Dongryeol Lee
bbee8bebaf
Fast code now computes confidence bands
2008-02-17 04:36:38 +00:00
Dongryeol Lee
2e9e51b9c8
No bug - just a misuse of the kernel function in two algorithms during comparison
2008-02-17 02:47:40 +00:00
Dongryeol Lee
56a601f145
Some bug fix
2008-02-17 02:39:00 +00:00
vasiloglou
3c3510610c
it doesn't give the expected results on the swiss roll
2008-02-17 00:16:49 +00:00
vasiloglou
3dbdd1d114
it is terminating now
2008-02-16 06:33:35 +00:00
vasiloglou
c441f3e714
it is running all bugsfixed
2008-02-16 06:24:37 +00:00
Dongryeol Lee
90ac6810c9
Need to fix the correctness of dense LPR
2008-02-16 05:48:51 +00:00
Dongryeol Lee
ff573bcc84
Little fix on the parameter
2008-02-16 02:57:31 +00:00
Dongryeol Lee
d3ba206c6b
Modified the tree-based code to do train/prediction phase separately
2008-02-16 02:24:25 +00:00
rriegel
ac83f3b760
LOO by index, not dist
2008-02-15 22:40:30 +00:00
Dongryeol Lee
5c42e89f7b
Minor addition to the driver file
2008-02-15 22:36:12 +00:00
Dongryeol Lee
7bce4938b7
Naive LPR confidence band fix
2008-02-15 22:12:23 +00:00
Dongryeol Lee
449e7c7f13
Minior fix to the do LOO nn search
2008-02-15 17:59:30 +00:00
vasiloglou
e809c275bc
Fixed the query=reference case
2008-02-15 16:32:55 +00:00
Dongryeol Lee
9846258b24
Variable bandwidth for naive algorithm implemented
2008-02-15 04:26:54 +00:00
Dongryeol Lee
104e4ad01a
Took out the template argument that specified the local polynomial order
2008-02-15 01:50:15 +00:00
Dongryeol Lee
b867b7c827
Verified the naive algorithm
2008-02-15 01:36:27 +00:00
vasiloglou
56129688ee
BFGS is not running properly but we are getting there
...
possibly the initial steps are not good
2008-02-15 00:25:04 +00:00
Dongryeol Lee
f4b7ddb67e
Naive algorithm now outputs confidence bands
2008-02-14 21:11:30 +00:00
Dongryeol Lee
d12ab9b3e9
Fixing the naive algorithm to compute confidence bands
2008-02-14 21:00:27 +00:00
vasiloglou
a8c3e1d9a7
ScaleRows works with a matrix too
2008-02-14 20:01:55 +00:00
vasiloglou
89fe436757
finished the BFGS
...
now I have to debug
2008-02-14 19:55:45 +00:00
hjiang
0dc8e7d83a
delete accidental insertion of a binary file
2008-02-14 19:11:22 +00:00
hjiang
97fa6032d2
Huijings code for getting the centroids and nearest neighbors
2008-02-14 19:10:38 +00:00
hjiang
91bff5516c
added 2 utilities to access the trees and made the tyupedef publicwq
2008-02-14 19:09:33 +00:00
Dongryeol Lee
2eb33b7e8f
Fixed parameter module
2008-02-13 01:59:57 +00:00
Dongryeol Lee
eb062b2da9
Confirmed the correctness of the pruning
2008-02-12 22:50:52 +00:00
Dongryeol Lee
27ef3809c0
More utility functions for testing correctness
2008-02-12 22:40:36 +00:00
Dongryeol Lee
ea3b948214
Added Deng and Moore's pruning rule, but needs to be filled in
2008-02-12 21:37:46 +00:00
Dongryeol Lee
e4a140f610
Minor fixes to output module
2008-02-12 20:28:36 +00:00
Dongryeol Lee
4df60ae9c8
Fix in the matrix util
2008-02-12 20:18:26 +00:00
Dongryeol Lee
641dbb08e3
Another fix to the naive lpr code
2008-02-12 19:07:23 +00:00
Dongryeol Lee
b79074358a
Minor fix to dense LPR code
2008-02-12 18:25:11 +00:00
Dongryeol Lee
13c84e3286
Bug fix in naive local polynomial
2008-02-12 16:56:43 +00:00
Bill March
c0a1108b95
changed stuff
2008-02-12 16:23:43 +00:00
Dongryeol Lee
1a1986b78b
Fix for NWR
2008-02-12 04:28:14 +00:00
Dongryeol Lee
963f9beda2
More fix
2008-02-12 02:55:05 +00:00
Dongryeol Lee
7f5a9d8edf
Fixed the compilation error
2008-02-12 02:52:44 +00:00
Dongryeol Lee
be5037c9dd
Bug fix
2008-02-12 02:15:01 +00:00
Dongryeol Lee
1e81b47fc4
More fixes
2008-02-12 01:05:23 +00:00
Dongryeol Lee
d8ef8d2ee4
Debuggin in process
2008-02-12 01:03:49 +00:00
Dongryeol Lee
1eee6feaac
Fixed the compilation error
2008-02-12 00:54:53 +00:00
Dongryeol Lee
63aaf1d852
Fixed the compilation error
2008-02-12 00:46:50 +00:00
Dongryeol Lee
aa010ab75a
Finished implementing the dense formulation of LPR
2008-02-12 00:38:00 +00:00
vasiloglou
59b86b4b0c
it seems to work
...
comments
The armijo rule is really slow
The statistical update of sigmas is good
We need to stay sometime in low sigmas otherwise we do not optimize
the trace
Next setp is BFGS method
2008-02-11 22:51:14 +00:00
Dongryeol Lee
f13a95e825
More fixes
2008-02-11 22:03:19 +00:00
Dongryeol Lee
5e3a004dc9
Algorithm reformulation
2008-02-11 21:35:37 +00:00
vasiloglou
38a107040e
it seems to me it is working, it is just a matter of tuning the parameters
2008-02-11 05:16:33 +00:00
Dongryeol Lee
259264690c
Finished the query tree initialization phase for dense version of LPR
2008-02-11 00:56:54 +00:00
Dongryeol Lee
68cfc553c0
Base case finished
2008-02-11 00:30:09 +00:00
Dongryeol Lee