From ad019b600d3f8600d2ae5c0d8ab548b30c7d1882 Mon Sep 17 00:00:00 2001
From: Andreas Mueller <amueller@ais.uni-bonn.de>
Date: Sun, 18 Dec 2011 00:13:03 +0100
Subject: [PATCH] DOC: Corrections and additions to the dataset docs. Also more
detailed docstrings for the functions loading the data.
---
doc/datasets/index.rst | 10 ++++++----
sklearn/datasets/base.py | 41 ++++++++++++++++++++++++++++++++--------
2 files changed, 39 insertions(+), 12 deletions(-)
diff --git a/doc/datasets/index.rst b/doc/datasets/index.rst
index f44ce2de9c..8f2a973305 100644
--- a/doc/datasets/index.rst
+++ b/doc/datasets/index.rst
@@ -28,7 +28,8 @@ that comes from the 'real world'.
General dataset API
===================
-There are three distinct kinds of dataset interfaces used at the moment.
+There are three distinct kinds of dataset interfaces for different types
+of datasets.
The simplest one is the interface for sample images, which is described
below in the :ref: _Sample_images section.
@@ -41,9 +42,10 @@ fetched from mldata.org have more sophisticated structure.
These functions return a ``bunch`` (which is a dictionary that is
accessible with the 'dict.key' syntax).
All datasets have at least two keys, ``data``, containg an array of shape
-``n_samples x n_features`` and ``target``, a numpy array of length ``n_features``,
-containing the targets.
-The datasets also contain a description in ``DESC`` and some contain
+``n_samples x n_features`` (except for 20newsgroups) and ``target``, a numpy
+array of length ``n_features``, containing the targets.
+
+The datasets also contain a description in ``DESCR`` and some contain
``feature_names`` and ``target_names``.
See the dataset descriptions below for details.
diff --git a/sklearn/datasets/base.py b/sklearn/datasets/base.py
index bfb3674cba..2e485027c8 100644
--- a/sklearn/datasets/base.py
+++ b/sklearn/datasets/base.py
@@ -180,13 +180,21 @@ def load_files(container_path, description=None, categories=None,
def load_iris():
"""Load and return the iris dataset (classification).
+ Classes: 3
+ Samples per class: 50
+ Samples total: 150
+ Dimensionality: 4
+ Features: real, positive
+
+ The iris dataset is a classic and very easy multi-class classification dataset.
Return
------
data : Bunch
Dictionary-like object, the interesting attributes are:
'data', the data to learn, 'target', the classification labels,
- 'target_names', the meaning of the labels, and 'DESCR', the
+ 'target_names', the meaning of the labels, 'feature_names', the
+ meaning of the features, and 'DESCR', the
full description of the dataset.
Examples
@@ -224,6 +232,13 @@ def load_iris():
def load_digits(n_class=10):
"""Load and return the digits dataset (classification).
+ Classes: 10
+ Samples per class: ~180
+ Samples total: 1797
+ Dimensionality: 64
+ Features: integers 0-16
+
+ Each datapoint is a 8x8 image of a digit.
Parameters
----------
@@ -234,7 +249,7 @@ def load_digits(n_class=10):
------
data : Bunch
Dictionary-like object, the interesting attributes are:
- 'data', the data to learn, `images`, the images corresponding
+ 'data', the data to learn, 'images', the images corresponding
to each sample, 'target', the classification labels for each
sample, 'target_names', the meaning of the labels, and 'DESCR',
the full description of the dataset.
@@ -276,11 +291,16 @@ def load_digits(n_class=10):
def load_diabetes():
"""Load and return the diabetes dataset (regression).
+ Samples total: 442
+ Dimensionality: 1r03
+ Features: real, -.2 < x < .2
+ Targets: integer 25 - 346
+
Return
------
data : Bunch
Dictionary-like object, the interesting attributes are:
- 'data', the data to learn and 'target', the labels for each
+ 'data', the data to learn and 'target', the regression target for each
sample.
"""
base_dir = join(dirname(__file__), 'data')
@@ -322,11 +342,16 @@ def load_linnerud():
def load_boston():
"""Load and return the boston house-prices dataset (regression).
+ Samples total: 506
+ Dimensionality: 13
+ Features: real, positive
+ Targets: real 5. - 50.
+
Return
------
data : Bunch
Dictionary-like object, the interesting attributes are:
- 'data', the data to learn, 'target', the classification labels,
+ 'data', the data to learn, 'target', the regression targets,
'target_names', the meaning of the labels, and 'DESCR', the
full description of the dataset.
@@ -361,14 +386,14 @@ def load_boston():
def load_sample_images():
"""Load sample images for image manipulation.
+ Loads both, ``china`` and ``flower``.
Return
------
data : Bunch
- Dictionary-like object, the interesting attributes are:
- 'data', the data to learn, `images`, the images corresponding
- to each sample, 'target', the classification labels for each
- sample, 'target_names', the meaning of the labels, and 'DESCR',
+ Dictionary-like object with the following attributes :
+ 'images', the two sample images, 'filenames', the file
+ names for the images, and 'DESCR'
the full description of the dataset.
Examples
--
GitLab