Modelo de Clasificación Logistica¶

En este libro se implementara el analisis de datos que permitira identificar los estudiantes con altas probabilidades de fracaso en la asignatura de matematicas utilizando herramienta de machine learning

Importación de Librerias¶

# Libreria para importar las funciones que facilitan la creacion del modelo usando clasificación
import turicreate as tr

Carga de los datos¶

Se realiza el cargue de la informacion del documento DataEstudiantes

# Se determina el archivo a trabajar y se carga en una variable
# ruta = 'DataEstudiantes.csv'
ruta = 'https://drive.google.com/uc?id=12y4Yirhk1B-Cy6jMTeMPa1SHHdxAuhFm&export=download&authuser=0'
students=tr.SFrame.read_csv(ruta, '&')

Finished parsing file https://drive.google.com/uc?id=12y4Yirhk1B-Cy6jMTeMPa1SHHdxAuhFm&export=download&authuser=0

Parsing completed. Parsed 100 lines in 0.043787 secs.

------------------------------------------------------
Inferred types from first 100 line(s) of file as 
column_type_hints=[str,str,str,int,str,int,int,float,int,float,float,float,float,float,float,float,float,float,float,float,float,float,float,str,str,str,str]
If parsing fails due to incorrect types, you can correct
the inferred type list above and pass it to read_csv in
the column_type_hints argument
------------------------------------------------------

Finished parsing file https://drive.google.com/uc?id=12y4Yirhk1B-Cy6jMTeMPa1SHHdxAuhFm&export=download&authuser=0

Parsing completed. Parsed 214 lines in 0.045633 secs.

Exploración de los Datos¶

Se visualizan las columnas de los datos que tiene la sábana de notas

# El nombre de las columnas cargadas en la variable head
students.head()

students['CALCULO']

dtype: float
Rows: 214
[75.66666667, 69.33333333, -1.0, 77.0, 70.33333333, -1.0, 73.66666667, -1.0, 64.0, -1.0, -1.0, 75.0, 74.33333333, -1.0, 70.3, 72.0, -1.0, 64.66666667, 66.0, 73.33333333, 72.0, 76.66666667, 70.0, 77.0, 69.33333333, 73.33333333, -1.0, 54.3, 54.3, 54.3, 54.3, -1.0, 79.33333333, 82.66666667, 68.0, -1.0, 61.33333333, 86.0, 73.33333333, 78.66666667, 71.33333333, 78.33333333, 92.0, 51.0, 66.66666667, 62.33333333, 74.66666667, 77.66666667, 77.66666667, 87.66666667, 65.66666667, 60.66666667, -1.0, 88.0, 85.66666667, 76.0, -1.0, 64.66666667, 54.3, 54.3, 54.3, 54.3, 54.3, 54.3, -1.0, 87.66666667, 67.66666667, 40.0, 67.0, 79.0, 62.0, 86.33333333, 72.0, 84.3, 65.33333333, 87.33333333, 75.0, 63.0, 65.0, 58.66666667, 78.0, 74.3, 71.66666667, 64.0, 61.0, 74.0, 67.7, 64.66666667, 69.66666667, 54.3, 54.3, 54.3, 54.3, 54.3, 54.3, 65.33333333, -1.0, 78.66666667, -1.0, 70.0, ... ]

Definir los valores para trabajar la regresion logistica¶

Se define el tipo de datos para la columna de calculo y se determina la cantidad de datos a trabajar

# Identidica el tipo de dato que tiene la columna
students['CALCULO'].dtype
# Se convierte la columna de string a float para facilitar su manipulación
students['CALCULO'] = students['CALCULO'].astype(float)

# Se define que los valores menores 60 el estudiante pierde la nota mayor de 60 la pasa para matematicas
students['logic_math'] = students['CALCULO'] <= 60

# Mostrar los Resultados
students['logic_math']

dtype: int
Rows: 214
[0, 0, 1, 0, 0, 1, 0, 1, 0, 1, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 0, ... ]

Entrenamiento de los Datos¶

Se establecen los datos de entrenamiento de y de prueba para usarlos en el modelo de regresion logistica

# Se crean la variables que tendran los datos de entrenamiento y los datos de prueba
train_data,test_data = students.random_split(.8,seed=0)

# Se establece el modelo de regresion logistica
student_model = tr.logistic_classifier.create(train_data,
                                              target='logic_math',
                                              features=['CALCULO'],
                                              validation_set=test_data)

Logistic regression:

--------------------------------------------------------

Number of examples          : 181

Number of classes           : 2

Number of feature columns   : 1

Number of unpacked features : 1

Number of coefficients      : 2

Starting Newton Method

--------------------------------------------------------

+-----------+----------+--------------+-------------------+---------------------+

Realizar una Evaluacion del Modelo¶

Se realizar la evaluacion del modelo creado

# Se evalua el desempeño del modelo
student_model.evaluate(test_data, metric='roc_curve')

{'roc_curve': Columns:
 	threshold	float
 	fpr	float
 	tpr	float
 	p	int
 	n	int
 
 Rows: 100001
 
 Data:
 +-----------+------+-----+---+----+
 | threshold | fpr  | tpr | p | n  |
 +-----------+------+-----+---+----+
 |    0.0    | 1.0  | 1.0 | 8 | 25 |
 |   1e-05   | 1.0  | 1.0 | 8 | 25 |
 |   2e-05   | 0.96 | 1.0 | 8 | 25 |
 |   3e-05   | 0.96 | 1.0 | 8 | 25 |
 |   4e-05   | 0.92 | 1.0 | 8 | 25 |
 |   5e-05   | 0.92 | 1.0 | 8 | 25 |
 |   6e-05   | 0.92 | 1.0 | 8 | 25 |
 |   7e-05   | 0.92 | 1.0 | 8 | 25 |
 |   8e-05   | 0.92 | 1.0 | 8 | 25 |
 |   9e-05   | 0.92 | 1.0 | 8 | 25 |
 +-----------+------+-----+---+----+
 [100001 rows x 5 columns]
 Note: Only the head of the SFrame is printed.
 You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.}

Aplicar el Clasificador de Estudiantes¶

Se crea una columna donde se establecen los datos de predichos de los estudiantes

students['predicted_students'] = student_model.predict(students, output_type = 'probability')
students

students_reviews = students[students['S']== 'F']
students_reviews

Ordenar los Estudiantes segun la nota predecida¶

Se ordenan los datos para facilitar su visualización

students_reviews = students_reviews.sort('predicted_students', ascending=False)
students_reviews['predicted_students']

dtype: float
Rows: 116
[0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9999999999976954, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.9150335288960079, 0.6776203092231332, 0.4649815813410904, 0.3930069114067918, 0.3286366658829335, 0.3253972994281613, 0.26435323232706737, 0.23673620018266578, 0.1876970556287616, 0.1876970556287616, 0.14502696179136373, 0.12936056544712063, 0.12936056544712063, 0.09965972287069574, 0.09834675162201748, 0.043764901351169797, 0.03344489579248053, 0.028182379146935152, 0.028182379146935152, 0.028182379146935152, 0.024772290156680934, 0.021765583056233676, 0.021765583056233676, 0.021765583056233676, 0.01830598763459913, 0.01830598763459913, 0.016071186568469967, 0.013902055916266066, 0.012022115947451966, 0.010393716029159982, 0.010393716029159982, 0.010393716029159982, 0.008983877801572069, 0.008983877801572069, 0.005795394780615419, 0.005006132847925033, 0.004323891287538867, 0.004323891287538867, 0.004323891287538867, 0.004323891287538867, 0.002784644607043176, 0.002404418434314465, 0.002404418434314465, 0.00207600172656795, 0.0017923624858579225, 0.0017923624858579225, 0.0015703254850705833, 0.0015703254850705833, 0.0015474161168096251, 0.0013358996021753738, 0.0011532619180817032, 0.0011532619180817032, 0.0011532619180817032, 0.0008594193333371009, 0.0008594193333371009, 0.0006498874369061764, 0.0006498874369061764, 0.0006498874369061764, 0.00047716666114851864, 0.00047716666114851864, 0.00035552681909932683, ... ]

students_reviews['predicted_students'].tail()

dtype: float
Rows: 10
[0.00017286927437613268, 0.00012690965662497222, 3.9095836660137104e-05, 1.0394782159676297e-05, 8.9720301387078e-06, 7.744011168062518e-06, 4.297999072149885e-06, 3.7097213236289467e-06, 6.342188947645615e-07, 4.078152044272551e-07]

Visualizacion de Resultados¶

# Estudiante con mas probabilidades de Perder Calculo
students_reviews[0]

{'ESTUDIANTE': 'CAÑON SUAREZ JESSICA VIVIANA',
 'R': 'N',
 'S': 'F',
 'CURSO': 1001,
 'J10': 'M',
 'Estrato': 2,
 'AÑO': 2016,
 'ECONOMIA': 42.7,
 'FÍSICA': 45,
 'QUÍMICA': 35.0,
 'SOCIALES': 36.66666667,
 'EDUCACION ARTISTICA': 40.0,
 'RECREACIÓN Y DEPORTES': 53.0,
 'RELIGION': 46.33333333,
 'ETICA': 41.66666667,
 'FILOSOFIA': 30.0,
 'INGLES': 42.33333333,
 'LENGUA CASTELLANA': 31.66666667,
 'TRIGONOMETRIA': 36.66666667,
 'ADMINISTRACION': 43.66666667,
 'MANEJO  ORDENADORES': 48.33333333,
 'PREPARACIÓN TECNOLÓGICA': 35.66666667,
 'CALCULO': -1.0,
 'CONVENCIONES MINISTERIO': 'BAJO',
 'C11': 'None',
 'J11': 'None',
 'A11': 'None',
 'logic_math': 1,
 'predicted_students': 0.9999999999976954}

# Estudiante con mas probabilidades de Pasar Calculo
students_reviews[-1]

{'ESTUDIANTE': 'LIZARAZO PINZON DEVI STHEFANY',
 'R': 'N',
 'S': 'F',
 'CURSO': 1003,
 'J10': 'M',
 'Estrato': 2,
 'AÑO': 2017,
 'ECONOMIA': 63.7,
 'FÍSICA': 71,
 'QUÍMICA': 70.0,
 'SOCIALES': 77.7,
 'EDUCACION ARTISTICA': 85.0,
 'RECREACIÓN Y DEPORTES': 72.0,
 'RELIGION': 65.7,
 'ETICA': 79.0,
 'FILOSOFIA': 73.0,
 'INGLES': 69.0,
 'LENGUA CASTELLANA': 75.0,
 'TRIGONOMETRIA': 88.0,
 'ADMINISTRACION': 0.0,
 'MANEJO  ORDENADORES': 96.0,
 'PREPARACIÓN TECNOLÓGICA': 82.0,
 'CALCULO': 93.0,
 'CONVENCIONES MINISTERIO': 'SUPERIOR',
 'C11': '1103',
 'J11': 'M',
 'A11': '2018',
 'logic_math': 0,
 'predicted_students': 4.078152044272551e-07}

students.dtype

dtype: int
Rows: 214
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 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, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 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, 0, 1, 0, 0, 0, 0, 0, ... ]

Realizar el Entrenamiento de los Datos¶

# Se crean la variables de entrenamiento y los datos de prueba
train_data,test_data = students.random_split(.8,seed=0)
# Se establece el modelo de regresion logistica
student_model = tr.logistic_classifier.create(train_data,
                                              target='logic_math',
                                              features=['CALCULO'],
                                              validation_set=test_data)

Logistic regression:

--------------------------------------------------------

Number of examples          : 181

Number of classes           : 2

Number of feature columns   : 1

Number of unpacked features : 1

Number of coefficients      : 2

Starting Newton Method

--------------------------------------------------------

+-----------+----------+--------------+-------------------+---------------------+

Realizar una Evaluacion del Modelo¶

student_model.evaluate(test_data, metric='roc_curve')

{'roc_curve': Columns:
 	threshold	float
 	fpr	float
 	tpr	float
 	p	int
 	n	int
 
 Rows: 100001
 
 Data:
 +-----------+--------------------+-----+----+----+
 | threshold |        fpr         | tpr | p  | n  |
 +-----------+--------------------+-----+----+----+
 |    0.0    |        1.0         | 1.0 | 11 | 22 |
 |   1e-05   | 0.9545454545454546 | 1.0 | 11 | 22 |
 |   2e-05   | 0.9545454545454546 | 1.0 | 11 | 22 |
 |   3e-05   | 0.8636363636363636 | 1.0 | 11 | 22 |
 |   4e-05   | 0.8636363636363636 | 1.0 | 11 | 22 |
 |   5e-05   | 0.8636363636363636 | 1.0 | 11 | 22 |
 |   6e-05   | 0.8636363636363636 | 1.0 | 11 | 22 |
 |   7e-05   | 0.8181818181818182 | 1.0 | 11 | 22 |
 |   8e-05   | 0.8181818181818182 | 1.0 | 11 | 22 |
 |   9e-05   | 0.8181818181818182 | 1.0 | 11 | 22 |
 +-----------+--------------------+-----+----+----+
 [100001 rows x 5 columns]
 Note: Only the head of the SFrame is printed.
 You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.}

Aplicar el Clasificador de Estudiantes¶

students['predicted_students'] = student_model.predict(students, output_type = 'probability')
students

students_reviews = students[students['S']== 'F']
students_reviews

Ordenar los Estudiantes segun la nota predecida¶

students_reviews = students_reviews.sort('predicted_students', ascending=False)
students_reviews['predicted_students']

dtype: float
Rows: 127
[0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.904920040373383, 0.6579719722539445, 0.555309151600646, 0.555309151600646, 0.555309151600646, 0.4476974541278717, 0.37799307185045816, 0.3161029420614644, 0.3130004071722924, 0.25460019669234774, 0.22824400339125306, 0.20386296122907488, 0.20386296122907488, 0.18148031696891703, 0.18148031696891703, 0.18148031696891703, 0.1407787402074131, 0.12581557506243843, 0.12581557506243843, 0.09738962666044967, 0.09613200686352698, 0.04348754503087077, 0.03787525361984694, 0.033424076737770675, 0.033424076737770675, 0.033424076737770675, 0.033424076737770675, 0.033424076737770675, 0.032962297414806274, 0.028268458251096328, 0.028268458251096328, 0.028268458251096328, 0.024917052469684457, 0.024917052469684457, 0.024917052469684457, 0.021954000437429597, 0.021954000437429597, 0.021954000437429597, 0.018533701127187323, 0.018533701127187323, 0.016317033203783967, 0.014159329212896758, 0.014159329212896758, 0.012282864820375829, 0.010652847132509309, 0.010652847132509309, 0.010652847132509309, 0.010652847132509309, 0.009237120642999315, 0.009237120642999315, 0.009237120642999315, 0.007894173106345907, 0.006941014835096334, 0.006015578627988955, 0.005212657809803537, 0.00513856885519102, 0.00513856885519102, 0.004516613047989859, 0.004516613047989859, 0.004516613047989859, 0.004516613047989859, 0.0039131453476086075, 0.002936517674931979, 0.002543628136603734, 0.002543628136603734, 0.0022030938592485414, 0.0019081445059617908, 0.0019081445059617908, 0.001676529937619111, 0.001676529937619111, 0.001652617455790608, 0.0014311981987690182, 0.0012394615676627003, ... ]

students_reviews['predicted_students'].tail()

dtype: float
Rows: 10
[0.0001430093000933575, 4.518036043937327e-05, 1.2358244076025793e-05, 1.0700581786827589e-05, 9.26526652268426e-06, 5.207409790988904e-06, 4.508914096159402e-06, 8.005663555127518e-07, 6.001757719799423e-07, 5.196708987177571e-07]

Visualizacion de Resultados¶

# Estudiante con mas probabilidades de Perder Calculo
students_reviews[0]

{'ESTUDIANTE': 'SOLARTE QUIÑONES LEYDY MARCELA',
 'R': 'N',
 'S': 'F',
 'CURSO': 1001,
 'AÑO': 2016,
 'ECONOMIA': 62.7,
 'FÍSICA': 60,
 'QUÍMICA': 55.33333333,
 'SOCIALES': 61.66666667,
 'EDUCACION ARTISTICA': 62.0,
 'RECREACIÓN Y DEPORTES': 61.66666667,
 'RELIGION': 60.0,
 'ETICA': 65.0,
 'FILOSOFIA': 60.66666667,
 'INGLES': 65.66666667,
 'LENGUA CASTELLANA': 54.33333333,
 'TRIGONOMETRIA': 46.33333333,
 'ADMINISTRACION': 62.66666667,
 'MANEJO ORDENADORES': 75.33333333,
 'PREPARACIÓN TECNOLÓGICA': 75.0,
 'CALCULO': 54.3,
 'C11': 1102,
 'A11': 2018,
 'JORNADA': 'MAÑANA',
 'ESTRATO': 2,
 'logic_math': 1,
 'predicted_students': 0.904920040373383}

ESTUDIANTE	R	S	CURSO	J10	Estrato	AÑO	ECONOMIA	FÍSICA	QUÍMICA	SOCIALES
AGREDO TORRES KAROL MELISA ...	N	F	1001	M	2	2016	69.7	73	60.66666667	72.0
ARIZA MARTINEZ CAMPVERLLY SAMANTHA ...	N	F	1001	M	2	2016	68.7	68	65.33333333	72.66666667
CAÑON SUAREZ JESSICA VIVIANA ...	N	F	1001	M	2	2016	42.7	45	35.0	36.66666667
CASALLAS HUERFANO LINA MARIA ...	N	F	1001	M	2	2016	74.7	63	62.66666667	73.33333333
CHUNZA GOMEZ LIZETH ANDREA ...	N	F	1001	M	2	2016	65.0	53	49.66666667	70.66666667
FARFAN PULIDO ANDRES FELIPE ...	N	M	1001	M	2	2016	43.3	43	33.33333333	23.33333333
GALINDO MORENO MARILYN DANIELA ...	N	F	1001	M	2	2016	60.7	60	57.33333333	64.66666667
GAVIRIA HERRERA SANTIAGO	N	M	1001	M	2	2016	62.3	52	40.33333333	59.0
GOMEZ ROMERO JONATHAN	N	M	1001	M	2	2016	63.0	55	51.33333333	63.33333333
GONZALEZ RODRIGUEZ HENRY JOEL ...	N	M	1001	M	2	2016	61.7	54	54.0	56.66666667

EDUCACION ARTISTICA	RECREACIÓN Y DEPORTES	RELIGION	ETICA	FILOSOFIA	INGLES	LENGUA CASTELLANA
78.33333333	62.33333333	68.33333333	74.66666667	80.33333333	65.66666667	71.66666667
79.33333333	71.66666667	69.33333333	73.66666667	62.33333333	60.0	74.0
40.0	53.0	46.33333333	41.66666667	30.0	42.33333333	31.66666667
78.33333333	67.66666667	76.0	75.33333333	71.0	69.0	72.33333333
57.0	52.66666667	66.66666667	70.33333333	61.0	58.66666667	64.66666667
56.0	47.66666667	48.66666667	23.33333333	38.33333333	42.0	36.66666667
81.66666667	51.0	68.0	69.66666667	61.66666667	59.0	63.33333333
67.33333333	48.33333333	56.33333333	56.66666667	60.0	53.33333333	68.66666667
61.0	66.66666667	60.33333333	73.33333333	60.0	61.0	56.66666667
67.0	66.0	64.33333333	57.0	64.0	60.66666667	58.33333333

TRIGONOMETRIA	ADMINISTRACION	MANEJO ORDENADORES	PREPARACIÓN TECNOLÓGICA	CALCULO	CONVENCIONES MINISTERIO
71.33333333	78.0	79.0	70.33333333	75.66666667	BASICO
65.33333333	62.33333333	85.66666667	68.66666667	69.33333333	BASICO
36.66666667	43.66666667	48.33333333	35.66666667	-1.0	BAJO
67.33333333	77.66666667	79.0	71.33333333	77.0	BASICO
54.0	66.66666667	78.0	67.33333333	70.33333333	BASICO
24.33333333	26.66666667	60.66666667	16.66666667	-1.0	BAJO
68.33333333	66.0	74.33333333	60.66666667	73.66666667	BASICO
54.33333333	65.66666667	63.0	51.0	-1.0	BAJO
66.0	49.33333333	76.66666667	65.0	64.0	BASICO
60.0	66.0	68.66666667	62.33333333	-1.0	BAJO

C11	J11	A11
1101	M	2017
1101	M	2017
None	None	None
1101	M	2017
1101	M	2017
None	None	None
1101	M	2017
None	None	None
1102	M	2017
None	None	None

C11	J11	A11	logic_math	predicted_students
1101	M	2017	0	0.0008594193333371009
1101	M	2017	0	0.013902055916266066
None	None	None	1	0.9999999999976954
1101	M	2017	0	0.00047716666114851864
1101	M	2017	0	0.008983877801572069
None	None	None	1	0.9999999999976954
1101	M	2017	0	0.00207600172656795
None	None	None	1	0.9999999999976954
1102	M	2017	0	0.12936056544712063
None	None	None	1	0.9999999999976954

ESTUDIANTE	R	S	CURSO	AÑO	ECONOMIA	FÍSICA	QUÍMICA	SOCIALES	EDUCACION ARTISTICA
AGREDO TORRES KAROL MELISA ...	N	F	1001	2016	69.7	73	60.66666667	72.0	78.33333333
AGUILERA GUERRERO JOSTEIN STEVEN ...	N	M	1003	2016	73.7	75	70.0	69.0	75.0
ALBARRACIN RODRIGUEZ JESID ALEXANDER ...	N	M	1002	2017	70.7	71	76.0	79.3	85.0
ALBARRACIN RODRIGUEZ JOSET DAVID ...	N	M	1002	2017	73.3	82	79.0	78.7	88.3
ALTUZARRA CASTILLO ANGIE KATHERIN ...	N	F	1001	2017	67.3	78	70.0	86.3	66.7
AMADO PARDO DUVAN CAMILO	N	M	1002	2016	70.3	74	79.0	75.0	78.0
AMAYA VELASQUEZ MIGUEL ANGEL ...	N	M	1003	2017	63.7	67	63.0	75.3	73.3
ANGARITA BELTRAN PAULA ANDREA ...	N	F	1001	2017	64.7	64	44.0	54.7	67.7
AREVALO RODRIGUEZ ASLY NATALY ...	N	F	1002	2016	73.0	73	83.0	73.0	83.3
AREVALO YAMA STIVEN ANDRES ...	N	M	1003	2017	60.0	61	52.0	68.0	55.0

RECREACIÓN Y DEPORTES	RELIGION	ETICA	FILOSOFIA	INGLES	LENGUA CASTELLANA	TRIGONOMETRIA
62.33333333	68.33333333	74.66666667	80.33333333	65.66666667	71.66666667	71.33333333
65.3	81.7	78.7	91.0	74.0	84.0	76.0
74.7	91.3	87.3	83.3	78.0	81.0	91.0
76.7	95.3	83.7	83.3	82.0	95.0	93.3
69.7	93.7	83.7	80.0	77.0	88.0	76.7
68.0	73.3	76.0	70.3	71.0	82.0	61.3
68.0	60.0	76.3	95.0	64.0	71.0	77.3
63.0	33.3	56.7	64.0	58.0	70.0	59.0
70.0	81.0	80.0	85.7	81.0	82.0	67.0
66.7	70.3	69.0	61.3	60.0	60.0	65.3

ADMINISTRACION	MANEJO ORDENADORES	PREPARACIÓN TECNOLÓGICA	CALCULO	C11	A11	JORNADA	ESTRATO
78.0	79.0	70.33333333	75.6667	1101	2017	MAÑANA	2
72.0	95.0	88.0	87.6667	1103	2017	MAÑANA	2
None	95.0	89.0	86.0	1102	2018	MAÑANA	1
None	94.0	91.0	85.7	1102	2018	MAÑANA	2
None	90.0	76.0	76.3	1101	2018	MAÑANA	2
78.0	78.0	70.0	79.3333	1102	2017	MAÑANA	2
None	81.0	67.0	68.3	1103	2018	MAÑANA	2
None	77.0	61.0	63.0	1101	2018	MAÑANA	1
75.0	83.0	88.0	82.6667	1102	2017	MAÑANA	2
None	74.0	65.0	67.0	1103	2018	MAÑANA	2

predicted_students
0.0009294988663506015
5.207409790988904e-06
1.0700581786827589e-05
1.2181687810479954e-05
0.0007071225793659439
0.00019074910389449252
0.021954000437429597
0.18148031696891703
4.518036043937327e-05
0.03787525361984694