forked from annoviko/pyclustering
-
Notifications
You must be signed in to change notification settings - Fork 0
/
CHANGES
executable file
·55 lines (37 loc) · 2.86 KB
/
CHANGES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
Change notes for 0.6.dev0 (start Jul 18, 2015)
GENERAL CHANGES (pyclustering):
- Implemented phase oscillatory network syncpr (pyclustering.nnet.syncpr).
See: https://github.com/annoviko/pyclustering/issues/208
- Feature for pyclustering.nnet.syncpr that allows to use ccore library for solving.
See: https://github.com/annoviko/pyclustering/issues/232
- Optimized simulation algorithm for sync oscillatory network (pyclustering.nnet.sync) when collecting results are not requested.
See: https://github.com/annoviko/pyclustering/issues/233
- Images of english alphabet 100x100.
See: https://github.com/annoviko/pyclustering/commit/aa28f1a8a363fbeb5f074d22ec1e8258a1dd0579
- Implemented feature to use rectangular network structures in oscillatory networks.
See: https://github.com/annoviko/pyclustering/issues/259
- Implemented CLARANS algorithm (pyclustering.cluster.clarans).
See: https://github.com/annoviko/pyclustering/issues/52
GENERAL CHANGES (ccore):
- Implemented phase oscillatory network for pattern recognition syncpr (ccore.syncpr).
See: https://github.com/annoviko/pyclustering/issues/232
- Implemented agglomerative algorithm for cluster analysis (ccore.agglomerative).
See: https://github.com/annoviko/pyclustering/issues/212
- Implemented feature to use rectangular network structures in oscillatory networks.
See: https://github.com/annoviko/pyclustering/issues/259
CORRECTED BUGS:
- Bug with callback methods in ccore library in syncnet (ccore.syncnet) and hsyncnet (ccore.hsyncnet) that may lead to loss of accuracy.
- Bug with division by zero in kmeans algorithm (ccore.kmeans, pyclustering.cluster.kmeans) when cluster after center updating is not able to capture object.
See: https://github.com/annoviko/pyclustering/issues/238
- Bug with stack overflow in KD tree in case of big data (pyclustering.container.kdtree, ccore.kdtree).
See: https://github.com/annoviko/pyclustering/pull/239
See: https://github.com/annoviko/pyclustering/issues/255
See: https://github.com/annoviko/pyclustering/issues/254
- Bug with incorrect clustering in case of the same elements in cure algorithm (pyclustering.cluster.cure).
See: https://github.com/annoviko/pyclustering/pull/239
- Bug with execution fail in case of wrong number of initial medians and in case of the same objects with several initial medians (pyclustering.cluster.kmedians).
See: https://github.com/annoviko/pyclustering/issues/256
- Bug with calculation synchronous ensembles near by zero: oscillators 2*pi and 0 are considered as different (pyclustering.nnet.sync, ccore.sync).
See: https://github.com/annoviko/pyclustering/issues/263
- Bug with cluster allocation in kmedoids algorithm in case of the same objects with several initial medoids (pyclustering.cluster.kmedoids).
See: https://github.com/annoviko/pyclustering/issues/269