Zeppelin_Python.json

{"paragraphs":[{"text":"%python\n\nimport sys\nsys.path.insert(0, '/opt/ana2/lib/python2.7/site-packages')\n\nimport csv\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport pandas as pd\nfrom scipy import stats as st\nfrom sklearn import preprocessing\nfrom sklearn import metrics\nfrom sklearn import linear_model\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.ensemble import RandomForestClassifier\nfrom sklearn.tree import DecisionTreeClassifier\nfrom sklearn import svm\nfrom sklearn.model_selection import StratifiedKFold\nfrom sklearn.model_selection import GridSearchCV\nfrom sklearn.metrics import roc_curve, auc\nimport statsmodels.api as sm\nimport seaborn as sns\nsns.set(style=\"white\")\nsns.set(style=\"whitegrid\", color_codes=True)","user":"anonymous","dateUpdated":"2019-01-09T18:39:26+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"ERROR","msg":[{"type":"TEXT","data":"java.lang.InterruptedException: sleep interrupted"}]},"apps":[],"jobName":"paragraph_1547055366102_2142605213","id":"20190109-183606_57902720","dateCreated":"2019-01-09T18:36:06+0100","dateStarted":"2019-01-09T18:39:26+0100","dateFinished":"2019-01-09T18:41:01+0100","status":"ABORT","progressUpdateIntervalMs":500,"focus":true,"$$hashKey":"object:261"},{"text":"%python\nimport warnings\nwarnings.filterwarnings(\"ignore\", category=FutureWarning)","user":"anonymous","dateUpdated":"2019-01-09T14:53:27+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547024394395_-1531815304","id":"20190109-095954_1335540854","dateCreated":"2019-01-09T09:59:54+0100","dateStarted":"2019-01-09T14:53:27+0100","dateFinished":"2019-01-09T14:53:27+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:262"},{"text":"%python\ninput1 = z.input(\"Indiquez le chemin de votre fichier d'entrée : \")\n#/home/tp-home008/mfigaro/Downloads/train.txt","user":"anonymous","dateUpdated":"2019-01-09T14:53:27+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{"Indiquez le chemin de votre fichier d'entrée : ":"/home/tp-home008/mfigaro/Downloads/train.txt","Indiquez la manière dont sont séparées vos variables (tabulation/espace/;) : ":"","Veuillez entrer tabulation, espace ou ; : ":""},"forms":{"Indiquez le chemin de votre fichier d'entrée : ":{"type":"TextBox","name":"Indiquez le chemin de votre fichier d'entrée : ","displayName":"Indiquez le chemin de votre fichier d'entrée : ","defaultValue":"","hidden":false,"$$hashKey":"object:960"}}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547026890616_26031311","id":"20190109-104130_556657884","dateCreated":"2019-01-09T10:41:30+0100","dateStarted":"2019-01-09T14:53:27+0100","dateFinished":"2019-01-09T14:53:27+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:263"},{"text":"%python\ninput2 = z.input(\"Indiquez la manière dont sont séparées vos variables (tabulation/espace/;) : \")   # ici espace\n# Passage en minuscule si l'utilisateur écrit en majuscule\ninput2 = input2.lower() ;","user":"anonymous","dateUpdated":"2019-01-09T14:53:27+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python","editorHide":false},"settings":{"params":{"Indiquez la manière dont sont séparées vos variables (tabulation/espace/;) : ":"espace","Veuillez entrer tabulation, espace ou ; : ":""},"forms":{"Indiquez la manière dont sont séparées vos variables (tabulation/espace/;) : ":{"type":"TextBox","name":"Indiquez la manière dont sont séparées vos variables (tabulation/espace/;) : ","displayName":"Indiquez la manière dont sont séparées vos variables (tabulation/espace/;) : ","defaultValue":"","hidden":false,"$$hashKey":"object:972"}}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547028992155_863439300","id":"20190109-111632_933632474","dateCreated":"2019-01-09T11:16:32+0100","dateStarted":"2019-01-09T14:53:27+0100","dateFinished":"2019-01-09T14:53:27+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:264"},{"text":"%python\n# Tant que l'utilisateur n'a pas rentré quelque chose de correct\nwhile not (input2 == 'tabulation' or input2 == 'espace' or input2 == ';') :\n    input2 =z.input(\"Veuillez entrer tabulation, espace ou ; : \")\n    input2 = input2.lower()","user":"anonymous","dateUpdated":"2019-01-09T14:53:27+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{"Veuillez entrer tabulation, espace ou ; : ":"espace"},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547029077826_-44643690","id":"20190109-111757_665784065","dateCreated":"2019-01-09T11:17:57+0100","dateStarted":"2019-01-09T14:53:27+0100","dateFinished":"2019-01-09T14:53:27+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:265"},{"text":"%python\n# On met les vrais caractères pour pouvoir importer le fichier\nif input2 == 'tabulation' :\n    input2 = '\\t'\nelif input2 == 'espace' :\n    input2 = ' '\n\n# Importation du fichier\nfile = open(input1, \"r\")\ndataset = pd.read_csv(file, sep=input2)\n\n# Nombre total d'instances\ntotal_rows = len(dataset)\n# Nombre total de variables\ntotal_columns = len(list(dataset))\n\n# Récupération du noms des différentes variables\ntitle = list(dataset.columns)\n\n# Variable d'intérêt\nreponse = z.input(\"Indiquez le nom de la variable d'intérêt : \")   # ici : y\n    ","user":"anonymous","dateUpdated":"2019-01-09T14:53:27+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{"Indiquez le nom de la variable d'intérêt : ":"y"},"forms":{"Indiquez le nom de la variable d'intérêt : ":{"type":"TextBox","name":"Indiquez le nom de la variable d'intérêt : ","displayName":"Indiquez le nom de la variable d'intérêt : ","defaultValue":"","hidden":false,"$$hashKey":"object:987"}}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547029121265_1307943006","id":"20190109-111841_1062807297","dateCreated":"2019-01-09T11:18:41+0100","dateStarted":"2019-01-09T14:53:27+0100","dateFinished":"2019-01-09T14:53:27+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:266"},{"text":"%python\nwhile (reponse not in title) :\n    reponse = z.input(\"Cette variable n'existe pas ! Veuillez vérifier son nom et essayer à nouveau : \")","user":"anonymous","dateUpdated":"2019-01-09T14:53:27+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python","editorHide":false},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547029961002_-1102085814","id":"20190109-113241_1450079534","dateCreated":"2019-01-09T11:32:41+0100","dateStarted":"2019-01-09T14:53:27+0100","dateFinished":"2019-01-09T14:53:28+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:267"},{"text":"%python\n# Vérifier s'il y a une variable identifiant\ninput3 = z.input(\"Vos données ont-elles une variable identifiant pour les individus ? (y/n) : \")     # ici : n\n# Passage en minuscule si l'utilisateur rentre Y ou N\ninput3 = input3.lower() ;","user":"anonymous","dateUpdated":"2019-01-09T14:53:28+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{"Vos données ont-elles une variable identifiant pour les individus ? (y/n) : ":"n"},"forms":{"Vos données ont-elles une variable identifiant pour les individus ? (y/n) : ":{"type":"TextBox","name":"Vos données ont-elles une variable identifiant pour les individus ? (y/n) : ","displayName":"Vos données ont-elles une variable identifiant pour les individus ? (y/n) : ","defaultValue":"","hidden":false,"$$hashKey":"object:1002"}}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547029688819_-70320342","id":"20190109-112808_2016985005","dateCreated":"2019-01-09T11:28:08+0100","dateStarted":"2019-01-09T14:53:28+0100","dateFinished":"2019-01-09T14:53:28+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:268"},{"text":"%python\n# Tant que l'utilisateur n'a pas rentré quelque chose de correct\nwhile not (input3 == 'y' or input3 == 'n') :\n    input3 = z.input(\"Veuillez entrer y ou n : \")\n    input3 = input3.lower()","user":"anonymous","dateUpdated":"2019-01-09T14:53:28+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547030002071_545581735","id":"20190109-113322_248670218","dateCreated":"2019-01-09T11:33:22+0100","dateStarted":"2019-01-09T14:53:28+0100","dateFinished":"2019-01-09T14:53:28+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:269"},{"text":"%python\n# S'il y en a une\nif input3 == 'y' :\n    input4 = z.input(\"Indiquez le nom de la variable identifiant : \")","user":"anonymous","dateUpdated":"2019-01-09T14:53:28+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547030030207_315108088","id":"20190109-113350_1916835350","dateCreated":"2019-01-09T11:33:50+0100","dateStarted":"2019-01-09T14:53:28+0100","dateFinished":"2019-01-09T14:53:28+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:270"},{"text":"%python\nif input3 == 'y' :\n    while input4 not in title or input4 == reponse :\n        if input4 not in title :\n            input4 = z.input(\"Cette variable n'existe pas ! Veuillez vérifier son nom et essayer à nouveau : \")\n        elif input4 == reponse :\n            input4 = z.input(\"Il s'agit de la variable d'intérêt ! Veuillez vérifier et essayer à nouveau : \")\n    # On supprime cette variable\n    dataset = dataset.drop(input4, 1)","user":"anonymous","dateUpdated":"2019-01-09T14:53:28+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547030081815_-2008879067","id":"20190109-113441_649096659","dateCreated":"2019-01-09T11:34:41+0100","dateStarted":"2019-01-09T14:53:28+0100","dateFinished":"2019-01-09T14:53:28+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:271"},{"text":"%python\n# Suppression des instances ayant des données manquantes\ndataset = dataset.dropna(how = 'any')\n\n# Forme des résultats\ninput6 = z.input(\"Choisissez la forme sous laquelle vous voulez les résultats (erreur test/matrice de confusion/courbe ROC) : \")\n# Passage en minuscule si l'utilisateur écrit en majuscule\ninput6 = input6.lower() ;","user":"anonymous","dateUpdated":"2019-01-09T14:53:28+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{"Choisissez la forme sous laquelle vous voulez les résultats (erreur test/matrice de confusion/courbe ROC) : ":"matrice ROC"},"forms":{"Choisissez la forme sous laquelle vous voulez les résultats (erreur test/matrice de confusion/courbe ROC) : ":{"type":"TextBox","name":"Choisissez la forme sous laquelle vous voulez les résultats (erreur test/matrice de confusion/courbe ROC) : ","displayName":"Choisissez la forme sous laquelle vous voulez les résultats (erreur test/matrice de confusion/courbe ROC) : ","defaultValue":"","hidden":false,"$$hashKey":"object:1023"}}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547030123710_165479693","id":"20190109-113523_1505587717","dateCreated":"2019-01-09T11:35:23+0100","dateStarted":"2019-01-09T14:53:28+0100","dateFinished":"2019-01-09T14:53:28+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:272"},{"text":"%python\n# Tant que l'utilisateur n'a pas rentré quelque chose de correct\nwhile not (input6 == 'erreur test' or input6 == 'matrice de confusion' or input6 == 'courbe roc') :\n    input6 = z.input(\"Veuillez entrer erreur test, matrice de confusion ou courbe ROC : \")\n    input6 = input6.lower()","user":"anonymous","dateUpdated":"2019-01-09T14:53:28+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{"Veuillez entrer erreur test, matrice de confusion ou courbe ROC : ":""},"forms":{"Veuillez entrer erreur test, matrice de confusion ou courbe ROC : ":{"type":"TextBox","name":"Veuillez entrer erreur test, matrice de confusion ou courbe ROC : ","displayName":"Veuillez entrer erreur test, matrice de confusion ou courbe ROC : ","defaultValue":"","hidden":false,"$$hashKey":"object:1035"}}},"results":{"code":"ERROR","msg":[{"type":"TEXT","data":"Traceback (most recent call last):\n  File \"/tmp/zeppelin_python-1652545266603803792.py\", line 320, in <module>\n    raise Exception(traceback.format_exc())\nException: Traceback (most recent call last):\n  File \"/tmp/zeppelin_python-1652545266603803792.py\", line 313, in <module>\n    exec(code, _zcUserQueryNameSpace)\n  File \"<stdin>\", line 2, in <module>\n  File \"/tmp/zeppelin_python-1652545266603803792.py\", line 67, in input\n    return self.z.input(name, defaultValue)\n  File \"/home/tp-home008/mfigaro/Documents/zeppelin-0.8.0-bin-all/interpreter/python/py4j-0.9.2/src/py4j/java_gateway.py\", line 827, in __call__\n    [get_command_part(arg, self.pool) for arg in new_args])\n  File \"/home/tp-home008/mfigaro/Documents/zeppelin-0.8.0-bin-all/interpreter/python/py4j-0.9.2/src/py4j/protocol.py\", line 274, in get_command_part\n    command_part = STRING_TYPE + escape_new_line(parameter)\n  File \"/home/tp-home008/mfigaro/Documents/zeppelin-0.8.0-bin-all/interpreter/python/py4j-0.9.2/src/py4j/protocol.py\", line 175, in escape_new_line\n    return smart_decode(original).replace(\"\\\\\", \"\\\\\\\\\").replace(\"\\r\", \"\\\\r\").\\\n  File \"/home/tp-home008/mfigaro/Documents/zeppelin-0.8.0-bin-all/interpreter/python/py4j-0.9.2/src/py4j/protocol.py\", line 198, in smart_decode\n    if isinstance(s, unicode):\n  File \"/tmp/zeppelin_python-1652545266603803792.py\", line 227, in handler_stop_signals\n    sys.exit(\"Got signal : \" + str(sig))\nSystemExit: Got signal : 2\n\n"}]},"apps":[],"jobName":"paragraph_1547030132382_-1600337625","id":"20190109-113532_6047975","dateCreated":"2019-01-09T11:35:32+0100","dateStarted":"2019-01-09T14:53:28+0100","dateFinished":"2019-01-09T14:55:00+0100","status":"ABORT","progressUpdateIntervalMs":500,"$$hashKey":"object:273"},{"text":"%python\ndataset.head()","user":"anonymous","dateUpdated":"2019-01-09T14:47:52+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547030136581_1580371885","id":"20190109-113536_10106922","dateCreated":"2019-01-09T11:35:36+0100","dateStarted":"2019-01-09T14:47:52+0100","dateFinished":"2019-01-09T14:47:52+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:274"},{"text":"%python\ny = dataset[reponse]\ny_values = list(y)","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547030899112_-1777762595","id":"20190109-114819_1622450722","dateCreated":"2019-01-09T11:48:19+0100","dateStarted":"2019-01-09T14:47:53+0100","dateFinished":"2019-01-09T14:47:53+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:275"},{"text":"Dimension de votre dataset","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":false,"results":{},"editorSetting":{"language":"scala","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/scala"},"settings":{"params":{},"forms":{}},"apps":[],"jobName":"paragraph_1547031054830_-1599127045","id":"20190109-115054_1749379458","dateCreated":"2019-01-09T11:50:54+0100","status":"READY","progressUpdateIntervalMs":500,"$$hashKey":"object:276"},{"text":"%python\nprint(dataset.shape)","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547031517748_-96383452","id":"20190109-115837_1246024645","dateCreated":"2019-01-09T11:58:37+0100","dateStarted":"2019-01-09T14:47:53+0100","dateFinished":"2019-01-09T14:47:53+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:277"},{"text":"Visualisation du résumé de la variable d'intérêt","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":false,"results":{},"editorSetting":{"language":"scala","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/scala"},"settings":{"params":{},"forms":{}},"apps":[],"jobName":"paragraph_1547038449195_-912462683","id":"20190109-135409_654072410","dateCreated":"2019-01-09T13:54:09+0100","status":"READY","progressUpdateIntervalMs":500,"$$hashKey":"object:278"},{"text":"%python\ny.describe()\nclasse = list(set(y))\nnb_classe =  len(list(set(y))) # Taille de la liste contenant la variable d'intérêt sans doublon\n\nprint \"Differentes classes de la variable d'interet : \", classe\nprint \"Nombre de classes : \", nb_classe","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547031537706_320552697","id":"20190109-115857_105294102","dateCreated":"2019-01-09T11:58:57+0100","dateStarted":"2019-01-09T14:47:53+0100","dateFinished":"2019-01-09T14:47:53+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:279"},{"text":"%python\nsns.countplot(x = reponse, data = dataset, palette = 'hls')\nplt.show()","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{"1":{"graph":{"mode":"table","height":300,"optionOpen":false}}},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[{"type":"TEXT","data":"   x10006_at  x10007_at  x100129361_at  x100"}]},"apps":[],"jobName":"paragraph_1547031566835_286425925","id":"20190109-115926_420033372","dateCreated":"2019-01-09T11:59:26+0100","dateStarted":"2019-01-09T14:47:53+0100","dateFinished":"2019-01-09T14:47:53+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:280"},{"text":"%python\n# Division du dataset\nxTrain, xTest = train_test_split(dataset, train_size=int(total_rows*(float(2)/3)), test_size=int(total_rows*(float(1)/3)))\n# Vérifications\n\nprint(\"Dimension du dataset d'apprentissage : \" + str(xTrain.shape))\nprint(\"Dimension du dataset de test : \" + str(xTest.shape))\n","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547031622603_-639976439","id":"20190109-120022_588468283","dateCreated":"2019-01-09T12:00:22+0100","dateStarted":"2019-01-09T14:47:53+0100","dateFinished":"2019-01-09T14:47:53+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:281"},{"text":"%python\n# Récupération des variables explicatives du dataset apprentissage et du dataset test\nXTrain = xTrain.iloc[:,:total_columns-1]\nXTest = xTest.iloc[:,:total_columns-1]\n# Vérification\nprint(\"Dataset d'apprentissage : \\n\")\nprint(XTrain.shape)\nprint(XTrain.head())\n\nprint(\"\\nDataset test : \\n\")\nprint(XTest.shape)\nprint(XTest.head())","user":"anonymous","dateUpdated":"2019-01-09T14:47:53+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[{"type":"TEXT","data":"4AAgCKI4AAgOIIIACgOAIIACiOAAIAiiOAAIDiCCAAoDgCCAAojgACAIojgACA4gggAKA4AggAKI4AAgCKI4AAgOIIIACgOAIIACiOAAIAiiOAAIDiCCAAoDgCCAAojgACAIojgACA4gggAKA4AggAKI4AAgCKI4AAgOIIIACgOAIIACjO6KoHVOHaa6/NE088kVqtlssvvzyHHHJI1ZMAgB2ouABavXp"}]},"apps":[],"jobName":"paragraph_1547031635136_1330602320","id":"20190109-120035_1547169020","dateCreated":"2019-01-09T12:00:35+0100","dateStarted":"2019-01-09T14:47:53+0100","dateFinished":"2019-01-09T14:47:53+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:282"},{"text":"%python\n# Isolation de la variable d'intérêt\nyTrain = xTrain.iloc[:,total_columns-1]\nyTrain_values = list(yTrain)\n\nyTest = xTest.iloc[:,total_columns-1]\nyTest_values = list(yTest)","user":"anonymous","dateUpdated":"2019-01-09T14:47:54+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033044863_-1081327947","id":"20190109-122404_1937421188","dateCreated":"2019-01-09T12:24:04+0100","dateStarted":"2019-01-09T14:47:54+0100","dateFinished":"2019-01-09T14:47:54+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:283"},{"text":"%python\n# Vérification que les classes soient bien réparties dans les 2 datasets\n\n# Dataset apprentissage\n    # /!\\ Représentation graphique ou juste afficher nb_occ, je sais pas\nsns.countplot(x = reponse, data = xTrain, palette = 'hls')\nplt.show()\nxTrain[reponse].value_counts()\n\n# Dataset test\nsns.countplot(x = reponse, data = xTest, palette = 'hls')\nplt.show()\nxTest[reponse].value_counts()","user":"anonymous","dateUpdated":"2019-01-09T14:47:54+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[{"type":"TEXT","data":"       11.37       3.48    ...       \n\n     x10224_at  x10226_at  x10227_at  x1022_at  x10231_at  x10234_at  \\\n123       6.35       9.38       9.63      6.72       2.67       2.86   \n84        3.55       8.94       7.20      7.35       4.73       2.35   \n126       4.26      10.10       7.58      7.45       3.05       4.74   \n63        2.68       9.14       8.17      8.75       2.93       4.06   \n0         4.26      10.55       7.82      7.87       2.97       4.48   \n\n     x10236_at  x10237_at  x10240_at  x10241_at  \n123       9.54       9.47       8.79       7.02  \n84        9.63       9.25       5.81       9.10  \n126       9.42      12.13       7.38      10.10  \n63        9.39       9.57       6.01       9.34  \n0         9.61      12.82       5.62      12.76  \n\n[5 rows x 100 columns]\n"},{"type":"HTML","data":"<div style='width:auto;height:auto'><img src=data:image/png;base64,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 style='width=auto;height:auto'><div>\n"},{"type":"HTML","data":"<div style='width:auto;height:auto'><img src=data:image/png;base64,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"}]},"apps":[],"jobName":"paragraph_1547033328232_-899068986","id":"20190109-122848_29715122","dateCreated":"2019-01-09T12:28:48+0100","dateStarted":"2019-01-09T14:47:54+0100","dateFinished":"2019-01-09T14:47:54+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:284"},{"text":"%python\n# Regression logistique\n# Construction d'un objet permettant de réaliser une régression logistique\nlogit_model=linear_model.LogisticRegression()\nlogit_model.fit(XTrain,yTrain)\npred_rl = logit_model.predict(XTest)\n\n# Performance du modèle\nscore_glm = logit_model.score(XTest, yTest)","user":"anonymous","dateUpdated":"2019-01-09T14:47:54+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033363629_293296651","id":"20190109-122923_1164325700","dateCreated":"2019-01-09T12:29:23+0100","dateStarted":"2019-01-09T14:47:54+0100","dateFinished":"2019-01-09T14:47:54+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:285"},{"text":"%python\n# Random Forest\n# Traitement des données.\n# S'il s'agit de str pour la variable d'intérêt, il faut la convertir en int pour appliquer RandomForestClassifier\nyTrain_ = [0]*len(yTrain_values)\nfor i in range(nb_classe):\n    for j in range(len(yTrain_values)):\n        if(yTrain_values[j]==classe[i]):\n            yTrain_[j]=i\n            \nyTest_ = [0]*len(yTest_values)\nfor i in range(nb_classe):\n    for j in range(len(yTest_values)):\n        if(yTest_values[j]==classe[i]):\n            yTest_[j]=i\n            \nrf_model = RandomForestClassifier()\n\n# Etablissement des différents paramètres à tester\nparameter_grid = {'n_estimators': [10, 25, 50, 100],\n                  'criterion': ['gini', 'entropy'],\n                  'max_features': [1, 2, 3, 4]}\n\n# Cross Validation (on fait tourner 10 fois le modele sur differents découpage)\ncross_validation = StratifiedKFold(n_splits=10)\n\n# Sélection des meilleurs paramètres\ngrid_search = GridSearchCV(rf_model,\n                           param_grid=parameter_grid,\n                           cv=cross_validation)\n\ngrid_search.fit(XTrain, yTrain_)\n\n# Visualisation des meilleurs paramètres\nprint('Best parameters: {}'.format(grid_search.best_params_))\n\n# Stockage des meilleurs paramètres\nnestim_best = grid_search.best_params_['n_estimators']\ncriterion_best = grid_search.best_params_['criterion']\nmax_features_best = grid_search.best_params_['max_features']\n\n# Mise en place du modèle avec les meilleurs paramètres\nrf_model = RandomForestClassifier(n_estimators = nestim_best, criterion = criterion_best, max_features = max_features_best)\nrf_model.fit(XTrain, yTrain_)\npred_rf = list(rf_model.predict(XTest))\npred_rf = np.array(pred_rf).reshape(-1,1)\nyTest_ = np.array(yTest_).reshape(-1,1)\n\n# Performance du modèle\nscore_rf = float(np.sum(pred_rf==yTest_))/len(pred_rf)","user":"anonymous","dateUpdated":"2019-01-09T14:48:32+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python","lineNumbers":false},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033380408_1803639707","id":"20190109-122940_479926315","dateCreated":"2019-01-09T12:29:40+0100","dateStarted":"2019-01-09T14:47:54+0100","dateFinished":"2019-01-09T14:48:25+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:286"},{"text":"%python\n# Arbre CART\ncart_model = DecisionTreeClassifier()\n\n# Etablissement des différents paramètres à tester\nparameter_grid = {'max_depth': [1, 2, 3, 4, 5],\n                  'max_features': [1, 2, 3, 4]}\n\n# Cross Validation (on fait tourner 10 fois le modele sur differents découpage)\ncross_validation = StratifiedKFold(n_splits=10)\n\n# Sélection des meilleurs paramètres\ngrid_search = GridSearchCV(cart_model,\n                           param_grid=parameter_grid,\n                           cv=cross_validation)\ngrid_search.fit(XTrain, yTrain_)\n\n# Visualisation des meilleurs paramètres\nprint('Meilleurs parametres: {}'.format(grid_search.best_params_))\n\n# Stockage des meilleurs paramètres\nmax_depth_best = grid_search.best_params_['max_depth']\nmax_features_best = grid_search.best_params_['max_features']\n\n# Mise en place du modèle avec les meilleurs paramètres\ncart_model = DecisionTreeClassifier(max_depth = max_depth_best, max_features = max_features_best)\ncart_model.fit(XTrain, yTrain_)\npred_cart = list(cart_model.predict(XTest))\npred_cart = np.array(pred_cart).reshape(-1,1)\n\n# Performance du modèle\nscore_cart = float(np.sum(pred_cart==yTest_))/len(pred_cart)","user":"anonymous","dateUpdated":"2019-01-09T14:48:25+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033407405_1607743568","id":"20190109-123007_101824441","dateCreated":"2019-01-09T12:30:07+0100","dateStarted":"2019-01-09T14:48:25+0100","dateFinished":"2019-01-09T14:48:26+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:287"},{"text":"%python\n# SVM\nsvm_model = svm.SVC()\n\n# Etablissement des différents paramètres à tester\nparameter_grid = {'gamma': [0.1, 1, 10, 100],\n                  'kernel': ['linear', 'rbf', 'poly'],\n                  'C': [0.1, 1, 10, 100, 1000]}\n\n# Cross Validation (on fait tourner 10 fois le modele sur differents découpage)\ncross_validation = StratifiedKFold(n_splits = 10)\n\n# Sélection des meilleurs paramètres\ngrid_search = GridSearchCV(svm_model,\n                           param_grid = parameter_grid,\n                           cv = cross_validation)\ngrid_search.fit(XTrain, yTrain_)\n\n# Visualisation des meilleurs paramètres\nprint('Meilleurs parametres: {}'.format(grid_search.best_params_))\n\n# Stockage des meilleurs paramètres\ngamma_best = grid_search.best_params_['gamma']\nkernel_best = grid_search.best_params_['kernel']\ncs_best = grid_search.best_params_['C']\n\n# Mise en place du modèle avec les meilleurs paramètres\nsvm_model = svm.SVC(gamma = gamma_best, kernel = kernel_best, C = cs_best, probability = True)\nsvm_model.fit(XTrain, yTrain_)\npred_svm = list(svm_model.predict(XTest))\npred_svm = np.array(pred_svm).reshape(-1,1)\n\n# Performance du modèle\nscore_svm = float(np.sum(pred_svm==yTest_))/len(pred_svm)","user":"anonymous","dateUpdated":"2019-01-09T14:48:26+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033485311_218147936","id":"20190109-123125_175208248","dateCreated":"2019-01-09T12:31:25+0100","dateStarted":"2019-01-09T14:48:26+0100","dateFinished":"2019-01-09T14:48:29+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:288"},{"text":"%python\n# Affichage des résultats\n\nif input6 == 'erreur test' :\n    print \"Precision de la Regression Logistique : \", round(score_glm*100,4), \"%\\n\"\n    print \"Precision du Random Forest : \", round(score_rf*100,4), \"%\\n\"\n    print \"Precision de l'arbre CART : \", round(score_cart*100,4), \"%\\n\"\n    print \"Precision de SVM : \", round(score_svm*100,4), \"%\\n\"\n    \n    err = [round(score_glm*100,4),round(score_rf*100,4),round(score_cart*100,4),round(score_svm*100,4)]\n    ind = err.index(max(err))\n    ","user":"anonymous","dateUpdated":"2019-01-09T14:48:29+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033496128_252703743","id":"20190109-123136_1864964957","dateCreated":"2019-01-09T12:31:36+0100","dateStarted":"2019-01-09T14:48:29+0100","dateFinished":"2019-01-09T14:48:29+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:289"},{"text":"%python\n# Affichage des résultats sous forme de matrice de confusion\n\nif input6 == 'matrice de confusion' : \n    print(\"Matrice de confusion de la Régression Logistique:\")\n    cm_rl = metrics.confusion_matrix(yTest, pred_rl)\n    plt.figure(figsize=(nb_classe+2,nb_classe+2))\n    sns.heatmap(cm_rl, annot=True, fmt=\".2f\", linewidths=.3, square = True, cmap = \"PiYG\");\n    plt.ylabel('Observés');\n    plt.xlabel('Prédis');\n    title = 'Score de prédiction: {0}'.format(round(score_glm*100,4))\n    plt.title(title, size = 13)\n    plt.show()\n    \n    print(\"\\nMatrice de confusion du Random Forest:\")\n    cm_rf = metrics.confusion_matrix(yTest_, pred_rf)\n    plt.figure(figsize=(nb_classe+2,nb_classe+2))\n    sns.heatmap(cm_rf, annot=True, fmt=\".2f\", linewidths=.3, square = True, cmap = \"PiYG\");\n    plt.ylabel('Observés');\n    plt.xlabel('Prédis');\n    title = 'Score de prédiction: {0}'.format(round(score_rf*100,4))\n    plt.title(title, size = 13)\n    plt.show()\n    \n    print(\"\\nMatrice de confusion de CART:\")\n    cm_cart = metrics.confusion_matrix(yTest_, pred_cart)\n    plt.figure(figsize=(nb_classe+2,nb_classe+2))\n    sns.heatmap(cm_cart, annot=True, fmt=\".2f\", linewidths=.3, square = True, cmap = \"PiYG\");\n    plt.ylabel('Observés');\n    plt.xlabel('Prédis');\n    title = 'Score de prédiction: {0}'.format(round(score_cart*100,4))\n    plt.title(title, size = 13)\n    plt.show()\n    \n    print(\"\\nMatrice de confusion SVM :\")\n    cm_svm = metrics.confusion_matrix(yTest_, pred_svm)\n    plt.figure(figsize=(nb_classe+2,nb_classe+2))\n    sns.heatmap(cm_svm, annot=True, fmt=\".2f\", linewidths=.3, square = True, cmap = \"PiYG\");\n    plt.ylabel('Observés');\n    plt.xlabel('Prédis');\n    title = 'Score de prédiction: {0}'.format(round(score_svm*100,4))\n    plt.title(title, size = 13)\n    plt.show()\n    \n    err = [round(score_glm*100,4),round(score_rf*100,4),round(score_cart*100,4),round(score_svm*100,4)]\n    ind = err.index(max(err))","user":"anonymous","dateUpdated":"2019-01-09T14:48:29+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{"3":{"graph":{"mode":"table","height":297,"optionOpen":false}}},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[{"type":"TEXT","data":"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 style='width=auto;height:auto'><div>\n-1    36\n 1    25\nName: y, dtype: int64\nBest parameters: {'max_features': 1, 'n_estimators': 50, 'criterion': 'gini'}\nMeilleurs parametres: {'max_features': 2, 'max_depth': 4}\nMeilleurs parametres: {'kernel': 'rbf', 'C': 0.1, 'gamma': 0.1}\nMatrice de confusion de la Régression Logistique:\n<matplotlib.figure.Figure object at 0x7f21f72b3590>\n<matplotlib.axes._subplots.AxesSubplot object at 0x7f21f72e2310>\n<matplotlib.text.Text object at 0x7f21f720e710>\n<matplotlib.text.Text object at 0x7f21f72e2a10>\n<matplotlib.text.Text object at 0x7f21f734c7d0>\n"},{"type":"HTML","data":"<div style='width:auto;height:auto'><img src=data:image/png;base64,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 style='width=auto;height:auto'><div>\n\nMatrice de confusion du Random Forest:\n<matplotlib.figure.Figure object at 0x7f21f71d1d90>\n<matplotlib.axes._subplots.AxesSubplot object at 0x7f21f746bbd0>\n<matplotlib.text.Text object at 0x7f21f72da850>\n<matplotlib.text.Text object at 0x7f21f72c9110>\n<matplotlib.text.Text object at 0x7f21f72da190>\n"},{"type":"HTML","data":"<div style='width:auto;height:auto'><img src=data:image/png;base64,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 style='width=auto;height:auto'><div>\n\nMatrice de confusion de CART:\n<matplotlib.figure.Figure object at 0x7f21f70797d0>\n<matplotlib.axes._subplots.AxesSubplot object at 0x7f21f72c9b90>\n<matplotlib.text.Text object at 0x7f21f70a3e10>\n<matplotlib.text.Text object at 0x7f21f7097590>\n<matplotlib.text.Text object at 0x7f21f7057850>\n"},{"type":"HTML","data":"<div style='width:auto;height:auto'><img src=data:image/png;base64,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 style='width=auto;height:auto'><div>\n\nMatrice de confusion SVM :\n<matplotlib.figure.Figure object at 0x7f21f70ca150>\n<matplotlib.axes._subplots.AxesSubplot object at 0x7f21f71734d0>\n<matplotlib.text.Text object at 0x7f21f731e890>\n<matplotlib.text.Text object at 0x7f21f7330610>\n<matplotlib.text.Text object at 0x7f21f7153a50>\n"},{"type":"HTML","data":"<div style='width:auto;height:auto'><img src=data:image/png;base64,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"}]},"apps":[],"jobName":"paragraph_1547033508751_419358345","id":"20190109-123148_816961333","dateCreated":"2019-01-09T12:31:48+0100","dateStarted":"2019-01-09T14:48:29+0100","dateFinished":"2019-01-09T14:48:31+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:290"},{"text":"%python\n# Affichage des résultats sous forme de courbe ROC\n\nif input6 == 'courbe roc' :\n    probas_rl = logit_model.predict_proba(XTest)\n    # probas est une matrice de deux colonnes avec la proabilités d'appartenance à chaque classe\n    fpr_rl, tpr_rl, thresholds_rl = roc_curve(yTest, probas_rl[:, 1])\n    roc_auc_rl = auc(fpr_rl, tpr_rl)\n    plt.plot(fpr_rl, tpr_rl, label='RL (aire = %0.2f)' % roc_auc_rl)\n    \n    probas_rf = rf_model.predict_proba(XTest)\n    fpr_rf, tpr_rf, thresholds_rf = roc_curve(yTest, probas_rf[:, 1])\n    roc_auc_rf = auc(fpr_rf, tpr_rf)\n    plt.plot(fpr_rf, tpr_rf, label='RF (aire = %0.2f)' % roc_auc_rf)\n    \n    probas_cart = cart_model.predict_proba(XTest)\n    fpr_cart, tpr_cart, thresholds_cart = roc_curve(yTest, probas_cart[:, 1])\n    roc_auc_cart = auc(fpr_cart, tpr_cart)\n    plt.plot(fpr_cart, tpr_cart, label='CART (aire = %0.2f)' % roc_auc_cart)\n    \n    probas_svm = svm_model.predict_proba(XTest)\n    fpr_svm, tpr_svm, thresholds_svm = roc_curve(yTest, probas_svm[:, 1])\n    roc_auc_svm = auc(fpr_svm, tpr_svm)\n    plt.plot(fpr_svm, tpr_svm, label='SVM (aire = %0.2f)' % roc_auc_svm)\n    \n    plt.plot([0, 1], [0, 1], 'k--')\n    plt.xlim([0.0, 1.0])\n    plt.ylim([0.0, 1.0])\n    plt.xlabel('Taux de faux positifs')\n    plt.ylabel('Taux de vrais positifs')\n    plt.title('Courbe ROC ')\n    plt.legend(loc=\"lower right\")\n    \n    plt.show()\n    \n    err = [round(roc_auc_rl*100,4),round(roc_auc_rf*100,4),round(roc_auc_cart*100,4),round(roc_auc_svm*100,4)]\n    ind = err.index(max(err))","user":"anonymous","dateUpdated":"2019-01-09T14:48:31+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033523812_512047809","id":"20190109-123203_1695114759","dateCreated":"2019-01-09T12:32:03+0100","dateStarted":"2019-01-09T14:48:31+0100","dateFinished":"2019-01-09T14:48:31+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:291"},{"text":"%python\nif ind == 0 :\n    print \"La methode utilisee pour la prediction de votre dataset sera la Regression Logistique avec une precision de \", round(score_glm*100,2), \"% !\\n\"\nelif ind == 1 :\n    print \"La methode utilisee pour la prediction de votre dataset sera Random Forest avec une precision de \", round(score_rf*100,2), \"% !\\n\"\nelif ind == 2 :   \n    print \"La methode utilisee pour la prediction de votre dataset sera CART avec une precision de \", round(score_cart*100,2), \"% !\\n\"\nelse :\n    print \"La methode utilisee pour la prediction de votre dataset sera SVM  avec une precision de \", round(score_svm*100,2), \"% !\\n\"","user":"anonymous","dateUpdated":"2019-01-09T14:48:32+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"results":{"code":"SUCCESS","msg":[]},"apps":[],"jobName":"paragraph_1547033533964_-699271228","id":"20190109-123213_1333759728","dateCreated":"2019-01-09T12:32:13+0100","dateStarted":"2019-01-09T14:48:32+0100","dateFinished":"2019-01-09T14:48:32+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:292"},{"text":"%python\n","user":"anonymous","dateUpdated":"2019-01-09T14:48:32+0100","config":{"colWidth":12,"fontSize":9,"enabled":true,"results":{},"editorSetting":{"language":"python","editOnDblClick":false,"completionKey":"TAB","completionSupport":true},"editorMode":"ace/mode/python"},"settings":{"params":{},"forms":{}},"apps":[],"jobName":"paragraph_1547033549134_1438036247","id":"20190109-123229_329787350","dateCreated":"2019-01-09T12:32:29+0100","status":"FINISHED","progressUpdateIntervalMs":500,"$$hashKey":"object:293"}],"name":"BD","id":"2E1JAB7HV","noteParams":{},"noteForms":{},"angularObjects":{"python:shared_process":[]},"config":{"isZeppelinNotebookCronEnable":false,"looknfeel":"default","personalizedMode":"false"},"info":{}}