Orange: Distributions - Revision history

Onnowpurbo at 02:35, 4 March 2020

2020-03-04T02:35:39Z

Onnowpurbo: /* Input */

2020-03-04T02:34:23Z

Input

Onnowpurbo at 01:01, 3 February 2020

2020-02-03T01:01:50Z

Onnowpurbo at 01:29, 28 January 2020

2020-01-28T01:29:06Z

Onnowpurbo at 03:32, 22 January 2020

2020-01-22T03:32:37Z

Onnowpurbo: Created page with "Sumber: https://docs.biolab.si//3/visual-programming/widgets/visualize/distributions.html Displays value distributions for a single attribute. Inputs Data: input datas..."

2020-01-12T06:46:58Z

Created page with "Sumber: https://docs.biolab.si//3/visual-programming/widgets/visualize/distributions.html Displays value distributions for a single attribute. Inputs Data: input datas..."

New page

Sumber: https://docs.biolab.si//3/visual-programming/widgets/visualize/distributions.html

Displays value distributions for a single attribute.

Inputs

Data: input dataset

The Distributions widget displays the value distribution of discrete or continuous attributes. If the data contains a class variable, distributions may be conditioned on the class.

For discrete attributes, the graph displayed by the widget shows how many times (e.g., in how many instances) each attribute value appears in the data. If the data contains a class variable, class distributions for each of the attribute values will be displayed as well (like in the snapshot below). In order to create this graph, we used the Zoo dataset.

../../_images/Distributions-Disc-stamped.png

A list of variables for distributions display

If Bin continuous variables is ticked, the widget will discretize continuous variables by assigning them to intervals. The number of intervals is set by precision scale. Alternatively, you can set smoothness for the distribution curves of continuous variables.

The widget may be requested to display value distributions only for instances of certain class (Group by). Show relative frequencies will scale the data by percentage of the dataset.

Show probabilities.

Save image saves the graph to your computer in a .svg or .png format.

Produce a report.

For continuous attributes, the attribute values are displayed as a function graph. Class probabilities for continuous attributes are obtained with Gaussian kernel density estimation, while the appearance of the curve is set with the Precision bar (smooth or precise).

For the purpose of this example, we used the Iris dataset.

../../_images/Distributions-Cont.png

In class-less domains, the bars are displayed in gray. Here we set Bin continuous variables into 10 bins, which distributes variables into 10 intervals and displays averages of these intervals as histograms (see 2. above). We used the Housing dataset.

../../_images/Distributions-NoClass.png

==Referensi==

* https://docs.biolab.si//3/visual-programming/widgets/visualize/distributions.html

==Pranala Menarik==

* [[Orange]]

@@ Line 23: / Line 23: @@
 Untuk atribut kontinu, nilai atribut ditampilkan sebagai grafik fungsi. Probabilitas class untuk atribut kontinu diperoleh dengan estimasi kepadatan kernel Gaussian, sedangkan tampilan kurva diatur dengan Precision bar (smooth atau precise).
-For the purpose of this example, we used the Iris dataset.
+Sebagai contoh, di bawah ini kita menggunakan dataset Iris.
 [[File:Distributions-Cont.png|center|200px|thumb]]

@@ Line 8: / Line 8: @@
   Data: input dataset
-Widget Distribusi menampilkan distribusi nilai atribut diskrit atau kontinu. Jika data berisi variabel class, distribusi dapat dikondisikan di class.
+Widget Distributions menampilkan distribusi nilai atribut diskrit atau kontinu. Jika data berisi variabel class, distribusi dapat dikondisikan di class.
 Untuk atribut diskrit, grafik yang ditampilkan oleh widget menunjukkan berapa kali (mis., berapa banyak instance) setiap nilai atribut muncul dalam data. Jika data berisi variabel class, distribusi class untuk setiap nilai atribut akan ditampilkan juga (seperti dalam snapshot di bawah). Untuk membuat grafik ini, kita menggunakan dataset Zoo.
@@ Line 30: / Line 30: @@
 [[File:Distributions-NoClass.png|center|200px|thumb]]
 ==Referensi==

@@ Line 2: / Line 2: @@
-Displays value distributions for a single attribute.
+Menampilkan distribusi nilai untuk satu atribut.
 ==Input==
@@ Line 8: / Line 8: @@
   Data: input dataset
-The Distributions widget displays the value distribution of discrete or continuous attributes. If the data contains a class variable, distributions may be conditioned on the class.
+Widget Distribusi menampilkan distribusi nilai atribut diskrit atau kontinu. Jika data berisi variabel class, distribusi dapat dikondisikan di class.
-For discrete attributes, the graph displayed by the widget shows how many times (e.g., in how many instances) each attribute value appears in the data. If the data contains a class variable, class distributions for each of the attribute values will be displayed as well (like in the snapshot below). In order to create this graph, we used the Zoo dataset.
+Untuk atribut diskrit, grafik yang ditampilkan oleh widget menunjukkan berapa kali (mis., berapa banyak instance) setiap nilai atribut muncul dalam data. Jika data berisi variabel class, distribusi class untuk setiap nilai atribut akan ditampilkan juga (seperti dalam snapshot di bawah). Untuk membuat grafik ini, kita menggunakan dataset Zoo.
 [[File:Distributions-Disc-stamped.png|center|200px|thumb]]
@@ Line 21: / Line 21: @@
 * Produce a report.
-For continuous attributes, the attribute values are displayed as a function graph. Class probabilities for continuous attributes are obtained with Gaussian kernel density estimation, while the appearance of the curve is set with the Precision bar (smooth or precise).
+Untuk atribut kontinu, nilai atribut ditampilkan sebagai grafik fungsi. Probabilitas class untuk atribut kontinu diperoleh dengan estimasi kepadatan kernel Gaussian, sedangkan tampilan kurva diatur dengan Precision bar (smooth atau precise).
 For the purpose of this example, we used the Iris dataset.
@@ Line 27: / Line 27: @@
 [[File:Distributions-Cont.png|center|200px|thumb]]
-In class-less domains, the bars are displayed in gray. Here we set Bin continuous variables into 10 bins, which distributes variables into 10 intervals and displays averages of these intervals as histograms (see 2. above). We used the Housing dataset.
+Dalam class-less domain, bar yang ditampilkan berwarna abu-abu. Di sini kita set variabel kontinu menjadi 10 bin, yang mendistribusikan variabel ke dalam 10 interval dan menampilkan rata-rata interval ini sebagai histogram (lihat 2. di atas). Kita menggunakan dataset Housing.
 [[File:Distributions-NoClass.png|center|200px|thumb]]

@@ Line 4: / Line 4: @@
 Displays value distributions for a single attribute.
-Inputs
+==Input==
-    Data: input dataset
+ Data: input dataset
 The Distributions widget displays the value distribution of discrete or continuous attributes. If the data contains a class variable, distributions may be conditioned on the class.
@@ Line 14: / Line 14: @@
 [[File:Distributions-Disc-stamped.png|center|200px|thumb]]
-    A list of variables for distributions display
+* A list of variables for distributions display
+* If Bin continuous variables is ticked, the widget will discretize continuous variables by assigning them to intervals. The number of intervals is set by precision scale. Alternatively, you can set smoothness for the distribution curves of continuous variables.
-    If Bin continuous variables is ticked, the widget will discretize continuous variables by assigning them to intervals. The number of intervals is set by precision scale. Alternatively, you can set smoothness for the distribution curves of continuous variables.
+* The widget may be requested to display value distributions only for instances of certain class (Group by). Show relative frequencies will scale the data by percentage of the dataset.
+* Show probabilities.
-    The widget may be requested to display value distributions only for instances of certain class (Group by). Show relative frequencies will scale the data by percentage of the dataset.
+* Save image saves the graph to your computer in a .svg or .png format.
+* Produce a report.
 For continuous attributes, the attribute values are displayed as a function graph. Class probabilities for continuous attributes are obtained with Gaussian kernel density estimation, while the appearance of the curve is set with the Precision bar (smooth or precise).