issues
search
tyler-tomita
/
RandomerForest
Discriminant Projection Forest results, datasets, etc.
44
stars
21
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
urgent: fig cigar figs
#65
jovo
closed
9 years ago
0
urgent: simulation fig
#64
jovo
closed
9 years ago
0
Profile RerF on Trunk
#63
tyler-tomita
closed
9 years ago
0
Trunk where RerF doesn't do worse
#62
tyler-tomita
closed
9 years ago
2
Redo Benchmark Data Analysis Using Updated Rank Procedure?
#61
tyler-tomita
closed
9 years ago
3
interpretability = sparsity
#60
jovo
closed
9 years ago
4
plot L(RF) - L(RerF) vs trace of covariance matrix
#59
jovo
opened
9 years ago
0
sample affine matrices maybe?
#58
jovo
closed
9 years ago
0
sample sparse matrices faster
#57
jovo
closed
9 years ago
0
figure out how to plot L(RF)-L(RerF) vs sparsity
#56
jovo
opened
9 years ago
1
imagenet classification
#55
jovo
opened
9 years ago
0
add methods section
#54
jovo
closed
9 years ago
0
add sections to paper (use your discretion, we will iterate)
#53
jovo
closed
9 years ago
0
1-sided mann-whitney test comparing RF vs. RerF(s+d+r) on benchmark data
#52
jovo
closed
9 years ago
7
fig 4 legend
#51
jovo
closed
9 years ago
0
fig 3 legend
#50
jovo
closed
9 years ago
0
R'er F vs. RerF (pronounced 'rerf')
#49
jovo
closed
9 years ago
2
fig 5: analysis of performance vs. RF
#48
jovo
closed
9 years ago
6
final code things to run
#47
jovo
closed
9 years ago
4
Final drafts of figs (without running anymore code)
#46
tyler-tomita
closed
9 years ago
0
fig 3: add 4th column for LOL (after finalizing other fig stuff)
#45
jovo
closed
9 years ago
7
fig captions
#44
jovo
closed
9 years ago
6
rapid fire fig changes
#43
jovo
closed
9 years ago
2
Fig1 Try Spiral
#42
tyler-tomita
closed
8 years ago
4
update fig2 with new dense code
#41
jovo
closed
9 years ago
2
fig2: is it using the newer code that only keeps what it needs?
#40
jovo
closed
9 years ago
1
fig1 make parity easier
#39
jovo
closed
9 years ago
0
2nd row to figure 3
#38
jovo
closed
9 years ago
0
Make even randomer by sampling from gaussians centered at -1 and 1 w/ overall variance of one
#37
tyler-tomita
opened
9 years ago
2
Update Randomer Forest so that only the split dimension for each node is stored
#36
tyler-tomita
closed
9 years ago
1
Bayes error for parity increases at d=10 and then drops
#35
tyler-tomita
closed
9 years ago
5
Tex file of manuscript
#34
tyler-tomita
closed
9 years ago
1
Fig 3 Changes
#33
tyler-tomita
closed
9 years ago
3
Fig 2 Changes
#32
tyler-tomita
closed
9 years ago
3
Fig 1: at least for parity, add some D's between 50 & 100
#31
tyler-tomita
closed
9 years ago
6
play around with switching projection matrix from only -1,0,1 to other stuff
#30
jovo
closed
9 years ago
1
supp fig 1: # of trees to stabilize
#29
jovo
closed
9 years ago
1
Fig 4b: benchmark in high-dimensional real data
#28
jovo
closed
9 years ago
0
Fig 4a: benchmark in real data
#27
jovo
closed
9 years ago
7
for each of the fig 1 simulations, also try a "rotation invariant" version
#26
jovo
closed
9 years ago
1
update fig 1 using "non-cheating" variants of RF
#25
jovo
closed
9 years ago
1
check 2D vs 3D LOL
#24
jovo
closed
9 years ago
2
Lhat of Randomer Forest w/ meandiff is lower than Bayes error
#23
tyler-tomita
closed
9 years ago
18
make all paths relative
#22
jovo
closed
9 years ago
0
v2 of Fig 1
#21
jovo
closed
9 years ago
9
plot Lhat vs. D for multimodal example
#20
jovo
closed
9 years ago
0
add Bayes error line to fig 1 for trunk
#19
jovo
closed
9 years ago
0
run LOL variants on rank transformed matrix
#18
jovo
closed
9 years ago
19
generate LOL_loocv function
#17
jovo
closed
9 years ago
0
check whether TF is rotation invariant on Trunk
#16
jovo
closed
9 years ago
1
Previous
Next