjpoHiTa3fhIDmc3J63uaKTAkGQdeB57I5x6qehXwU12k1OkHLRplrppqSvbCZBTlJwePVO1MBzQ1D12KIunJ82rqvgRjwBPnYqcvQ3IhNmgK4UBaRWsainuQnKSzgcbumXWwSXZ3eCboOd87vinshM+gWLf08Ui5cspdGUtvZrIu2Yh19Z1Hg1PpvoS7Wt74WWrzhEeCLhz8OQl0tyE9usEsLEMU8ip7RMdnSrR3I3Nwsi903jxLuSxJqjmN6W+fe+kLYNmP1RCOZR21CU5yzATW3Gom0LFuJzFIkCyILtT0Xum2xUzCBWhu4c6gRy6QoyKYp4ggtMHWBRs4boyxlZ1SiyzRcqG0qPLpOw2vM69iUogkNph5ycxNrp6TWu+Mq9EcVF+H4QOg3zvC0Y6+Ketl5CvblwsK8NGI5Fjw0ZLG0Zin1qAlsi2hiqDuZYFVEGtGDx0ICw1NDCeQobrLewGpBSo5lp4VEYTVyFB2HWinRpRXuXx9FRRaxa5OjX8Y94hSI05WuxRIYurLopGzWUHpDhopkwVUxU/KiUnNLFLSORc+BfmYwcjJqVTot1C76ucSg9k7uwhcKd6rdG3OihdRbjGex2KXFbi1LiPsOL7m13lIAha6rYTE93C7RqAF5hrqImKnuOXWUlrYPkg5P58mSnSPQuU3kzrGFNFiGE5mwSlAAaSykui3mibQCYdQfvPXT69xII9quGBmjWEgiCqsFcrhBrKKFXXcmrXLdU+lTIYmFfx+uP6oBbXGhKXGwpQSoMRhkEOgk4oyc4qfT8GQudCmq9KoNaiLRnKbZAG8Bebg2fap0ueDi9NVQA0OYLaHi7OnUlKHFbrQuSZFd9b2dLNaFBbDMaRdr4xCTZ/Z9tdQ6tA3KW7KiD6tLjThGkzWOgLDEmGzf+LxSXG/6uB3ljrIw5XUBAUldsLt0tTLpgFchDYWNdrCqMAfquOZEPpjJXaHsd6Q+0M62VqoKOVXclepOSo6YoukI3VcExUVYvHpiqpm0qKznjN5VWPRzMMmseuYueIDDckVGjW7rRnlbbEIRDbdrzpGSzgG58c6ZSDEW1SPWqqIkcQxFkkDXCoy9R4Ds8f6SHHrDNkLqK+bKcZe59PoNRRRR6JKyunvCZ6HoQC1CojIws/YO7UKpeyvMU4625axQnKGLSr5lYdwTuMuQRfQKUYxKQlPFNTpWa9cz3F2QpVMWPaI1UNNnfa8vQuHy5ZI7SmFG7UAt+LI2glgXKWSNXdZUMW2cYpkgkkjCcT/AFBV7KVFnYZOwmoJIb4pCnXqH7iv22ITds8B65em0RMVZlIIMxrgeWHdBSDFZY6sRb2SAID3I1CydhrXx5KgR0Pps8bke6W9xCxolBCp4Qz1bajWdCl63qICouTiJ6kKpmdISCianFvapftFoo8JlOtYcUPwU/GyTKkc54aRIXDicSI8kb2iFVmMR8GVYvY5ClUTpGy2URrG2asZNEXO0VtbWM82RubS9c9Utb1trcjO5oxRGAKZGyidRs6gegf3k3emi8kDQVuKL9AK5pTpNIJ1o1FncWjYMUnPDBMXuWaCPTfilFYKQC5CczfUBX2fSmkZ24VCkQVMkugvHWNQCkAgF7cE9in5WQ9m8kU04goqgLliJLJksCCbNlikTiYSCWkM4uEAN5hgcshilJKQ3aknko1jw3kV8IAloRB5pTXsumttiUwERwyRavwOFDLggG8cGwSWhs5CuRdEmz0H/5Nmj5ycRGDgk3iuCnGgQHZ4hFxbmJRJpbC3WR1ekd46sA2a/5RrbthVHtuo0EeAaPjetLz6NIL3vLAE5dmMXiRrDpRXDH63xSwMk5eTGknTYFqJDOoFsRDEQhbLtDqTRKLUA2gnGlalZoCkjDqWhAgBsk5DBIw1rIJOQJ8AE3URaWKdQvnw9AKh0MC8StgE/yfh+Ra9nOGqV9sa7JiLRXpC3wEuYDuI6Oo8iqTfOPee0Ur9Dc0uQ/RVJJG9ZvFbYdAIio9WhOPno9Lqt29Vyz5QLhXmJZHjcmS8FZ1i+AToFtsn62H3TScDO01zpRqPuCRw1qMwI816genVq9RqUupCGto0Vs3j1xNNpiSD4pYHFZ9bM9y0ZrAbF0VFCnw4079HrG8R/ETSwpKjuW6dYF8G+tmo7E4Fx6wxOEuRIAfvCqWpBVVQN1kLaKxSLHHXS2BjcgQoyOOSgkRq8shkESyVaCZIga1g+0hRbYbripGtCWhvzfliw4emoxu94DSSspamT18FbljYQCYfYdOqVipykcClLUEIliYLl6vMVMI7ekBmerHTXhaP7Myhs9s/+XuuFS/bSiC0kiMGn5j7l+MLyCNIWay6RDVPCxUlqEVwTLlwkt5yaBOtbUbIj6JIcpppR8eaGKfN9S/LhCaKKzT3uTofjy7AGOkfVXMawVCIEkXdbjKm5XQEdcbR36hhuotSo2SQVWIRl8RIxSm6WLK/Bp7ACGJGulgb/SZGiZlSYtCEZPNqmAbQVIof4M2mAR5PFeXjLoiFAL2HdhmjJloZk3lb7c8PtCXGfUz3lP5svhysmHlAa4hQDmbA5x/d6B6WV7yiF2Xy5sViN7K82jIcDlqAcD6Aw9DMn64HVamS0xNCNbKxnoROq0c+yGGaOjwNuuxl7ur6AGjKH29OLkYbCUAqSnJMbYVlEEzatqfdG0XFZR6bjjqoLWDjmii8N1NHBYCNI6460TUBIfIx6x2wKS6Pbn6PHPxnTmENhF4BGQVK1xmChTlksJ2ZP1CmRugpDZTVMZHVSB0d5wcFqzbgO1LO92SJzVyJ5sL/YYLMyW4dtEsu9DT2Fk2nYxT4JZy4Joe6wXSLOop+Zpp6xV9KVQuoqWSvz2EWB1oV5Hco4lg7NlcmgLip5UYMH4QypFy7ZSyNdN1PHxGFZokXomcPfnpSN9WQP+iFmiS+qJE58gEU0flWCCqkcJbrFzGoxhuJlGGrUWdZzRgZjc30gHQo+CDZ31HvBf7+SXpM4Plkyk7AutRxDy3Bl8DEF4cU6431rA6inhHlCnO/U0keTRxHTox7aPEWnWBQfZRTWZcEW9VPXCkPiugz0exMmAkW5Nh4gMyynSjmMuE4zuCWqJ6YpU8eMzco49WzoqetM6gMsmnOhmjIWCQvTxdSB4/UCyVGYTbMxH/XkYUI2yiwpiAfdMYKaSTQULRUoxwk9YzYMXCjMSyb6ZI4i3SKCTWpCctrRkmpJQSBRHaaMLsHHDJtQmFm65jIpaRTq9QXVM2JOIeHSoXcVxnVkwxCQoxSuVIL0moR8fmKSgyh2vqoF2H30+iMg2YIaSVpBM0UgvC3i+dBimKhPwmCkXAPAuclogTQUZtXmsiUgENeKw8rpcgGBPZ84HmLWTNqO1zBDn+gi5snAIIxHy0h1DzFzxtfRBJeqtO5OKJJQFdI62sBzMaSltoNx07HjjLiySR15LahG3NYdNRjOq6A/kYiD9oXsMJ9jkkm9cMleGvEbKXBcHcEYb4peblzHpkEJ1PpkWCd8FvINqMvU2oqj+as/Dl979mBbEQuQoVti0c+sOyWtG+bqaaHDWdaR45Mlkxyw+P0N9eqSvNfagDsPrmQH5qAhwoRqCc0VMaXXQrFEJzPTHHxjGLgb/dKY2zkwKto1ZM0cWS2vkSJWQHtDK0wlk30Osj8EXyfyorCcK+nRPjLPC8H2orVYZ0f2I8uVjyOzVTuoS40qfY3PzSfOvII0KfRCvzbKniKXZ/wok6pjjYctp0i4pOuRvu/uqtHDc02prS27nCOtfCdZmDvnTGkAyE6Yl4GTioxXFP1irASUPnY9zdEWbAMBNbHg4cICzqFje2+KdK7rNr0aLJG0hqjuOILj6aRjJiGioSxPrtHRYtHPsoOPaI6mrpqjiq/qMWipRkAvo4YCtBEZJkKdEz5GNk+kKfxGA6xZt5iuoJwtx5myCeTmSD5tvtqvlClxPA7BPDlEoxdb8oourJRr8I6Vpeyax3Y8z9JIyNuGL+Y7eP7CSwyKWjgs2u6gUb/qDyv9YSUdO/2R0R1abE7LU2Ds84mZ3PRxO8odZWE29xDp2MGY+gBfyqQ7gCMKtoRZNUj2iEYyW0SvvCSLwufc4FR9oJjrgqADUphWPZMpOjvZ4Oj1kT2qssC6BK/ysCz3DsiNEfoladYoPPbgmxQkF0tHR434RaMmpC7YuosM3iIUU+fwnaS1BOtGsE1HVyQoawm0MB6kGL6IwUxqxsmQong4CakLPFvJwvGXNReQQED7osKkSI3GOxsii6cTwLNrRsk9MnW6rUtFVnGsQ4wUaUww5dIWqyYcv66NE0E4vgf0IFqbbYimsrPkPK85S+Z55gd+4Af41//6XyMi/NW/+lf50Ic+RM43X+K/9Eu/xPve9z4+/elPc/nyZd73vvftyDKeT+4ohakrAn3bg2iMaijSwIUqsBCkrxgaqdVC4xOORi3VhpOaA5av2aKfJQk+hDWY23AmSlAS2aK1BixOm8RUwJYK/RJdH1O6PYQGTSHgMFos6gsWLcge+huLvxVZ1XwXNHsj0CA1ZR4sRuPNSq6BgQMh7YCesLLCcZ8iDiMGGq0WhfXeArYFyQocOLUQlf0qbYIA5MOmGP0p8sBc0TFQC5bC0lZxmBO6iFZk1YZ3m0NJpoNm1dehKNKz42KLTMYZ3+uLYE2eSUYO8G3f9m188IMfvCm38q/8yq/w7ne/m3/1r/4Vf/bP/llu3LjBo48+eq7PuaNcMlt4wEwGoy6h7EVXJYkYsrQIhhPrYnerPdAbaVVgYfh+vM868N7wpWMLYkzFyrE9i87HLkCNdK0oOcSu7i5I3163dOiEcmmP7pENetJcqdZbVV0D+k/030dbb+DC0hgKX/bbLMxZSUcKRQOmUsJd9FlIJVqYKa2mctKsVoF1CkyZdXFcN2G0jB8ER5ovgiBQlvG79+0erZy6F2QYddHu69JigS8Dq2dLhy6uUxa2Y4WJe0c7VsSE1kZ6oOFq1lW4m/RR9znzezW96eNW5FZG2r/3ve/lfe97H9/8zd9MSomrV6/yNV/zNef6nDtKYSgCswRZ9yZFCreNpfBRw2UZNYqIm+AKszkxz0HFWtcp2F1mgVmpk8IY75epMTa253xK4Zp4izu2qeNRqWPCSlMgV+xgSbq2Dr6uVszUEilhnWmdjg3O3yagMUfgTIPzeIoYbUsjG2Mz2rVNDQlZYgwGEH+boA2YWU3xRoZRNwkbNYjSR2WaMjYF/axsudBG3Y26YBR8jDjKxwRF8VGxOY5R5hSQou29nnR336VdF3PcH2mUtjLJ6TmcIeZy0wds4UYvHhn58fExv/mbv8nnPvc53vjGN/Ka17yGv/yX/zJf+MIXzrUE7yiF8T6CWMmt70Larqbh6lgmELfpdBHGqL5TfJQn3wWi23EQMb24/b8tSBu2FK20yWK+8+l3GLA+jml7Qrl3SXpijVKhiwyALgzbc5L57nMddlbQ28AlkmMHMXDJ99qcl0s1Cp9NPzzHeWytWF4UNMcIcp3CvZQpuMCs4eIsx33xFAyd1rJatPtRh9P7tk0I1BRZP0+n1xc8Zb7jMfPt79qs+fb3/hm/54Z2OIfT7yY3fUCAVreP51KYs8jInylPP/007s4v/uIv8qu/+qv83u/9HsMw8N3f/d1nnyh3WAyz7Td3C587iVFTYksciQSLo+C7Md/qMZtlu9i8QTiks0ZYESBC9Qiwa5JA2I4BpBTCrdDBwqK1OkvKhjcOYYfoW3n1gN5Y48MqKJeO4hjutIxYBNy6ifhFZnbtCYxEHedEY8TemKNNYS+6OItquIjLiFXKlNCuRgpbnXQs2L7RUZExghxPEG+zaGduiGmZI4uoM9ElKnHvJEV8Iq6oGmXbctA70ggvzFvsRUM0a0CEPDmuGli5NlFK1YMM5Ax5oRmxWxlpv33t93//9/PAAw8A8KM/+qN81Vd9FcfHx+zt7T3vZ91RCiOblhHTGO3tNXrLdQ7D0JI2kSYu4deLOFZSFATNsbEF52OgE3VuO5xrzJVvMyip0lDHiswE3MV9V2fxqbk0jQJJZ4Jdclih19dwaRWLWaL/PrXeecTRWXc8Xk5Ux2nMNXlqSj8TLcQSi19r9L3UPi5Uq2NzJBUoEmt0IkaCVyLhsW13KOGKaQ0CEUsNFd2oZTHHqpDmSNdr6+fRlhHTOWpB23S7iOBtsK5Yc8VE4hrX2uZ6tqLuOWZc+gtUmFsZaX/lyhVe//rX3/w8zjFa445yyaz3Bq+XaBiTZvItFoE3F2Rrhba1ke2jSgT0W/ehJtm5KduJYbtWYzhlz08BSnRrWmkSbdHLhhxOzVWpgmXFLq3onjiJ4mUH4q0uw+lUtG1qOyhpIxiPAauRnRsPonVAGj+xGNEkVttne2I5B/mFbCemWeN69taardIyX+2ztn3+7f5sR587sqvD0O6V6elz1jXXK7W/Wyu2azDTWNd+5vg+ah91LVrT3Jniz/G4BTkvGTnA3/ybf5MPf/jDfO5zn2O9XvNjP/Zj/Lk/9+d21uf55I6yMDsFSGHGXcONsD6YYXZj8hofltSWrlWPesSku16NbQzkqrsv1doicKVB9YlUcEeM/u4D7lItRVFy1F02zFzQhSFHzRJcWdE9foK/ehmp466xUQ5gc1xLd0jUjjRqNNs+nbL0sJzJqJcEPYpzSBj1UmTjdC3MxAAmL0o/G5uDVvvoPGbvdduYpd2+PiwFLWaxJJAj5ovhTy0+2RKNb+9DtkgAqLTpyq1/pkaM2NrlWpwW3acicR5+DoV5oRYGbo2M/Id+6Id46qmn+JN/8k8C8C3f8i38y3/5L8/1OeLnsUO3gYgIb/zxfxJxRW+RWk2G38jh2gzNNVl5sNzvGV50N3IiTRKxxBT8xTWDdobeUEonMERsUHPLcDX3rx4Y+USwq4W6yUgXFE6qDtfyDjGpc8QbMsair12D6ZyssYNl8+Wj+U0mwZbB4h/db89kkSRqTduERqs/Vo26R70cFq0fLVK6CeqU6YpR95yhGuUox2zM7HRrsLsrZU7okZJOwJdOWTn5ukaCoefU9VJIx4Kvol+n5iDusOO86+1RMcx0N7LdiZ/ddWG+FC6lGjAYMgr/14/+b8/p7ogID/zzm8+X/Mz/+vfO5Sa9nHJul2wcR773e7+XN7zhDRwcHPA1X/M1fOQjH9k9f+PGDb7ru76LS5cu8epXv5of//Eff9b7z3r+POI9bOsa3hhOPEnECUmihhIRbDwnYChFlUK09FpqHGASbtKWg9gb51f0vrNjvd+O6N5yGBvgptH8pRF/ePv8dBIpVRNBZ8VzpJz12gZq6+7008ldvnWH/NTlsS7+Vxe0VHZYELfmgs6RPi6qwbcsodxWYmLZ6ImienptLpQUxdnt9ZsKvr0f2p7Lgj3jHhUVSrufc4pJAK7NrRTdXXPpZZdp3DLQWBeWvopGDenML1Zu/rgN5dwuWSmF++67j49//OP8iT/xJ/jkJz/Jt33bt3H//ffzrd/6rTz44IM89dRTfPazn+Wxxx7j7W9/Ow888ADf8z3fA3Dm8+cRmQIzltTIjTG+TkGsl2q02yYznMBl1e2k39JAh11ki1KNYlnSlvGZ48vvpFJEwzVxj7biiWiWGqNbExN6LUHMZ4213oOc21rKdWvBpGup3isL9NoG31ugKTjSSs+uU1Smlrmbm6WsEkQT0oLyhUXmqkQygwQ517azh+ujfQHXZyQ8An8m7nTWLIJ5Iza0AJa2FgJSIKG9tSFou2cY4bKNArZNj0NHpZTAs6VGoI4GgnwboyEt9qvP9W0+Q25T3NjN5NwWZm9vjx/7sR/jK77iKxARvv7rv55v+ZZv4ROf+AQnJyf83M/9HB/4wAe4cuUKb3zjG3nwwQd3ldaznj+veCNmcA8k8EwKP5zYlSHIxE05JfSG3U5et7xbWwZHInngrSelEFRIZnJKoOdhAZ6ZQSiWKDWFBWjZoeAUeEawbdLOTfCc8L1g2JQars+WfyDwNqccYtvFtw3Mddvb35Replb4nJRSNQqWvQeNk0n8vfXjLK67bBGVfnqvbGvhPFLbZu06Pa69Zdnj+olzqO3ebeH420kF1Hhe/Bn/Q85tKNxu/rgd5UvOkm02G37jN36DN7/5zXzqU59imqYvqrT+9m//NsCZz99M3v/+9z+rygtEQXBVSQcz3WpmWI0Rx3Qx/9F6J/cBg8l7c+zwi5gBKYtK2itBsbSwSO8u6g42IqvogkydkZYFVjH4WjpD+9ihBWAwur2JYTGTCMbNAG9GZsmGYKAse05ZtExbdkhKvXtJ99gaSzG2z/uAp1ibdyne+JBbt5lskcQL32XyZPv0qsZ1dhWdIe3PJIPODPYrvqox+m9h9PsTuh/XbguHSxXZr9R9w1dRLE2rEmTiq4ofGLJnsGf4XkU7Q5KjXQyPyosahdDed5AZFrHCPcV4QXpr0JizYxBpA3L/x8ftKF9Slszdeec738lXfdVX8R3f8R38+q//Ont7e89Chl65cmVXZT06Onre528m73//+59V2RWRYIhBmCelTkJdR61EJkACMTymjrQW5qptpxRqjZpJ2SgyK74BEsGifxJtx1aFSYJkz2pCm5vESaKOgbHySeEkMc0p6hKtsJfGsDB1W5Sct6TiRONWmxcpVRjvW5KfWmP9AuvTzmWRlmbWdUuBE/6/CLDWIMRsc2OqC37UxXSCRbRC1+s95lHf0Wux4NzDKm2eXkRvzUlrECPDEETnCDBBTVGrcfWAFg3BOKPpdEqbzKGxU+ni/rRBBbpuKGYBnRRZt5EabcLa2QvqHK+5TeSWFcbdefe7382nPvUpPv7xj6Oq7O/vc3JyQillpxTXr1/fVVnPev7cn519V0WOOSmtBmCR99/BYbZwFtpPabuzBa7FWs1g+3/PkWZ2FzRHitiTU3KkXb1Atz9HW7ETzV85dtIYnxHV/0jNeizaljq2ZWSNSk97LVi/IG2OsbQfC633sJ6tuGhd7NTitU1Ec7wT6hDARnK4Zh2VKk5Wo04p6lTJ4DionDxFHERzKbej+KLQ0oowjYJ2m0KG9tlCmx8KZIsC7sm2PtVQSRJB/rYm43thXdMU95XsLQlzhtym1uRmcksumbvzfd/3fXzyk5/kYx/72K6K+tVf/dV0Xcdv/dZv7V778MMP86Y3velcz59X0qaBBUsD/TUgo5ZYoGlqPn4DPMoWDDjH6DtpvTBSZPee+Lv9b5LGMtk4vGbaZ5ximyLwBxx0DOTxNo4SAu6SN6Ec3WG8Rsd4/TaUsD4x7+0zfGFN2hhJbOfr60mc/zZuse50QUttZIEeyYZcDRuV6aiPfnqcLtV2/myDt8aYqbtrkiqnMzYtrDCtKCsNILqNS7Zsm2ndmu5mkFF2k9+24NLgQdt6k9K4r6PweqbYczxuQ7klhXnPe97Dr//6r/Orv/qrXL16dff/1WrFd37nd/Le976X69ev8+lPf5oPf/jDu0rrWc+fV+pB3Y36ZmGwqrt0sK2iLsEiYhoWFt2BgyEHBdp4b1u2eGcwWBnW2gJsZVG76Vr1vosYw9so8JQCm+aD4yuLgiBRqNN1W2AtgI4gud1dD6uoFtxobBdzEso9C9LTG/QI0pHu6Ga1CPk4oCoRSMcxLDlpalCZGYpGydB6p5juyCe8xSe2MsiGLCusasDwFy2FvjBKg+Yz+C62s/0aDDZ9awlYxn22Zevrb/HKFtJf9j1aLZZhbbbtAzY4vl8iljlD7qQY5twK85nPfIaf+qmf4lOf+hQPPPAA+/v77O/v76qoP/mTP8nly5e5//77edvb3sbf+Bt/41kp47OeP49IH19+tyz0y5nFYgqurc5IOVC+uYugf1jMaF/JfaHrC3koLPY3SG6s+BJkE6mraDJ0qPR7I6mrdH2h7wq5BogzaVAhCU7KlWE50acS2ENvWS9nV/iLzk+nXDK26GeZYudOJ3Et3jt+2bF7BvKTG9IYIzq895hvM4cFSTOw3N4AidHkNZILaVHJe3Oklg8svKfJ6RdzXHeqdLmyt7+mb/dDkpFXM/1qjmuXQDx3QyF3hS7X4CZb1Pj/cqYTCxBrFwmAviuoxATpZB7zODHymsCx4dA7qa8BxjxLXgRozMsl545hHnjggeetul66dImPfvSjX/Lz5xHLsXsnDTdIszCPkBDSwjFzGMCKMA0SrSUZtGWeZjQsg4Z/XjuBocVDWWIiMmAE17EMQRTBgh0k3gdvsyUbXCYHCBGJSj/buZg9aPT37nx+sQbVmYHOKSKkTuHqErm2we9dRBG22yqfwNIoGvFWShYBfxeu39gpHbG4fdSYCEArHLaUMUZcl0Vhlz4GItEbDBKjQLqIbbJE1s8mYAgXrWSl00o1Py3cdnHPvLa2hRLFXpZxf7yFRt6BnwN8ebtak5vJHYUl802kdr3BV7wTdFRkA5UU4/BabGHrBJtEXbYRFBsNt2rWmFbsghnoWgNbpUF8R2sUs9KKgDWKj7KIOLluAg4jHgGxtCBd5viMbblDSlMqiNigLZy0BV3O4U65gi0Ev3dBvnHCfHkFLtgypjqzhpTiID6HElMJxHaBuVfYKFksGDdN2Mz9bhCSbKCMHfOcSKO2UelQx6CagmhBdpwySrQaF6E2Jk8fiWzd2OIRDaTyrnGsJVWEiLnq3P5XoKQu3NUzv9gXe6W8dHJHKUzatMJb1higug04HZii+FiLkEp0BapFtd7RCOw3gSMTou+DxhwjJSr/aPPnSxQM1YC1hCK1haGF6FiUGLlBitjCMxEvnSh5CuuVxrBworHzCpEt0+a6aMs6eQ+IMF9e0T2+od69iBGA6vhAdDMiwbA/BN2tTEKdEtSgxhVr6GkV8o2wBtsCKCdKNwmUoKdqqkpahykIlDW7kRzaxvFBIBZS50GFNINqQzrXLZcC0TOTTu+P7P5/Csx8PpHbNMC/mdxRCpOHEu5T1ziUszGXHlGiCFlbWriDYZiomuiGVsDbaKMRymEVTNC+YnPGO8iL0LwqkU2SxnWc9gpOotZYXGkoaJs2bJuuWRPCipwEysBbmrUsHW2Zuu0CTjUWvHXgq4CqbFPJuFDvXpCurfFLizhQi4sM0Cr4LDHpq4tRf5aEMnd4X6EqAxO+Z7HAnbA++yNzydSjHMq356RVxT1HDDI4OUW85J3jNQdoNQOds6SwPlmgXUyEztqgNq2jFZr7eE3JuUJPbDpdhXO4ZLdrRuxmckcpzHpoZBJdLBRPCXoQFaQPQKQuC6YwL2N6l3UJeiGIuUGKtlF8AstYVF51h/1K0mDvs5InKJbJAqqVYolZNcoVk5KLNPLuqF1wqWJjRuYgh0ibiFtsEUOItg0wMbEswJg6E81qM9jSwrJcWtB/foO8NjNdUSZRZK10i/CzzCJNPi9BamS88BgryCwcLRtRoAu6gLHPFM9oH/06nqJTVTuhNih+zTH1zDNI1mgN0Bg0VUWoS0ijRqZ5EHwjqIYrG8FiGOhRg/ZGOkP2ArZ6llzEMC+RpEr4z9Iaq7pgr98GyOHGNHKJFOhdqQGTl0lIahQX0qZV3jtgjoA11VaIQ3c8yDt6V4fZY5Z9EFjIbiRf8qAaMhfkOEWwK5H+1RQUMjo27JZFs5eJIl4jG5acjLXCq7Qxf4K8NqPHJ3i3jwzh+s1zjvitj9V5ICMbMjamcPtSzNLUTRetDRIoBNsSfYyRuvYq0YLd0AyeCWqpWdFGMeVbjFtri9BRYw6Mx322VgPbqUMbOBv+YgBi/SgHcPMMuXDJXiLRI41xF5VgcsmCHLdYoe3UPmuwy5RgOpEu4hpdCyxiZ86HkRWyEvCQOjSXqc2U3CqFboSkMXJi7hobjaVG39qsxaykAnlu8Ps9b0OIiOavQgATx9amnKLPv4qeEkt4gqXBmhazwHRF8W6f/NkJXrOIDWGKDJmuHBmMMmZkFZgzloYcKuPcoestuRkxfj1n1JR8FFRPFYEJuuMGTm0T0XQMa0OJUSGkcEPFUhsvCCrPKAR7Awd4bFjd4am1sMyOQ/osuVCYl0i2bcDVwi3yqpHabUyN29SvzLQW24aD0khxUoPIoS4Fb8OAthgoaF924/gSkR2/lk/CYjkFiz6RJfIUlftcjdpth8KyI+qQKTol6SJmYdHm1Cwc1gF3SaUVJZdRhEyJsF5KuGGDwGsW6NEJ5fIKXygyGLWHNAudOSWlqO883dDRK8MnWtDTYD40S5m2sRKRvtZWJnJ2sCLTaH/YiRBMNkUiAULcS3EP9EOD9ABRxBzC+gtQh0a7e5ZcKMxLI7bFiOUGP1LDRSNL00bE2RDZHlu2TFGKYqABqoZNKVwvoYEu2dGibtHB22GqOoU75glmj90y+k8ibRyYtmbRiHRtUY3gvFEpoRKW0ANAaVVajaIpuUX2apd/JmodstZG8SSUyyv02oZ6eYnNCaIhm5qFIpHJIgW0Zyw50MKFNoYvqJystRRIbT36fWslborlGax1eVqVHZLBOpAU90H79pq0LdQGVm0LF5MieJSSohlu4ecqjZ8D0HzbyB1FgpE2QipO9pbRKUEop1O4O+lEdiwyOkasssOMbaSNuYhGp7QRZBNul85+SnBXwj3T2iYjt4C2ToHUVdrYCW1Ml42RMqrz0WSVLEBVCSPNMTQ1NdxV2rTUtPjuJxpFSTdBMLQ6fVfC9ZwERKmXG1mgR4V9GGY8Q1YjTaDZYpHS2GzmuC8yx0Pr6f/UG+FgaalkwnXTKfBiqfgpRqzxB8j6lLxPR2mEhOEOR1o7fqbxFGOmJy2ePEtehEr/PM+85z3v4erVq9x11108+OCDlPL8QLb1es1XfuVXPovP7Cy5syzMPVH1y0MJXFYS5pMuSLIXAWDMy0JZZ9JQqSVcNl1WfKUxd+W4w6eoVOe+UjYabttg9KkyW8Q+Ngv5OlEoTAELqWuFlaO9UY4zvmgYsxPFhpZqXgYVklSolzxojWaNrs+pVSXSFm8VXANsoilLhobBmiOe8hxpXhkMmxP17iX9I2vqXQumSxld1OAl7uN96Rpodsrdc2trBp2F7q6WVu4SvhBkWfEeSh9JAEkNOzZFpbWMEpPVADpHJKYzS+tU7fpC2eRo9NIAY0JYmLpqlf7cqHjP4ZK9GDHMrXArb+V973sfDzzwAE888cS5P+eOsjC5D3xTykbqK0M3o12NkXM5GsVyV9GhMiwmcl/pFzPLfmK1GOlTPJ8WNZ7rC6kzNDt9Mrpc6bpCSnE8TUH2J11gzRiMLhdULZqpGsujN/5hloY0Agvr21jxRPj5DezJMvid2aIVBNgCHfuIdzTFCHDpDV3VHfxfFk69a4Ecr8Fsm5DaZd8kRQPdapjoh5nFciKvZpbDxHIxkRcl7lVfWQ4zXV/jnvWFfjHHYzWRhko/zAwNr7dYTnGfukrq496lvqLZyCnuC32Qc+hQY4ZNZ/TLGV2c3RCzQzr/D49bkVvhVgb4zd/8TX7lV36Fv/t3/+4tfc4dpTDFg+d3nDObGg/TaDMuptGqYkotiU3pMBemmphLpljiZOyjVdk1poNJcBKbRZPZeuowV4pLtP9uaVARFsO04+OaSo6elx0sP9DEeRFuVKoR7OYbQr8JgGKXK1mMtD/HDJgpXEA9bi3SLeukU2SqAqkc1iWVSFD0XcEuQX31wPLJY3QT1sgTyNMZ66EMsCmZucbAqFIT66lnM2WsNv7lMTFOmXlOuIFVZSqJYspUcxBqVKGUxGbqIiWu0ZptVRktRTeAB6uMibS+It9xw9mW++wcNZbnU5gXm1sZgp/ie7/3e3nooYfo+3NMfHqG3FEu2cFUwEHM0ALZnUMFkpIw3KDzmSrOcp450Y5slW6qeFUOvHBiHSIx8Wt5YmQv1Cp0WuirUZJTmiLOi8TgFU9GVqffm9jziewzo2ROhsTKCushRV9JdrSr2Kys5pGZRF06SYKwoijQBzdZRyVXo3RCv6iMXaSn65SQoTIv4UBGypjpLAL8MaUAWgLjsOQrr3+Bz77mHta5Z8gxRPbq8gQ/zsymkeZG2BtHqitr6SiSGbyQS2UmU1CSGalYW6jOJB1dqZiElbVqeJ6RKUEXMdysKVg/8Z1Cr1cdgxm6ATpnUWbW5/lin8eanIdm6Sxu5f+R/fKf/JN/wlvf+la+8Ru/kf/4H//jec5wJ3eUwmz2oFalGyKVqWpMNWobmqJQWDtlIlEHoxjUHOyTbkBRZvNG0eRsgM3cKt0dTBYo3tIKcLaBzSBBo9rFThozJZXZFa/a5rMElSuZNq7POUlBsicpWgk8B5o6d+F+VXHmMUVM01c6LICUNZAGUp0NUWcpKVEkAvzayCaKKp99zT2UpybqwcA6BUznSHr8slOqBgdzUXQhFIRxjHthSaBXyhysMgqkTKu9OLYOboEqgdtbDCWKpg7mMcm5ukOO+2Xa0mbVGQfwBagp8zJR5nNYmBeYJbsVbuXf+73f46d/+qf5r//1v35Jn3VHKczUsFsptyA0C3XdRSW9EyoJ7aJdeMoJmxJOZZZwuZJ7FDarUGqiLCpykvChUoEZwTbBXJnF8JOMpQKjcpQXMQZCw0UTiYJf9dzms4CfpMCHlIgpvCg25Rg93hrSynFCcrQVT5sUhBnrPnz9TVAplbnDFtYq+JGtqxlYawxorZHCXueeejCw+v0N8wM9xTvWXRfzbKqSMdgkxlWK8eBzVN7ryoKDbEqBfkhRjImx405y2aXRq8E4dsxH/a49nNrcsDZhLWClQjoKxbYEuUA56YMQ8Sx5gUH/rXArf+ITn+DRRx/ljW98IxDZtcPDQ+655x5++Zd/mT/zZ/7M837WHaUwqYtFNK1jGrJVCQb+Rs6gGkF3nSPWSQ51jlHj4uBTQiUmg/XDTHBYBLGGWGSAUl8xCf/d91tAOykHqzXXxoOYVrxOsF9JXYzdUIu4iGUNFn2AWehnwzqLGlBfKKKku2bq9Z46peD32ih+ELxnWSwgJX1gTkSBpZGelqizLAOtrOrI05khV9apY36gJ31+ZPoK0JVhmxxKhqJFmDZdEIGUsLR2mGMKwBg8BFSYiOKUFwnkt7S6VhHYr/T9HLM1K1FXsqhbyVBDESVGdDDEjE/GLXv/ebBkL3xtbLmV3/a2twE8J7fyO97xDt7+9rfv/v5P/+k/8c53vpOHH36Ye++998zPuaMUxhR8Wem6GpB8EYrl6OHoLOAdXYzs6xaFMmuglXO4MmmIdKimSqlK0qACIjm1d7LM2BYz1ht6PTd8lDOu++iPSRHcl5MUSOVGFKHiyFpj8tcUmavNAaTs+ByldK+CP9XYXXpHOkck6Jw0R1NYXTpUpadGMH+obR6lYDWRrhFNXn3g21ygeMf0FbD4zAiv7Sh3l0ipe7hH/WqKhIl0oIr0NcZ3dA3yItG16hb0sngbW4FE1jEZJWm0PnROHmrUpTxSydr4A/INoV4KayMSmT2ms7/XF0NhzsutvFqtWK1Wu/e96lWvQkS4//77z/U5dxS38lv+0QeoQ1iaYaokN67PK8yU3BV8VBbLmXVNHMjEET29FnI2ZIJ+KFybVqQNTJJYDWNkvNrC2JOJ4z7jItSS4HpCUuz6dk+Bo4RmYzlXjjcDJQur5cRomdoojrquYHN0P9rg5BFGTzF8tiqdG7MmWFW6VAOHNkUMZTXoawerUIOJc9x0+MraeSrdFKnjOsCV5QlH0rOeOlhV+qfg6lPXmYclEx1IIBEuLU6YSuKkDPiskU7XyjjHMVWNTiysh4e1zFKpGt2WZsJ6leg3MadmlWaOpz5YewjkMgA1Ibki7qTeSX1h8szv/O/PzZEsIrzpB/6/N33u//zQ377tuJXvKAuz+k+ZutAYPHroMFWWV1K4CLWjv+7M+0v2rJBv9CyvZrrrAdLMx4YNS5Z7Sn9olGUCX6JZSFPgX9x7Lr1+w1P9gnwUqdLlIwIL8Dcb86GiT3SkR3v2+phWvN5b4AcWY/JyFPG0smslLuatmxPEhbLv6DWB44zVjrpfYTFjHWzmnnwjiPWOlgndBJDSp4CZ6CiUu2f6oTCVjB9n/HKDvmwy5e7CfLJk9fOF5V6HD8rm7oRe22O1qaS7UqAR5kQ6NvY6KAeCSyafVGrvdEfOdEXJxwFQTRuwXin/C/SPZLoMppn9NY1P2Vk9GgOrrn21sPqCsXjSOb4vkzeJ8tqzY5gL8OVLJJ4b8bV649eKEQynEqTaWgTrYyaDb3v1B/AuIO+WZMdbdspZ1ly8RrSNtC+yHb6ggStrbS07HJg1KPy2duBbbFj89CqQA7AJESPIlla1NqsyR5+JFKCBS5mDhLwxqscxTfCizCnqIbM1utgaFftalMm7UJYyY8sUpOHbydAS52UJpE/RTtBqLNbpblyFJ8EGbdzNGgBVk3avG4y/DZcRO70fwfumWNqSf/j5IC4XCvMSSQNCeo4vY0tBtBv+syXxM2L1dlsFa70uuUHP+6g34JE+1RLPRSDbwJddvCZYYGKhawYy1EWkfWlM/7S16DTEbwN00sEWAy9FwDw6Lp024q69vzH4b0fg7QihtwdV340PdGt4Nm1k5Rr9NHXbxyOCDxrKMo1h3bS1M3QNwpJakqPdm1CEqNRbY+kMthFi9o5ClwtoF38nx3OQaojH9xDtyIZUQ2cDb9PS/piBL+8ohZkua/vinXlPqCvBkkZXY++UZTyvpswHMfKh7oWCWJ92jJfdUbMy2cGEqYu+DwS6FBgu86guT1dixqSl6IVhEGxPohEkN3DiCIyRTlVlxz1mfTSmWVNsJ7Jd24QF3oCgmbBQG0JRNoouGpCxbCH6Hk1zc1gMr9FhWorCJgUr/yJqJ5u7U7McipeR8e4BmROWNZSeuFdSQ/ktQ90LEnPrnLpU6oJWW4kNqqiQFrpjFa1DQ0EkYf2qtmugzJeg9kpZ6i4ZcJZcuGQvkaS178bQ5U0gaqd7jO1Id2lsLp6N/tBh2eAnBbrjWPRlr7EyiqMn0UqskyAWx1vdNXGsOSDtPaRrAgPsLzZUT4xHS/Jx9H34osJBNG9NU8aT07lB0WBizQHWL0noLCaX9ftTcB07oVhLI+9v8KyUsYOT6MEf+4yNguUchdYhsmjdXSOLYWI99eyNI7oQxlVi2nT0q4lL64Je2wulUhjvHuj/aIOmnuneSGjkEydtjLIXGcG0ge44SPjyOtyy7gjK0Ghf1UhfPlF8sSOKCVctCrL5BBDYXK6kQ+gOY75NGp3p6tnf64XCvESSZsc3DtLG1k0ehbFMS3HG4tcKsonBrXkT7b86GnRKGo20dmg7rE+QNwbVw52ZWyFuBLKQ1vFt2hzz7mWKPhAAJqUW4j2zYuIxrXg+bUorLlSTYKVBGGsjoiD6dbwK09hRTZjnRDdJYN88B5S+kVlYg+jPJZOzReziSmnt0bUqxQMTttpUpEScJ3NCU0+6vkb39xBT0rqSTwpiGanR+aazg1t0ivZxv3KFNMXmsCkpGHcM0EgGSI70dl43d25WGI00GlJbu/N5+vUvFOalkXlfY7zcQMDaF3I6uz5v/XQAYd5PmCp1FV+49RIEdD0tdmmEeBXKMoJYVyg6RFDbuIznfSWpM1tHHTM+KLIfO6vU6KnBlDxJ9M57MMVgYaGkSBBwtMmsPmb0JCr1UoPzq9YctEybaIarR0FYIaPusnXSoPq1S0ylgxrYsHGMCr4WKNJxUjPprijkWhekF9O9Gdnfo/qE5gE7UOoygym13cMI8qMOU5eRWfQk1Mmog1KLIynqPp5AGq0SCPNBQGrSiVAHZbocylq21c8z5MLCvESic8QDaFgTt2inFQcZjbQRrA+aJCyyTkGKHYsfgeThZqCO1dhZg+m+LZASHZFp3Rqspkj42CZFfaWRl2ON9rWG+yEzSB+fk+YY9+Db4zdyPAh6p7SOxaxzKL5MCj0wCWkMnJzlyHzFVK84rk7gC8GJnp0iOZg5x2CwRBWfopFtS1GrmeAYMMXzQL62wfcGcCGNBhqbTFiO1gS2Dgu6Jf1IG0eOlTS2CdQ1ztOlxVljoJT7Gx6E5RPPos89S8TunKj/jlIYaEmnDeF2FWc+iO7ImhSZolVPi5E2UJYJnTyC3RKFubICzHe0s6i0KVyGFEhUJo3083QgDE9DMme5t2GcenzdkY+hrNpE5JVhK49uzVatt20MnBvkXSFwvQKXasxn8Rb4JyetSvACZIL4bs/xZAHFb/341hE9McvoRaljYvASMztXhh1mpK90vZHmFKnjhhHLJ05aV+xA8b2B/NQaTR3W50jGNffWOmKj8fh7G4fMe4IcFPwohUX30wxfWcb34QrT3ZCvQ2qsmbtjnuM7vVPkjlIYKZH0j3500Kn1wHfPTPd70BzNjrqSxmhgymuL1PFC0GJI0R01kY6+o1YamDnKCW+zGtPG0N7pKWzowzVqbJOWhDrQMkoNNZ09ZmRac9to6OjmzktP4y2ONmFbBAsMvVHHyOemVaWmFBRFE1E76WOGp/fQDcYsSi4Veg0g5dLRwchaSceG9DG/siyVtDHySaEuMrigqUOvn+CXV8gq+kHSbIg5eQ3TkNE5Yr90UnBNjClMkDgt8WK7ekwaI81sfcSZqfFLhwVLZ3+vFy7ZSyPeRaFNCB/ZUoqFT1iHXeFQlbKQaDXOgQQoy3TaIanahpc6+bgF+xLHXHuHk5CN0HljlTE4mQbqOpNqQOTrEC5SPozP0TGUJ+ZXtmFPKayfqaAW5BvViMlfRLauuOAl0tVpFNJacM9oF8wz3XFjd2kzbUofMY/PKfpZZsGmFLRTXWKchb0uFrV17TP2FLG4+DRaWJbLK9K1NTYo1iVqJ3in+CY2mLrcTglIkRG7kdrsmIhpdPJTUo0cG0/3lCKzY53vOOK2IzyeTy4U5iUSy60I1ytpvR0G3oqLXezAoQiCDBF0l70IRNPoOyqksifNogTlkk7gKXr7tcUW9LTjtfeUROoNNkpdbit7z0YMRPLBY/KxxO9AxFwGGKQtQ4wGf5kIkS3bLhoNIo1aGskeEY6ItUJ9iVgrYZQ2LTqp7eIcQxvcJc6tLlpBtibqIhIQLiCrHhuUWicYFqcKthfunFoULeuQ2n1pCIAc/AU2beM+3xVblZbBbNX/6ZLs7vnzyYXCvETi4rvRfOEDBem3EJAVtlXqLb/x//BgG9xrU7JWho7fW6CrUYm3hhporR5BEjG2IoR5y0I53kuQUHgU8qJovi1SxrEtASmyaj6cInhdCBRA52jXWPSrwBC8AJ4JYopW7TclhrOqYR6dkgrRz7KtLSXDJe8q+JYjxthepzVeaCBodIcFPo2wXOyu1dt4jS0syFOMC5SGsLAcmxfPQE+IB4BUVU6D+JZRO0sugv6XSPKJU/ZaBqwNHVLzoFfVCFDrQtFq5Al8qZGlwsgn4XLVLqrn277xOjQyigR57fRWOREQMURafaIXEk7KhSoJqeHLCwFxIRteU1gUcaxu8W7R30+O3dcEUq7UpA2LBt45abCAqXgsxpyMmttsyxTZKN+ed2+kRaUWSMUCJYAzEXzGHUY+qVgXH1D3lO64cYgNUajUyUmzhRumAssgQOfyEBmuLjJj1kvUtgpUKmphhVN1SNIweYZOFt9Br6ht3WNFx/NZj60C3wlyRykMWEsPt/ksBCOj0rJIHtxazODFGq9vY560IP0TQGZHdPueBiRsmZ95StFvIpFFM40gfi6JaoqqoN2pxZI2K8UnAliZTmfAaMs4MW+5iiM9LXOD1mxT3wJZnDIJavH+JB5cx6M0Rkpvn7Ot68RitKLMpXEKmEI1at8ULRFw/UUUJR1lN+DJHO+2sRxweYD1GksD1txTy4TLm2GuiT4gYrg13jYnmG70NN1sApZ8l9TQC27lV060gDdivt3QUQs3JU2x0CJECBpWM8hjxCJaAkBpGq+ri5ZOdd9Rmwbe0cIiHMkuaySDIVTGEjWUfOJYipqORZ41FmIiQJHNTfEuju26BYpKUNu2ERFCsOOT4hq2SuSNc1lLo2zS5gpu3aQUfSjSxuUJjS43xyTo7ijgLZYDG5bXTlpH1k7baI68Bt9UZK91q05gaSA/tcHzKkj/qrfUsJCLI9VJLbZJU1zvNiWNOFJSYOMaIiBtedjOktus5+X55I5SmLJqwMheIkuVDW+ZIMsN35UJiqBVVPfn/XhtTu05EcoyXDFpWP26XYw4ntt8GIsdej4QFKF4QhHqQphX8XovjZnfIhuknQcl7NQAx7MEPCS1Sv+2vXeMYFkKaNUYMzhFcVM34DXHyIkiUKI4azVYP8sY3AVSlEk6bA3JBSYBj5GD05UGyVda1T6Qw3XZkiUeqWNosVqzmpYUzytsHpF+oC4S7LownS3tLRL3GiKGqUvFspBKfEc6R5p5GuRc6P4LC/MSST6JQpnm8LF1cizFbq/QajOxM3drD5bHMXbB7rilonNUwreTestCAnC4Ze3f9o00vuXATMXnS2fkYqTpNFlgg8eE4ZaFU7XoQWkBuvdQNSyjuyNDNHyJNMvTG7KoMWtzbNm7wXak6Fbis7azOVk4aahoNbpSqX3wPou085MazV9DYNDqAroj0NFaW0RYBJ2NutSgipUWs+SA30g/kK5t4PJy1wYw9WFhvG+GziK2QiL2c3HW9yW6605/aMx7jaB97zwu2YWFeUnEW3+Gdd76NCLNSYoslWYhCPYVITJo2tHg6EIZ5FlZILHY8Wt/ms0ZugjO8UbUTSMo13C/JAn0shvwaq2a7hI+fyEon+JzWpyRIlGxHQshIrs5MNZJa9bZZutCUU1bai9tF5zvJpEFqZ5iYtRt9k8AhNpqTA7h4olTBic3OiqxUKIt6tv6OFfr29i+6mFZLi/xzQaGITYAj/S8a1xn3qaV5XQCQkwJEGp/mt63dB6FedGWyEsudxTzZfjJzUK4x86kLYskDaJfWwZsBKqQNxYE5KOTx3h9XteWEoXtgFct8fAaRA9iFiMxLMjEF90cdRkP1EDssHG82kUgvVU6z2H1LAWzvreazLalN5rH2D2kt+gIba+vXfT3eCL6dFJk3HbwmGxUDUpZz/Eea1myLJW0MVIrLFIbUfvk6GRIMfKJ0d0ogZRo61mKIyUyXtvuURkG9MYm3qftXhSL7GSJ+yKN/dMSMUG51Wu2WUs/xwqT6jd93Iqcl4x8HEe+93u/lze84Q0cHBzwNV/zNXzkIx859+fcWQpj4It4WItlgOhITAFpd42Kuw1RtymLZpFa05O3XU/m5mJIW9yNFrZmiZrHELutaxBTbKYeyQ59pLatl20Vh4wF84oE95jk6JGRVSWnimRrZZ7WX5LAB8N7g8Y93GkNitjOWFLIYqgHGyWDBYsNThIjqbFcTtjSWAwTnVaG/YmuK7gFAHXbyu05ftZhayGEeU8pB6nFHpzGfrl1XObmvnaKHSyQ4w26CVf19Hi6I133RBC2j77ji966a1rOXvgvBrfyM8nIf+d3fodf+7Vf44Mf/OAXva6Uwn333cfHP/5xbty4wT/7Z/+Mv/N3/g4f+9jHzvU5d5RLdvLlAT7UoTJfSXQ6U9c9FKEkoajC5YqZ46+qFFdqDvqlcSPIolDmTDmJeszQzfhJYuqh64KCte4JHQWXRJFEvVLJFcY++px94cjlmYUXxjpASdhJzISxLqYGeyX4v2ZFToiCn0XcRJsrY31Qw9qUKJ6o5sgY/fzrkwV1QYxUbwNlHfASvNHrzdBm2szMc2Y+6oM3LCm2Esr/EqnfLldKa/7alEStjhwpclAYk+E3Ugx6HZxKZa6JXCLAn/ro9BQF3QzUzYRfVUpO9FqYasZN0arUgxK0TZue49caokbpAiBbz7PCXoQs2Uc+8hE+9KEPcd999wHwwz/8w/zgD/7gF7H37+3t8WM/9mO7v7/+67+eb/mWb+ETn/gE3/qt33rm53xJFuZmczVu3LjBd33Xd3Hp0iVe/epX8+M//uPPes9Zz59H+ieU9GjCHu/QxxLyuZ70tJIOhXwk9Dec9ITSP67kRxL905AeTciTHd1jifxHmf4JoXsKhscFHs3IDSU/nvBHOngkw9OJeq2D40QqkK4l5FoiVehno3ta8Ud7Nk8s8ZMU48O3rluN2Sp5jLRtXkd2LBWnK9HqnMTiuCdtJOBayNVINWAz2tqS0ySk2lzPtewa41Jtlm/UYO3cRGBTNplaEv3oDI8khicUfawjXUuURxakz3fkJ5T+upO/kMh/1NFfE9JTcf36aE//WCI9nkhPJbrPJfKTmfxoRm906N4CfXQKVpsbie660F93dAP5SUUfzShOdygMX1D6JyHfEPLJ+WKYl5OM/Jmy2Wz4jd/4Dd785jefaw1+SRbmZnM1HnzwQZ566ik++9nP8thjj/H2t7+dBx54gO/5nu851/PnkXJABLhtbkodWpA9RwBaVq2+0tK11mn45NmxVWxkdQjMVV2BLwQ9jkBWPbJIsnBK76RrGnD/bFH66CrzUY+4BpHFFlulLX7YFjpbv457sLOwRbwlwFtT1XYNbUlhpLFlqqDNBdwmHnR7XGnFVtP4kDYuzzyxpW/1zimidK36H4QVDSX9DHaXbelIikXSpMFd3CG1RIj37XUl3LqSE/lggVwfsbuHgI9ppMfrKgqiVUF6mlsc9FWm51CY54lXXgoy8mce+53vfCdf9VVfxXd8x3ec+TnwJViYm83VODk54ed+7uf4wAc+wJUrV3jjG9/Igw8+uJvPcdbz55VnNoFt6ZUC/OcBeqy+S1GqPQOYKRINXg3CLzVQjdv0Ki321TnwXrm2YHaIlgDLkLWyHCZY1gjwuygOqhoptaA4G5otYpgugnWVeK7Timpl6Erg1bqInchOzkE7q8nwFDGKtLhFxKCzKFCK0/WFfijIUFF1UlfRruJDMFKu0tyAlKeEFSK0Gsop9isWdcKGqC1txwNar7vCquWozZhAryUU9u4BWZ9E8qIjmHE00t7a1Vb5335hEQ+e+b2a3/RxXnkmGflWnouMfCvuzrvf/W4+9alP8Yu/+Iuonk8Vbklhnmuuxqc+9Smmafoik/jbv/3b53r+ZvL+97//WeYYiH76TcBBdALZhMmXWeg2LUawbVdkwOHzCLpu8BVvI/1KjPNjVjCNWsuoQZVa2vFnIa/D3UtHMI8dZZPJh/HedCNG2FlJUSspcTyvitUGhmwTu9yEMgUkfxy7mEQ2NU6vKsxjos6pjUmPQqhvNI476+71VKFsMvMmeAHmGpMDtgw0dUocT32gEY6dfBzX6yVG56XNljxESGunOzS6G0Z3A3QT9ajuqPEibKINIR87OsXcGIqik+LDHunJNenYSMeQD6F/2vF1Ctf4OjtEgcv56jAvRGGeSUa+leciI4dQlu/7vu/jk5/8JB/72Mee0wLdTG5JYZ45V+OZcnR0xN7eHjmfenhXrlzh8PDwXM/fTN7//vfj7rsHsCOes236ti0UMbB5+3t7LYHfMpUAO6Y23CdQ+TsWfmhQMou/C6lhvxRr3GCeg8hukkRRbSnjWPBpDl4AabgwGhZMt/FIg/HQ2n13cyNLpHu1RHC8TYfr9r3t/xL+2O7/YoRyToEzY2xxTBVoQ5m80R/VLogNLbWelS2Ke8s3lgVPDXWcWl2r1U3idcGD4JnAqbV7hCgsl+iNdZCHeEvrl20tprmA50Aqw4uTVt6SkT/yyCM88sgjz0lGDvCe97yHX//1X+dXf/VXuXr16i19zrljmOebq7G/v8/JyQmllJ1SXL9+fWcOz3r+vCLm1BWBx+qB6nhNeAoErczEF5x8N1U5epoNr5Fa1XkLugwcWRQE26JIjqZKlYxnD8vTwB3mjccshfuRCuzik4aWtuaeiLViqEYCoPbR+74dF7Et/tFiKuwZfSVKAzPqlksvYhUJiEy4WOHyGQ69BV3TUJESA5Qst1rUtgGusbtADHEtjXNMdqjqQB17GzNouW0U4liyptANlpQ8UAidUK8s0etr6Bah1CXQzXj0+ngbG3OmvAiV/vOSkX/mM5/hp37qpxiGgQceeGD3/u/+7u/mp3/6p8/8nHMrzPPN1fi3//bf0nUdv/Vbv8Wf+lN/CgiT+KY3vQmAr/7qr37e588rq0cr4+XEfCD01xqcf9n6MNzJx1AHpTsxusmDzWQMor98Er3pWhoQsgaidt4T+qMa1e3qsXBM0VYEXX2+Ml9W5nUswO4I0nWlP6wcv06ZDuQUnu6BUGZLvLFtUPYdVC3imnVDGlTQKRaWl5YM2B6ryu6YzDHaOwqyUU8xC2yZacJpzDGNsGL1qO1A2OtXJfJJFFvnA0XH2AjSGIH+1sfYQvR1CtaYLVxoa5HqQSEfaQT4NWBKuEC3YP93r+GLnutfu2Lx+MTi8YK8YUVdKOPVc7hkL0Jaues6HnroIR566KEveu6ZivDAAw+8IILzc7tk73jHO/i93/s9Hn74YR5++GH+6T/9pxwcHPDwww/zDd/wDXznd34n733ve7l+/Tqf/vSn+fCHP7wziavV6nmfP69sOZEDWt8AgXJqTWKH9521oPPdbr6txLsSE8lo1qVj153oKcZ/C9vuzNZCgJPU0eStl8YbjWxYHG/wG2if+4xslGdiRkp7rWTfnYc3NEB8XjtWarAYadfyjG9o97uHlZHW4cjWkvrp6yJ5ts1ytOtJ23vT3MzU3uKnEH2kEYS04uO2kU4HC3OnW1xbO1YWfNHDegQ3xGpwL9A+5xzzYah+88dtKOe2MGfN1fjJn/xJ3vWud3H//fezXC55z3ve86yU8VnPn0cOv1zxZVTRx4PgFeaGgEXlvSwEv2SsTdl04HOCuwNTlkYC9HjSfndBFoau4fhAqHveRv71aI1pYPlIOHpDRhzG0qETTK+C7q7KfBz+uq5bz0ejhnXVgLeXcLPSeNr1KA5cb9xee81bHCGvFZYtUBbgmqIeMUR3CGXfd0mMOG7Cc8yUpDrpKOGXKvmGQFWufXUjPI9RTmwux9zNdBK1qunuONfuKUUJC00fyQopUX9a35d2jJtpdHzTo5c9IDldWHopkUC5/rUr8AWLP9qwuXfB+l6h7CllL3gKzhKxOwdM9iVX+r/5m7+Za9eu7f6+dOkSH/3oR5/z9Wc9fx5ZPGbMV4KxPj8dhcIqLaVZnHQszFnpaiXPFoBCE3o30sYpq4RMEQtt+cbKEvqNB1IYYbi7sLYeiMr98GSl7CU0V0ShP4ksWX/kHN8TtE1bUj/LxGCitUbaNQfExhrJYPCfRdap9uGq1YVji1DmOrfMXq6MmlrwLI2WNpSrrhwdKp5hMGMcIgfAYNRLEeesvmABQK3GfAnSITA28oqxUSHNjsxhfVQFNYuO0E1A9Lvrgej0LuAux6810qEgPZhqgC8JUvbFYxNilc29C/ITx3RzYv26FVoaR/NZcptak5vJHQWN6Q+JHdiU7lolrT3Irxt4Mq+j7yNPRr92yiWLDkeXSI+uQ7E8NfqkCn6XMDztUYtwQZbONAd4UIDueuyQk4UblNeQryn5UNCDKNRZzw6NvM2GbRn7XbauTjzvHci6kQl6JBtoaGcUpBFw7IIQ2HVl7oqPBAhUNwSuLm0zfoJYzGexFCz6tVe6w6A8mi7HRpHGsLLWbVu1ozfHWiOezgHRr73s+mpEjXQSWDQ3IR+zc4sXjxekGOt7hW5OLD9zyHxpAIkxI2fJ/xQW5pWQzd1h5q2DzV3RW+95+4XGmO26EKzLlMGxQfBlpKPzqjU/eSMj34v0p3WCvaoxpphTFn3QyUp8ztH9GRGoi5gWNu8Hud50KYV16dti3kLs20LXNgrPheBSbjB42cYIidNCa+9R3ytElqwE8lisWS2aUhhojgJnBNyOmpILMMaxUx/DjKLhJlj6o7tUd2ybWxIRcdkxuUpDEGxbHua91r7cYsbSge6xaxwbr2wVWJA3hKte9pT161bMlwbmPGM5R7H5LLlQmJdGtrGtioM0ate+LVbVYDVqqVSZ2DHVe4r6hmmkePu1U1Zt0c5QFxpwfhVEK7YHcqK7XZ8EeVEpxyl4LgrUZTSP0UcTmJGCh0xbgG1RzaeCS9RuXBx6TumYiuD7wWTpPZTjiEm0q8F+eZThJFEHxxaOn0QMl5cFMWFR5hjtfdLHANbOSbWQN0G9WfuGPxujJaKI7Fgt02i4KtOlSDBsCSvS1DolEyBCaihlrYC2aW0iUZNpCYHaxmCUPUGLgiiWM4vPn7C5f/Wc3+dO7hx9ubMUZjoQyh7hsnQtVVslSCeGSlmlWMQI0wIwYfOqiDF0LQHl2E7Salk17Xw35wQn5shIuFoAm33IG5hrRkWYD6D0rTjoDYU8KjpJ3M0UMBwsLIJUUGkweAm0ArVlwgxko4FunkHXEUcxR5rcxzgnnWKn1wI2CVWDVG8NlFligkFymMJ1LK+NbssAfEmMnLCwrrs6lKaWQOCU7HyUrecIe7IrZmpxagZZSOMX8F0FXxzGq6cJjro4xett7l+hNzZnfq8XLtlLJIvHC/MmUYcgetDq1FX7imcjX3dmU7TW+JJXih8BCt2h7XBS0ggdtG7JMKIqnibHXlOYSx/p0ArLx0I50n6N0d3HQn46qvzT1SACtN5PJ3E118pFWjRuqAZ/mJtE+/EYAX3QF0m4fyJByDEJ9E6dE+pxje6CrhuHdGtb9llZFMMkiPbYIjpHYfX52pANTtkXdBNMlGXZyM4bp9u20cuJBjMIXJ3niHFq6zcSjLl39DAx70XvPjWu1dVYPhpgvPW9ie6wxqzRSwnroFw5R9RfLxTmJZHf+OgPvdKncCEvhVxYmJdGbrcR1C+HiMgf/+uudw6T3x2lMBfyx1QuXLILuZBbkAuX7MWXH/mRH3mlT+EVkf8prvsOcjnF/9g7yBdyO4uI8P949f/7ps/9yqP/v9sufrtjLMyF/DGWixjmQi7k/OIXWbILuZBbkAuFuZALOb9cWJgLuZBbkdsssH8+ue25lc9LMn0nyVmE2N/8zd/MMAzs7+/vHp///Od3z78YLKK3ldR688ctyK2skxeypm57hTkvyfSdJOchxP5H/+gfcXR0tHu89rWv3T33TBbRX/u1X+NnfuZn+Bf/4l+8EpfyoojXetPHrcitrJMXtKb8Npf777/ff+EXfmH398///M/761//+lfwjF4a+fZv/3Z/73vf6+7u3/RN3+Qf+tCHbvq64+Nj7/ve//N//s+7//3jf/yP/Ru/8RtfjtN80QXwt8v/66aPW1met7JOXsiauq0V5qmnnnLAP/3pT+/+97u/+7sO+LVr117BM3txZb1e++te97rdl/hN3/RNfvfdd/vVq1f9LW95i//zf/7Pd6/9L//lvzjg8zzv/vexj33Mr1y58rKf94shDzzwwLYp4osely9fftbfP/IjP3LTY9zKOnmha+q2Dvq/VJLpO0n8JoTYP/ETP8HXfd3XsVqt+A//4T/wjne8g4ODA77927/9S2IRvZ3lD/7gD17wMW5lnbzQNXVbxzBfCsn0nST+HITY3/AN38Dly5fpuo6/8Bf+Au9617v4N//m3wDPZhHdypfCIvrHSW5lnbzQNXVbK8ytkkzfSeK3QIj9TGb5Z7KIbuVLYRH94yS3sk5e8Jr6EtzOl1Xe+973+lvf+lb/whe+4F/4whf8rW99q//oj/7oK31aL1je/e53+5vf/GZ/4oknnvX/p59+2n/5l3/Zj4+PvZTiH//4x/3y5cv+8z//87vX/LW/9tf8277t2/zatWv+u7/7u/7617/+WXHO/4xyK+vkhayp215hpmnyd7/73X7lyhW/cuWKv+c973lWwHsnyh/8wR844MMw+N7e3u7xrne9yx977DH/03/6T/vBwYEfHBz4m970Jv/Zn/3ZZ73/+vXr/lf+yl/x/f19f9WrXvXHYgN5ofJ86+Rd73qXv+td7zrXa8+SC3j/hVzILchtHcNcyIXcbnKhMBdyIbcgFwpzIRdyC3KhMBdyIbcgFwpzIRdyC3KhMBdyIbcgFwpzIRdyC3KhMBdyIbcgFwpzIRdyC3KhMBdyIbcg/38015ewDc50CAAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from mne_rsa import plot_rdms\n", "plot_rdms(pixel_rdm);" ] }, { "cell_type": "markdown", "id": "24f71fc6-a5d3-4657-9822-f6c1776f1952", "metadata": {}, "source": [ "Staring deeply into this RDM will reveal to you which images belonged to the \"scrambled faces\" class, as those pixels are quite different from the actual faces and each other.\n", "We also see that for some reason, the famous faces are a little more alike than the unknown faces.\n", "\n", "The diagonal is all zeros. Take a moment to ponder why that would be." ] }, { "cell_type": "markdown", "id": "86471704-d428-47b5-9728-377b67dffb4c", "metadata": {}, "source": [ "
\n", " IMPLEMENTATION DETAIL
\n", " The compute_rdm function is a wrapper around scipy.spatial.distance.pdist.\n", " This means that all the metrics supported by pdist are also valid for compute_dsm.\n", " This also means that in MNE-RSA, the native format for an RDM is the so-called \"condensed\" form.\n", " Since RDMs are symmetric, only the upper triangle is stored.\n", " The scipy.spatial.distance.squareform function can be used to go from a square matrix to its condensed form and back.\n", "
" ] }, { "cell_type": "markdown", "id": "e2f5d3a3-e405-4b67-afc2-df27e03626c5", "metadata": {}, "source": [ "## Your second RDM\n", "\n", "There are many sensible representations possible for images.\n", "One intriguing one is to create them using convolutional neural networks (CNNs).\n", "For example, there is the [FaceNet](https://github.com/davidsandberg/facenet) model by [Schroff et al. (2015)](http://arxiv.org/abs/1503.03832) that can generate high-level representations, such that different photos of the same face have similar representations.\n", "I have run the stimulus images through FaceNet and recorded the generated embeddings for you to use:" ] }, { "cell_type": "code", "execution_count": 10, "id": "3b745df5-d5dc-499f-9724-4c486568628d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "For each of the 450 images, the embedding is a vector of length 512: (450, 512)\n" ] } ], "source": [ "store = np.load(f\"{data_path}/stimuli/facenet_embeddings.npz\")\n", "filenames = store[\"filenames\"]\n", "embeddings = store[\"embeddings\"]\n", "print(f\"For each of the 450 images, the embedding is a vector of length 512: {embeddings.shape}\")" ] }, { "cell_type": "markdown", "id": "8286c0e7-517c-4a2c-9fcc-f9bff07bf0a6", "metadata": {}, "source": [ "
\n", "EXERCISE:\n", " \n", "I leave it up to you to construct the RDM based on the FaceNet embedding vectors using the [`compute_rdm`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.compute_rdm.html) function.\n", "Use Pearson correlation as dissimility metric and store the RDM in a variable called `facenet_rdm`.\n", "Make sure that the stimuli are in the same order as the pixel RDM we created earlier!\n", "
" ] }, { "cell_type": "code", "execution_count": 13, "id": "7d33f777-c2d0-4b81-9793-60a4cb9df884", "metadata": {}, "outputs": [], "source": [ "facenet_rdm = compute_rdm(embeddings) # write your code here" ] }, { "cell_type": "markdown", "id": "863f3e6c-c2a9-4e99-bcbe-609a6a32919b", "metadata": {}, "source": [ "If you created the FaceNet RDM correctly, executing the cell below should plot both RDMs side-by-side:" ] }, { "cell_type": "code", "execution_count": 14, "id": "6c2886c2-0e27-4555-826a-ef9cc8997dd8", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plot_rdms([pixel_rdm, facenet_rdm], names=[\"pixels\", \"facenet\"]);" ] }, { "cell_type": "markdown", "id": "2d811553-b0c0-42f0-bd55-a1ed370e9009", "metadata": {}, "source": [ "## A look at the brain data\n", "\n", "We've seen how we can create RDMs using properties of the images or embeddings generated by a model.\n", "Now it's time to see how we create RDMs based on the MEG data.\n", "For that, we first load the epochs from a single participant." ] }, { "cell_type": "code", "execution_count": 15, "id": "d14358f7-6f47-4032-8164-1d0a438c0d66", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Reading /home/vanvlm1/projects/neuroscience_tutorials/rsa/data/sub-02/sub-02-epo.fif ...\n", " Found the data of interest:\n", " t = -200.00 ... 2900.00 ms\n", " 0 CTF compensation matrices available\n", "Adding metadata with 2 columns\n", "879 matching events found\n", "No baseline correction applied\n", "0 projection items activated\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", " \n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "
\n", " \n", " \n", " General\n", "
Filename(s)\n", " \n", " sub-02-epo.fif\n", " \n", " \n", "
MNE object typeEpochsFIF
Measurement date2009-04-09 at 11:04:14 UTC
Participant
ExperimenterMEG
\n", " \n", " \n", " Acquisition\n", "
Total number of events879
Events counts\n", " \n", " face/famous/first: 147\n", "
\n", " \n", " face/famous/immediate: 78\n", "
\n", " \n", " face/famous/long: 66\n", "
\n", " \n", " face/unfamiliar/first: 149\n", "
\n", " \n", " face/unfamiliar/immediate: 65\n", "
\n", " \n", " face/unfamiliar/long: 79\n", "
\n", " \n", " scrambled/first: 150\n", "
\n", " \n", " scrambled/immediate: 71\n", "
\n", " \n", " scrambled/long: 74\n", " \n", " \n", "
Time range-0.200 – 2.900 s
Baseline-0.200 – 0.000 s
Sampling frequency220.00 Hz
Time points683
Metadata879 rows × 2 columns
\n", " \n", " \n", " Channels\n", "
Magnetometers\n", " \n", "\n", " \n", "
Gradiometers\n", " \n", "\n", " \n", "
EOG\n", " \n", "\n", " \n", "
ECG\n", " \n", "\n", " \n", "
Stimulus\n", " \n", "\n", " \n", "
Head & sensor digitization137 points
\n", " \n", " \n", " Filters\n", "
Highpass1.00 Hz
Lowpass40.00 Hz
" ], "text/plain": [ "" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import mne\n", "epochs = mne.read_epochs(f\"{data_path}/sub-02/sub-02-epo.fif\")\n", "epochs" ] }, { "cell_type": "markdown", "id": "026759bf-b8b6-47da-b88c-126096d42e0b", "metadata": {}, "source": [ "Each epoch corresponds to the presentation of an image, and the signal across the sensors over time can be used as the neural representation of that image.\n", "Hence, one could make a neural RDM, of for example the gradiometers, like this:" ] }, { "cell_type": "code", "execution_count": 16, "id": "64cb2f7a-504b-4cd8-99a0-921426669d4a", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "neural_rdm = compute_rdm(epochs.copy().pick(\"grad\").crop(0.1, 0.2).get_data())\n", "plot_rdms(neural_rdm);" ] }, { "cell_type": "markdown", "id": "45db1fe4-828b-455b-af03-97a6d281fa17", "metadata": {}, "source": [ "To compute RSA scores, we want to compare the resulting neural RDM with the RDMs we've created earlier.\n", "However, if we inspect the neural RDM closely, we see that its rows and column don't line up with those of the previous RDMs.\n", "There are too many (879 vs. 450) and they are in the wrong order. Making sure that the RDMs match is an important and sometimes tricky part of RSA.\n", "\n", "To help us out, a useful feature of MNE-Python is that epochs have an associated [`epochs.metadata`](https://mne.tools/stable/auto_tutorials/epochs/30_epochs_metadata.html) field.\n", "This metadata is a [Pandas DataFrame](https://pandas.pydata.org/docs/user_guide/dsintro.html#dataframe) where each row contains information about the corresponding epoch.\n", "The epochs in this tutorial come with some useful `.metadata` already:" ] }, { "cell_type": "code", "execution_count": 17, "id": "9a4e8d42-39f3-46e4-83c2-4da33008b7f8", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
triggerfile
013u032.bmp
114u032.bmp
213u088.bmp
313u084.bmp
45f123.bmp
.........
8825f016.bmp
8836f016.bmp
8845f002.bmp
8856f002.bmp
8867f150.bmp
\n", "

879 rows × 2 columns

\n", "
" ], "text/plain": [ " trigger file\n", "0 13 u032.bmp\n", "1 14 u032.bmp\n", "2 13 u088.bmp\n", "3 13 u084.bmp\n", "4 5 f123.bmp\n", ".. ... ...\n", "882 5 f016.bmp\n", "883 6 f016.bmp\n", "884 5 f002.bmp\n", "885 6 f002.bmp\n", "886 7 f150.bmp\n", "\n", "[879 rows x 2 columns]" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "epochs.metadata" ] }, { "cell_type": "markdown", "id": "9efd1716-631f-421c-a911-fb20c9273b4f", "metadata": {}, "source": [ "While the trigger codes only indicate what type of stimulus was shown, the `file` column of the metadata tells us the exact image.\n", "Couple of challenges here: the stimuli where shown in a random order, stimuli were repeated twice during the experiment, and some epochs were dropped during preprocessing so not every image is necessarily present twice in the `epochs` data. 😩\n", "\n", "Luckily, MNE-RSA has a way to make our lives easier.\n", "Let's take a look at the [`rdm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rdm_epochs.html) function, the Swiss army knife for computing RDMs from an MNE-Python `epochs` object:" ] }, { "cell_type": "code", "execution_count": 19, "id": "b8452c77-33e8-4e5f-b0d3-54fb79ce005b", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "\u001b[0;31mSignature:\u001b[0m\n", "\u001b[0mrdm_epochs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mnoise_cov\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mspatial_radius\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mtemporal_radius\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mdist_metric\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'correlation'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mdist_params\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mn_folds\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mpicks\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mtmin\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mtmax\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mdropped_as_nan\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mDocstring:\u001b[0m\n", "Generate RDMs in a searchlight pattern on epochs.\n", "\n", "Parameters\n", "----------\n", "epochs : instance of mne.Epochs\n", " The brain activity during the epochs. The event codes are used to distinguish\n", " between items.\n", "noise_cov : mne.Covariance | None\n", " When specified, the data will by normalized using the noise covariance. This is\n", " recommended in all cases, but a hard requirement when the data contains sensors\n", " of different types. Defaults to None.\n", "spatial_radius : floats | None\n", " The spatial radius of the searchlight patch in meters. All sensors within this\n", " radius will belong to the searchlight patch. Set to None to only perform the\n", " searchlight over time, flattening across sensors. Defaults to None.\n", "temporal_radius : float | None\n", " The temporal radius of the searchlight patch in seconds. Set to None to only\n", " perform the searchlight over sensors, flattening across time. Defaults to None.\n", "dist_metric : str\n", " The metric to use to compute the RDM for the epochs. This can be any metric\n", " supported by the scipy.distance.pdist function. See also the\n", " ``epochs_rdm_params`` parameter to specify and additional parameter for the\n", " distance function. Defaults to 'correlation'.\n", "dist_params : dict\n", " Extra arguments for the distance metric used to compute the RDMs. Refer to\n", " :mod:`scipy.spatial.distance` for a list of all other metrics and their\n", " arguments. Defaults to an empty dictionary.\n", "y : ndarray of int, shape (n_items,) | None\n", " For each Epoch, a number indicating the item to which it belongs. When\n", " ``None``, the event codes are used to differentiate between items. Defaults to\n", " ``None``.\n", "n_folds : int | sklearn.model_selection.BaseCrollValidator | None\n", " Number of cross-validation folds to use when computing the distance metric.\n", " Folds are created based on the ``y`` parameter. Specify ``None`` to use the\n", " maximum number of folds possible, given the data. Alternatively, you can pass a\n", " Scikit-Learn cross validator object (e.g. ``sklearn.model_selection.KFold``) to\n", " assert fine-grained control over how folds are created.\n", " Defaults to 1 (no cross-validation).\n", "picks : str | list | slice | None\n", " Channels to include. Slices and lists of integers will be interpreted as channel\n", " indices. In lists, channel *type* strings (e.g., ``['meg', 'eeg']``) will pick\n", " channels of those types, channel *name* strings (e.g., ``['MEG0111',\n", " 'MEG2623']`` will pick the given channels. Can also be the string values \"all\"\n", " to pick all channels, or \"data\" to pick data channels. ``None`` (default) will\n", " pick all MEG and EEG channels, excluding those maked as \"bad\".\n", "tmin : float | None\n", " When set, searchlight patches will only be generated from subsequent time points\n", " starting from this time point. This value is given in seconds. Defaults to\n", " ``None``, in which case patches are generated starting from the first time\n", " point.\n", "tmax : float | None\n", " When set, searchlight patches will only be generated up to and including this\n", " time point. This value is given in seconds. Defaults to ``None``, in which case\n", " patches are generated up to and including the last time point.\n", "dropped_as_nan : bool\n", " When this is set to ``True``, the drop log will be used to inject NaN values in\n", " the RDMs at the locations where a bad epoch was dropped. This is useful to\n", " ensure the dimensions of the RDM are the same, irregardless of any bad epochs\n", " that were dropped. Make sure to use ``ignore_nan=True`` when using RDMs with\n", " NaNs in them during subsequent RSA computations. Defaults to ``False``.\n", "\n", " .. versionadded:: 0.8\n", "\n", "Yields\n", "------\n", "rdm : ndarray, shape (n_items, n_items)\n", " A RDM for each searchlight patch.\n", "\u001b[0;31mFile:\u001b[0m ~/micromamba/lib/python3.11/site-packages/mne_rsa/sensor_level.py\n", "\u001b[0;31mType:\u001b[0m function" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from mne_rsa import rdm_epochs\n", "rdm_epochs?" ] }, { "cell_type": "markdown", "id": "401134f9-5dee-48a4-8c95-86aea9509297", "metadata": {}, "source": [ "In MNE-Python tradition, the function has a lot of parameters, but all-but-one have a default so you only have to specify the ones that are relevant to you.\n", "For example, to redo the neural RDM we created above, we could do something like:" ] }, { "cell_type": "code", "execution_count": 20, "id": "334ff430-9744-4831-a892-160cb3f1af27", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "neural_rdm_gen = rdm_epochs(epochs, tmin=0.1, tmax=0.2)\n", "\n", "# dsm_epochs returns a generator of RDMs\n", "# unpacking the first (and only) RDM from the generator\n", "neural_rdm = next(neural_rdm_gen)\n", "plot_rdms(neural_rdm);" ] }, { "cell_type": "markdown", "id": "cdf11484-5a2a-4238-bd9e-84ee5ced89fc", "metadata": {}, "source": [ "Take note that [`rdm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rdm_epochs.html) returns a [generator](https://wiki.python.org/moin/Generators) of RDMs.\n", "This is because one of the main use-cases for MNE-RSA is to produce RDMs using sliding windows (in time and also in space), which can produce a large amount of RDMs that can take up a lot of memory of you're not careful.\n", "\n", "## The `y` parameter that solves ~~all~~ most alignment problems\n", "Looking at the neural RDM above, something is clearly different from the one we made before.\n", "This one has 9 rows and columns.\n", "Closely inspecting the docstring of [`rdm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rdm_epochs.html) reveals that it is the `y` parameter that is responsible for this:\n", "\n", "```\n", "y : ndarray of int, shape (n_items,) | None\n", " For each Epoch, a number indicating the item to which it belongs. When\n", " None, the event codes are used to differentiate between items.\n", " Defaults to None.\n", "```\n", "\n", "Instead of producing one row per epoch, [`rdm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rdm_epochs.html) produced one row per event type, averaging across epochs of the same type before computing dissimilarity.\n", "This is not quite what we want though.\n", "If we want to match `pixel_rdm` and `facenet_rdm`, we want every single one of the 450 images to be its own stimulus type.\n", "\n", "
\n", "EXERCISE:\n", " \n", "Turning it over to you: in the cell below, write the code necessary to construct the desired neural RDM.\n", "This is your first real challenge in this workshop.\n", "Keep the following in mind:\n", "\n", " - Each of the 450 images should be on a row by itself\n", " - We will achieve this by setting the `y` parameter of [`rdm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rdm_epochs.html) to a list that assigns each of the 879 epochs to a number from 1-450 (or 0-449) indicating which image was shown. Take care to assign number according to the order in which they appear in `pixel_rdm` and `facenet_rdm`.\n", " - An image is identified by its filename, and we have the `files` and `filenames` variables left over from earlier that contain all the images in the proper order.\n", " - The `epochs.metadata[\"file\"]` column contains the filenames corresponding to the epochs.\n", " - Tell [`rdm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rdm_epochs.html) to only consider data from 0.1 to 0.2 seconds.\n", " - The result will be a generator. Use `next()` to unpack the RDM from it.\n", "
" ] }, { "cell_type": "code", "execution_count": 21, "id": "74f8f6b7-1766-4c00-ae6d-31752d33cd04", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
triggerfile
013u032.bmp
114u032.bmp
213u088.bmp
313u084.bmp
45f123.bmp
.........
8825f016.bmp
8836f016.bmp
8845f002.bmp
8856f002.bmp
8867f150.bmp
\n", "

879 rows × 2 columns

\n", "
" ], "text/plain": [ " trigger file\n", "0 13 u032.bmp\n", "1 14 u032.bmp\n", "2 13 u088.bmp\n", "3 13 u084.bmp\n", "4 5 f123.bmp\n", ".. ... ...\n", "882 5 f016.bmp\n", "883 6 f016.bmp\n", "884 5 f002.bmp\n", "885 6 f002.bmp\n", "886 7 f150.bmp\n", "\n", "[879 rows x 2 columns]" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "epochs.metadata" ] }, { "cell_type": "code", "execution_count": 25, "id": "7e1672d1-72cb-4c34-a6fa-f2790d173fa2", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "y = [list(filenames).index(f) for f in epochs.metadata.file]# compute y here\n", "neural_rdm = next(rdm_epochs(epochs, y=y, tmin=0.1, tmax=0.2)) # compute the RDM here\n", "\n", "# This plots your RDM\n", "plot_rdms(neural_rdm);" ] }, { "cell_type": "markdown", "id": "912423dd-7e29-4d8c-a176-fb3272919b7d", "metadata": {}, "source": [ "If you've done it correctly, the cell below will compure RSA between the neural RDM and the pixel and FaceNet RDMs we created earlier.\n", "The RSA score will be the Spearman correlation between the RDMs, which is the default metric used in the [original RSA paper](https://www.frontiersin.org/articles/10.3389/neuro.06.004.2008/full)." ] }, { "cell_type": "code", "execution_count": 26, "id": "840fc74e-2803-4fbf-a477-f0b32c3885b9", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "RSA score between neural RDM and pixel RDM: 0.07869920694906636\n", "RSA score between neural RDM and FaceNet RDM: 0.07529582461337744\n" ] } ], "source": [ "from mne_rsa import rsa\n", "rsa_pixel = rsa(neural_rdm, pixel_rdm, metric=\"spearman\")\n", "rsa_facenet = rsa(neural_rdm, facenet_rdm, metric=\"spearman\")\n", "\n", "print(\"RSA score between neural RDM and pixel RDM:\", rsa_pixel)\n", "print(\"RSA score between neural RDM and FaceNet RDM:\", rsa_facenet)" ] }, { "cell_type": "markdown", "id": "6c71d57f-e011-4afa-9b7d-6c6e5da3e7fe", "metadata": {}, "source": [ "## Slippin' and slidin' across time\n", "\n", "The neural representation of a stimulus is different across brain regions and evolves over time.\n", "For example, we would expect that the pixel RDM would be more similar to a neural RDM that we computed across the visual cortex at an early time point, and that the FaceNET RDM might be more similar to a neural RDM that we computed at a later time point.\n", "\n", "For the remainder of this notebook, we'll restrict the `epochs` to only contain the sensors over the left occipital cortex.\n", "\n", "
\n", " IMPORTANT NOTE
\n", " Just because we select sensors over a certain brain region, does not mean the magnetic fields originate from that region.\n", " This is especially true for magnetometers. To make it a bit more accurate, we only select gradiometers.\n", "
" ] }, { "cell_type": "code", "execution_count": 27, "id": "f753bc22-826e-4586-98af-55c804b4218f", "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", " \n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", " \n", " \n", " \n", "\n", "\n", "\n", "
\n", " \n", " \n", " General\n", "
Filename(s)\n", " \n", " sub-02-epo.fif\n", " \n", " \n", "
MNE object typeEpochsFIF
Measurement date2009-04-09 at 11:04:14 UTC
Participant
ExperimenterMEG
\n", " \n", " \n", " Acquisition\n", "
Total number of events879
Events counts\n", " \n", " face/famous/first: 147\n", "
\n", " \n", " face/famous/immediate: 78\n", "
\n", " \n", " face/famous/long: 66\n", "
\n", " \n", " face/unfamiliar/first: 149\n", "
\n", " \n", " face/unfamiliar/immediate: 65\n", "
\n", " \n", " face/unfamiliar/long: 79\n", "
\n", " \n", " scrambled/first: 150\n", "
\n", " \n", " scrambled/immediate: 71\n", "
\n", " \n", " scrambled/long: 74\n", " \n", " \n", "
Time range-0.100 – 1.000 s
Baseline-0.200 – 0.000 s
Sampling frequency220.00 Hz
Time points243
Metadata879 rows × 2 columns
\n", " \n", " \n", " Channels\n", "
Gradiometers\n", " \n", "\n", " \n", "
Head & sensor digitization137 points
\n", " \n", " \n", " Filters\n", "
Highpass1.00 Hz
Lowpass40.00 Hz
" ], "text/plain": [ "" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "picks = mne.channels.read_vectorview_selection(\"Left-occipital\")\n", "picks = [\"\".join(p.split(\" \")) for p in picks]\n", "epochs.pick(picks).pick(\"grad\").crop(-0.1, 1)" ] }, { "cell_type": "markdown", "id": "c74e9d13-f266-49d3-b75d-b286eaafc67d", "metadata": {}, "source": [ "
\n", "EXERCISE:\n", " \n", "In the cell below, use [`rdm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rdm_epochs.html) to compute RDMs using a sliding window by setting the `temporal_radius` parameter to `0.1` seconds.\n", "Use the entire time range (`tmin=None` and `tmax=None`) and leave the result as a generator (so no `next()` calls).\n", "Store the resulting generator in a variable called `neural_dsms_gen`.\n", "
" ] }, { "cell_type": "code", "execution_count": 30, "id": "c3aa6b81-3515-4c6a-8d12-ad7593ea6131", "metadata": {}, "outputs": [], "source": [ "neural_rdms_gen = rdm_epochs(epochs, y=y, temporal_radius=0.1) # write your call to rdm_epochs() here" ] }, { "cell_type": "markdown", "id": "87567d32-1bd9-4692-928f-7ddd5af6504c", "metadata": {}, "source": [ "If you did it correctly, the cell below will consume the generator (with a nice progress bar) and plot a few of the generated RDMs:" ] }, { "cell_type": "code", "execution_count": 31, "id": "ba4fce9d-0b6c-4777-b543-11e06b19b550", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ " 0%| | 0/199 [00:00" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from tqdm import tqdm\n", "times = epochs.times[(epochs.times >= 0) & (epochs.times <= 0.9)]\n", "neural_rdms_list = list(tqdm(neural_rdms_gen, total=len(times)))\n", "plot_rdms(neural_rdms_list[::10], names=[f\"t={t:.2f}\" for t in times[::10]]);" ] }, { "cell_type": "markdown", "id": "65a5b1ec-3b73-4639-acbc-061c5fea3eda", "metadata": {}, "source": [ "## Putting it altogether for sensor-level RSA\n", "\n", "Now all that is left to do is compute RSA scored between the neural RDMs you've just created and the pixel and FaceNet RDMs.\n", "We could do this using the [`rsa_gen`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rsa_gen.html) function, but I'd rather directly show you the [`rsa_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rsa_epochs.htm) function that combines computing the neural RDMs with computing the RSA scores:" ] }, { "cell_type": "code", "execution_count": 32, "id": "eadd2456-cefd-4b4d-a04e-fad5bea9d3eb", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "\u001b[0;31mSignature:\u001b[0m\n", "\u001b[0mrsa_epochs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mrdm_model\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mnoise_cov\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mspatial_radius\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mtemporal_radius\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mepochs_rdm_metric\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'correlation'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mepochs_rdm_params\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mrsa_metric\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'spearman'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mignore_nan\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mn_folds\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mpicks\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mtmin\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mtmax\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mdropped_as_nan\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n", "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mDocstring:\u001b[0m\n", "Perform RSA in a searchlight pattern on epochs.\n", "\n", "The output is an Evoked object where the \"signal\" at each sensor is the RSA,\n", "computed using all surrounding sensors.\n", "\n", "Parameters\n", "----------\n", "epochs : instance of mne.Epochs\n", " The brain activity during the epochs. The event codes are used to distinguish\n", " between items.\n", "rdm_model : ndarray, shape (n, n) | (n * (n - 1) // 2,) | list of ndarray\n", " The model RDM, see :func:`compute_rdm`. For efficiency, you can give it in\n", " condensed form, meaning only the upper triangle of the matrix as a vector. See\n", " :func:`scipy.spatial.distance.squareform`. To perform RSA against multiple\n", " models at the same time, supply a list of model RDMs.\n", "\n", " Use :func:`compute_rdm` to compute RDMs.\n", "noise_cov : mne.Covariance | None\n", " When specified, the data will by normalized using the noise covariance. This is\n", " recommended in all cases, but a hard requirement when the data contains sensors\n", " of different types. Defaults to None.\n", "spatial_radius : floats | None\n", " The spatial radius of the searchlight patch in meters. All sensors within this\n", " radius will belong to the searchlight patch. Set to None to only perform the\n", " searchlight over time, flattening across sensors. Defaults to None.\n", "temporal_radius : float | None\n", " The temporal radius of the searchlight patch in seconds. Set to None to only\n", " perform the searchlight over sensors, flattening across time. Defaults to None.\n", "epochs_rdm_metric : str\n", " The metric to use to compute the RDM for the epochs. This can be any metric\n", " supported by the scipy.distance.pdist function. See also the\n", " ``epochs_rdm_params`` parameter to specify and additional parameter for the\n", " distance function. Defaults to 'correlation'.\n", "epochs_rdm_params : dict\n", " Extra arguments for the distance metric used to compute the RDMs. Refer to\n", " :mod:`scipy.spatial.distance` for a list of all other metrics and their\n", " arguments. Defaults to an empty dictionary.\n", "rsa_metric : str\n", " The RSA metric to use to compare the RDMs. Valid options are:\n", "\n", " * 'spearman' for Spearman's correlation (the default)\n", " * 'pearson' for Pearson's correlation\n", " * 'kendall-tau-a' for Kendall's Tau (alpha variant)\n", " * 'partial' for partial Pearson correlations\n", " * 'partial-spearman' for partial Spearman correlations\n", " * 'regression' for linear regression weights\n", "\n", " Defaults to 'spearman'.\n", "ignore_nan : bool\n", " Whether to treat NaN's as missing values and ignore them when computing the\n", " distance metric. Defaults to ``False``.\n", "\n", " .. versionadded:: 0.8\n", "y : ndarray of int, shape (n_items,) | None\n", " For each Epoch, a number indicating the item to which it belongs. When\n", " ``None``, the event codes are used to differentiate between items. Defaults to\n", " ``None``.\n", "n_folds : int | sklearn.model_selection.BaseCrollValidator | None\n", " Number of cross-validation folds to use when computing the distance metric.\n", " Folds are created based on the ``y`` parameter. Specify ``None`` to use the\n", " maximum number of folds possible, given the data. Alternatively, you can pass a\n", " Scikit-Learn cross validator object (e.g. ``sklearn.model_selection.KFold``) to\n", " assert fine-grained control over how folds are created.\n", " Defaults to 1 (no cross-validation).\n", "picks : str | list | slice | None\n", " Channels to include. Slices and lists of integers will be interpreted as channel\n", " indices. In lists, channel *type* strings (e.g., ``['meg', 'eeg']``) will pick\n", " channels of those types, channel *name* strings (e.g., ``['MEG0111',\n", " 'MEG2623']`` will pick the given channels. Can also be the string values \"all\"\n", " to pick all channels, or \"data\" to pick data channels. ``None`` (default) will\n", " pick all MEG and EEG channels, excluding those maked as \"bad\".\n", "tmin : float | None\n", " When set, searchlight patches will only be generated from subsequent time points\n", " starting from this time point. This value is given in seconds. Defaults to\n", " ``None``, in which case patches are generated starting from the first time\n", " point.\n", "tmax : float | None\n", " When set, searchlight patches will only be generated up to and including this\n", " time point. This value is given in seconds. Defaults to ``None``, in which case\n", " patches are generated up to and including the last time point.\n", "dropped_as_nan : bool\n", " When this is set to ``True``, the drop log will be used to inject NaN values in\n", " the RDMs at the locations where a bad epoch was dropped. This is useful to\n", " ensure the dimensions of the RDM are the same, irregardless of any bad epochs\n", " that were dropped. Make sure to use ``ignore_nan=True`` when using RDMs with\n", " NaNs in them during subsequent RSA computations. Defaults to ``False``.\n", "\n", " .. versionadded:: 0.8\n", "n_jobs : int\n", " The number of processes (=number of CPU cores) to use. Specify -1 to use all\n", " available cores. Defaults to 1.\n", "verbose : bool\n", " Whether to display a progress bar. In order for this to work, you need the tqdm\n", " python module installed. Defaults to False.\n", "\n", "Returns\n", "-------\n", "rsa : Evoked | list of Evoked\n", " The correlation values for each searchlight patch. When spatial_radius is set to\n", " None, there will only be one virtual sensor. When temporal_radius is set to\n", " None, there will only be one time point. When multiple models have been\n", " supplied, a list will be returned containing the RSA results for each model.\n", "\n", "See Also\n", "--------\n", "compute_rdm\n", "\u001b[0;31mFile:\u001b[0m ~/micromamba/lib/python3.11/site-packages/mne_rsa/sensor_level.py\n", "\u001b[0;31mType:\u001b[0m function" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from mne_rsa import rsa_epochs\n", "rsa_epochs?" ] }, { "cell_type": "markdown", "id": "7b150008-0cbc-4c3f-81fd-e8ad4bc263de", "metadata": {}, "source": [ "The signature of [`rsa_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rsa_epochs.htm) is very similar to that of [`dsm_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.dsm_epochs.html).\n", "The main difference is that we also give it the \"model\" RDMs, in our case the pixel and FaceNet RDMs.\n", "[`rsa_epochs`](https://users.aalto.fi/~vanvlm1/mne-rsa/functions/mne_rsa.rsa_epochs.htm) will return the RSA scores as a list of `mne.Evoked` objects: one for each model RDM we gave it.\n", "\n", "
\n", "EXERCISE:\n", " \n", "Go ahead and:\n", " - compute the RSA scores for `epochs` gainst `[pixel_rdm, facenet_rdm]`\n", " - do this in a sliding windows across time, with a temporal radius of 0.1 seconds\n", " - optionally set `verbose=True` to activate a progress bar\n", " - optionally set `n_jobs=-1` to use multiple CPU cores to speed things up\n", " - store the result in a variable called `ev_rsa`\n", "
" ] }, { "cell_type": "code", "execution_count": 33, "id": "14ace896-6e73-4489-8fcb-5b583776e401", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Performing RSA between Epochs and 2 model RDM(s)\n", " Temporal radius: 22 samples\n", " Time interval: None-None seconds\n", "Creating temporal searchlight patches\n" ] } ], "source": [ "ev_rsa = rsa_epochs(epochs, [pixel_rdm, facenet_rdm], y=y, temporal_radius=0.1)" ] }, { "cell_type": "markdown", "id": "716b28ba-b8cb-4c20-8435-f72491592b52", "metadata": {}, "source": [ "If you did it correctly, executing the cell below will create a nice plot of the result." ] }, { "cell_type": "code", "execution_count": 34, "id": "d3dd18e0-8d32-4283-9e54-297d092a08d8", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ev_rsa[0].comment = \"pixels\"\n", "ev_rsa[1].comment = \"facenet\"\n", "mne.viz.plot_compare_evokeds(ev_rsa, picks=[0], ylim=dict(misc=[-0.02, 0.2]), show_sensors=False);" ] }, { "cell_type": "markdown", "id": "f54e5e97-187c-46b6-a54d-c1978efb4822", "metadata": {}, "source": [ "If you've made it this far, you have successfully completed your first sensor-level RSA! 🎉\n", "This is the end of this notebook.\n", "I invite you to join me in the [next notebook](fast_source_level.ipynb) where we will do source level RSA.\n", "\n", "
\n", "
\n", ">>>>> Continue to source-level RSA >>>>>\n", "
" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "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": "3.11.11" } }, "nbformat": 4, "nbformat_minor": 5 }