Richard Townsend
45ca6063f1
Not sure if this bagging version is better or not
...
More more similar to "Attribute bagging:improving accuracy of classifier ensembles by using random feature subsets" (Brill)
2014-05-18 11:49:35 +01:00
Richard Townsend
1b0e2dce7c
Correction to randomisation and train-test split
2014-05-18 11:23:32 +01:00
Stephen Whitworth
5d1039f3d0
Merge pull request #32 from sjwhitworth/travis
...
Travis
2014-05-17 22:32:28 +01:00
Stephen Whitworth
6cc6900da7
Shut up Travis
2014-05-17 21:50:53 +01:00
Stephen Whitworth
7c3e86d06c
Added travis build command
2014-05-17 21:48:42 +01:00
Richard Townsend
26660e1470
Corrected a problem with pruning, actual ID3 decision tree type
...
Going to modify Bagging to select attributes on its own
2014-05-17 21:45:26 +01:00
Stephen Whitworth
b70e2a318a
Fixing up stuff for Travis
2014-05-17 21:41:10 +01:00
Stephen Whitworth
60cff544b8
Merge branch 'docs' into travis
2014-05-17 21:38:01 +01:00
Stephen Whitworth
22ab4092e7
Fixing up stuff for Travis
2014-05-17 21:33:48 +01:00
Stephen Whitworth
829f98b0a0
Update README.md
2014-05-17 21:20:34 +01:00
Stephen Whitworth
f5cd616f47
Update README.md
2014-05-17 21:19:31 +01:00
Stephen Whitworth
0fc6fb7cfa
Update README.md
2014-05-17 21:19:18 +01:00
Stephen Whitworth
cc4d6a60ef
Added some starter documentation.
2014-05-17 21:17:57 +01:00
Richard Townsend
12ace9def5
Identified source of the low accuracy
2014-05-17 20:37:19 +01:00
Stephen Whitworth
697fbde085
Rename travis.yml to .travis.yml
2014-05-17 20:34:15 +01:00
Stephen Whitworth
085b8e2a30
Create travis.yml
2014-05-17 20:32:40 +01:00
Stephen Whitworth
f22c62f1ed
Update README.md
2014-05-17 20:32:04 +01:00
Richard Townsend
13c0dc3eba
Reduced-error pruning
2014-05-17 18:06:01 +01:00
Richard Townsend
c516907b13
Passes all the tests
2014-05-17 17:35:10 +01:00
Richard Townsend
db3ac3c695
ID3 algorithm working
2014-05-17 17:28:51 +01:00
Richard Townsend
cf165695c8
ChiMerge seems to improve accuracy
2014-05-17 16:20:56 +01:00
Stephen Whitworth
c68e2eed28
Merge pull request #29 from Sentimentron/discretisation-upstream
...
Equal-width and Chi-merge discretisation
2014-05-16 08:05:12 +01:00
Richard Townsend
fdb67a4355
Initial work on decision trees
...
Random Forest has occasional disastrous accuracy:
never seen that happen in WEKA
2014-05-14 14:00:22 +01:00
Richard Townsend
f59cac6345
Equal-width and Chi-merge discretisation
2014-05-13 22:45:52 +01:00
Richard Townsend
c6972988ee
Merge pull request #27 from Sentimentron/instances
...
Adds Attributes and Instances
2014-05-13 22:22:17 +01:00
Richard Townsend
a2c67592df
Adds Instances and Attributes type
...
* Refactors KNNClassifier to use them
* csv handling moved back into base due to a circular dependency
* Also adds the datasets used to test CSV handling
2014-05-13 22:08:11 +01:00
npbool
a6fffa0dda
add make.go; fix windows support
2014-05-09 21:29:45 +08:00
Ningping Wang
b87006a324
fix linux blas linking -fpic
2014-05-09 10:29:35 +00:00
Ningping Wang
e3fbbfd44d
change .gitignore to include c/c++ types
2014-05-09 18:26:45 +08:00
Ningping Wang
dca2e11c02
move liblinear to ext/
2014-05-09 18:19:55 +08:00
Ningping Wang
7f1e235323
Merge remote-tracking branch 'ups/master'
2014-05-09 17:42:28 +08:00
Stephen Whitworth
9f3d9eaa64
Update linear_regression.go
...
Removing merge conflict.
2014-05-06 23:13:02 +01:00
Ningping Wang
0d62bc67e5
change test to goconvey; modify makefile
2014-05-06 12:55:58 +08:00
Stephen Whitworth
d069d68d05
Merge branch 'master' of https://github.com/sjwhitworth/golearn
2014-05-05 22:49:06 +01:00
Stephen Whitworth
1ade0afca6
Refactored KNN to implement the estimator interface
2014-05-05 22:41:55 +01:00
Stephen Whitworth
3850153945
Merge pull request #20 from sjwhitworth/feature/ifesdjeen/distances2
...
Remeake of my previous distances branch
2014-05-05 22:17:23 +01:00
Stephen Whitworth
3eaeafd0dc
Merge branch 'feature/ifesdjeen/distances2' of https://github.com/sjwhitworth/golearn into optimisation
2014-05-05 21:57:08 +01:00
Stephen Whitworth
7e8eb70cd1
Implemented batch and stochastic gradient descent
2014-05-05 21:40:51 +01:00
Stephen Whitworth
cd43a6565d
Implemented batch and stochastic gradient descent
2014-05-05 21:40:33 +01:00
Alex Petrov
93894245ad
Change wording for Eucledian Distance comments
2014-05-05 22:30:04 +02:00
Bert Chang
1e21802d2e
Use AlmostEqual for numerical comparison.
2014-05-05 23:28:49 +08:00
npbool
be85fca03c
cleanup
2014-05-05 21:57:53 +08:00
npbool
05782e9ccc
logistic regression, using liblinear
2014-05-05 21:50:13 +08:00
Alex Petrov
1d250aab6b
Remove unused subVector calls
2014-05-05 08:45:29 +02:00
Alex Petrov
13327e9250
Run goftm on all files ti fix indentation
2014-05-05 08:32:38 +02:00
Stephen Whitworth
dc96e818d8
Made a start on gradient descent
2014-05-05 02:39:00 +01:00
Alex Petrov
ad77c0dc7e
Add Chebyshev and Cranberra distances, improve Manhattan distance
...
Previously, Manhattan distance measure was only able to calculate
vector distances. Now, it's possible to calculate distances on
matrices, too, without unnecessary overhead of copying vectors,
doing an in-place lookups.
2014-05-04 23:11:37 +02:00
Alex Petrov
1c22641a3b
Fix manhattan distance calculation
2014-05-04 22:58:32 +02:00
Stephen Whitworth
b4abf54c07
Changing comments to play nicely with godoc
2014-05-04 09:57:57 +01:00
Stephen Whitworth
73e7df1ef0
Refactored KNN
2014-05-04 09:53:53 +01:00
