diff --git a/Jupyter notebook/Notebook_Classification_needs_wants.ipynb b/Jupyter notebook/Notebook_Classification_needs_wants.ipynb new file mode 100644 index 0000000..b174cb6 --- /dev/null +++ b/Jupyter notebook/Notebook_Classification_needs_wants.ipynb @@ -0,0 +1,2479 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "fe7ea532", + "metadata": {}, + "source": [ + "Instal and import nimare version 0.0.10rc2 " + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "4b2fb426", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: duecredit in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (0.9.1)\n", + "Requirement already satisfied: six in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from duecredit) (1.16.0)\n", + "Requirement already satisfied: requests in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from duecredit) (2.25.1)\n", + "Requirement already satisfied: citeproc-py>=0.4 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from duecredit) (0.6.0)\n", + "Requirement already satisfied: lxml in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from citeproc-py>=0.4->duecredit) (4.6.3)\n", + "Requirement already satisfied: certifi>=2017.4.17 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->duecredit) (2021.5.30)\n", + "Requirement already satisfied: chardet<5,>=3.0.2 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->duecredit) (4.0.0)\n", + "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->duecredit) (1.26.6)\n", + "Requirement already satisfied: idna<3,>=2.5 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->duecredit) (2.10)\n", + "Note: you may need to restart the kernel to use updated packages.\n" + ] + } + ], + "source": [ + "pip install duecredit" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "106d7a55", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: nimare==0.0.10rc2 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (0.0.10rc2)\n", + "Requirement already satisfied: nilearn>=0.7.1 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (0.8.0)\n", + "Requirement already satisfied: pandas in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (1.3.0)\n", + "Requirement already satisfied: fuzzywuzzy in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (0.18.0)\n", + "Requirement already satisfied: scikit-learn in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (0.24.2)\n", + "Requirement already satisfied: pymare>=0.0.2 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (0.0.2)\n", + "Requirement already satisfied: numpy in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (1.20.3)\n", + "Requirement already satisfied: tqdm in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (4.61.2)\n", + "Requirement already satisfied: cognitiveatlas in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (0.1.9)\n", + "Requirement already satisfied: scipy in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (1.6.2)\n", + "Requirement already satisfied: nibabel>=3.0.0 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (3.2.1)\n", + "Requirement already satisfied: requests in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (2.25.1)\n", + "Requirement already satisfied: statsmodels in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (0.12.2)\n", + "Requirement already satisfied: indexed-gzip>=1.4.0 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nimare==0.0.10rc2) (1.6.1)\n", + "Requirement already satisfied: packaging>=14.3 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nibabel>=3.0.0->nimare==0.0.10rc2) (21.0)\n", + "Requirement already satisfied: joblib>=0.12 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from nilearn>=0.7.1->nimare==0.0.10rc2) (1.0.1)\n", + "Requirement already satisfied: pyparsing>=2.0.2 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from packaging>=14.3->nibabel>=3.0.0->nimare==0.0.10rc2) (2.4.7)\n", + "Requirement already satisfied: python-dateutil>=2.7.3 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from pandas->nimare==0.0.10rc2) (2.8.2)\n", + "Requirement already satisfied: pytz>=2017.3 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from pandas->nimare==0.0.10rc2) (2021.1)\n", + "Requirement already satisfied: sympy in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from pymare>=0.0.2->nimare==0.0.10rc2) (1.8)\n", + "Requirement already satisfied: wrapt in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from pymare>=0.0.2->nimare==0.0.10rc2) (1.12.1)\n", + "Requirement already satisfied: six>=1.5 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from python-dateutil>=2.7.3->pandas->nimare==0.0.10rc2) (1.16.0)\n", + "Requirement already satisfied: idna<3,>=2.5 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->nimare==0.0.10rc2) (2.10)\n", + "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->nimare==0.0.10rc2) (1.26.6)\n", + "Requirement already satisfied: certifi>=2017.4.17 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->nimare==0.0.10rc2) (2021.5.30)\n", + "Requirement already satisfied: chardet<5,>=3.0.2 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from requests->nimare==0.0.10rc2) (4.0.0)\n", + "Requirement already satisfied: threadpoolctl>=2.0.0 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from scikit-learn->nimare==0.0.10rc2) (2.2.0)\n", + "Requirement already satisfied: future in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from cognitiveatlas->nimare==0.0.10rc2) (0.18.2)\n", + "Requirement already satisfied: patsy>=0.5 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from statsmodels->nimare==0.0.10rc2) (0.5.1)\n", + "Requirement already satisfied: mpmath>=0.19 in /Users/laurencegrenier/miniconda3/lib/python3.9/site-packages (from sympy->pymare>=0.0.2->nimare==0.0.10rc2) (1.2.1)\n", + "Note: you may need to restart the kernel to use updated packages.\n" + ] + } + ], + "source": [ + "pip install nimare==0.0.10rc2" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "8f94c9d7", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:numexpr.utils:NumExpr defaulting to 4 threads.\n" + ] + } + ], + "source": [ + "import nimare" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "86129dd5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'0.0.10rc2'" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "nimare.__version__" + ] + }, + { + "cell_type": "markdown", + "id": "2fb68cd6", + "metadata": {}, + "source": [ + "# Import librairies" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "4768e9cb", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "import pandas as pd\n", + "import nimare as nm \n", + "\n", + "from nilearn.image import math_img\n", + "from nilearn.plotting import plot_stat_map\n", + "\n", + "from nimare.correct import FWECorrector\n", + "from nimare.io import convert_sleuth_to_dataset\n", + "from nimare.meta.cbma import ALE, ALESubtraction\n", + "from nimare.tests.utils import get_test_data_path" + ] + }, + { + "cell_type": "markdown", + "id": "619b8020", + "metadata": {}, + "source": [ + "Import files containing the data of the included experiments " + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "6987d544", + "metadata": {}, + "outputs": [], + "source": [ + "need_file = ('/Users/laurencegrenier/Desktop/Fasted-satieted_hunge in response to relevant stimuli - copie (1) - copie.txt')\n", + "\n", + "want_file = ('/Users/laurencegrenier/Desktop/coordonnées désir-wants all (1) - copie.txt')\n" + ] + }, + { + "cell_type": "markdown", + "id": "ebda1465", + "metadata": {}, + "source": [ + "Convert the files into sleuth files so the ALE meta-analysis could be operate " + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "ad815e2b", + "metadata": {}, + "outputs": [], + "source": [ + "need_set = convert_sleuth_to_dataset(need_file)\n", + "\n", + "want_set = convert_sleuth_to_dataset(want_file)" + ] + }, + { + "cell_type": "markdown", + "id": "8862402c", + "metadata": {}, + "source": [ + "Explore the Data" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "8ab3fc41", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Index(['id', 'study_id', 'contrast_id', 'x', 'y', 'z', 'space', 'i', 'j', 'k'], dtype='object')" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "need_set.coordinates.columns" + ] + }, + { + "cell_type": "markdown", + "id": "148d25ff", + "metadata": {}, + "source": [ + "# Generating MA map for each study" + ] + }, + { + "cell_type": "markdown", + "id": "5d573193", + "metadata": {}, + "source": [ + "MA maps of the need_set " + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "17154542", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "data shape (91, 109, 91)\n", + "affine: \n", + "[[ -2. 0. 0. 90.]\n", + " [ 0. 2. 0. -126.]\n", + " [ 0. 0. 2. -72.]\n", + " [ 0. 0. 0. 1.]]\n", + "metadata:\n", + " object, endian='<'\n", + "sizeof_hdr : 348\n", + "data_type : b''\n", + "db_name : b''\n", + "extents : 0\n", + "session_error : 0\n", + "regular : b''\n", + "dim_info : 0\n", + "dim : [ 3 91 109 91 1 1 1 1]\n", + "intent_p1 : 0.0\n", + "intent_p2 : 0.0\n", + "intent_p3 : 0.0\n", + "intent_code : none\n", + "datatype : float64\n", + "bitpix : 64\n", + "slice_start : 0\n", + "pixdim : [-1. 2. 2. 2. 1. 1. 1. 1.]\n", + "vox_offset : 0.0\n", + "scl_slope : nan\n", + "scl_inter : nan\n", + "slice_end : 0\n", + "slice_code : unknown\n", + "xyzt_units : 0\n", + "cal_max : 0.0\n", + "cal_min : 0.0\n", + "slice_duration : 0.0\n", + "toffset : 0.0\n", + "glmax : 0\n", + "glmin : 0\n", + "descrip : b''\n", + "aux_file : b''\n", + "qform_code : unknown\n", + "sform_code : aligned\n", + "quatern_b : 0.0\n", + "quatern_c : 1.0\n", + "quatern_d : 0.0\n", + "qoffset_x : 90.0\n", + "qoffset_y : -126.0\n", + "qoffset_z : -72.0\n", + "srow_x : [-2. 0. 0. 90.]\n", + "srow_y : [ 0. 2. 0. -126.]\n", + "srow_z : [ 0. 0. 2. -72.]\n", + "intent_name : b''\n", + "magic : b'n+1'\n", + "\n", + "data shape (91, 109, 91)\n", + "affine: \n", + "[[ -2. 0. 0. 90.]\n", + " [ 0. 2. 0. -126.]\n", + " [ 0. 0. 2. -72.]\n", + " [ 0. 0. 0. 1.]]\n", + "metadata:\n", + " object, endian='<'\n", + "sizeof_hdr : 348\n", + "data_type : b''\n", + "db_name : b''\n", + "extents : 0\n", + "session_error : 0\n", + "regular : b''\n", + "dim_info : 0\n", + "dim : [ 3 91 109 91 1 1 1 1]\n", + "intent_p1 : 0.0\n", + "intent_p2 : 0.0\n", + "intent_p3 : 0.0\n", + "intent_code : none\n", + "datatype : float64\n", + "bitpix : 64\n", + "slice_start : 0\n", + "pixdim : [-1. 2. 2. 2. 1. 1. 1. 1.]\n", + "vox_offset : 0.0\n", + "scl_slope : nan\n", + "scl_inter : nan\n", + "slice_end : 0\n", + "slice_code : unknown\n", + "xyzt_units : 0\n", + "cal_max : 0.0\n", + "cal_min : 0.0\n", + "slice_duration : 0.0\n", + "toffset : 0.0\n", + "glmax : 0\n", + "glmin : 0\n", + "descrip : b''\n", + "aux_file : b''\n", + "qform_code : unknown\n", + "sform_code : aligned\n", + "quatern_b : 0.0\n", + "quatern_c : 1.0\n", + "quatern_d : 0.0\n", + "qoffset_x : 90.0\n", + "qoffset_y : -126.0\n", + "qoffset_z : -72.0\n", + "srow_x : [-2. 0. 0. 90.]\n", + "srow_y : [ 0. 2. 0. -126.]\n", + "srow_z : [ 0. 0. 2. -72.]\n", + "intent_name : b''\n", + "magic : b'n+1'\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from nimare.meta.kernel import ALEKernel\n", + "\n", + "kernel = nm.meta.kernel.ALEKernel() \n", + "\n", + "need_MA = kernel.transform(need_set, return_type=\"image\")\n", + "\n", + "plot_stat_map(need_MA[2], title=\"MA map\")\n", + "plot_stat_map(need_MA[3], title=\"MA map\")\n", + "\n", + "type(need_MA)\n", + "\n", + "print(need_MA[2])\n", + "\n", + "print(need_MA[3])" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "56249a90", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "nibabel.nifti1.Nifti1Image" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "type(need_MA[3]) #verify it's in the good format" + ] + }, + { + "cell_type": "markdown", + "id": "401d458b", + "metadata": {}, + "source": [ + "MA maps of the want_set" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "1f67af6d", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "kernel = nm.meta.kernel.ALEKernel() \n", + "\n", + "want_MA = kernel.transform(want_set, return_type=\"image\")\n", + "\n", + "plot_stat_map(want_MA[2], title=\"MA map of the third study -want\")\n", + "plot_stat_map(want_MA[3], title=\"MA map of the fourth study -want\")" + ] + }, + { + "cell_type": "markdown", + "id": "f8a664a9", + "metadata": {}, + "source": [ + "# Prepare the data for machine learning\n", + "-Mask to select the grey matter" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "de4860df", + "metadata": {}, + "outputs": [], + "source": [ + "from nilearn.image import smooth_img\n", + "\n", + "from nilearn.datasets import load_mni152_brain_mask\n", + "from nilearn.input_data import NiftiMasker\n", + "\n", + "from nilearn import plotting\n", + "import matplotlib.pyplot as plt\n", + "\n", + "from nilearn.plotting import plot_anat, plot_stat_map, plot_img, plot_roi" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "d5cbac4d", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "nilearn.input_data.nifti_masker.NiftiMasker" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#creating the mask \n", + "mask_img = load_mni152_brain_mask()\n", + "masker = NiftiMasker(mask_img=mask_img, standardize = False, mask_strategy = 'template')\n", + "mask = masker.fit()\n", + "type(mask)" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "4a5dfbc4", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "38\n", + "38\n" + ] + } + ], + "source": [ + "#Apply the mask on all of the studies in the need_set \n", + "print(len(need_MA))\n", + "\n", + "\n", + "need_mask = []\n", + "\n", + "for i in range (len(need_MA)): \n", + " need_img = need_MA[i]\n", + " need_MA_mask = mask.fit_transform(need_img)\n", + " \n", + " need_mask.append(need_MA_mask) \n", + "print(len(need_mask))" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "2bf22fa8", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "38\n", + "33\n" + ] + } + ], + "source": [ + "#Apply the mask on all of the studies in the want_set \n", + "print(len(need_MA))\n", + "\n", + "\n", + "want_mask = []\n", + "\n", + "for i in range (len(want_MA)): \n", + " want_img = want_MA[i]\n", + " want_MA_mask = mask.fit_transform(want_img)\n", + " \n", + " want_mask.append(want_MA_mask) \n", + "print(len(want_mask))" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "c9173407", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "[array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n", + " 1.37026548e-12, 7.86974837e-12, 0.00000000e+00]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[1.05694523e-06, 9.49356996e-07, 6.87956167e-07, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[6.35488070e-11, 4.45746201e-10, 2.41140708e-09, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]]), array([[0., 0., 0., ..., 0., 0., 0.]])]\n" + ] + }, + { + "data": { + "text/plain": [ + "38" + ] + }, + "execution_count": 16, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "print(type(want_mask))\n", + "print(need_mask) \n", + "len(need_mask)" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "bdfe2e8c", + "metadata": {}, + "outputs": [], + "source": [ + "#Fix the 3D problem, convert 3D array of the 38 element of the list into 1D array (concatenate all the dimensions) " + ] + }, + { + "cell_type": "markdown", + "id": "1bcabc98", + "metadata": {}, + "source": [ + "Reducing dimension of the need_mask by transforming each of the need_mask[i] element into an array of 1D " + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "0a31191d", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0.00000000e+00 0.00000000e+00 0.00000000e+00 ... 1.37026548e-12\n", + " 7.86974837e-12 0.00000000e+00]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[1.05694523e-06 9.49356996e-07 6.87956167e-07 ... 0.00000000e+00\n", + " 0.00000000e+00 0.00000000e+00]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[6.35488070e-11 4.45746201e-10 2.41140708e-09 ... 0.00000000e+00\n", + " 0.00000000e+00 0.00000000e+00]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n", + " 1.37026548e-12, 7.86974837e-12, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([1.05694523e-06, 9.49356996e-07, 6.87956167e-07, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([6.35488070e-11, 4.45746201e-10, 2.41140708e-09, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.])]\n" + ] + } + ], + "source": [ + "need_reduce_mask = []\n", + "for i in range (len(need_mask)): \n", + " a = need_mask[i] #create a variable for the array of the item in the need_mask list \n", + " b = a.view() #.view to make sure that the initial data doesn't get modify \n", + " c = b.reshape(228453) #reshape the b which means reshape the need_mask[i] into 1D of 228 453 voxels \n", + " need_reduce_mask.append(c) #create a new array containing the 38 element of need_mask reshape into 1D \n", + " print(c) \n", + "\n", + "print(need_reduce_mask)" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "c187e69b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "1\n", + "228453\n" + ] + } + ], + "source": [ + "print(len(need_mask[1]))\n", + "print(len(need_reduce_mask[1]))\n", + "#check if the reduce dimension worked " + ] + }, + { + "cell_type": "markdown", + "id": "986b4d5d", + "metadata": {}, + "source": [ + "Reducing dimension of the want_mask by transforming each of the need_mask[i] element into an array of 1D" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "80b333f7", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[1.87374494e-05 3.65598204e-05 5.45984360e-05 ... 0.00000000e+00\n", + " 0.00000000e+00 0.00000000e+00]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[0. 0. 0. ... 0. 0. 0.]\n", + "[array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([1.87374494e-05, 3.65598204e-05, 5.45984360e-05, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.])]\n" + ] + } + ], + "source": [ + "want_reduce_mask = []\n", + "for i in range (len(want_mask)): \n", + " a = want_mask[i] #create a variable for the array of the item in the need_mask list \n", + " b = a.view() #.view to make sure that the initial data doesn't get modify \n", + " c = b.reshape(228453) #reshape the b which means reshape the need_mask[i] into 1D of 228 453 voxels \n", + " want_reduce_mask.append(c) #create a new array containing the 38 element of need_mask reshape into 1D \n", + " print(c) \n", + "\n", + "print(want_reduce_mask)" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "c7fac058", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "228453" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "len(want_reduce_mask[1])" + ] + }, + { + "cell_type": "markdown", + "id": "94453fee", + "metadata": {}, + "source": [ + "Creating the variables of the model \n", + "X = features (71)\n", + "y = condition (need/want) " + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "86acaf41", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0.0\n", + "71\n" + ] + } + ], + "source": [ + "#Making the X variable of the model, concatenante the need_reduce_mask and the want_reduce_mask \n", + "\n", + "X = need_reduce_mask + want_reduce_mask \n", + "#print(X)\n", + "print(X[1][228433])\n", + "\n", + "print(len(X))" + ] + }, + { + "cell_type": "markdown", + "id": "c34c74d0", + "metadata": {}, + "source": [ + "Creating the variable Y : 0 = needs, 1 = wantings " + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "488fa2d8", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n", + "[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]\n" + ] + }, + { + "data": { + "text/plain": [ + "71" + ] + }, + "execution_count": 24, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "y = []\n", + "for i in range (38): \n", + " y_i = 0 \n", + " y.append(y_i)\n", + "print(y)\n", + "len(y)\n", + "\n", + "for i in range (33): \n", + " y_i = 1 \n", + " y.append(y_i)\n", + "print(y) \n", + "len(y)\n", + " " + ] + }, + { + "cell_type": "markdown", + "id": "289af3bd", + "metadata": {}, + "source": [ + "# Creating the machine learning model " + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "d0bf8fbe", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n", + " 1.37026548e-12, 7.86974837e-12, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([1.05694523e-06, 9.49356996e-07, 6.87956167e-07, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([6.35488070e-11, 4.45746201e-10, 2.41140708e-09, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([1.87374494e-05, 3.65598204e-05, 5.45984360e-05, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.])]\n", + "[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]\n" + ] + }, + { + "data": { + "text/plain": [ + "list" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Make sure that X and y are ok \n", + "print(X)\n", + "print(y)\n", + "type(X)" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "4567efdb", + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "np.save(\"test_data/X.npy\", X)\n", + "np.save(\"test_data/y.npy\", y)" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "id": "4f72b755", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "MA_plt_mask = mask.inverse_transform(X[0])\n", + "MA_mask_map = plot_stat_map(MA_plt_mask)\n", + "plt.show()\n", + "\n", + "#Make sure that the operations went well, retrieve the data of the map by using X " + ] + }, + { + "cell_type": "markdown", + "id": "c9642564", + "metadata": {}, + "source": [ + "Separating the data into training data and test (validation) data " + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "id": "8434d86d", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "training: 53 testing: 18\n", + "[array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([6.35488070e-11, 4.45746201e-10, 2.41140708e-09, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([1.87374494e-05, 3.65598204e-05, 5.45984360e-05, ...,\n", + " 0.00000000e+00, 0.00000000e+00, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n", + " 1.37026548e-12, 7.86974837e-12, 0.00000000e+00]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.]), array([0., 0., 0., ..., 0., 0., 0.])]\n" + ] + } + ], + "source": [ + "#split the sample to training/test set \n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "\n", + "X_train1, X_test, y_train1, y_test = train_test_split(\n", + " X, # x\n", + " y, # y\n", + " test_size = 0.25, # 75%25% split \n", + " random_state = 123, #same shuffle each time \n", + " shuffle = True, # shuffle dataset before splitting # before splitting\n", + " stratify = y,# keep\n", + " # distribution\n", + " # of conditions\n", + " # consistent\n", + " # betw. train\n", + " # & test sets.\n", + " )\n", + "\n", + "# print the size of our training and test (validation) groups\n", + "print('training:', len(X_train1),\n", + " 'testing:', len(X_test))\n", + "print (X_train1)" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "bb8744f7", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "training: 31 testing: 22\n" + ] + } + ], + "source": [ + "#Split into training/validation set \n", + "X_train2, X_val, y_train2, y_val = train_test_split(\n", + " X_train1, # x\n", + " y_train1, # y\n", + " test_size = 0.4, # 60%/40% split \n", + " random_state = 123, #same shuffle each time \n", + " shuffle = True, # shuffle dataset before splitting \n", + " stratify = y_train1,# keep\n", + " # distribution\n", + " # of conditions\n", + " # consistent\n", + " # betw. train\n", + " # & test sets.\n", + " )\n", + "\n", + "# print the size of our training and test (validation) groups\n", + "print('training:', len(X_train2),\n", + " 'testing:', len(X_val))" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "79be64d9", + "metadata": {}, + "outputs": [], + "source": [ + "# 71 = 31 (training) + 22 (validation) + 18 (testing)" + ] + }, + { + "cell_type": "markdown", + "id": "04cfb97e", + "metadata": {}, + "source": [ + "Let's create the model " + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "ddbb6057", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "SVC(kernel='linear')" + ] + }, + "execution_count": 31, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from sklearn.svm import SVC\n", + "\n", + "l_svc = SVC(kernel='linear') # define the model\n", + "\n", + "l_svc.fit(X_train2, y_train2) # fit the model" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "id": "712c642a", + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.metrics import mean_absolute_error\n", + "\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import seaborn as sns" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "abe77c15", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "accuracy = 0.5454545454545454\n", + "MAE = 0.45454545454545453\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/laurencegrenier/miniconda3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n", + " warnings.warn(\n" + ] + }, + { + "data": { + "text/plain": [ + "Text(0.5, 0, 'Predicted Conditions')" + ] + }, + "execution_count": 33, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# predict the *validation* data based on the model trained on X_train2\n", + "y_pred = l_svc.predict(X_val) \n", + "\n", + "# caluclate the model accuracy\n", + "acc = l_svc.score(X_val, y_val) \n", + "mae = mean_absolute_error(y_true=y_val,y_pred=y_pred)\n", + "\n", + "# print results\n", + "print('accuracy = ', acc)\n", + "print('MAE = ',mae)\n", + "\n", + "sns.regplot(y_pred,y_val)\n", + "plt.xlabel('Predicted Conditions')" + ] + }, + { + "cell_type": "markdown", + "id": "d1e8086e", + "metadata": {}, + "source": [ + "Confusion matrix " + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "54c979f6", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[[12 0]\n", + " [10 0]]\n" + ] + } + ], + "source": [ + "from sklearn.metrics import confusion_matrix\n", + "\n", + "matrix_con_val1 = confusion_matrix(y_true = y_val, y_pred = y_pred) \n", + "\n", + "print(matrix_con_val1)" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "id": "38a17e22", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "from pandas import DataFrame\n", + "\n", + "cmdf1 = DataFrame(matrix_con_val1, index = ['Needs','Wants'], columns = ['Needs','Wants'])\n", + "\n", + "import seaborn as sns\n", + "\n", + "ax_val1 = sns.heatmap(cmdf1, cmap = \"black_blue_r\", vmin = 0, vmax = 14, annot = True) #cmap = 'red_transparent_full_alpha_range', center = 8, annot = True)\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Observed')\n", + "ax_val1.figure.savefig(\"Val_Confusion_matrice_1.png\")" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "id": "282ed6f4", + "metadata": {}, + "outputs": [], + "source": [ + "#Not good...maybe because most of the values are 0 and the non-zero values are still near to 0... \n", + "#Try to change the scale and see \n", + "#exp = np.exp(10*x)-1 " + ] + }, + { + "cell_type": "markdown", + "id": "6293a13f", + "metadata": {}, + "source": [ + "Transform the X data into exponential " + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "id": "76589eeb", + "metadata": {}, + "outputs": [], + "source": [ + "X = np.array(X)" + ] + }, + { + "cell_type": "code", + "execution_count": 38, + "id": "09e5bd2e", + "metadata": {}, + "outputs": [], + "source": [ + "X_exp = np.exp(np.array(X)*100)-1" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "id": "9cb93702", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[0., 0., 0., ..., 0., 0., 0.],\n", + " [0., 0., 0., ..., 0., 0., 0.],\n", + " [0., 0., 0., ..., 0., 0., 0.],\n", + " ...,\n", + " [0., 0., 0., ..., 0., 0., 0.],\n", + " [0., 0., 0., ..., 0., 0., 0.],\n", + " [0., 0., 0., ..., 0., 0., 0.]])" + ] + }, + "execution_count": 39, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "X_exp" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "602f3e3d", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([0., 0., 0., ..., 0., 0., 0.])" + ] + }, + "execution_count": 40, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "X_exp[1]" + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "id": "976c1a2b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "training: 53 testing: 18\n", + "[[0. 0. 0. ... 0. 0. 0.]\n", + " [0. 0. 0. ... 0. 0. 0.]\n", + " [0. 0. 0. ... 0. 0. 0.]\n", + " ...\n", + " [0. 0. 0. ... 0. 0. 0.]\n", + " [0. 0. 0. ... 0. 0. 0.]\n", + " [0. 0. 0. ... 0. 0. 0.]]\n" + ] + } + ], + "source": [ + "#split the sample to training/test set...again...\n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "\n", + "X_train1, X_test, y_train1, y_test = train_test_split(\n", + " X_exp, # x\n", + " y, # y\n", + " test_size = 0.25, # 75%25% split \n", + " random_state = 123, #same shuffle each time \n", + " shuffle = True, # shuffle dataset before splitting # before splitting\n", + " stratify = y,# keep\n", + " # distribution\n", + " # of conditions\n", + " # consistent\n", + " # betw. train\n", + " # & test sets.\n", + " )\n", + "\n", + "# print the size of our training and test (validation) groups\n", + "print('training:', len(X_train1),\n", + " 'testing:', len(X_test))\n", + "print (X_train1)" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "id": "4b05cc99", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "training: 31 testing: 22\n" + ] + } + ], + "source": [ + "#Split into training/validation set \n", + "X_train2, X_val, y_train2, y_val = train_test_split(\n", + " X_train1, # x\n", + " y_train1, # y\n", + " test_size = 0.4, # 60%/40% split \n", + " random_state = 123, #same shuffle each time \n", + " shuffle = True, # shuffle dataset before splitting \n", + " stratify = y_train1,# keep\n", + " # distribution\n", + " # of conditions\n", + " # consistent\n", + " # betw. train\n", + " # & test sets.\n", + " )\n", + "\n", + "# print the size of our training and test (validation) groups\n", + "print('training:', len(X_train2),\n", + " 'testing:', len(X_val))" + ] + }, + { + "cell_type": "code", + "execution_count": 43, + "id": "a43e3712", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "SVC(kernel='linear')" + ] + }, + "execution_count": 43, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from sklearn.svm import SVC\n", + "\n", + "l_svc = SVC(kernel='linear') # define the model\n", + "\n", + "l_svc.fit(X_train2, y_train2) # fit the model" + ] + }, + { + "cell_type": "code", + "execution_count": 44, + "id": "387bbec5", + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.metrics import mean_absolute_error\n", + "\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import seaborn as sns" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "id": "c4077e62", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "accuracy = 0.8181818181818182\n", + "MAE = 0.18181818181818182\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/laurencegrenier/miniconda3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n", + " warnings.warn(\n" + ] + }, + { + "data": { + "text/plain": [ + "Text(0.5, 0, 'Predicted Conditions')" + ] + }, + "execution_count": 45, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# predict the *validation* data based on the model trained on X_train2\n", + "y_pred = l_svc.predict(X_val) \n", + "\n", + "# caluclate the model accuracy\n", + "acc = l_svc.score(X_val, y_val) \n", + "mae = mean_absolute_error(y_true=y_val,y_pred=y_pred)\n", + "\n", + "# print results\n", + "print('accuracy = ', acc)\n", + "print('MAE = ',mae)\n", + "\n", + "sns.regplot(y_pred,y_val)\n", + "plt.xlabel('Predicted Conditions')" + ] + }, + { + "cell_type": "code", + "execution_count": 46, + "id": "b58de8a7", + "metadata": {}, + "outputs": [], + "source": [ + "matrix_con_val2 = confusion_matrix(y_true = y_val, y_pred = y_pred) " + ] + }, + { + "cell_type": "code", + "execution_count": 47, + "id": "0eb80d9c", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[[12 0]\n", + " [ 4 6]]\n" + ] + } + ], + "source": [ + "print(matrix_con_val2)" + ] + }, + { + "cell_type": "code", + "execution_count": 48, + "id": "2f921398", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "from pandas import DataFrame\n", + "cmdf2 = DataFrame(matrix_con_val2, index = ['Needs','Wants'], columns = ['Needs','Wants'])\n", + "\n", + "import seaborn as sns\n", + "\n", + "ax_val2 = sns.heatmap(cmdf2, cmap = \"black_blue_r\", vmin = 0, vmax = 14, annot = True) #cmap = 'red_transparent_full_alpha_range', center = 8, annot = True)\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Observed')\n", + "ax_val2.figure.savefig(\"Val_Confusion_matrice_2.png\")" + ] + }, + { + "cell_type": "markdown", + "id": "4c328048", + "metadata": {}, + "source": [ + "Retrieve the features which contributes the most to the classification " + ] + }, + { + "cell_type": "code", + "execution_count": 49, + "id": "fb5a4a8e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n", + " -1.78683124e-12, -1.02621773e-11, 0.00000000e+00]])" + ] + }, + "execution_count": 49, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "l_svc.coef_" + ] + }, + { + "cell_type": "code", + "execution_count": 50, + "id": "d44131c0", + "metadata": {}, + "outputs": [], + "source": [ + "coef = np.abs(l_svc.coef_)" + ] + }, + { + "cell_type": "code", + "execution_count": 51, + "id": "456eb9c1", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n", + " 1.78683124e-12, 1.02621773e-11, 0.00000000e+00]])" + ] + }, + "execution_count": 51, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "coef" + ] + }, + { + "cell_type": "code", + "execution_count": 52, + "id": "a463c6e0", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "228453" + ] + }, + "execution_count": 52, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "len(coef[0]) #chaque voxel a un coeficient de corrélation " + ] + }, + { + "cell_type": "code", + "execution_count": 53, + "id": "40e364e7", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "0.0" + ] + }, + "execution_count": 53, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "coef[0][1]" + ] + }, + { + "cell_type": "code", + "execution_count": 54, + "id": "e677c0ef", + "metadata": {}, + "outputs": [], + "source": [ + "essai = mask.inverse_transform(coef)" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "id": "34cc5b52", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "essai_map = plot_stat_map(essai)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 56, + "id": "b1af71f3", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "essai_map = plot_stat_map(essai, threshold=0.001)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 57, + "id": "31b2a3ed", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "" + ], + "text/plain": [ + "" + ] + }, + "execution_count": 57, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Save the image to show in the presentation \n", + "from nilearn import plotting\n", + "from nilearn.plotting import view_img\n", + "\n", + "feature_img = plotting.view_img(essai, threshold=0.001) \n", + "feature_img" + ] + }, + { + "cell_type": "code", + "execution_count": 58, + "id": "f8f19b54", + "metadata": {}, + "outputs": [], + "source": [ + "#Save the image \n", + "feature_img.save_as_html(\"Validation Model coeficient features image.html\")\n", + "essai_map.savefig('Validation Model coeficient features map.png')" + ] + }, + { + "cell_type": "markdown", + "id": "e8a842c6", + "metadata": {}, + "source": [ + "Test the model on test set " + ] + }, + { + "cell_type": "code", + "execution_count": 59, + "id": "e93a6da3", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "SVC(kernel='linear')" + ] + }, + "execution_count": 59, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "l_svc.fit(X_train1,y_train1)" + ] + }, + { + "cell_type": "code", + "execution_count": 60, + "id": "93b0943a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "accuracy = 0.8333333333333334\n", + "MAE = 0.16666666666666666\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/laurencegrenier/miniconda3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n", + " warnings.warn(\n" + ] + }, + { + "data": { + "text/plain": [ + "Text(0.5, 0, 'Predicted Conditions')" + ] + }, + "execution_count": 60, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAEGCAYAAABmXi5tAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAApsUlEQVR4nO3deZRc5Xnn8e9T3VXqfdXe1UIsYhGgrUvgBWyMjVlss0si9mTxic14xk4mmTPnJCeZk0wmf4xzMhnHPnaCCXEcZzGS2AmyseMYgw0YdQuhDQNCLF3d2gUS6r2q3vnj3uq+aqrVpVZV1/b7nKND161bVW9fqp6+dX/3fa455xARkfIXKvQARERkdqjgi4hUCBV8EZEKoYIvIlIhVPBFRCpEdaFeeO7cuW7p0qWFenkRkZLU09NzxDk3byaPLVjBX7p0Kd3d3YV6eRGRkmRmb830sTqkIyJSIVTwRUQqhAq+iEiFUMEXEakQKvgiIhVi2oJvZt8xs0NmtmuK+83MvmFme81sh5mtyeaFd/Yd5/w/2sLv37/tTMcsIlJxbvjaUyz9wyeILLyga6bPkc0e/neBG05z/43AMv/f3cDfZvviyZTj4e37VfRFRE7jhq89xa8ODpz180xb8J1zTwPHTrPKLcD3nOd5oMXMFmXz4mbefx/bcSCb1UVEKlIuij3k5hh+B9AbuB33l72Pmd1tZt1m1p0cPD6+PJlST34RkcmGx5L88/Mznmf1PrmYaWsZlmWs4M65e4F7AeYsWja+TlUo01OIiFSmoydH+Mdn3+Sfnn+LdwbHcva8uSj4caAzcDsK9GfzwPTFtm5esTAHwxARKW17D77Ht5/ex6Mv9TOaSAEQrjLqwlUcH06c9fPnouA/BnzFzO4HrgSOO+f2Z/PAqpBx84qFfO2urE7sEREpO6OJFM+9foS//8WbPPPq4fHDI0011fynD5zDb31oKfObanIS3Np017Q1s+8D1wBzgYPAnwJhAOfcPWZmwDfxzuQZBD7vnJu2K1osFnNqniYilWh4LMmJoTF+vOcg//LLt9mz/8T4fYuaa/jCVedy1xVLqJ/z/n1yM+txzsVm8rrT7uE7535tmvsd8OWZvLiISCVwzjE0lmRgJMmxgRG27DzAAz1x+t4dGl/n4oWN3P2R87h55WKqq/IzJ7Zg7ZFFRMpZKuUYGE0wOJpkaNQr9I9s7+fR7f0cH5oIYj94XjtfuOpcPnbxPEKh/DY/UMEXEcmRRDLFwGiSwdEEw2MpnHP0HhvkgZ44T+45eEoQ+4lLFvCbHzqHtUvbZ+1MRRV8EZGzMJJIMjiSZGA0MV7QnXPs7j/Bxq29PPv60fEgtrGmmptXLmbD2k4uXNBITbhqVseqgi8icoaGx5IMjHiHa8aSqfHlyZTjF3uPsKm7lz373xtfvrCphju7onxqxUI6WutoqgkXYtgq+CIi0wmGroOjifd1BxgeS/Lk7gNs7onT/+7w+PKLFjayvivKRy6cR2t9hLa6CKECTjRVwRcRySCZcgwGQtdUhlPY3xkc5ZEX+3h0ez8nAhOjPnheO+vXRlnR0UxtpJr2hghzqmf38E0mKvgiIr5MoWsmb6eD2N0HGEt664SrjE8uX8idXR2c015PdShEW0OEhgzn0hdK8YxERKQAMoWumTjn2NV3gk3dpwaxTTXV3LxqMbeu6qCtPoKZ0VwbpqU2XNDDN5mo4ItIxZkqdM1kqiB2UXMN67qiXH/ZQmr9s23q/MM34TxNnDpbKvgiUvaccwyOenvxQ6PJrFqyD48l+eEuL4jdf3wiiL14YSMb1nZy1QVzx8+fD1eFaG+IUBcp7pJa3KMTEZmhYOg6OJqc8nj8ZMcGRnl0e+YgdsPaKJd3NGP+1ZvMjNa6MM214fFlxUwFX0TKxlgyxeBIksExb0/+TLx9dJDNPXF+tOf9Qey6rihL2utOWb9hTjVt9ZG89b3JBxV8ESlp2YaumTjn2Nl3nI1b4zy37+j48slBbFCkOkR7/RxqI4U/zfJMqeCLSMkZ8k+dzCZ0zSSZcvzcD2JfniaITQuZ0Vofobm2MLNkc0EFX0SK3kxC10yG/CD2gUlB7CWLGtkQ6+TDgSA2qLEmTFt9pOQvx6qCLyJFaaahaybHBkZ5+MU+Hnupn/f8INaAD53fzvpYJ5d1NGUMXeeEq2ivj8x6k7N8UcEXkaKRDl0HRhMMj51Z6JrJW0cH2NwT58d7Dp4SxF5/6ULu7IqypK0u4+OqQkZbfYTGAjU5yxcVfBEpqLMJXTNxzrGj7zgbt/by/L5j48ubaqq5dVUHt6xeTGtdJONjzYymmmpaC9zkLF9U8EVk1p1t6JpJMuV45rXDbOyO88qBiSB2cUsNd66JcsNlC097aKY2UkV7/Rwi1aVzmuWZUsEXkbzLVeiaydBokh/s2s+D2/pOCWKXL2pk/dpOPnx+5iA2rRibnOVL+f+GIlIQ6dB1YCTJ0NjZha6ZTBnEXtDOhlgnl3U0n/bx6SZnrXWlMUs2F1TwRSRnch26ZvLW0QE2d8f58csTQWykOsT1yxdwZ1eUzimC2KBib3KWLyr4InJWch26ZuKcY0f8OBu7MwSxqzu4ZdXUQWxQqTQ5y5fK/K1FZMaccwyPpRgYTTA4kiSRyk+RB++w0NOvHmZTd5xXDk4EsR0ttayLRfnk8gVZnSMfMqOlhJqc5YsKvohMK5Xyr+mah9A1k3QQ+0BPHwdOTASxly5uYl0sOm0QG1SKTc7yRQVfRDJKptz4Xnw+QtdMjp4c8YPY/ZwcmQhiP3zBXNbHotMGsUGR6hBzG+aUzSzZXFDBF5FxsxG6ZvLm0QE2bY3zk1+dGsTecKl3jdho6/RBbFpVyGipK+0mZ/migi9S4YbHvF41AyOJnE2CyoZzjpfi3ozYX74xEcQ214a5ZdVibl21mJYsgtigcmlyli8q+CIVZjZD10zSQezG7l5ePXhyfHk6iL1++QLmnOFhmHJrcpYvKvgiFSCVcgyOJRn0L9ydmoXj8ZMNjSbZsms/D/TEOXhiZHz5ZYubWB/r5IPnt5/xnnl1KERrfbjsmpzliwq+SJkqROiayemC2A1ro1y6OPsgNq3cm5zlS1YF38xuAL4OVAH3Oee+Oun+ZuCfgSX+c/5f59w/5HisIjKNQoWumbxxxJsR++8vHyThn8Y5ZzyIjdLRWjuj562EJmf5Mm3BN7Mq4FvAdUAc2Gpmjznn9gRW+zKwxzn3GTObB7xiZv/inBvNy6hFZFyhQtdMnHNs732Xjd1xXggEsS21YW5b3cHNKxfTXDezwy/hqhBt9RHqK6DJWb5ks+WuAPY65/YBmNn9wC1AsOA7oNG8KWwNwDEgkeOxigiFD10zSaYcT71ymE3dvbx2aCKIjbbWsj4W5bpLzjyITTMzWmrDtFRQk7N8yabgdwC9gdtx4MpJ63wTeAzoBxqBDc65970Lzexu4G6AJUuWzGS8IhWpGELXTAZHE2zZ6V0j9tB7E0Hs5R0TQWzoLIp0vT9LttKanOVLNgU/0/+tye+264HtwLXA+cCPzewZ59yJUx7k3L3AvQCxWKw43rEiRapYQtdMjpwc4aFtfTy+o5+BES8rMODqZXNZH+tk+eKms3r+cJU3S7Y2otMscymbgh8HOgO3o3h78kGfB77qvHfkXjN7A7gYeCEnoxSpEKOJlNdDfjTJSIFD10zeODLApu5efvLyoVOD2Mv8ILZlZkFsWsiM1roITbXVOnyTB9kU/K3AMjM7F+gD7gI+O2mdt4GPA8+Y2QLgImBfLgcqUq6KKXTNxDnHi2+/y6buXl54853x5bkIYoMaaqppq1OTs3yatuA75xJm9hXgSbzTMr/jnNttZl/y778H+HPgu2a2E++b3R84547kcdwiJcs5v/PkSJKh0eIIXTNJJFP87FXvGrF7JwWx67q81sQzDWKD1ORs9mR1fpNzbguwZdKyewI/9wOfzO3QRMpHsYaumQyOJnhi5wEefF8Q28z6WPSsg9i0qpDRWh+hSbNkZ41OaBXJk0Qy5Rf54gtdMzn8njcjNhjEhgyuWjaX9V1nH8QGqclZYajgi+RQsYeumew7fJLNPfFTgtia6hDX5yiIDaoJV9HeEGFOtQ7fFIIKvshZKvbQNRPnHNv8IHZrIIhtrQtzazqIzWE/eTU5Kw4q+CJnqFRC10wSyRRPvXqYTVvj7D08EcR2ttayLtbJJ5cvyGmPGjU5Ky4q+CJZKKXQNZOBkQRbdu7nwW19pwSxK6LNrOvKXRAbVBfxZsmqyVnxUMEXmUIimWJgNMngaILhsVTRh66ZjAexL/UzMDoRxF69bB7rY1EuWZS7IDZNTc6Kl/6PiASUYuiayeuHT7K5O85PfnWIZCCIvfHyRdyxpoPFOQxi09TkrPip4EvFGx7zAtfB0WTJhK6ZpIPYjVt76X4r/0FskJqclQYVfKk4pRy6ZpJIpvip35r49cMD48uXtNWxPhblE5fkNogNUpOz0qKCLxWh1EPXTAZGEjyxcz8P9vRx+OSpQeyGWCdXnteW8yA2TU3OSpMKvpStcghdMzl0YpiHXuzjiR37TwliP3rhPNbFoly8MPdBbJCanJUuFXwpK+USumby+qGTbOzu5aevHJ4IYsMhbrpsEXd0dbCoOfdBbNCccBXt9RE1OSthKvhS8soldM3EOUfPW++wsTtOz6Qg9vY1HXxmxWKa8hTEpqnJWflQwZeSEwxdB0cT43u75WQsEMTuCwSx5/hB7MfzGMQGNdWGaa1Tk7NyoYIvJSHlX+5vaDRZNqFrJidHEjyxYz8Pbotz5OTo+PKV0WY2rO3kinPzF8QGqclZeVLBl6JVrqFrJodODPPgtj6e2LmfwUlB7PpYJxctbJyVcVSHQrQ1RGjQLNmypP+rUlRGEt658SdHEowmyut4fCZ7D51kU6Yg9vJF3LkmysLmmlkZh5nRXBumpTasJmdlTAVfCq6cQ9dMnHN0v/UOm7b20vP2u+PL2+oj3L66g8+sXDSrbYTV5KxyqODLrKuE0DWTsWSKn/7qEJu64+w7Eghi2+tYH+vk4xfPn9WiG64K0d4QoS6iMlAp9H9aZkUy5Rgc9fbih8o4dM3k5EiCf9uxn4cmBbGrOltYH4vOWhCbZma01oVprlWTs0qjgi95U0mhayYHTwzz0BRB7Ia1nVy4YHaC2KAGv8mZZslWJhV8yamRhHfR7oHRyghdM3nt4Hts6o7z01cOkT5aVRMO8anLF3HHLAaxQWpyJqCCLzlQaaFrJs45tr75Dpu6e9kWCGLb6yPcvqaDT6+Y3SA2TU3OJEgFX86Yc867aLc/EapSQtdMxpIp/sMPYt8IBLFL/SD22lkOYoMaa8K01WuWrExQwZesBEPXwdFkxR2Pn+zkcILHd/Tz0It9HJ0UxG5YG+WKpW0F26NWkzOZigq+TGksmWJwJMngmLcnL3DgxDAPbYvzxI4DDI1NBLHXXDSf9bFoQYLYNDU5k+mo4MspFLpm9trB99jYHeepQBBbG67iUysWcvuaKAubZj+ITTMzGv0e9ZolK6ejgi8KXafgnOOFN4+xqTvOi5OC2Nv8a8Q21BT2I1QbqaKtXk3OJDsq+BVIoevpjSa8IHZzz6lB7Llz61kfi3LtxfMLfrFuNTmTmdC7pUIodJ3eyeEEj73Uz8Mv9nF0YCKIXbOkhfWxTtYubS34qY3pJmetdZolK2cuq4JvZjcAXweqgPucc1/NsM41wF8DYeCIc+6jORulzEg6dB0YTTBcZpf7y6UDJ4Z5sCfOlp3FF8QG1UWqaW+IFPzbhZSuaQu+mVUB3wKuA+LAVjN7zDm3J7BOC/A3wA3OubfNbH6exivTUOiavVcPvsfGrb387NXDpwSxn16xiNvXdLCggEFskJqcSa5k8w66AtjrnNsHYGb3A7cAewLrfBZ4yDn3NoBz7lCuBypTG/L71Sh0nV7KOV54wwtit/e+O768vSHCHas7+PSKwgexaSEzWtTkTHIom3d2B9AbuB0Hrpy0zoVA2MyeAhqBrzvnvjf5iczsbuBugCVLlsxkvIJC15kYTaT4ycsH2dQT562jg+PLiymIDVKTM8mHbAp+pl2LyRWmGugCPg7UAs+Z2fPOuVdPeZBz9wL3AsRiMVWpM5AOXQdGkgyNKXTN1nvDYzz+0n4eerGPY4EgtmtJC+vXdhI7p/BBbFCk2mtyplmykg/ZFPw40Bm4HQX6M6xzxDk3AAyY2dPASuBVZMYUus7cgePDPLAtzpad+xke8w5zVYWMj13kXSP2gvkNBR7hqapCRktdhOZazZKV/Mmm4G8FlpnZuUAfcBfeMfugR4Fvmlk1EME75PO1XA60Uih0PTuZgti6SJXfmriD+UUSxAapyZnMlmkLvnMuYWZfAZ7EOy3zO8653Wb2Jf/+e5xzL5vZD4EdQArv1M1d+Rx4uXDOMTyWYmA0weBIkkRKRf5MTQSxvWzvPT6+fG5DhDvWRPnUikVFOUFJTc5ktlmhjgXHYjHX3d1dkNcutFTKv6arQtezMlUQe968etZ3RflYkQWxaVUho60+UpD++FL6zKzHORebyWOLb7enTCVTbnwvXqHr2TkxNMbjO/p5+MX+U4PYc1rZEIvSVWRBbJqZ0VRTTauanEmBqODnkULX3Np/fIgHevr4wc79DCcmgthrL57P+q4o5xdZEBtUG6mivX5OwS6GIgIq+Dk3POb1qhkYSWgSVI786sAJNm2N8/Rrpwaxn16xiNtXF2cQm1Yd8mbJ1hdhhiCVR+/Cs6TQNT9SzvHLfcfY2N3LjnjpBLFpZkZLbZgWNTmTIlK8n5gilko5BseSDPo95FM6Hp8zo4kU//7yQTZ3x3nr2EQQe/68etbHOrnmonlFGcQG1fuzZIt9nFJ5VPCzpNA1v04MjY23Jn5ncGx8eeycVtYXcRAbpCZnUuz0zjwNha751//uEA9ue38Q+/GL57MuFuX8ecUbxKaFzGiti9BUW130f5SksqngT6LQdXa8vP8Em7rjPBMIYusjVXxm5WJuW93BvMY5hR1glhr8a8mqyZmUgoov+ApdZ0/KOZ7fd5SNW+Ps7JsIYuc3zuH2NR186vJFJXM2i5qcSSkqjU9Xjil0nV2jiRQ/3nOQzT1x3g4EsRfMa2D92ijXXDivZPaQ1eRMSlnFFHyFrrPvuB/EPjIpiF27tJUNsU5WL2kpqWPeanImpa6sC/5oIjV+JSiFrrOn/90hNvfE+eGuA4z4QWy1PyO2VILYoJpwFe0NEeZU6/CNlLayK/gKXQvn5f0n2Njdy89fO1LSQWxadShEa31YTc6kbJR8wXfO6zw5OJpU6FoAKed47vWjbOp+fxB7R1eUmy5bWDJBbJqanEm5Kq1Pok+ha+GNJlL8aM9BNnf30vvO0PjyC+Y3sCEW5aMlFMQGqcmZlLOSKfiJZMov8gpdC+n4oB/Ebj81iL3i3DbWx6Ks7iytIDYtXBWirV5NzqS8FfW7Ox26DowmGVHoWlB97w7xQHecH+4+NYj9+CXzWdcV5bwSC2LT1ORMKknRFXyFrsVlT/8JNnX38sxrR0h/p6qfU8VnVizm9jUdzG0orSA2SE3OpNIUvOCnQ9eBkSRDowpdi0E6iN24tZdd/SfGl89vnMOdXVFuunxhSTcIC1d5s2RrIzrNUipLwT61Kec4dGJYoWsRGRlLekFsT5x4GQWxaWpyJpWuYAU/kXKcHEkU6uUl4PjgGI++1McjL/bz7tCpQeyGWJRVJRrEBqnJmUgRHNKRwul7Z4gHet4fxH7ikgWsi0U5d259gUd49tTkTGSCCn4F2tM/MSM2fTCtYU41n1m5iNtWl3YQm1YVMlrrIzRplqzIOBX8CpFMpWfEnhrELmjygtgbLyvtIDaoqTZMa52anIlMVh6fcJnSyFiSJ/cc5IFJQeyFCxrYEOvkIxfOK5vCqCZnIqengl+m3h0c5ZHt/Ty6vZ/jgSD2A+e1sT7Wycpoc8kHsWlqciaSHRX8MhN/Z5DNPXGe3H2Q0UAQe93yBdzZVR5BbJqZ0VwbpqU2rCZnIllQwS8Tu/uPs3FrnF/sPTWIvdkPYtvLIIgNqot4s2TV5Ewkeyr4JSyZcjzrB7G7A0HswqYa7ujq4KbLFpXdbFI1OROZOX1qStCwPyO2EoLYNDOjtS5Mc62anInMlAp+CTldELsh1smKMgpigxr8JmeaJStydrIq+GZ2A/B1oAq4zzn31SnWWws8D2xwzj2Qs1FWuN5jgzzQE+fJPRNBbLjKuO6SBdwZi7K0vXyC2CA1ORPJrWkLvplVAd8CrgPiwFYze8w5tyfDen8BPJmPgVYa5xy7/Rmxz+49Oh7ENtZUc7N/jdi2+khBx5gvanImkh/Z7OFfAex1zu0DMLP7gVuAPZPW+x3gQWBtTkdYYZIpxy/2HmFTdy979r83vnxRcw13dkW54bKF1JZxX5iGmmra6+eUXQYhUgyyKfgdQG/gdhy4MriCmXUAtwHXcpqCb2Z3A3cDdEQ7z3SsZW14LMmTuw+wuSdO/7vD48svWtjIhliUq5eVXxAbNCdcRXt9RE3ORPIom4KfqcpMbmD/18AfOOeSp/sK7py7F7gXYMXqNWqCD7wzOMqjL3rXiD0xPNEu+oPntbN+bZQVHeUZxKapyZnI7Mmm4MeB4O54FOiftE4MuN8vTHOBm8ws4Zx7JBeDLEdvp4PY3QcYS3p/+8JV3ozYdV1RzinTIDaoqTZMW11Es2RFZkk2BX8rsMzMzgX6gLuAzwZXcM6dm/7ZzL4L/JuK/fs559jV510j9tnXJ4LYpppqPlPmQWyQmpyJFMa0Bd85lzCzr+CdfVMFfMc5t9vMvuTff0+ex1jyKj2ITasOhWhriNCgWbIiBZHVJ885twXYMmlZxkLvnPutsx9WeRgaS/LkLi+I3X98Ioi9eGEj62OdXL1sblkHsWlqciZSHLSrlQfHBkZ5dHsfj27vPyWI/dD57ayPRbm8zIPYoLpINe0NEcKaJStScCr4OfT2Ua818Y/2nBrEfnL5QtZ1RVnSXlfgEc6ecFWI9oZI2VxFS6Qc6NN4lpxz7OzzWhM/t+/o+PKmmmpuXrWYW1dVRhCbFjKjRU3ORIqSCv4MJVOOn+89wsatvfzqQOUGsUFqciZS3FTwz9DpgtgNazu56oLKCGKDItUh2uvV5Eyk2KngZ+nYwCiPbO/jsUAQa6SD2E4u62iquEMYIfNmyTbXapasSClQwZ/GaYPYWJQlbZUTxAY11oRpq49U3LcZkVKmgp+Bc44dfcfZuLWX5/cdG1/eVFPNras6uGX1YlrrKieIDVKTM5HSpYIfkEw5nnntMBu747wSCGIXt9SwrivK9ZcurNhCpyZnIqVPBR8viP3BzgM8uO3UIPaSRY1siHXy4QoMYtPMjKaaalrV5Eyk5FV0wT82MMrDL/bx2Ev9vBcMYi9oZ0Osk0sXV14QG1QbqaK9fg6Rap1mKVIOKrLgv3V0gM09cX685+ApQewNly7kjq7KDWLT1ORMpDxVzCfaOceO+HE2diuInUq6yVlrnWbJipSjsi/4yZTj6VcPs6k7zisHJ4LYjpZa7uyKcv2lCyo2iA2q92fJqsmZSPkq24I/NJrkB7v280BPHwdOTASxyxc1sX5tlA+fX7lBbJCanIlUjrL7lB89OeIHsfs5OfL+IPayjubCDrBIqMmZSOUpm4L/5tEBNnfH+feXJ4LYSHWI6y/1rhEbba3sIDZITc5EKlNJF3znHC/Fj7NpUhDbXBvmlpWLFcROEqkOMbdhjjILkQpVkgU/mXL87NXDbOru5dWDJ8eXd7TUsi4W5ZPLFcQGVYWMljo1OROpdCVV8AdHE/xg1wEe6Ilz8MTI+PJLFzexPtbJh85vVxA7iZqciUhaSRT8oydHeOjFPh6fFMRetWwu67qiCmIzUJMzEZmsqAv+G0cmgthEaiKIveFS7xqxHa21BR5h8akKGW31ERrV5ExEJim6gu+cY3vvu2zsjvPCGxNBbEttmFtXL+aWlR0016mYTaYmZyIynaIp+MmU46lXvCD2tUMTQWy0tZZ1XV4QO0eHJzJSkzMRyUbBC/7gaIItO70g9tB7E0HsZYub2LC2kw+e305IE4MyCleFaKuPUK8mZyKShYJVikTSce/T+3h8Rz8DI0nAC2KvXjaXdbEoly5WEDsVM6OlNkyLmpyJyBkoWMHfd/gk92/tBWCOH8TeqSB2WmpyJiIzVbCC7/CC2NtWd3DzysUKYqcRrvJmydZGlGOIyMwUrOAvaKrh+1+8UkHsNEJmtNZFaKqt1uEbETkrBSv4LXVhFftpNNRU01anJmcikhtZVRIzu8HMXjGzvWb2hxnu/5yZ7fD/PWtmK3M/1MoRqQ6xuKWW+Y01KvYikjPT7uGbWRXwLeA6IA5sNbPHnHN7Aqu9AXzUOfeOmd0I3AtcmY8Bl7OqkNFaH6FJs2RFJA+yOaRzBbDXObcPwMzuB24Bxgu+c+7ZwPrPA9FcDrISqMmZiORbNgW/A+gN3I5z+r333wZ+kOkOM7sbuBugI9qZ5RDLW024ivaGCHOqlWeISH5lU/Az7XK6jCuafQyv4F+V6X7n3L14h3tYsXpNxueoFNWhEK31YTU5E5FZk03BjwPB3fEo0D95JTNbAdwH3OicO5qb4ZUfNTkTkULJpuBvBZaZ2blAH3AX8NngCma2BHgI+HXn3Ks5H2WZUJMzESmkaQu+cy5hZl8BngSqgO8453ab2Zf8++8B/gRoB/7GnxyUcM7F8jfs0qImZyJSDLKqQM65LcCWScvuCfz8BeALuR1a6VOTMxEpJtrlzBM1ORORYqOCn2NqciYixUoFP0fU5ExEip0Kfg401FTTXj9Hs2RFpKip4J+FOeEq2usj1Kjrp4iUABX8GVCTMxEpRSr4Z6ipNkxrnZqciUjpUcHPkpqciUipU8GfRnUoRFtDhAbNkhWREqcqNgUzo7k2TEttWE3ORKQsqOBnUBfxZsmqyZmIlBMV/IBwVYj2hgh1EW0WESk/qmx4h29a68I016rJmYiUr4ov+A1+k7NqNTkTkTJXsQU/Uh2ivV5NzkSkclRcwVeTMxGpVBVV8BtrwrTVa5asiFSmiij4anImIlLmBV9NzkREJpRtwW+qDdNWF9EsWRERX9kV/NpIFW31anImIjJZ2RR8NTkTETm9kq+O6SZnrXWaJSsicjolXfDrItW0N0QIa5asiMi0SrLgq8mZiMiZK6mKGTKjRU3ORERmpGQKvpqciYicnaIv+JHqEHMb5miWrIjIWSragl8VMlrqIjTXapasiEguFGXBV5MzEZHcK6qCryZnIiL5k1UCamY3mNkrZrbXzP4ww/1mZt/w799hZmume85XDrzHtX/1M677q59RFTLmNc6ho6VWxV5EJE+mLfhmVgV8C7gRWA78mpktn7TajcAy/9/dwN9mO4AkcO1fPkWjOlqKiORVNnv4VwB7nXP7nHOjwP3ALZPWuQX4nvM8D7SY2aJsB5HMergiIjJT2RT8DqA3cDvuLzvTdTCzu82s28y6k4PHz3SsIiJyFrIp+JlOlXEzWAfn3L3OuZhzLlZV15zN+EREJEeyKfhxoDNwOwr0z2CdKVXr7EsRkbzLpuBvBZaZ2blmFgHuAh6btM5jwG/4Z+t8ADjunNufzQCqDfb+n0+d0aBFROTMTXsevnMuYWZfAZ4EqoDvOOd2m9mX/PvvAbYANwF7gUHg89M97+UdzXR/VYVeRGS2ZDXxyjm3Ba+oB5fdE/jZAV/O7dBERCSX1HpSRKRCqOCLiFQIFXwRkQqhgi8iUiHMy1sL8MJm7wGvFOTFi89c4EihB1EktC0maFtM0LaYcJFzrnEmDyxke+RXnHOxAr5+0TCzbm0Lj7bFBG2LCdoWE8yse6aP1SEdEZEKoYIvIlIhClnw7y3gaxcbbYsJ2hYTtC0maFtMmPG2KFhoKyIis0uHdEREKoQKvohIhch7wc/HBdBLVRbb4nP+NthhZs+a2cpCjHM2TLctAuutNbOkmd05m+ObTdlsCzO7xsy2m9luM/vZbI9xtmTxGWk2s8fN7CV/W0zbmbcUmdl3zOyQme2a4v6Z1U3nXN7+4bVTfh04D4gALwHLJ61zE/ADvKtmfQD4ZT7HVKh/WW6LDwGt/s83VvK2CKz3H3idWu8s9LgL+L5oAfYAS/zb8ws97gJuiz8C/sL/eR5wDIgUeux52BYfAdYAu6a4f0Z1M997+Hm/AHoJmXZbOOeedc694998Hu/KYeUom/cFwO8ADwKHZnNwsyybbfFZ4CHn3NsAzrly3R7ZbAsHNJqZAQ14BT8xu8PMP+fc03i/21RmVDfzXfBzdgH0MnCmv+dv4/0FL0fTbgsz6wBuA+6hvGXzvrgQaDWzp8ysx8x+Y9ZGN7uy2RbfBC7Bu4TqTuC/OedSszO8ojKjupnv1go5uwB6Gcj69zSzj+EV/KvyOqLCyWZb/DXwB865pLczV7ay2RbVQBfwcaAWeM7MnnfOvZrvwc2ybLbF9cB24FrgfODHZvaMc+5EnsdWbGZUN/Nd8PN+AfQSktXvaWYrgPuAG51zR2dpbLMtm20RA+73i/1c4CYzSzjnHpmVEc6ebD8jR5xzA8CAmT0NrATKreBnsy0+D3zVeQey95rZG8DFwAuzM8SiMaO6me9DOnm9AHqJmXZbmNkS4CHg18tw7y1o2m3hnDvXObfUObcUeAD4r2VY7CG7z8ijwNVmVm1mdcCVwMuzPM7ZkM22eBvvmw5mtgC4CNg3q6MsDjOqm3ndw3d5ugB6KcpyW/wJ0A78jb9nm3Bl2CEwy21REbLZFs65l83sh8AOIAXc55zLeLpeKcvyffHnwHfNbCfeYY0/cM6VXdtkM/s+cA0w18ziwJ8CYTi7uqnWCiIiFUIzbUVEKoQKvohIhVDBFxGpECr4IiIVQgVfRKRCqOBLzvhdLbeb2S4z2+yfMz7T5/puukOmmd1nZstPs+41ZvahGbzGm2Y2N8PyBjP7tpm97ndkfNrMrjzT58/wvEvT3Q/NLGZm38g0fjP7Uhm3T5ACyvdMW6ksQ865VQBm9i/Al4D/l77TzKqcc8kzfVLn3BemWeUa4CTw7Jk+9xTuA94AljnnUmZ2Hl7/lpxxznUD3f7NawiMv5LmIcjs0h6+5MszwAX+3utPzexfgZ1mVmVmf2lmW/0+3v8Zxvt7f9PM9pjZE8D89BP5TcNi/s83mNk2vx/6T8xsKd4flt/3v11cbWbzzOxB/zW2mtmH/ce2m9mPzOxFM/s2GfqRmNn5eDNZ/2e6KZffvfEJ//7/7n+D2WVmv+cvW2pmL5vZ3/nfCH5kZrX+fV3+WJ8Dvhx4nWvM7N+mGP//MrP/4a+3ysye97fVw2bWGtgmf2FmL5jZq2Z2tb/8Un/Zdv8xy3Lzv1PKgQq+5JyZVeP189/pL7oC+GPn3HK8pnDHnXNrgbXAF83sXLzOmBcBlwNfxLs2wOTnnQf8HXCHc24lsM459yZeR82vOedWOeeeAb7u314L3IG3xw7ebMWfO+dW401NX5Jh+JcC2zN9EzGzLrwZjVfi9SD/opmt9u9eBnzLOXcp8K7/ugD/APyuc+6DmbbVFOMP+h7ebNIVeNvzTwP3VTvnrgB+L7D8S8DX/W9aMbyeKyKADulIbtWa2Xb/52eAv8cr3C84597wl38SWGETV7BqxiuWHwG+7xfafjP7jwzP/wHg6fRzOeem6hf+CWC5TXTZbDKzRv81bvcf+4SZvTPF46dyFfCw38QMM3sIuBrvj8cbzrn0794DLDWzZqDFOZe+QtU/4f0hzEqGx/8jsDmwykPB1/N/fg74YzOL4vXQfy37X0/KnQq+5NL4Mfw0v+gOBBcBv+Oce3LSejcxfXtXy2Id8L65ftA5N5RhLNM9fjew0sxCGfqsn65P80jg5yReG+NsxztT6ddM4n+WnXP/ama/BD4FPGlmX3DOZfrjKRVIh3Rktj0J/BczCwOY2YVmVg88DdzlH+NfBHwsw2OfAz7qHwLCzNr85e8BjYH1fgR8JX3DzFb5Pz4NfM5fdiPQOvkFnHOv44Wpf2b+XwgzW2Zmt/iPv9XM6vwx34b3TSYj59y7wHEzS1/X4HNTrDp5/OnHHwfeSR+fB34dOO31bP2AeZ9z7ht43zxWnG59qSwq+DLb7sO7Pus2/xTFb+PtnT4MvIZ3nPpvyVDYnHOHgbuBh8zsJWCjf9fjwG3p0BP4XSDmh5Z78I5rA/wZ8BEz24Z3aOntKcb4BWAhXr/1nXi5Qb9zbhvwXbze67/E61r54jS/7+eBb/mh7dAU60wef9BvAn9pZjuAVcD/nub1NgC7/ENrF+NlACKAumWKiFQM7eGLiFQIFXwRkQqhgi8iUiFU8EVEKoQKvohIhVDBFxGpECr4IiIV4v8Dds+5RvyqAuIAAAAASUVORK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# predict the *validation* data based on the model trained on X_train2\n", + "y_pred = l_svc.predict(X_test) \n", + "\n", + "# caluclate the model accuracy\n", + "acc = l_svc.score(X_test, y_test) \n", + "mae = mean_absolute_error(y_true=y_test,y_pred=y_pred)\n", + "\n", + "# print results\n", + "print('accuracy = ', acc)\n", + "print('MAE = ',mae)\n", + "\n", + "sns.regplot(y_pred,y_test)\n", + "plt.xlabel('Predicted Conditions')" + ] + }, + { + "cell_type": "code", + "execution_count": 61, + "id": "6cafdd1c", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEGCAYAAAB7DNKzAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAuNklEQVR4nO3deXxc9Xnv8c8z0oy1r941MjZgFhu8aQxJWEIggCEBs1kmIW1vbhPKvaVN29e9Tbrc9ia999WkuW0amoUSQmjaNHjBgAkkzkogAYIl23gjgDGLRvK+W5Ilzei5f5wz0rEsjcaSzqzP+/Xyy5ozZ0Y/HVvzzJzv+T0/UVWMMcaYkQQyPQBjjDHZzQqFMcaYpKxQGGOMScoKhTHGmKSsUBhjjEmqONMDOFuTJ0/W2bNnZ3oYxhiTU1pbWw+q6pSxPDbnCsXs2bNpaWnJ9DCMMSaniMi7Y32snXoyxhiTlBUKY4wxSVmhMMYYk5QVCmOMMUlZoTDGGJOUb4VCRB4Rkf0isn2E+0VEHhCRXSKyVUSWpPK829qPcd5fPsufPrZpYgdsjDF5aNlXnmP2554hNP38prE+h5+fKB4FliW5/yZgrvvnXuCbqT5xvF95YsseKxbGGJPEsq88x2/3dY77eXwrFKr6PHA4yS7Lge+q42WgRkRmpPLcIs7f67fuHe8wjTEmb01EkYDMZhQNQJvndtTddgYRuVdEWkSkJd51bGB7vN/W0jDGmKFO9cX5j5fHPL/uDJmcmS3DbBv2lV9VHwIeApg0Y+7APkWB4Z7CGGMK06GTPTz64jv8x8vvcqSrb8KeN5OFIgo0em6HgY5UHphYlO/WBdMnfFDGGJNLVJXX953gW8/v5umte+iN9QMQLBLKgkUcOxUb9/fIZKFYD9wvIo8BlwPHVHVPKg8sCgi3LpjOV+5O6UIpY4zJK/F+pas3xm92H+bfXnqHX715cOB0TFVJMZ943zn8lw/MZmpVyYQE2uLXmtki8n3gGmAysA/4WyAIoKoPiogAX8O5MqoL+KSqjtrtLxKJqDUFNMYUmp5YnO7eOCdOxfj5b/ezamMbO/ccH7h/RnUJn7pyDndfNovySWd+BhCRVlWNjOV7+/aJQlU/Nsr9CvyhX9/fGGNymarS3Rens8cpECd7+tiwYx9rW6O0H+0e2O+i6ZXce/W53LpwJsVF/lyflHNtxo0xJl/F4v109jqFobsvjqpytKuXJ7d08NSWDo51DwbU7z+3nk9dOYcPXTSFQMDfC1itUBhjTAad6ovT1Runqzc2EEQDtB3uYm1rlA07950WUH/44mn83gfOYens+rRd+WmFwhhj0qi/X+nqcwpDd2/8tPlgqsqOjuOs2tjGi28dGgioK0uKuXXhTFYubeSCaZWUBIvSOmYrFMYY47PeWD/dvXE6e2P0xPoZehFRvF/59a6DrG5pY+eeEwPbp1eVcFdTmI8smE5DbRlVJcF0Dx2wQmGMMRMuEUR3uXlDX7x/2P1O9cXZsGMva1qjdBw9NbD9wumVNDeFufqCKdSWh6grCxHI4ARjKxTGGDMBYvF+uvrcILo3Tn+SqQdHunp5cnM7T23p4LhnQtz7z62neWmYBQ3VlIaKqa8IMak4vaeZhmOFwhhjxuiUWxg6hwTRI3kvEVDv2Etf3CkkwSLhhnnTuaupgXPqyykOBKirCFExzFyITMmekRhjTJbr73fnNgwTRI9EVdnefpzVLacH1FUlxdy6aCa3LWqgrjyEiFBdGqSmNJjR00zDsUJhjDFJJILorr4Yp/rODKJHMlJAPaO6hBVNYW68ZDql7tVLZe5ppqBPE+bGywqFMcZ4qCqn+vrp6o3RlSSIHsmpvjg/2u4E1HuODQbUF02vZOXSRq48f/LA/IdgUYD6ihBloex+Kc7u0RljTBokmux19zpXKiULokdyuLOXp7YMH1CvXBrm0oZqxF11TUSoLQtSXRoc2JbNrFAYYwpSTyxOV0+crr44PX3xMT/Pe4e6WNMa5cc7zwyoVzSFmVVfdtr+FZOKqSsP+daXyQ9WKIwxBSERRCfmNsT6z+6Ukpeqsq39GKs2Rnlp96GB7UMDaq9QcYD68kmUhjJ/uevZskJhjMlbffH+gcKQaLI3HvF+5VduQP3aKAF1QkCE2vIQ1aWZmVU9EaxQGGPyhqrSE3OKQ2dP7KyD6JF0uwH12iEB9cUzKlkZaeQKT0DtVVkSpK48lPPLNluhMMbkNG8Q3d2X2tyGVB3u7OWJze2sf7WDE25ALcAHzqunOdLIJQ1Vw4bRk4JF1JeH0t68zy9WKIwxOSex2ltXb5xT4wiiR/LuoU7WtEb5yc59pwXUN86fzl1NYWbVlQ37uKKAUFceojJDzfv8YoXCGJP1Um2yN97vsbX9GKs2tvHy7sMD26tKirltUQPLF8+ktiw07GNFhKqSYmoz3LzPL1YojDFZKdFkr6tnYoLokcT7lRfePMCqliiv7x0MqGfWlHDXkjDLLpme9BRSaaiI+vJJhIpz53LXs2WFwhiTNUZa7c0P3b1xfrh9D49vaj8toJ43o5LmpY1ccd7wAXVCNjbv80v+/4TGmKyVbLU3v4wYUJ9fz8pII5c0VCd9fKJ5X21ZbsyqnghWKIwxaTXWJnvj9e6hTta0RPnJa4MBdag4wI3zpnFXU5jGEQJqr2xv3ucXKxTGGF+Nt8neeL/31ugxVrUME1AvbmD5opEDaq9cad7nl8L8qY0xvjqb1d78EO9Xnn/jAKtbory+bzCgbqgpZUUkzA3zpqU0xyEgQk0ONe/zixUKY8yESKz2Nt4me+ORCKjXtraz9/hgQD1/ZhUrIuFRA2qvXGze5xcrFMaYMZnIJnvjdehkjxtQ7+Fkz2BAfcX5k2mOhEcNqL1CxQEmV0zKm1nVE8EKhTEmZX3xfrc1d3qD6JG8c6iT1Ruj/Oy3pwfUy+Y7a1CHa0cPqBOKAkJNWW437/OLFQpjzIgyGUQnG9OrUWcG9W/eHgyoq0uDLF80k9sWzaQmhYDaK1+a9/nFCoUx5jQTsdqbHxIB9aqWNt7Yd3JgeyKgvnHeNCad5emifGve5xcrFMaYgSZ7nb2ZC6JH0t0b59nte1jbGmXf8Z6B7ZfMrKI50sj7z6s/608CxYEAteXBvGve5xcrFMYUoESTvc6ezAfRI0kWUK9cGmb+zNQD6oR8b97nF18LhYgsA74KFAEPq+oXh9xfDfwHMMsdy/9T1e/4OSZjClUs3k/nBK725pe3DzozqH/62j5ibkuPSQMBdZiG2tIxPW8hNO/zi2+FQkSKgK8D1wNRYKOIrFfVnZ7d/hDYqaq3iMgU4HUR+Z6q9vo1LmMKSTqb7I2HqrKl7SirWqK84gmoa0qD3L64gVsXzqS6bGyniYJFAerKQ5QXQPM+v/h55C4DdqnqbgAReQxYDngLhQKV4kx5rAAOAzEfx2RMXosPzG1IX5O98Yj3K8+9foDVLW28uX8woA7XltIcCXP9xWcfUCeICDWlQWoKqHmfX/wsFA1Am+d2FLh8yD5fA9YDHUAlsFJVz3jbIyL3AvcCzJo1y5fBGpOremODl6/6sdqbH7p6Yzy7zVmDev+JwYD60obBgDowjhf3cndWdaE17/OLn4ViuH/loW9vbgS2ANcC5wE/EZEXVPX4aQ9SfQh4CCASiWT3WyRjfJaO1d78cvBkD+s2tfP01g46e5yiJsBVcyfTHGlk3syqcT1/sMiZVV0asstdJ5KfhSIKNHpuh3E+OXh9EviiOqnaLhF5G7gIeMXHcRmTczLdZG+83j7YyeqWNn722v7TA+pL3IC6ZmwBdUJAhNqyEFWlxXaayQd+FoqNwFwRmQO0A3cDHx+yz3vAdcALIjINuBDY7eOYjMkZuRJEj0RV2fzeUVa3tPHKO0cGtk9EQO1VUVJMXZk17/OTb4VCVWMicj+wAefy2EdUdYeI3Ofe/yDwd8CjIrIN5xPoZ1X1oF9jMiabZWK1Nz/E4v388g1nDepdQwLqFU1Oi++xBtRe1rwvfXy9XkxVnwWeHbLtQc/XHcANfo7BmGyWqdXe/NDVG+OZbXt5/IyAuprmSHjcAXVCUUCoLQ9RZbOq08YuLDYmjbKxyd54HTjhzKD2BtQBgSvnTqa5afwBtZc178sMKxTG+CzRZK8rR4Pokew+cJI1rdHTAuqS4gA3TlBA7VUSLKK+IsSkYjvNlAlWKIzxQU8s7q7bkH1N9sZDVdnkBtQbPQF1bVmQ2xIB9QSu52DN+7KDFQpjJkA2rfbmh1i8n+feOMDqjVF2HRgMqBtrS1kRaeSGedMmtIeSNe/LLlYojBmjvnj/QGHI5iZ749HZE+PZbXt4fFP7aQH1gnA1K5omLqD2Kgs5s6qteV/2sEJhTIpUlZ5YP509+RNEj2QgoH61g87ewYD6qrlTaI6EuXjGxAXUCda8L3vZv4gxSXhXe+vuy925Dal668BJ1rRE+dlv9w/8rCXFAW66dAZ3Lmlg5gQG1AnWvC/7WaEwZojEam+51GRvPBIB9aqNbbS8639A7WXN+3KDFQpT8HK5yd54xOL9/MJt8f3Wgc6B7bPqymiOhPnwxRMbUHtZ877cYoXCFKREk72unvwNokfS2RPjmW17eLy1nQMnTw+oV0YaufzcugkPqBOseV9uskJhCkauN9kbr/3HT7FuczvPbN1zWkD9wQumsCIS5qLpEx9Qe1nzvtxlhcLkrXxpsjdeb+0/yaqWNn7x+oHBgDoY4OZLZnBnUwMzqic+oPaaFCyivjxkzftymBUKk1fyqcneeKgqre8eYVVLlNYhAfUdSxq4ZcFMqnwKqBOseV/+sEJhclqiyV6n+6mhUILokfR5AurdnoD6HDegvs7HgNqrqjRIbZk178sXVihMzsn11d78cLInxjNb9/D4pigHT/YObF8Yrmbl0kYum+NfQO1lzfvykxUKkxNOuYUh35rsjdf+46d4fFM7z2zbQ9eQgLo50siF0yvTMo7iQIC6ihAVNqs6L436ryoiVwBbVLVTRD4BLAG+qqrv+j46U7ASTfY6e2Oc6u3PuyZ747Vr/0lWDxdQXzqDu5aEmV5dkpZxiAjVpUFqSoPWvC+PpVL+vwksFJGFwJ8D3wa+C3zQz4GZwtMX73dbcxd2ED0SVaXl3SOs3thG63tHB7bXlYe4Y3EDtyyckdZ23Na8r3CkUihiqqoishznk8S3ReT3/B6YyX/5uNqbH/ri/fzit/tZ3RJl90FPQF1fRnOkkesumprWF+tgUYD6ihBlITvNVChS+Zc+ISJ/AXwCuFpEigC73s2MibfJXpcF0Umd7Inxg617WDckoF7UWENzJJy2gDpBRKgtC1Jdas37Ck0qhWIl8HHg91V1r4jMAr7s77BMPkk02evstSA6FfuOn2LdCAH1yqWNXDAtPQG1V4XbvM9mVRemUQuFqu4F/slz+z2cjMKYYSWa7HX25Odqb355c98JVrdE+cXr+0lMIi8JBvjIpTO4M40BtZc17zOQ2lVPdwBfAqYC4v5RVfW3MYzJKYWw2psfVJWN7xxhdUsbmzwBdX15iDuWNPDRBekNqBOseZ/xSuXU0z8At6jqa34PxuSWQm+yNx598X5+7gbUb3sC6tluQH1tmgNqr8qSIHXlNqvaDEqlUOyzImHACaK7++J09cQKYrU3P5w8FePprR2s29zOoSEB9cqlYS6bXZexd/DWvM+MJJVC0SIiq4AngYHm9aq6zq9BmezRGxu8fLUQVnvzy97jp1i3KcozW/fS3TcYUF9z4VSaI+GMBNQJ1rzPjCaVQlEFdAE3eLYpYIUiDxXqam9+eXPfCVa1RHnOE1CXBov4yILp3LEkzPSq9AfUCSJCpbtGhM2qNsmkctXTJ9MxEJM5hbzamx9UlVfeOczqliibhwTUt7trUFeUZHayWmmoiLpya95nUpPKVU9h4F+AK3A+SfwK+IyqRn0em/GRBdETrzfmBNRrWk8PqOdMLqc5Eubai6YSzPA8BGveZ8Yilf8t3wH+E1jh3v6Eu+16vwZlJp6t9uafk6dirH+1gyc2t3OoczCgXjKrhuZII0tn12b8EtNE877aMptVbc5eKoViiqp+x3P7URH5k1SeXESWAV8FioCHVfWLw+xzDfDPOG1BDqqqNRucILbam7/2Hj/F461Rnt2WfQG1V1momPqKUMY/zZjclUqhOOi2F/++e/tjwKHRHuT2hPo6ziePKLBRRNar6k7PPjXAN4BlqvqeiEw9y/EbD2uylx5v7DvBqo1t/PKNA6cF1B9dMIM7ljQwLYMBtZc17zMTJZX/Qf8V+BrwFZyM4kV322guA3ap6m4AEXkMWA7s9OzzcWCd2xYEVd2f+tANDDbZ67LV3nzVr8orbzsB9Za2owPb6ytC3Lm4gY8uyHxAnRAQocaa95kJlMpVT+8Bt47huRuANs/tKHD5kH0uAIIi8hxQidPG/Iw+UiJyL3AvwKxZs8YwlPxiq72lT2+sn5+9to/VrVHePdQ1sD2bAmova95n/DBioRCRP1fVfxCRf8H5JHEaVf3jUZ57uLcyQ5+nGGgCrgNKgZdE5GVVfWPI93oIeAggEokU3FvmxGpviU8N1mTPfydO9fH0q3tYt7mdw56AumlWDc1LG4mck/mA2itU7DTvs1nVxg/JPlEk2na0jPG5o0Cj53YY6Bhmn4Oq2gl0isjzwELgDQpcosleV68F0em099gp1m6K8uy2PZzqcwpyUUD40IXOGtTnT63I8AhPVxQQaspCVJfarGrjnxELhao+7X7ZpaprvPeJyIphHjLURmCuiMwB2oG7cTIJr6eAr4lIMRDCOTX1lRTHnldUlZ5YP509FkRnwnABdVmoyG3x3cDULAmovax5n0mXVNK3vwDWpLDtNKoaE5H7gQ04l8c+oqo7ROQ+9/4HVfU1EfkRsBXox7mEdvvZ/hC5ylZ7y6zBgLqNLW3HBrZPrghx55IwH1kwIysnplnzPpNuyTKKm4CbgQYRecBzVxUQS+XJVfVZ4Nkh2x4ccvvLFNCKeYnV3qzJXuaMFFCfO6Wc5qYwH8qygDqhKCDUlYcysj6FKWzJ3i514OQTtwKtnu0ngD/1c1D5xFZ7yx7Hu/t4emsHT2zuOD2gPqeWlZEwTVkWUCeICFUlxdRa8z6TIckyileBV0Xke6qa0icI44jF++m01d6yxp5j3axtbeeH2/ZwKjYYUF970VSam8Kcl2UBtVdpqIj68kkZW8TIGEh+6mm1qjYDm0VkuMtjF/g6shxjTfayz2/3Hmf1xijPv3l6QP3RBTO4Y3F2BtQJxQFnVnV5FmYkpvAk+1/4Gffvj6ZjILnGmuxlp35VfrP7MKta2tgazZ2AOkFEqCkNUmPN+0wWSXbqaY/797vpG052SzTZ6+yN0ROzuQ3ZpDfWz09f28ealijvHh4MqM+bUk5zpJFrLpySlQG1V7k7qzrbx2kKT7JTTycYZkZ2gqpW+TKiLGKrvWW/4919Ay2+j3T1DWyPnFNLcxYH1F7WvM9ku2SfKCoBROQLwF7g33HactyD05cpLyVWe+u2JntZreNoN49vOjOgvu6iqayIhDlvSvYG1AkBEWrLQlSVFmd9MTOFLZW3MDeqqreZ3zdF5DfAP/g0prRLNNnrtCA667225zirW6K84Amoy0NF3LJwJrcvbmBK5aTMDjBFFe5a1da8z+SCVApFXETuAR7DORX1MSCnZ4olmux1WhCdE/pVeXn3IVZtjLKtfTCgnlo5iTuWNPCRS2fkzNVB1rzP5KJUfrs+jrNK3VdxCsWvObNnU9bri/fT1WOrveWS3lg/P9m5jzWtUd7zBNTnT6mgeWmYay6YkjPvyK15n8llqaxH8Q7OgkM5xVZ7y13H3ID6ySEB9dLZtayMNLJ4Vk1OndO35n0m141aKERkCvBpYLZ3f1VNZZW7tLIme7mt42g3a1qj/Gj7XnrcrKjYnUGdKwG1V0mwiPqKEJOK7TSTyW2pnHp6CngB+ClZmE0kmux19tpqb7nqtT3HWdXSxq/ePJjTAXVCcSBAbXnQmveZvJFKoShT1c/6PpKzkFizwZrs5a5+VV566xCrW84MqO9sCnPzJdNzJqBOsOZ9Jl+l8pv4AxG52W0ZnnF98X72HT+V6WGYMeqN9fPjnftY09JG25Huge3nT61gZSTMB3MooPay5n0mn6VSKD4D/KWI9AKJZFEzNTPbUofcdKzLDai3nB5QXzanjuZImMWNuRVQJwSLAtSVW/M+k99Sueopb2dhG/+1H+1mbUuUH+04PaC+7uKprGgKc26OBdQJ1rzPFJKU3gaJyK3A1e7N51T1B/4NyeSDnR3HWd3SxgtvHhz4FFg+qYhbFszkjiUNTK7IrYDay5r3mUKTyuWxXwSWAt9zN31GRK5U1c/5OjKTcxIB9aqNbWzvOD6wfWrlJO5qCnPzpdNzuvFdsMiZVV0asstdTWFJ5bf2ZmCRqvYDiMi/AZsBKxQGgJ6+uBNQt0aJ5lFAnWDN+0yhS/XtXQ1w2P262p+hmFxzrKuPp15t58nNHRztPj2gXhkJsyhHA2ova95nTGqF4u9xlkP9BU6b8auBv/B1VCartR/pZm3rmQH1hy+exopImDmTyzM8wvGz5n3GDErlqqfvi8hzODmFAJ9V1b1+D8xkn50dgzOoEwF1xaRiblk4g9sX53ZAnVAUEGrLQ1TZrGpjBiRb4e5GoFJV17rLoq53t98jIvtV9SfpGqTJnHh/Ygb16QH1tConoL7pktwOqL2qSoPUllnzPmOGSvYb/nnglmG2/wx4ArBCkcd6+uJs2LmPtUMC6gumVbAy0sjVF0zJmxdUa95nTHLJCkWZqh4YulFV94pI7p+ENsM62tXLk1s6eGpLB8c8AfX7zq2jOdLIwnB1zgfUCda8z5jUJCsUJSJSrKox70YRCQKl/g7LpFv0SBdrWqNs2LFvYDnY4oBw/bxp3NWUHwF1gohQXRqkpjRozfuMSUGyQrEO+JaI3K+qnQDuJ4kH3PtMHtjRcYxVG6P8etfpAfWtbkBdnwcBtVdZyJlVbc37jEldskLx18D/Ad4VkXfdbbOAbwP/y++BGf/E+5UX3YB6hyegnl5Vwp1NDdx8yYy8m31szfuMGbsRf2vcU06fE5HPA+e7m3epavdIjzHZ7ZQ7g7oQAuoEEaG2LEh1qTXvM2asUplH0Q1sS8NYjE+SBdQrI40syKOA2qvCbd5ns6qNGR9fP4eLyDLgq0AR8LCqfnGE/ZYCLwMrVXWtn2MqJG2Hu1jbGmXDzsGAOlgkXH/xNO6KhJldnz8BtZc17zNmYvlWKESkCPg6cD0QBTaKyHpV3TnMfl8CNvg1lkKiquxwZ1C/uOvQQEBdWVLMre4a1HXloYyO0S/WvM8Yf6TSZlyAe4BzVfULIjILmK6qr4zy0MtwMo3d7vM8BiwHdg7Z74+Ax3FahJgxivcrv951kNUtbezcc2Jg+4zqEu5qCrPskumU5nHfooqSYurLJ+VdxmJMNkjlE8U3gH7gWuALwAlSe2FvANo8t6PA5d4dRKQBuN197hGfT0TuBe4FaAg3pjDkwnGqL86GHXtZ0xql4+jgWuIXTq9kZSTMVXPzL6D2mhQsor48ZM37jPFRKoXiclVdIiKbAVT1iIikcu5iuFenoUte/zNOk8F4slMFqvoQ8BDAgsVLbNls4EhXL09tdtagPn5qcE7k+8+tp3lpmAUN+RlQJ1jzPmPSJ5VC0efmCAogIlNwPmGMJgp43/6HgY4h+0SAx9wXtMnAzSISU9UnU3j+gvReIqDesZe+uFMzg0XODOoVTWHOydOA2quqNEhdWchmVRuTJqkUigdwmgBOFZH/C9yFMxlvNBuBuSIyB2gH7gY+7t1BVeckvhaRR4EfWJE4k6qyvd1Zg/rFtwYD6qqSYm7J84Day5r3GZMZqcyj+J6ItALX4ZxOuk1VX0vhcTERuR/naqYi4BFV3SEi97n3Pzi+oee/Qg+oE4oDAeoqQlTYrGpjMkJUk5/yd69yOoOqvufLiEaxYPESffLHz2fiW6dNd1+cDdudgHrPscGA+qLplTRHGrlq7uS8DqgTrHmfMRNHRFpVNTKWx6byFu0ZnHxCgBJgDvA6MH8s39CM7HBnL09taeepLR2nBdQfOK+e5kiYS/M8oPYqCxVTXxEiaLOqjcm4VE49Xeq9LSJLgD/wbUQF6L1DTovvH+88PaC+Yd50VjSFmVVfluERpk+wKEB9RShvVs0zJh+c9W+jqm5yW26YcVBVtrU7Lb5f2n1oYHtVSTG3LprJbYsKI6BOCIhQY837jMlKqczM/jPPzQCwBDhj5TuTmni/8qtdB1m1sY3f7i3cgNrLmvcZk91S+URR6fk6hpNZPO7PcPJXsoB65dJGrjy/MAJqr1BxgPpya95nTLZLWijciXYVqvo/0zSevHO4s5cnt7Sz3hNQC4mAupFLGqoK7lRLQJxZ1dWlNqvamFwwYqFIrJfthtfmLCUNqCNhZtUVTkDtVVkSpK48VHCfnozJZck+UbyCk0dsEZH1wBqgM3Gnqtq62UOoKlvbj7FqYxsv7z48sL2qpJjbFjWwfPFMassKJ6D2suZ9xuSuVDKKOuAQTofXxHwKBaxQuOL9ygtvHmBVS5TXPQH1zJoSVjSFuXH+9IJ9gbTmfcbkvmSFYqp7xdN2BgtEgnVwxQmof7htL49vOj2gvnhGJSsjjVxRgAF1gohQVVJMrTXvMybnJSsURUAFqbULLyiHO3t5YnM761/t4IQ3oD6/npWRRubPLLyA2qs0VER9+SRCxXa5qzH5IFmh2KOqX0jbSHLAu4c6WdMa5Sc7950WUC+bP507mwo3oE6w5n3G5Kdkv9GF+5bYQ1XZGj3GqhYLqEeSaN5XW2azqo3JR8kKxXVpG0UWGimgbqgp5a6mMDfOn1awAbVXuTur2pr3GZO/RiwUqnp4pPvyWXdvnB9u38Pa1nb2Hh8MqOfPrGJFJMwV5xVuQO1lzfuMKRz2W+46dLLHDaj3cLJnMKC+4vzJNEfCXNJQndkBZglr3mdM4Sn4QvHOoU7WtET56WuDAXWoOMCN8501qMO1hR1Qe1nzPmMKU0EWClXl1egxVg8JqKtLgyxfNJPbFs2kpsADaq9QcYDJFZMskzGmQBVUoYj3K7984wCrW9p4Y9/Jge0NNaWsiIS5YZ4F1F5FAaGmzJr3GVPoCqJQdPXG+OH2vaxtjbLveM/A9vkzq2iONPKB8+otoB7CmvcZYxLyulAcOtnDus3tPD0koL5y7mRWNFlAPRxr3meMGSovC8XbBwcD6lj/YEC9bL6zBnVDbWmGR5h9igJCXXmISmveZ4wZIm8Khaqype0oq1qivPL2YEBdUxrktsUzWb6wgeoyexEcypr3GWNGk/OFIt6vPPe6E1C/uX8woA7XlrKiyQmoJ9lplGFZ8z5jTCpytlB09cZ4dpsTUO8/MRhQXzKzipVLG3n/efUEbELYsIJFAerKQ5Rb8z5jTApy7pUiFle+9cLuMwLqq+ZOZkUkzPyZFlCPRESoKQ1SY837jDFnIecKxe4DJ/n+K20ATHID6rssoB6VNe8zxoxVzhUKxQmob1/cwK0LZ1pAPYpgkTOrujRkOY0xZmxyrlBMqyrh+5++3ALqUQREqC0LUVVabKeZjDHjknOFoqYsaEViFBUlxdSVWfM+Y8zE8PWVRESWicjrIrJLRD43zP33iMhW98+LIrLQz/Hku1BxgJk1pUytLLEiYYyZML59ohCRIuDrwPVAFNgoIutVdadnt7eBD6rqERG5CXgIuNyvMeWrooBQWx6iymZVG2N84Oepp8uAXaq6G0BEHgOWAwOFQlVf9Oz/MhD2cTx5yZr3GWP85mehaADaPLejJP+08PvAD4e7Q0TuBe4FaAg3TtT4clpJsIj6ihCTii2vMcb4y89CMdxbXB12R5EP4RSKK4e7X1UfwjktxYLFS4Z9jkJRHAhQWx605n3GmLTxs1BEAe/b/zDQMXQnEVkAPAzcpKqHfBxPTrPmfcaYTPGzUGwE5orIHKAduBv4uHcHEZkFrAN+R1Xf8HEsOc2a9xljMsm3QqGqMRG5H9gAFAGPqOoOEbnPvf9B4G+AeuAb7qSwmKpG/BpTrrHmfcaYbODrK5CqPgs8O2Tbg56vPwV8ys8x5CJr3meMySb2VjXLWPM+Y0y2sUKRJax5nzEmW1mhyDBr3meMyXZWKDKooqSY+vJJNqvaGJPVrFBkwKRgEfXlIUqsC64xJgdYoUgja95njMlFVijSpKo0SG2ZNe8zxuQeKxQ+s+Z9xphcZ4XCJ8WBAHUVISpsVrUxJsfZq9gEExGqS4PUlAateZ8xJi9YoZhAZSFnVrU17zPG5BMrFBMgWBSgviJEWcgOpzEm/9gr2ziICLVlQapLrXmfMSZ/WaEYowq3eV+xNe8zxuQ5KxRnKVQcoL7cmvcZYwqHFYoUWfM+Y0yhskKRgsqSIHXlNqvaGFOYrFAkYc37jDHGCsWwrHmfMcYMskIxRFVpkLqykM2qNsYYlxUKV2moiLpya95njDFDFXyhsOZ9xhiTXMG+Oiaa99WW2axqY4xJpiALRVmomPqKEEGbVW2MMaMqqEJhzfuMMebsFcQrZkCEGmveZ4wxY5L3hcKa9xljzPjkbaEIFQeYXDHJZlUbY8w45V2hKAoINWUhqkttVrUxxkyEvCoU1rzPGGMmXl4UCmveZ4wx/vE14RWRZSLyuojsEpHPDXO/iMgD7v1bRWTJaM/5+t4TXPuPv+T6f/wlRQFhSuUkGmpKrUgYY4xPfCsUIlIEfB24CZgHfExE5g3Z7SZgrvvnXuCbqT5/HLj2y89RaR1ejTHGV35+orgM2KWqu1W1F3gMWD5kn+XAd9XxMlAjIjNS/QbxiRurMcaYEfhZKBqANs/tqLvtbPdBRO4VkRYRaYl3HZvwgRpjjBmZn4ViuEuPdAz7oKoPqWpEVSNFZdUTMjhjjDGp8bNQRIFGz+0w0DGGfUZUbFfBGmOM7/wsFBuBuSIyR0RCwN3A+iH7rAd+17366X3AMVXdk8qTFwvs+vuPTOyIjTHGnMG3eRSqGhOR+4ENQBHwiKruEJH73PsfBJ4FbgZ2AV3AJ0d73ksbqmn5ohUIY4xJF18n3KnqszjFwLvtQc/XCvyhn2MwxhgzPtZS1RhjTFJWKIwxxiRlhcIYY0xSViiMMcYkJU6enDtE5ATweqbHkSUmAwczPYgsYcdikB2LQXYsBl2oqpVjeWAuthl/XVUjmR5ENhCRFjsWDjsWg+xYDLJjMUhEWsb6WDv1ZIwxJikrFMYYY5LKxULxUKYHkEXsWAyyYzHIjsUgOxaDxnwsci7MNsYYk165+InCGGNMGlmhMMYYk1TWFgoRWSYir4vILhH53DD3i4g84N6/VUSWZGKc6ZDCsbjHPQZbReRFEVmYiXGmw2jHwrPfUhGJi8hd6RxfOqVyLETkGhHZIiI7ROSX6R5juqTwO1ItIk+LyKvusRi1U3UuEpFHRGS/iGwf4f6xvW6qatb9wWlL/hZwLhACXgXmDdnnZuCHOKvkvQ/4TabHncFj8QGg1v36pkI+Fp79fo7TufiuTI87g/8vaoCdwCz39tRMjzuDx+IvgS+5X08BDgOhTI/dh2NxNbAE2D7C/WN63czWTxSXAbtUdbeq9gKPAcuH7LMc+K46XgZqRGRGugeaBqMeC1V9UVWPuDdfxlkpMB+l8v8C4I+Ax4H96RxcmqVyLD4OrFPV9wBUNV+PRyrHQoFKERGgAqdQxNI7TP+p6vM4P9tIxvS6ma2FogFo89yOutvOdp98cLY/5+/jvGPIR6MeCxFpAG4HHiS/pfL/4gKgVkSeE5FWEfndtI0uvVI5Fl8DLsZZankb8BlV7U/P8LLKmF43s7WFx3CrYQ+9jjeVffJByj+niHwIp1Bc6euIMieVY/HPwGdVNe68ecxbqRyLYqAJuA4oBV4SkZdV9Q2/B5dmqRyLG4EtwLXAecBPROQFVT3u89iyzZheN7O1UESBRs/tMM47gbPdJx+k9HOKyALgYeAmVT2UprGlWyrHIgI85haJycDNIhJT1SfTMsL0SfV35KCqdgKdIvI8sBDIt0KRyrH4JPBFdU7U7xKRt4GLgFfSM8SsMabXzWw99bQRmCsic0QkBNwNrB+yz3rgd90U/33AMVXdk+6BpsGox0JEZgHrgN/Jw3eLXqMeC1Wdo6qzVXU2sBb473lYJCC135GngKtEpFhEyoDLgdfSPM50SOVYvIfzyQoRmQZcCOxO6yizw5heN7PyE4WqxkTkfmADzhUNj6jqDhG5z73/QZwrWm4GdgFdOO8Y8k6Kx+JvgHrgG+476ZjmYcfMFI9FQUjlWKjqayLyI2Ar0A88rKrDXjaZy1L8f/F3wKMisg3n9MtnVTXv2o+LyPeBa4DJIhIF/hYIwvheN62FhzHGmKSy9dSTMcaYLGGFwhhjTFJWKIwxxiRlhcIYY0xSViiMMcYkZYXCZJzb5XWLiGwXkTXuNf9jfa5HEx1jReRhEZmXZN9rROQDY/ge74jI5GG2V4jIv4rIW26H0udF5PKzff5hnnd2ohuoiERE5IHhxi8i9+Vxmw6TQVk5j8IUnG5VXQQgIt8D7gP+KXGniBSpavxsn1RVPzXKLtcAJ4EXz/a5R/Aw8DYwV1X7ReRcnP5CE0ZVW4AW9+Y1eMZfSPNITHrZJwqTbV4AznffLf9CRP4T2CYiRSLyZRHZ6PbR/wMY6K//NRHZKSLPAFMTT+Q2w4u4Xy8TkU3uegQ/E5HZOAXpT91PM1eJyBQRedz9HhtF5Ar3sfUi8mMR2Swi/8ow/XJE5Dycmc9/nWg253Yzfca9/8/cT0zbReRP3G2zReQ1EfmW+wnkxyJS6t7X5I71JeAPPd/nGhH5wQjj/98i8j/c/RaJyMvusXpCRGo9x+RLIvKKiLwhIle52+e727a4j5k7Mf+cJh9YoTBZQ0SKcdbT2OZuugz4K1Wdh9Ps8JiqLgWWAp8WkTk4nWIvBC4FPo2zNsfQ550CfAu4U1UXAitU9R2cDrNfUdVFqvoC8FX39lLgTpxPCODMbv2Vqi7GaYEwa5jhzwe2DPfJR0SacGbAXo6zBsCnRWSxe/dc4OuqOh846n5fgO8Af6yq7x/uWI0wfq/v4sw+XoBzPP/Wc1+xql4G/Iln+33AV91PdhGcnkDGAHbqyWSHUhHZ4n79AvBtnBf8V1T1bXf7DcACGVyxrhrnRfZq4PvuC3SHiPx8mOd/H/B84rlUdaR+/R8G5slg19kqEal0v8cd7mOfEZEjIzx+JFcCT7jN+RCRdcBVOEXnbVVN/OytwGwRqQZqVDWxIt2/4xTQlAzz+H8D1nh2Wef9fu7XLwF/JSJhnDUs3kz9xzP5zgqFyQYDGUWC+2Ld6d0E/JGqbhiy382M3iZZUtgHnE/Y71fV7mHGMtrjdwALRSQwzDoHyfqd93i+juO0A091vGOV+J5x3NcAVf1PEfkN8BFgg4h8SlWHK7qmANmpJ5MrNgD/TUSCACJygYiUA88Dd7sZxgzgQ8M89iXgg+6pKkSkzt1+Aqj07Pdj4P7EDRFZ5H75PHCPu+0moHboN1DVt3BC5s+LW1lEZK6ILHcff5uIlLljvh3nk9OwVPUocExEEuuK3DPCrkPHn3j8MeBIIn8AfgdIul62G7zvVtUHcD7pLEi2vyksVihMrngYZ/3nTe6lov+K8274CeBNnPPw32SYF0RVPQDcC6wTkVeBVe5dTwO3J8Jg4I+BiBvm7sQ5bw/weeBqEdmEcwrsvRHG+ClgOs56B9twcpEOVd0EPIqz9sFvcLq4bh7l5/0k8HU3zO4eYZ+h4/f6PeDLIrIVWAR8YZTvtxLY7p4CvAgn4zAGsO6xxhhjRmGfKIwxxiRlhcIYY0xSViiMMcYkZYXCGGNMUlYojDHGJGWFwhhjTFJWKIwxxiT1/wGEdEz+qrJ1dgAAAABJRU5ErkJggg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "#save the image to show in the presentation \n", + "import pandas as pd\n", + "x, y2 = pd.Series(y_pred, name=\"Predicted Conditions\"), pd.Series(y_test, name=\"True Conditions\")\n", + "accuracy = sns.regplot(x=x, y=y2)\n", + "accuracy.figure.savefig(\"accuracy.png\")" + ] + }, + { + "cell_type": "code", + "execution_count": 62, + "id": "1d98a56a", + "metadata": {}, + "outputs": [], + "source": [ + "matrix_con_test = confusion_matrix (y_true = y_test, y_pred = y_pred) " + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "id": "9fcb946d", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[[10 0]\n", + " [ 3 5]]\n" + ] + } + ], + "source": [ + "print(matrix_con_test)" + ] + }, + { + "cell_type": "code", + "execution_count": 64, + "id": "f42b68d3", + "metadata": {}, + "outputs": [], + "source": [ + "from pandas import DataFrame\n", + "cmdf3 = DataFrame(matrix_con_test, index = ['Needs','Wants'], columns = ['Needs','Wants'])" + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "id": "cf6559f9", + "metadata": {}, + "outputs": [], + "source": [ + "import seaborn as sns" + ] + }, + { + "cell_type": "code", + "execution_count": 66, + "id": "c3f2f5a0", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "ax = sns.heatmap(cmdf3, cmap = \"black_blue_r\", vmin = 0, vmax = 14, annot = True)\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Observed')\n", + "ax.figure.savefig(\"Confusion_matrice_test.png\")" + ] + }, + { + "cell_type": "code", + "execution_count": 67, + "id": "3926ba23", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ 3.80989403e-06, 7.44035510e-06, 1.11214602e-05, ...,\n", + " -1.41961701e-12, -8.15318262e-12, 0.00000000e+00]])" + ] + }, + "execution_count": 67, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "l_svc.coef_" + ] + }, + { + "cell_type": "code", + "execution_count": 68, + "id": "c1421ebe", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[[3.80989403e-06 7.44035510e-06 1.11214602e-05 ... 1.41961701e-12\n", + " 8.15318262e-12 0.00000000e+00]]\n" + ] + } + ], + "source": [ + "coef2 = np.abs(l_svc.coef_)\n", + "print(coef2)" + ] + }, + { + "cell_type": "code", + "execution_count": 69, + "id": "bb6d20d8", + "metadata": {}, + "outputs": [], + "source": [ + "coef3 = l_svc.coef_" + ] + }, + { + "cell_type": "code", + "execution_count": 70, + "id": "f404d0b7", + "metadata": {}, + "outputs": [], + "source": [ + "essai3= mask.inverse_transform(coef3)\n", + "essai2 = mask.inverse_transform(coef2)" + ] + }, + { + "cell_type": "code", + "execution_count": 71, + "id": "86719376", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "essai2_map = plot_stat_map(essai2, threshold=0.001)\n", + "essai3_map = plot_stat_map(essai3, threshold=0.001)\n", + "essai3_map.savefig('Coeficient map')\n", + "essai2_map.savefig('Abs Coeficient map')" + ] + }, + { + "cell_type": "code", + "execution_count": 72, + "id": "3506267c", + "metadata": {}, + "outputs": [], + "source": [ + "from nilearn import plotting \n", + "from nilearn.plotting import view_img_on_surf \n", + "\n", + "features_plot = plotting.view_img_on_surf(essai3)" + ] + }, + { + "cell_type": "code", + "execution_count": 73, + "id": "17e0d0d3", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "" + ], + "text/plain": [ + "" + ] + }, + "execution_count": 73, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "features_plot" + ] + }, + { + "cell_type": "code", + "execution_count": 74, + "id": "170439ed", + "metadata": {}, + "outputs": [], + "source": [ + "features_plot.save_as_html(\"Model coeficient features interactive plot.html\")" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "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": "3.9.5" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/README.md b/README.md index 0087c83..f323c2b 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,147 @@ -# lgrenier_project -BIO -My name is Laurence, I am a master student in psychology at Montreal University. -I am interested in social neurosciences and cognitive neuroscience. +# Using ALE algorithm and machine learning to classify need and desire states + +### Personal background + +##### My name is Laurence and I've completed a bachelor's degree in cognitive neuroscience at Montreal University. I am currently a master student in psychology at Montreal University. I am particularly interested in social neurosciences, thereby working at the NeSC (Neuroscience en contextes sociaux)lab. My master project is a coordinate based meta-analysis on social hierarchy. + +##### I really wish to contribute to scientific findings by adopting approaches that consider the current issues of reproductibility and social inclusion in neuroscience! I've decided to do brainhack school in order to familiarize myself with coding and neuroimaging datas. + + + + + +
Laurence Grenier + + +* * * + +# Project Definition +* * * + +## Using ALE algorithm and machine learning to classify need and desire states + + +### Theoritical background +* * * +##### Overconsumption is associated with several environmental, social and individual problems (Lipschutz, 2001). One explanation suggested for overconsumption is that "we consume what we want belong what we need" (Stearns, 2006), (Juvénal et al., 2021). Some stimuli are sought for their survival value (food, water,...), while others are sought for their reward value (money, interesting objects,...). While motivation towards the former seems to increase, or even be triggered, by a state of deprivation, motivation towards the latter seems to increase as a function of the previously learned value of the stimulus, corresponding to the association between this stimulus and a reward. However, since a stimulus can be pursued for its *needing* value **AND** for its *wanting* value, for example some kind of food, or even some activities, we can wonder what are the similarities and the differences between those two conditions at a neural level? + + +##### To answer this question, we previously made a coordinate-based meta-analysis with the ALE (Activation likelihood estimation) algorithm in order to find the consistent activations reported in the litterature for each of those conditions. Therefore, we identified regions more commonly reported in the experiments regarding *needs* than in the expriment of *wantings*, vice versa. + +#### ALE +* * * +##### The ALE approach is used to show the convergence between the regions reported in the litterature of a specified topic. For each of the selected studies, the coordinates of the significant voxels are taken (activation peaks). Then, the algorithm is making a gaussian matrice representing the spatial uncertainty, based on the sample of the corresponding study, around the activation peaks reported and thus making what is call a *MA map (main activation map)*.Therefore, each of the experiment has it own *MA map* where each voxels is associated with an activation probability (that increase in the closest to the peaks).Then, the union of the MA maps are taken and the algorithm is making permutations to identify which activations peaks are significantly reported across the experiments, thus representing the convergence of activations. + +###### [Acar, F et al.,2018](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0208177) + + +[![ALE description!](https://github.com/PSY6983-2021/lgrenier_project/blob/lalou97-patch-iss1-add-bio/images/Step-by-step-overview-of-the-ALE-algorithm-From-an-entire-Statistical-Parametric-Map.png)](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0208177) + + + +### Objectives +* * * + +##### The goal of the present project is to build a machine learning model which will make the distinction between the studies focusing on the *needs* and the studies focusing on the *wantings* based on the MA maps generated by ALE algorithm during the first step. + +##### The puposes of this project are : +##### - Build a machine learning model which will classify the *needs* and the *wants* state based on studies included in the meta-analysis to train the model. +##### - Retrieve the most contributing features in the machine learning classification model and interpret the coefficient's value in comparison of ALE meta-analysis results. +##### - Familiarize myself with open science pratices, programming and machine learning. + + +### Tools +* * * +##### - Python +##### - Jupyter notebooks +##### - [nilearn](https://nilearn.github.io/index.html#) +##### - [scikit-learn](https://scikit-learn.org/stable/index.html) +##### - github +##### - [nimare](https://nimare.readthedocs.io/en/latest/about.html) + + +### Data +* * * + +#### The dataset is composed of all the activation peaks of included studies in the meta-analysis. + +### *needs* : 38 studies +##### Contrasts of interest were meeting those criterions: +##### - Perception of a food stimulus during a hunger state > perception of a stimulus during a satiety state. +##### - Perception of a food stimulus during a hunger state > perception of a non-food stimulus during a hunger state. + +### *wantings* : 33 studies +##### Contrasts of interest were meeting those criterions: +##### - Perception of a cue announcing a reward (or a larger reward) > perception of a cue not announcing a reward (or a smaller reward). + +#### Here's all the coordinates of the sample in the brain. +### - Green : Needs +### - Red : Wants +![coordinates gif!](https://github.com/PSY6983-2021/lgrenier_project/blob/lalou97-patch-iss1-add-bio/images/All_coordinates_Enregistrement%20d%E2%80%99e%CC%81cran%2C%20le%202021-09-02%20a%CC%80%2015.32.05%20(1).gif) + + +### Deliverables +* * * +#### At the end of this project, I'll get : +##### - Markdown README.md for the description of the present project +##### - Jupyter notebook including the code, and the brain images + + +# Results +* * * +#### Summary +![Brief description!](https://github.com/PSY6983-2021/lgrenier_project/blob/lalou97-patch-iss1-add-bio/images/Model_description.png) + +### Data preparation +#### - Transform all of the activation peaks of each study into a MA map with [nimare](https://nimare.readthedocs.io/en/latest/about.html).This ended up with 71 MA map, because 71 studies had been included. +#### - Apply a mask to select the voxels of the grey matter. I used the [MNI152 mask](https://nilearn.github.io/modules/generated/nilearn.datasets.load_mni152_brain_mask.html) + +##### Here is an example of a MA maps with the mask: +![Ma map example!](https://github.com/PSY6983-2021/lgrenier_project/blob/lalou97-patch-iss1-add-bio/images/MA%20map%20of%20the%20first%20study-need.png) + + +### Machine learning +##### I used a [linear support vector classification](https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html) to build the model. + +##### At first, the model classified all the data into *needs*. This problem occurred *maybe* because the value of most of the voxels is zero, or close to zero (as you can notice in the above MA map). By arranging the data in order to increase the gap between the null voxels and the non-zero ones, the classification has been improved. + +##### Finally, the model predicted the data with an accuracy of 83.3 ± 16.6 %. +#### There are the confusion matrices that shows the results: +![Confusion matrice](https://github.com/PSY6983-2021/lgrenier_project/blob/lalou97-patch-iss1-add-bio/images/Confusion%20Matrices.png) + +#### Interpretation +##### Even if the accuracy is quite good, because the dataset contains more *needs* than *wants*, the model have initially more than 50% of accuracy if it classifies all the datas into *needs*. Therefore, we should be careful with the interpretation of that accuracy value. This also explains why the model never classifies the *needs* into *wants* and why it is only making errors by classifying the *wants* into *needs*. A solution to overcome this problem would be to balance the weight of the different conditions according to the proportions in the dataset. + +##### Moreover, I've split the data into training set and test set by myself with a proportion of 75% for the training set and 25% for the test set, and I've split again the training set into 60% training set and 40% validation set. Thus, the model is not optimal, because it is only learning with 45% of the initial dataset, which is quite small when having 71 datas. A cross-validation method would have been more efficient in that case. + + + +#### Features importance +##### There you can see the most contributing features on the brain. The more the region is *yellow clear*, the more changes in the value of that voxel influence the decision of the model. +![Feature importance](https://github.com/PSY6983-2021/lgrenier_project/blob/lalou97-patch-iss1-add-bio/images/feature_importance.gif) + + +## Tools I learned during this project +* * * +###### - Python +###### - Packages to run machine learning with scikit-learn +###### - Packages ro run meta-analysis on python with nimare +###### - Tools to create great brain images and data visualization (matplotlib, seaborn...) + +# Conclusion +* * * +##### In conclusion, the objectives of my project are mostly reach. Even if the model is not optimal and few modifications should be applied to improve it's validity, I've succeeded in building a machine learning model and in formatting the initial datas so the model can process them. + +##### Finally, I've started this course without a significant background in coding and a little fear about it (I admit). One of my motivation to do brainhack school was to demystify this field in order to be ready to make great neuroscience! I'm glad I did it, learned python and many useful packages to analyse and play with neuroimaging datas. + +##### I really want to thank BrainHack School team for their remarkable competence and exceptional support during all the course! +![Thank you!](https://github.com/PSY6983-2021/lgrenier_project/blob/lalou97-patch-iss1-add-bio/images/Capture%20d%E2%80%99e%CC%81cran%2C%20le%202021-09-03%20a%CC%80%2017.18.42.png) + + +### References +* * * +###### [L. Charbonnier, F. van Meer, A.M. Johnstone, D. Crabtree, W. Buosi, Y. Manios, O. Androutsos, A. Giannopoulou, M.A. Viergever, P.A.M. Smeets, Effects of hunger state on the brain responses to food cues across the life span,NeuroImage,Volume 171,2018,Pages 246-255,ISSN 1053-8119,https://doi.org/10.1016/j.neuroimage.2018.01.012.](https://neurovault.org/collections/3235/) + +###### [Simon B. Eickhoff, Danilo Bzdok, Angela R. Laird, Florian Kurth, Peter T. Fox, Activation likelihood estimation meta-analysis revisited, NeuroImage, Volume 59, Issue 3, 2012,Pages 2349-2361,ISSN 1053-8119, https://doi.org/10.1016/j.neuroimage.2011.09.017.](https://www.sciencedirect.com/science/article/abs/pii/S1053811911010627?via%3Dihub) + +###### [Assessing robustness against potential publication bias in Activation Likelihood Estimation (ALE) meta-analyses for fMRI, Acar F, Seurinck R, Eickhoff SB, Moerkerke B (2018) Assessing robustness against potential publication bias in Activation Likelihood Estimation (ALE) meta-analyses for fMRI. PLOS ONE 13(11): e0208177. https://doi.org/10.1371/journal.pone.0208177](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0208177) diff --git "a/images/All_coordinates_Enregistrement d\342\200\231e\314\201cran, le 2021-09-02 a\314\200 15.32.05 (1).gif" "b/images/All_coordinates_Enregistrement d\342\200\231e\314\201cran, le 2021-09-02 a\314\200 15.32.05 (1).gif" new file mode 100644 index 0000000..0581419 Binary files /dev/null and "b/images/All_coordinates_Enregistrement d\342\200\231e\314\201cran, le 2021-09-02 a\314\200 15.32.05 (1).gif" differ diff --git "a/images/Capture d\342\200\231e\314\201cran, le 2021-09-03 a\314\200 17.18.42.png" "b/images/Capture d\342\200\231e\314\201cran, le 2021-09-03 a\314\200 17.18.42.png" new file mode 100644 index 0000000..d9b717d Binary files /dev/null and "b/images/Capture d\342\200\231e\314\201cran, le 2021-09-03 a\314\200 17.18.42.png" differ diff --git a/images/Confusion Matrices.png b/images/Confusion Matrices.png new file mode 100644 index 0000000..b1aaa5e Binary files /dev/null and b/images/Confusion Matrices.png differ diff --git a/images/MA map of the first study-need.png b/images/MA map of the first study-need.png new file mode 100644 index 0000000..81fe5d8 Binary files /dev/null and b/images/MA map of the first study-need.png differ diff --git a/images/Model_description.png b/images/Model_description.png new file mode 100644 index 0000000..4d91d6b Binary files /dev/null and b/images/Model_description.png differ diff --git a/images/Step-by-step-overview-of-the-ALE-algorithm-From-an-entire-Statistical-Parametric-Map.png b/images/Step-by-step-overview-of-the-ALE-algorithm-From-an-entire-Statistical-Parametric-Map.png new file mode 100644 index 0000000..106b992 Binary files /dev/null and b/images/Step-by-step-overview-of-the-ALE-algorithm-From-an-entire-Statistical-Parametric-Map.png differ diff --git a/images/feature_importance.gif b/images/feature_importance.gif new file mode 100644 index 0000000..8b4e54a Binary files /dev/null and b/images/feature_importance.gif differ