Orange: The Data - Revision history

Onnowpurbo: /* Data Selection and Sampling */

2020-01-09T08:28:00Z

Data Selection and Sampling

← Older revision		Revision as of 08:28, 9 January 2020
Line 337:		Line 337:
	203		203

	==Data Selection ~~and~~ Sampling==		==Data Selection dan Sampling==

	~~Besides the name of the~~ data ~~file~~, Orange.data.Table ~~can accept the~~ data ~~domain and a~~ list of data ~~items and returns a new~~ dataset. ~~This is useful for any~~ data ~~subsetting~~:		Selain nama file data, Orange.data.Table dapat menerima domain data dan list item data dan mengembalikan sebagai dataset baru. Ini berguna untuk setiap subset data:

	data = Orange.data.Table("iris.tab")		data = Orange.data.Table("iris.tab")
Line 346:		Line 346:
	print("Subset size:", len(subset))		print("Subset size:", len(subset))

	~~The~~ code ~~outputs~~:		Output code ini:

	Dataset instances: 150		Dataset instances: 150
	Subset size: 99		Subset size: 99

	~~and inherits the~~ data ~~description~~ (domain) ~~from the original~~ dataset. ~~Changing the~~ domain ~~requires setting up a new~~ domain ~~descriptor~~. ~~This feature is useful for any kind of feature selection~~:		dan inherit deskripsi data (domain) dari dataset asli. Mengubah domain memerlukan pengaturan deskriptor domain baru. Fitur ini berguna untuk segala jenis pemilihan fitur:

	data = Orange.data.Table("iris.tab")		data = Orange.data.Table("iris.tab")
Line 363:		Line 363:
	print(new_data[0])		print(new_data[0])

	~~We could also construct a random sample of the~~ dataset:		Kita juga dapat membuat sampel acak dataset:

	>>> sample = Orange.data.Table(data.domain, random.sample(data, 3))		>>> sample = Orange.data.Table(data.domain, random.sample(data, 3))
Line 372:		Line 372:
	]		]

	~~or randomly sample the attributes~~:		atau sampel acak dari atribut:

	>>> atts = random.sample(data.domain.attributes, 2)		>>> atts = random.sample(data.domain.attributes, 2)
Line 379:		Line 379:
	>>> new_data[0]		>>> new_data[0]
	[5.100, 1.400 \| Iris-setosa]		[5.100, 1.400 \| Iris-setosa]







	==Referensi==		==Referensi==

Onnowpurbo: /* Missing Values */

2020-01-09T08:24:35Z

Missing Values

← Older revision		Revision as of 08:24, 9 January 2020
Line 306:		Line 306:
	Untuk membangun domain classless, kita bisa memasukan None untuk nilai class.		Untuk membangun domain classless, kita bisa memasukan None untuk nilai class.

	==Missing ~~Values~~==		==Missing Value==

	~~Consider the following exploration of the~~ dataset ~~on votes of the~~ US ~~senate~~:		Pertimbangkan eksplorasi dataset berikut tentang vote pada senat US:

	>>> import numpy as np		>>> import numpy as np
Line 319:		Line 319:
	False		False

	~~The particular~~ data ~~instance included missing~~ data (~~represented with~~ ‘?’) ~~for the first and the fourth attribute~~. ~~In the original~~ dataset ~~file~~, ~~the missing values are~~, by default, ~~represented with a blank space~~. ~~We can now examine each attribute and report on proportion of~~ data ~~instances for which this feature was undefined~~:		Contoh data termasuk didalamnya data yang hilang (diwakili dengan ‘?’) Untuk atribut pertama dan keempat. Dalam file dataset asli, nilai-nilai yang hilang, secara default, diwakili dengan ruang kosong. Kita sekarang dapat memeriksa setiap atribut dan melaporkan proporsi instance data yang fitur ini tidak terdefinisi:

	data = Orange.data.Table("voting.tab")		data = Orange.data.Table("voting.tab")
Line 326:		Line 326:
	print("%4.1f%% %s" % (100.0 * n_miss / len(data), x.name))		print("%4.1f%% %s" % (100.0 * n_miss / len(data), x.name))

	~~First three lines of the~~ output ~~of this script are~~:		Tiga baris pertama dari output skrip ini adalah:

	2.8% handicapped-infants		2.8% handicapped-infants
Line 332:		Line 332:
	2.5% adoption-of-the-budget-resolution		2.5% adoption-of-the-budget-resolution

	~~A single-liner that reports on number of~~ data ~~instances with at least one missing value is~~:		Satu baris yang melaporkan jumlah instance data dengan setidaknya satu nilai yang hilang adalah:

	>>> sum(any(np.isnan(d[x]) for x in data.domain.attributes) for d in data)		>>> sum(any(np.isnan(d[x]) for x in data.domain.attributes) for d in data)

Onnowpurbo: /* Meta Attributes */

2020-01-09T08:21:42Z

Meta Attributes

← Older revision		Revision as of 08:21, 9 January 2020
Line 246:		Line 246:
	[big, 1.400, square \| 13.400]		[big, 1.400, square \| 13.400]

	==Meta ~~Attributes~~==		==Meta Attribute==

	~~Often~~, ~~we wish to include descriptive fields in the~~ data ~~that will not be used in any computation~~ (~~distance estimation~~, ~~modeling~~), ~~but will serve for identification or additional information~~. ~~These are called~~ meta ~~attributes~~, ~~and are marked with~~ meta ~~in the third~~ header ~~row~~:		Seringkali, kita ingin memasukkan deskriptive field dalam data yang tidak akan digunakan dalam perhitungan apa pun (estimasi jarak, pemodelan), tetapi akan berfungsi untuk identifikasi atau informasi tambahan. Ini disebut atribut meta, dan ditandai dengan meta di baris header ketiga:

	name hair eggs milk backbone legs type		name hair eggs milk backbone legs type
Line 258:		Line 258:
	bear 1 0 1 1 4 mammal		bear 1 0 1 1 4 mammal

	~~Values of~~ meta ~~attributes and all other~~ (non-meta) ~~attributes are treated similarly in~~ Orange, ~~but stored in separate~~ numpy ~~arrays~~:		Nilai atribut meta dan semua atribut lainnya (non-meta) diperlakukan serupa di Orange, tetapi disimpan dalam array numpy yang terpisah:

	>>> data = Orange.data.Table("zoo")		>>> data = Orange.data.Table("zoo")
Line 281:		Line 281:
	['bear']], dtype=object))		['bear']], dtype=object))

	~~Meta attributes may be passed to~~ Orange.data.Table ~~after providing arrays for attribute and~~ class ~~values~~:		Atribut meta akan diteruskan ke Orange.data.Table setelah menyediakan array untuk nilai atribut dan class:

	from Orange.data import Table, Domain		from Orange.data import Table, Domain
Line 299:		Line 299:
	print(data)		print(data)

	~~The~~ script ~~outputs~~:		Output script:

	[[2.200, 1625.000 \| no] {houston, 10},		[[2.200, 1625.000 \| no] {houston, 10},
	[0.300, 163.000 \| yes] {ljubljana, -1}		[0.300, 163.000 \| yes] {ljubljana, -1}

	~~To construct a~~ classless ~~domain we could pass~~ None ~~for the~~ class ~~values~~.		Untuk membangun domain classless, kita bisa memasukan None untuk nilai class.

	==Missing Values==		==Missing Values==

Onnowpurbo: /* Orange Dataset dan NumPy */

2020-01-09T08:16:46Z

Orange Dataset dan NumPy

← Older revision		Revision as of 08:16, 9 January 2020
Line 195:		Line 195:
	==Orange Dataset dan NumPy==		==Orange Dataset dan NumPy==

	Dataset Orange sebenarnya di wrapped oleh NumPy array. Wrapping dilakukan untuk menyimpan informasi tentang nama-nama dan nilai-nilai fitur, dan NumPy ~~arraydigunakan~~ untuk kecepatan dan kompatibilitas dengan machine learning toolbox yang berbeda, seperti scikit-learning, yang menjadi sandaran Orange. Mari kita tampilkan nilai array ini untuk tiga contoh data pertama dari dataset iris:		Dataset Orange sebenarnya di wrapped oleh NumPy array. Wrapping dilakukan untuk menyimpan informasi tentang nama-nama dan nilai-nilai fitur, dan NumPy array digunakan untuk kecepatan dan kompatibilitas dengan machine learning toolbox yang berbeda, seperti scikit-learning, yang menjadi sandaran Orange. Mari kita tampilkan nilai array ini untuk tiga contoh data pertama dari dataset iris:

	>>> data = Orange.data.Table("iris")		>>> data = Orange.data.Table("iris")
Line 205:		Line 205:
	array([ 0., 0., 0.])		array([ 0., 0., 0.])

	~~Notice that we access the arrays for attributes and~~ class ~~separately~~, ~~using~~ data.X ~~and~~ data.Y. ~~Average values of attributes can then be computed efficiently by~~:		Perhatikan bahwa kita mengakses array untuk atribut dan class secara terpisah, menggunakan data.X dan data.Y. Nilai rata-rata atribut kemudian dapat dihitung secara efisien dengan:

	>>> import np as numpy		>>> import np as numpy
Line 211:		Line 211:
	array([ 5.84333333, 3.054 , 3.75866667, 1.19866667])		array([ 5.84333333, 3.054 , 3.75866667, 1.19866667])

	~~We can also construct a~~ (classless) ~~dataset from a~~ numpy array:		Kita juga dapat membuat dataset (classless) dari numpy array:

	>>> X = np.array([[1,2], [4,5]])		>>> X = np.array([[1,2], [4,5]])
Line 218:		Line 218:
	[Feature 1, Feature 2]		[Feature 1, Feature 2]

	~~If we want to provide meaninful names to attributes~~, ~~we need to construct an appropriate~~ data ~~domain~~:		Jika kita ingin memberikan nama yang bermakna bagi atribut, kita perlu membuat domain data yang sesuai:

	>>> domain = Orange.data.Domain([Orange.data.ContinuousVariable("lenght"),		>>> domain = Orange.data.Domain([Orange.data.ContinuousVariable("lenght"),
Line 226:		Line 226:
	[lenght, width]		[lenght, width]

	~~Here is another example~~, ~~this time with the construction of a~~ dataset ~~that includes a numerical~~ class ~~and different types of attributes~~:		Berikut adalah contoh lain, kali ini dengan konstruksi dataset yang mencakup numeric class dan berbagai jenis atribut:

	size = Orange.data.DiscreteVariable("size", ["small", "big"])		size = Orange.data.DiscreteVariable("size", ["small", "big"])
Line 232:		Line 232:
	shape = Orange.data.DiscreteVariable("shape", ["circle", "square", "oval"])		shape = Orange.data.DiscreteVariable("shape", ["circle", "square", "oval"])
	speed = Orange.data.ContinuousVariable("speed")		speed = Orange.data.ContinuousVariable("speed")
			domain = Orange.data.Domain([size, height, shape], speed)
	domain = Orange.data.Domain([size, height, shape], speed)

	X = np.array([[1, 3.4, 0], [0, 2.7, 2], [1, 1.4, 1]])		X = np.array([[1, 3.4, 0], [0, 2.7, 2], [1, 1.4, 1]])
Line 241:		Line 240:
	print(data)		print(data)

	~~Running of this scripts yields~~:		Menjalankan script ini akan menghasilkan:

	[[big, 3.400, circle \| 42.000],		[[big, 3.400, circle \| 42.000],

Onnowpurbo: /* Orange Dataset dan NumPy */

2020-01-09T04:32:55Z

Orange Dataset dan NumPy

← Older revision		Revision as of 04:32, 9 January 2020
Line 195:		Line 195:
	==Orange Dataset dan NumPy==		==Orange Dataset dan NumPy==

	Orange ~~datasets are actually~~ wrapped NumPy ~~arrays~~. Wrapping ~~is performed to retain the information about the feature names and values~~, ~~and~~ NumPy ~~arrays are used for speed and compatibility with different~~ machine learning ~~toolboxes~~, ~~like~~ scikit-~~learn~~, ~~on which~~ Orange ~~relies~~. ~~Let us display the values of these arrays for the first three~~ data ~~instances of the~~ iris ~~dataset~~:		Dataset Orange sebenarnya di wrapped oleh NumPy array. Wrapping dilakukan untuk menyimpan informasi tentang nama-nama dan nilai-nilai fitur, dan NumPy arraydigunakan untuk kecepatan dan kompatibilitas dengan machine learning toolbox yang berbeda, seperti scikit-learning, yang menjadi sandaran Orange. Mari kita tampilkan nilai array ini untuk tiga contoh data pertama dari dataset iris:

	>>> data = Orange.data.Table("iris")		>>> data = Orange.data.Table("iris")

Onnowpurbo: /* Orange Datasets and NumPy */

2020-01-09T04:31:13Z

Orange Datasets and NumPy

← Older revision		Revision as of 04:31, 9 January 2020
Line 193:		Line 193:
	print(Counter(str(d.get_class()) for d in data))		print(Counter(str(d.get_class()) for d in data))

	==Orange ~~Datasets and~~ NumPy==		==Orange Dataset dan NumPy==

	Orange datasets are actually wrapped NumPy arrays. Wrapping is performed to retain the information about the feature names and values, and NumPy arrays are used for speed and compatibility with different machine learning toolboxes, like scikit-learn, on which Orange relies. Let us display the values of these arrays for the first three data instances of the iris dataset:		Orange datasets are actually wrapped NumPy arrays. Wrapping is performed to retain the information about the feature names and values, and NumPy arrays are used for speed and compatibility with different machine learning toolboxes, like scikit-learn, on which Orange relies. Let us display the values of these arrays for the first three data instances of the iris dataset:

Onnowpurbo: /* Data Instances */

2020-01-09T04:30:50Z

Data Instances

← Older revision		Revision as of 04:30, 9 January 2020
Line 118:		Line 118:
	Values of attribute 'fuel-type': diesel, gas		Values of attribute 'fuel-type': diesel, gas

	==Data ~~Instances~~==		==Data Instance==

	~~Data table stores~~ data ~~instances~~ (~~or examples~~). ~~These can be indexed or~~ traversed ~~as any~~ Python list. ~~Data instances can be considered as vectors~~, ~~accessed through element index~~, ~~or through feature name~~.		Tabel data menyimpan instance data (atau contoh). Ini dapat diindeks atau dilalui (traversed) seperti Python list. Instance data dapat dianggap sebagai vektor, diakses melalui indeks elemen, atau melalui nama fitur.

	import Orange		import Orange
Line 136:		Line 136:
	print("The 3rd value of the 25th data instance:", data[24][2])		print("The 3rd value of the 25th data instance:", data[24][2])

	~~The script above displays the following~~ output:		Script di atas menghasilkan output berikut:

	First three data instances:		First three data instances:
Line 147:		Line 147:
	The 3rd value of the 25th data instance: 1.900		The 3rd value of the 25th data instance: 1.900

	~~The~~ Iris ~~dataset we have used above has four continuous attributes~~. ~~Here’s a script that computes their mean~~:		Dataset Iris yang kami gunakan di atas memiliki empat atribut kontinu. Berikut ini skrip yang menghitung rata-ratanya:

	average = lambda x: sum(x) / len(x)		average = lambda x: sum(x) / len(x)
Line 156:		Line 156:
	print("%-15s %.2f" % (x.name, average([d[x] for d in data])))		print("%-15s %.2f" % (x.name, average([d[x] for d in data])))

	~~The above script also illustrates indexing of~~ data ~~instances with objects that store features~~; in d[x] ~~variable~~ x ~~is an~~ Orange ~~object~~. ~~Here’s the output~~:		Script di atas juga menggambarkan pengindeksan instance data dengan objek yang menyimpan fitur; dalam d [x] variabel x adalah objek Orange. Inilah hasilnya:

	Feature Mean		Feature Mean
Line 164:		Line 164:
	petal width 1.20		petal width 1.20

	~~A slightly more complicated~~, ~~but also more interesting~~, ~~code that computes~~ per-class ~~averages~~:		Kode yang sedikit lebih rumit, tetapi juga lebih menarik, yang menghitung rata-rata per-class:

	average = lambda xs: sum(xs) / float(len(xs))		average = lambda xs: sum(xs) / float(len(xs))
Line 177:		Line 177:
	print("%-15s" % a.name, " ".join(dist))		print("%-15s" % a.name, " ".join(dist))

	~~Of the four features~~, ~~petal width and length look quite discriminative for the type of~~ iris:		Dari empat fitur, kelopak lebar dan panjang terlihat cukup diskriminatif untuk jenis iris:

	Feature Iris-setosa Iris-versicolor Iris-virginica		Feature Iris-setosa Iris-versicolor Iris-virginica
Line 185:		Line 185:
	petal width 0.24 1.33 2.03		petal width 0.24 1.33 2.03

	~~Finally~~, ~~here is a quick code that computes the class distribution for another~~ dataset:		Akhirnya, berikut adalah kode cepat yang menghitung distribusi kelas untuk dataset lainnya:

	import Orange		import Orange

Onnowpurbo: /* Explorasi Data Domain */

2020-01-09T04:21:00Z

Explorasi Data Domain

← Older revision		Revision as of 04:21, 9 January 2020
Line 84:		Line 84:
	==Explorasi Data Domain==		==Explorasi Data Domain==

	~~Data table stores information on~~ data ~~instances as well as on~~ data domain. Domain ~~holds the names of attributes~~, ~~optional classes~~, ~~their types and~~, ~~and if categorical~~, ~~the value names~~. ~~The following code~~:		Tabel data menyimpan informasi tentang instance data serta domain data. Domain menyimpan nama atribut, opsional class, jenisnya, dan, jika kategorikal, nama nilai. Kode berikut:

	import Orange		import Orange
Line 101:		Line 101:
	print("Class:", data.domain.class_var.name)		print("Class:", data.domain.class_var.name)

	~~outputs~~:		output:

	25 attributes: 14 continuous, 11 discrete		25 attributes: 14 continuous, 11 discrete
Line 107:		Line 107:
	Class: price		Class: price

	~~Orange’s objects often behave like~~ Python ~~lists and dictionaries~~, ~~and can be indexed or accessed through feature names~~:		Objek Orange sering berperilaku seperti Python list atau dictionary, dan dapat diindeks atau diakses melalui nama fitur:

	print("First attribute:", data.domain[0].name)		print("First attribute:", data.domain[0].name)
Line 113:		Line 113:
	print("Values of attribute '%s': %s" % (name, ", ".join(data.domain[name].values)))		print("Values of attribute '%s': %s" % (name, ", ".join(data.domain[name].values)))

	~~The~~ output ~~of the above~~ code is:		output code di atas adalah:

	First attribute: symboling		First attribute: symboling

Onnowpurbo: /* Exploration Data Domain */

2020-01-09T04:18:36Z

Exploration Data Domain

← Older revision		Revision as of 04:18, 9 January 2020
Line 82:		Line 82:
	Kami telah membuat tabel data baru dengan mengirimkan informasi tentang struktur data (data.domain) dan subset instance data.		Kami telah membuat tabel data baru dengan mengirimkan informasi tentang struktur data (data.domain) dan subset instance data.

	==~~Exploration~~ Data Domain==		==Explorasi Data Domain==

	Data table stores information on data instances as well as on data domain. Domain holds the names of attributes, optional classes, their types and, and if categorical, the value names. The following code:		Data table stores information on data instances as well as on data domain. Domain holds the names of attributes, optional classes, their types and, and if categorical, the value names. The following code:

Onnowpurbo: /* Exploration of the Data Domain */

2020-01-09T04:02:41Z

Exploration of the Data Domain

← Older revision		Revision as of 04:02, 9 January 2020
Line 82:		Line 82:
	Kami telah membuat tabel data baru dengan mengirimkan informasi tentang struktur data (data.domain) dan subset instance data.		Kami telah membuat tabel data baru dengan mengirimkan informasi tentang struktur data (data.domain) dan subset instance data.

	==Exploration ~~of the~~ Data Domain==		==Exploration Data Domain==

	Data table stores information on data instances as well as on data domain. Domain holds the names of attributes, optional classes, their types and, and if categorical, the value names. The following code:		Data table stores information on data instances as well as on data domain. Domain holds the names of attributes, optional classes, their types and, and if categorical, the value names. The following code: