te_rex 0.1.0 → 0.1.1
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/te_rex/bayes_data.rb +3 -3
- data/lib/te_rex/version.rb +1 -1
- data/test/bayes_data_test.rb +1 -1
- data/test/corpus_test.rb +0 -1
- data/test/trained_bayes_provider_errors_test.rb +7 -3
- metadata +2 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA1:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: bae6e503a7f674b25ede59c7f9ef1c5de54b6a78
|
|
4
|
+
data.tar.gz: 71e7da095d199a9bc51419a3432912fb461b23c4
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 84220db22732d965161f8e677662b4cb9392f02762d4cf89ef027cdf9622fdc10f8eb26fea72e91301151f6fe24cc71a09029b4d31c77a753b5f8005bfff1868
|
|
7
|
+
data.tar.gz: 3fb8d3e66ff969b967075d13a898b8a12e81c5456996e890cb77fe6b21ed076be59d4749ca3d3715d429a298cc48969b516ccfc42548fd51f700bd638dc9aada
|
data/lib/te_rex/bayes_data.rb
CHANGED
|
@@ -39,8 +39,8 @@ module TeRex
|
|
|
39
39
|
# Each word in the string is interned and shows count in the document.
|
|
40
40
|
def index_frequency(text)
|
|
41
41
|
cfi = clean_stemmed_filtered_index(text)
|
|
42
|
-
|
|
43
|
-
cfi
|
|
42
|
+
cni = clean_filtered_index(text)
|
|
43
|
+
cfi.merge(cni)
|
|
44
44
|
end
|
|
45
45
|
|
|
46
46
|
# Return text with datetime and moneyterms replaced, remove cardinal terms (1st, 23rd, 42nd), remove punctuation.
|
|
@@ -88,7 +88,7 @@ module TeRex
|
|
|
88
88
|
idx = Hash.new(0)
|
|
89
89
|
word_array.each do |word|
|
|
90
90
|
word.downcase!
|
|
91
|
-
if !TeRex::StopWord::LIST.include?(word)
|
|
91
|
+
if !TeRex::StopWord::LIST.include?(word) && word.length > 3
|
|
92
92
|
idx[word.intern] += 1
|
|
93
93
|
end
|
|
94
94
|
end
|
data/lib/te_rex/version.rb
CHANGED
data/test/bayes_data_test.rb
CHANGED
data/test/corpus_test.rb
CHANGED
|
@@ -8,7 +8,7 @@ class TrainedBayesProviderErrorsTest < PryTest::Test
|
|
|
8
8
|
@@credit_data = TeRex::Train::CREDITDATA
|
|
9
9
|
@@credit_decline = TeRex::Train::CREDITDECLINE
|
|
10
10
|
@@credit_service = TeRex::Train::CREDITSERVICE
|
|
11
|
-
|
|
11
|
+
#@@unexpected = TeRex::Train::UNEXPECTED
|
|
12
12
|
#@@unk = TeRex::Train::UNKNOWNERROR
|
|
13
13
|
|
|
14
14
|
@@cls = TeRex::Classifier::Bayes.new(
|
|
@@ -19,7 +19,7 @@ class TrainedBayesProviderErrorsTest < PryTest::Test
|
|
|
19
19
|
{:tag => "CreditDataError", :msg => "Credit Card data is invalid"},
|
|
20
20
|
{:tag => "CreditDeclineError", :msg => "Waht? Credit Card declined!"},
|
|
21
21
|
{:tag => "CreditServiceError", :msg => "External service problem processing"},
|
|
22
|
-
{:tag => "UnexpectedResponseError", :msg => "Unexpected response"}
|
|
22
|
+
#{:tag => "UnexpectedResponseError", :msg => "Unexpected response"}
|
|
23
23
|
#{:tag => "UnknownError", :msg => "Unexpected response"},
|
|
24
24
|
)
|
|
25
25
|
@@avail.each {|txt| @@cls.train("AvailabilityError", txt) }
|
|
@@ -29,7 +29,7 @@ class TrainedBayesProviderErrorsTest < PryTest::Test
|
|
|
29
29
|
@@credit_data.each {|txt| @@cls.train("CreditDataError", txt) }
|
|
30
30
|
@@credit_decline.each {|txt| @@cls.train("CreditDeclineError", txt) }
|
|
31
31
|
@@credit_service.each {|txt| @@cls.train("CreditServiceError", txt) }
|
|
32
|
-
|
|
32
|
+
#@@unexpected.each {|txt| @@cls.train("UnexpectedResponseError", txt) }
|
|
33
33
|
#@@unk.each {|txt| @@cls.train("UnknownError", txt) }
|
|
34
34
|
|
|
35
35
|
|
|
@@ -63,6 +63,10 @@ class TrainedBayesProviderErrorsTest < PryTest::Test
|
|
|
63
63
|
#assert s_unexpected1 == ["UnexpectedResponseError", "Unexpected response"]
|
|
64
64
|
end
|
|
65
65
|
|
|
66
|
+
test "total word counts correct" do
|
|
67
|
+
assert @@cls.total_words == 5696
|
|
68
|
+
end
|
|
69
|
+
|
|
66
70
|
|
|
67
71
|
# test "Training Data Set Test: cancel-cacancelical examples should classify correctly" do
|
|
68
72
|
#
|
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: te_rex
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.1.
|
|
4
|
+
version: 0.1.1
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Joshua Bowles
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: bin
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2015-01-
|
|
11
|
+
date: 2015-01-20 00:00:00.000000000 Z
|
|
12
12
|
dependencies:
|
|
13
13
|
- !ruby/object:Gem::Dependency
|
|
14
14
|
name: fast-stemmer
|