pca 0.1.0 → 0.2.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/lib/pca.rb +20 -6
- metadata +4 -4
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 437676549bb33931b4565d105e4a73d5acfc1f35
|
4
|
+
data.tar.gz: da5c93ae3bcee85fa79a7aab62dc14d6da729856
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: bce79847afa2af0cba8305a5eeeccaae29ae651dd9c26a9cef9154971e0680225e61801e179b769dd893544bc4b145b0fea8456cc159984ad882690665a08b13
|
7
|
+
data.tar.gz: a5b103e1813788e5235c04b7cc90cbee3141d9ec49bfb342a2ba0a7e1019307464b1ea65914b1254b70f274d9021aeb0980b7933323f0f9bf6ec4e2e163b4f2f
|
data/lib/pca.rb
CHANGED
@@ -1,10 +1,11 @@
|
|
1
1
|
require 'gsl'
|
2
2
|
|
3
3
|
class PCA
|
4
|
-
attr_reader :components, :singular_values, :mean, :explained_variance, :explained_variance_ratio
|
4
|
+
attr_reader :components, :singular_values, :mean, :std, :explained_variance, :explained_variance_ratio
|
5
5
|
|
6
6
|
def initialize opts = {}
|
7
7
|
@n_components = opts[:components]
|
8
|
+
@scale_data = opts[:scale_data]
|
8
9
|
end
|
9
10
|
|
10
11
|
def fit x
|
@@ -14,7 +15,7 @@ class PCA
|
|
14
15
|
end
|
15
16
|
|
16
17
|
def transform x
|
17
|
-
x = prepare_data x,
|
18
|
+
x = prepare_data x, use_saved_mean_and_std: true
|
18
19
|
_transform x
|
19
20
|
end
|
20
21
|
|
@@ -26,7 +27,8 @@ class PCA
|
|
26
27
|
|
27
28
|
def inverse_transform x
|
28
29
|
x = ensure_matrix x
|
29
|
-
out = x * @components
|
30
|
+
out = x * @components
|
31
|
+
out.size2.times {|col| out.col(col).mul! @std[col] } if @scale_data
|
30
32
|
out.size2.times {|col| out.col(col).add! @mean[col] }
|
31
33
|
out
|
32
34
|
end
|
@@ -34,22 +36,26 @@ class PCA
|
|
34
36
|
private
|
35
37
|
def prepare_data x, opts = {}
|
36
38
|
x = ensure_matrix x
|
37
|
-
@mean = calculate_mean(x) unless opts[:
|
39
|
+
@mean = calculate_mean(x) unless opts[:use_saved_mean_and_std]
|
38
40
|
mean_normalize x
|
41
|
+
if @scale_data
|
42
|
+
@std = calculate_std(x) unless opts[:use_saved_mean_and_std]
|
43
|
+
scale(x)
|
44
|
+
end
|
39
45
|
x
|
40
46
|
end
|
41
47
|
|
42
48
|
def _fit x
|
43
49
|
covariance_matrix = (x.transpose * x) / x.size1
|
44
50
|
u, v, s = covariance_matrix.SV_decomp
|
45
|
-
@components = slice_n
|
51
|
+
@components = slice_n(u).transpose
|
46
52
|
@singular_values = slice_n s
|
47
53
|
@explained_variance = @singular_values**2 / x.size1
|
48
54
|
@explained_variance_ratio = @explained_variance / @explained_variance.sum
|
49
55
|
end
|
50
56
|
|
51
57
|
def _transform x
|
52
|
-
x * @components
|
58
|
+
x * @components.transpose
|
53
59
|
end
|
54
60
|
|
55
61
|
def ensure_matrix x
|
@@ -71,6 +77,14 @@ class PCA
|
|
71
77
|
x.size2.times {|col| x.col(col).sub! @mean[col] }
|
72
78
|
end
|
73
79
|
|
80
|
+
def calculate_std x
|
81
|
+
x.size2.times.map {|col| x.col(col).sd }
|
82
|
+
end
|
83
|
+
|
84
|
+
def scale x
|
85
|
+
x.size2.times {|col| x.col(col).div! @std[col] }
|
86
|
+
end
|
87
|
+
|
74
88
|
def slice_n x
|
75
89
|
return x unless @n_components
|
76
90
|
case x
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: pca
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.2.0
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Geoff Buesing
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2015-
|
11
|
+
date: 2015-04-01 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: rb-gsl
|
@@ -24,7 +24,7 @@ dependencies:
|
|
24
24
|
- - ">="
|
25
25
|
- !ruby/object:Gem::Version
|
26
26
|
version: '0'
|
27
|
-
description: Principal Component Analysis
|
27
|
+
description: Principal Component Analysis (PCA). Uses GSL for calculations.
|
28
28
|
email: gbuesing@gmail.com
|
29
29
|
executables: []
|
30
30
|
extensions: []
|
@@ -54,6 +54,6 @@ rubyforge_project:
|
|
54
54
|
rubygems_version: 2.4.5
|
55
55
|
signing_key:
|
56
56
|
specification_version: 4
|
57
|
-
summary: Principal Component Analysis
|
57
|
+
summary: Principal Component Analysis (PCA)
|
58
58
|
test_files: []
|
59
59
|
has_rdoc:
|