eps 0.3.5 → 0.3.9
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +16 -0
- data/LICENSE.txt +1 -1
- data/README.md +15 -21
- data/lib/eps/base_estimator.rb +21 -11
- data/lib/eps/evaluators/lightgbm.rb +2 -8
- data/lib/eps/lightgbm.rb +7 -4
- data/lib/eps/pmml/loader.rb +1 -1
- data/lib/eps/version.rb +1 -1
- metadata +8 -92
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 49da7dab8a04a1404446f56051ea9faaf5b2260888ef9e2eccf12ec0f5ece887
|
4
|
+
data.tar.gz: 56a85a477e48cb07b39253e182878a947371ec30375b4570987c885a0e6bb93a
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 4a611604c4172110f67e3b39147ddb77346fe98390ee9a31791f61e1da2b4a6f1f6e83f63c4ab722c30d2247797c66328b9acefb51a907f76808793014207dd2
|
7
|
+
data.tar.gz: 1aca1429a26a1eafee5717680507cf04794f639bb13c4ce416709bf0e64ca6017525413f74bcfa7728bd5272ff5b845dfb2da17cdee781eb03ea98a56c3441b3
|
data/CHANGELOG.md
CHANGED
@@ -1,3 +1,19 @@
|
|
1
|
+
## 0.3.9 (2021-10-14)
|
2
|
+
|
3
|
+
- Fixed error with `lessOrEqual` operator
|
4
|
+
|
5
|
+
## 0.3.8 (2021-02-08)
|
6
|
+
|
7
|
+
- Fixed error with categorical and text features
|
8
|
+
|
9
|
+
## 0.3.7 (2020-11-23)
|
10
|
+
|
11
|
+
- Fixed error with LightGBM summary
|
12
|
+
|
13
|
+
## 0.3.6 (2020-06-19)
|
14
|
+
|
15
|
+
- Fixed error with text features for LightGBM
|
16
|
+
|
1
17
|
## 0.3.5 (2020-06-10)
|
2
18
|
|
3
19
|
- Added `learning_rate` option for LightGBM
|
data/LICENSE.txt
CHANGED
data/README.md
CHANGED
@@ -7,7 +7,7 @@ Machine learning for Ruby
|
|
7
7
|
|
8
8
|
Check out [this post](https://ankane.org/rails-meet-data-science) for more info on machine learning with Rails
|
9
9
|
|
10
|
-
[![Build Status](https://
|
10
|
+
[![Build Status](https://github.com/ankane/eps/workflows/build/badge.svg?branch=master)](https://github.com/ankane/eps/actions)
|
11
11
|
|
12
12
|
## Installation
|
13
13
|
|
@@ -134,7 +134,7 @@ For text features, use strings with multiple words.
|
|
134
134
|
{description: "a beautiful house on top of a hill"}
|
135
135
|
```
|
136
136
|
|
137
|
-
This creates features based on word count
|
137
|
+
This creates features based on [word count](https://en.wikipedia.org/wiki/Bag-of-words_model).
|
138
138
|
|
139
139
|
You can specify text features explicitly with:
|
140
140
|
|
@@ -147,12 +147,12 @@ You can set advanced options with:
|
|
147
147
|
```ruby
|
148
148
|
text_features: {
|
149
149
|
description: {
|
150
|
-
min_occurences: 5,
|
151
|
-
max_features: 1000,
|
152
|
-
min_length: 1,
|
153
|
-
case_sensitive: true,
|
154
|
-
tokenizer: /\s+/,
|
155
|
-
stop_words: ["and", "the"]
|
150
|
+
min_occurences: 5, # min times a word must appear to be included in the model
|
151
|
+
max_features: 1000, # max number of words to include in the model
|
152
|
+
min_length: 1, # min length of words to be included
|
153
|
+
case_sensitive: true, # how to treat words with different case
|
154
|
+
tokenizer: /\s+/, # how to tokenize the text, defaults to whitespace
|
155
|
+
stop_words: ["and", "the"] # words to exclude from the model
|
156
156
|
}
|
157
157
|
}
|
158
158
|
```
|
@@ -218,7 +218,7 @@ Build the model with:
|
|
218
218
|
PriceModel.build
|
219
219
|
```
|
220
220
|
|
221
|
-
This saves the model to `price_model.pmml`.
|
221
|
+
This saves the model to `price_model.pmml`. Check this into source control or use a tool like [Trove](https://github.com/ankane/trove) to store it.
|
222
222
|
|
223
223
|
Predict with:
|
224
224
|
|
@@ -343,9 +343,7 @@ df = Daru::DataFrame.from_csv("houses.csv")
|
|
343
343
|
Eps::Model.new(df, target: "price")
|
344
344
|
```
|
345
345
|
|
346
|
-
|
347
|
-
|
348
|
-
When importing data from CSV files, be sure to convert numeric fields. The `table` method does this automatically.
|
346
|
+
When reading CSV files directly, be sure to convert numeric fields. The `table` method does this automatically.
|
349
347
|
|
350
348
|
```ruby
|
351
349
|
CSV.table("data.csv").map { |row| row.to_h }
|
@@ -375,7 +373,11 @@ Eps::Model.new(data, learning_rate: 0.01)
|
|
375
373
|
|
376
374
|
### Linear Regression
|
377
375
|
|
378
|
-
|
376
|
+
By default, an intercept is included. Disable this with:
|
377
|
+
|
378
|
+
```ruby
|
379
|
+
Eps::Model.new(data, intercept: false)
|
380
|
+
```
|
379
381
|
|
380
382
|
To speed up training on large datasets with linear regression, [install GSL](https://github.com/ankane/gslr#gsl-installation). With Homebrew, you can use:
|
381
383
|
|
@@ -391,14 +393,6 @@ gem 'gslr', group: :development
|
|
391
393
|
|
392
394
|
It only needs to be available in environments used to build the model.
|
393
395
|
|
394
|
-
#### Options
|
395
|
-
|
396
|
-
By default, an intercept is included. Disable this with:
|
397
|
-
|
398
|
-
```ruby
|
399
|
-
Eps::Model.new(data, intercept: false)
|
400
|
-
```
|
401
|
-
|
402
396
|
## Probability
|
403
397
|
|
404
398
|
To get the probability of each category for predictions with classification, use:
|
data/lib/eps/base_estimator.rb
CHANGED
@@ -3,6 +3,7 @@ module Eps
|
|
3
3
|
def initialize(data = nil, y = nil, **options)
|
4
4
|
@options = options.dup
|
5
5
|
@trained = false
|
6
|
+
@text_encoders = {}
|
6
7
|
# TODO better pattern - don't pass most options to train
|
7
8
|
train(data, y, **options) if data
|
8
9
|
end
|
@@ -209,29 +210,38 @@ module Eps
|
|
209
210
|
[data, target]
|
210
211
|
end
|
211
212
|
|
212
|
-
def prep_text_features(train_set)
|
213
|
-
@text_encoders = {}
|
213
|
+
def prep_text_features(train_set, fit: true)
|
214
214
|
@text_features.each do |k, v|
|
215
|
-
|
216
|
-
|
215
|
+
if fit
|
216
|
+
# reset vocabulary
|
217
|
+
v.delete(:vocabulary)
|
218
|
+
|
219
|
+
# TODO determine max features automatically
|
220
|
+
# start based on number of rows
|
221
|
+
encoder = Eps::TextEncoder.new(**v)
|
222
|
+
counts = encoder.fit(train_set.columns.delete(k))
|
223
|
+
else
|
224
|
+
encoder = @text_encoders[k]
|
225
|
+
counts = encoder.transform(train_set.columns.delete(k))
|
226
|
+
end
|
217
227
|
|
218
|
-
# TODO determine max features automatically
|
219
|
-
# start based on number of rows
|
220
|
-
encoder = Eps::TextEncoder.new(**v)
|
221
|
-
counts = encoder.fit(train_set.columns.delete(k))
|
222
228
|
encoder.vocabulary.each do |word|
|
223
229
|
train_set.columns[[k, word]] = [0] * counts.size
|
224
230
|
end
|
231
|
+
|
225
232
|
counts.each_with_index do |ci, i|
|
226
233
|
ci.each do |word, count|
|
227
234
|
word_key = [k, word]
|
228
235
|
train_set.columns[word_key][i] = 1 if train_set.columns.key?(word_key)
|
229
236
|
end
|
230
237
|
end
|
231
|
-
@text_encoders[k] = encoder
|
232
238
|
|
233
|
-
|
234
|
-
|
239
|
+
if fit
|
240
|
+
@text_encoders[k] = encoder
|
241
|
+
|
242
|
+
# update vocabulary
|
243
|
+
v[:vocabulary] = encoder.vocabulary
|
244
|
+
end
|
235
245
|
end
|
236
246
|
|
237
247
|
raise "No features left" if train_set.columns.empty?
|
@@ -19,13 +19,7 @@ module Eps
|
|
19
19
|
# sparse matrix
|
20
20
|
@text_features.each do |k, v|
|
21
21
|
encoder = TextEncoder.new(**v)
|
22
|
-
|
23
|
-
values = data.columns.delete(k)
|
24
|
-
counts = encoder.transform(values)
|
25
|
-
|
26
|
-
encoder.vocabulary.each do |word|
|
27
|
-
data.columns[[k, word]] = [0] * values.size
|
28
|
-
end
|
22
|
+
counts = encoder.transform(data.columns[k])
|
29
23
|
|
30
24
|
counts.each_with_index do |xc, i|
|
31
25
|
row = rows[i]
|
@@ -52,7 +46,7 @@ module Eps
|
|
52
46
|
@trees.each_slice(num_trees).each do |trees|
|
53
47
|
tree_scores << sum_trees(rows, trees)
|
54
48
|
end
|
55
|
-
|
49
|
+
rows.size.times.map do |i|
|
56
50
|
v = tree_scores.map { |s| s[i] }
|
57
51
|
if probabilities
|
58
52
|
exp = v.map { |vi| Math.exp(vi) }
|
data/lib/eps/lightgbm.rb
CHANGED
@@ -10,7 +10,7 @@ module Eps
|
|
10
10
|
str << "Model needs more data for better predictions\n"
|
11
11
|
else
|
12
12
|
str << "Most important features\n"
|
13
|
-
@importance_keys.zip(importance).sort_by { |k, v| [-v, k] }.first(10).each do |k, v|
|
13
|
+
@importance_keys.zip(importance).sort_by { |k, v| [-v, display_field(k)] }.first(10).each do |k, v|
|
14
14
|
str << "#{display_field(k)}: #{(100 * v / total).round}\n"
|
15
15
|
end
|
16
16
|
end
|
@@ -57,7 +57,7 @@ module Eps
|
|
57
57
|
|
58
58
|
# text feature encoding
|
59
59
|
prep_text_features(train_set)
|
60
|
-
prep_text_features(validation_set) if validation_set
|
60
|
+
prep_text_features(validation_set, fit: false) if validation_set
|
61
61
|
|
62
62
|
# create params
|
63
63
|
params = {
|
@@ -71,7 +71,7 @@ module Eps
|
|
71
71
|
end
|
72
72
|
|
73
73
|
# create datasets
|
74
|
-
categorical_idx =
|
74
|
+
categorical_idx = train_set.columns.keys.map.with_index.select { |k, _| @features[k] == "categorical" }.map(&:last)
|
75
75
|
train_ds = ::LightGBM::Dataset.new(train_set.map_rows(&:to_a), label: train_set.label, weight: train_set.weight, categorical_feature: categorical_idx, params: params)
|
76
76
|
validation_ds = ::LightGBM::Dataset.new(validation_set.map_rows(&:to_a), label: validation_set.label, weight: validation_set.weight, categorical_feature: categorical_idx, params: params, reference: train_ds) if validation_set
|
77
77
|
|
@@ -144,7 +144,10 @@ module Eps
|
|
144
144
|
end
|
145
145
|
|
146
146
|
if bad_observations.any?
|
147
|
-
|
147
|
+
bad_observations.each do |obs|
|
148
|
+
p obs
|
149
|
+
end
|
150
|
+
raise "Bug detected in evaluator. Please report an issue."
|
148
151
|
end
|
149
152
|
end
|
150
153
|
|
data/lib/eps/pmml/loader.rb
CHANGED
@@ -222,7 +222,7 @@ module Eps
|
|
222
222
|
else
|
223
223
|
operator = xml_predicate.attribute("operator").value
|
224
224
|
value = xml_predicate.attribute("value").value
|
225
|
-
value = value.to_f if operator == "greaterThan"
|
225
|
+
value = value.to_f if operator == "greaterThan" || operator == "lessOrEqual"
|
226
226
|
field = xml_predicate.attribute("field").value
|
227
227
|
field = derived_fields[field] if derived_fields[field]
|
228
228
|
{
|
data/lib/eps/version.rb
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: eps
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.3.
|
4
|
+
version: 0.3.9
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Andrew Kane
|
8
|
-
autorequire:
|
8
|
+
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date:
|
11
|
+
date: 2021-10-14 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: lightgbm
|
@@ -38,92 +38,8 @@ dependencies:
|
|
38
38
|
- - ">="
|
39
39
|
- !ruby/object:Gem::Version
|
40
40
|
version: '0'
|
41
|
-
|
42
|
-
|
43
|
-
requirement: !ruby/object:Gem::Requirement
|
44
|
-
requirements:
|
45
|
-
- - ">="
|
46
|
-
- !ruby/object:Gem::Version
|
47
|
-
version: '0'
|
48
|
-
type: :development
|
49
|
-
prerelease: false
|
50
|
-
version_requirements: !ruby/object:Gem::Requirement
|
51
|
-
requirements:
|
52
|
-
- - ">="
|
53
|
-
- !ruby/object:Gem::Version
|
54
|
-
version: '0'
|
55
|
-
- !ruby/object:Gem::Dependency
|
56
|
-
name: daru
|
57
|
-
requirement: !ruby/object:Gem::Requirement
|
58
|
-
requirements:
|
59
|
-
- - ">="
|
60
|
-
- !ruby/object:Gem::Version
|
61
|
-
version: '0'
|
62
|
-
type: :development
|
63
|
-
prerelease: false
|
64
|
-
version_requirements: !ruby/object:Gem::Requirement
|
65
|
-
requirements:
|
66
|
-
- - ">="
|
67
|
-
- !ruby/object:Gem::Version
|
68
|
-
version: '0'
|
69
|
-
- !ruby/object:Gem::Dependency
|
70
|
-
name: minitest
|
71
|
-
requirement: !ruby/object:Gem::Requirement
|
72
|
-
requirements:
|
73
|
-
- - ">="
|
74
|
-
- !ruby/object:Gem::Version
|
75
|
-
version: '0'
|
76
|
-
type: :development
|
77
|
-
prerelease: false
|
78
|
-
version_requirements: !ruby/object:Gem::Requirement
|
79
|
-
requirements:
|
80
|
-
- - ">="
|
81
|
-
- !ruby/object:Gem::Version
|
82
|
-
version: '0'
|
83
|
-
- !ruby/object:Gem::Dependency
|
84
|
-
name: numo-narray
|
85
|
-
requirement: !ruby/object:Gem::Requirement
|
86
|
-
requirements:
|
87
|
-
- - ">="
|
88
|
-
- !ruby/object:Gem::Version
|
89
|
-
version: '0'
|
90
|
-
type: :development
|
91
|
-
prerelease: false
|
92
|
-
version_requirements: !ruby/object:Gem::Requirement
|
93
|
-
requirements:
|
94
|
-
- - ">="
|
95
|
-
- !ruby/object:Gem::Version
|
96
|
-
version: '0'
|
97
|
-
- !ruby/object:Gem::Dependency
|
98
|
-
name: rake
|
99
|
-
requirement: !ruby/object:Gem::Requirement
|
100
|
-
requirements:
|
101
|
-
- - ">="
|
102
|
-
- !ruby/object:Gem::Version
|
103
|
-
version: '0'
|
104
|
-
type: :development
|
105
|
-
prerelease: false
|
106
|
-
version_requirements: !ruby/object:Gem::Requirement
|
107
|
-
requirements:
|
108
|
-
- - ">="
|
109
|
-
- !ruby/object:Gem::Version
|
110
|
-
version: '0'
|
111
|
-
- !ruby/object:Gem::Dependency
|
112
|
-
name: rover-df
|
113
|
-
requirement: !ruby/object:Gem::Requirement
|
114
|
-
requirements:
|
115
|
-
- - ">="
|
116
|
-
- !ruby/object:Gem::Version
|
117
|
-
version: '0'
|
118
|
-
type: :development
|
119
|
-
prerelease: false
|
120
|
-
version_requirements: !ruby/object:Gem::Requirement
|
121
|
-
requirements:
|
122
|
-
- - ">="
|
123
|
-
- !ruby/object:Gem::Version
|
124
|
-
version: '0'
|
125
|
-
description:
|
126
|
-
email: andrew@chartkick.com
|
41
|
+
description:
|
42
|
+
email: andrew@ankane.org
|
127
43
|
executables: []
|
128
44
|
extensions: []
|
129
45
|
extra_rdoc_files: []
|
@@ -156,7 +72,7 @@ homepage: https://github.com/ankane/eps
|
|
156
72
|
licenses:
|
157
73
|
- MIT
|
158
74
|
metadata: {}
|
159
|
-
post_install_message:
|
75
|
+
post_install_message:
|
160
76
|
rdoc_options: []
|
161
77
|
require_paths:
|
162
78
|
- lib
|
@@ -171,8 +87,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
171
87
|
- !ruby/object:Gem::Version
|
172
88
|
version: '0'
|
173
89
|
requirements: []
|
174
|
-
rubygems_version: 3.
|
175
|
-
signing_key:
|
90
|
+
rubygems_version: 3.2.22
|
91
|
+
signing_key:
|
176
92
|
specification_version: 4
|
177
93
|
summary: Machine learning for Ruby. Supports regression (linear regression) and classification
|
178
94
|
(naive Bayes)
|