pyrox.dev
shogun: fix compile errors with Eigen 3.4
Fixes #142811.
+3
pkgs/applications/science/machine-learning/shogun/default.nix
+3
pkgs/applications/science/machine-learning/shogun/default.nix
+74
pkgs/applications/science/machine-learning/shogun/eigen-3.4.patch
+74
pkgs/applications/science/machine-learning/shogun/eigen-3.4.patch
···
···
1
+
From: Sebastián Mancilla <smancill@smancill.dev>
2
+
Subject: [PATCH] Fix compile errors when using Eigen 3.4
3
+
4
+
---
5
+
.../machine/gp/MultiLaplaceInferenceMethod.cpp | 18 +++++++++---------
6
+
1 file changed, 9 insertions(+), 9 deletions(-)
7
+
8
+
diff --git a/src/shogun/machine/gp/MultiLaplaceInferenceMethod.cpp b/src/shogun/machine/gp/MultiLaplaceInferenceMethod.cpp
9
+
index 2e27678d2..60050afea 100644
10
+
--- a/src/shogun/machine/gp/MultiLaplaceInferenceMethod.cpp
11
+
+++ b/src/shogun/machine/gp/MultiLaplaceInferenceMethod.cpp
12
+
@@ -84,9 +84,9 @@ class CMultiPsiLine : public func_base
13
+
float64_t result=0;
14
+
for(index_t bl=0; bl<C; bl++)
15
+
{
16
+
- eigen_f.block(bl*n,0,n,1)=K*alpha->block(bl*n,0,n,1)*CMath::exp(log_scale*2.0);
17
+
- result+=alpha->block(bl*n,0,n,1).dot(eigen_f.block(bl*n,0,n,1))/2.0;
18
+
- eigen_f.block(bl*n,0,n,1)+=eigen_m;
19
+
+ eigen_f.segment(bl*n,n)=K*alpha->segment(bl*n,n)*CMath::exp(log_scale*2.0);
20
+
+ result+=alpha->segment(bl*n,n).dot(eigen_f.segment(bl*n,n))/2.0;
21
+
+ eigen_f.segment(bl*n,n)+=eigen_m;
22
+
}
23
+
24
+
// get first and second derivatives of log likelihood
25
+
@@ -272,7 +272,7 @@ void CMultiLaplaceInferenceMethod::update_alpha()
26
+
{
27
+
Map<VectorXd> alpha(m_alpha.vector, m_alpha.vlen);
28
+
for(index_t bl=0; bl<C; bl++)
29
+
- eigen_mu.block(bl*n,0,n,1)=eigen_ktrtr*CMath::exp(m_log_scale*2.0)*alpha.block(bl*n,0,n,1);
30
+
+ eigen_mu.segment(bl*n,n)=eigen_ktrtr*CMath::exp(m_log_scale*2.0)*alpha.segment(bl*n,n);
31
+
32
+
//alpha'*(f-m)/2.0
33
+
Psi_New=alpha.dot(eigen_mu)/2.0;
34
+
@@ -316,7 +316,7 @@ void CMultiLaplaceInferenceMethod::update_alpha()
35
+
36
+
for(index_t bl=0; bl<C; bl++)
37
+
{
38
+
- VectorXd eigen_sD=eigen_dpi.block(bl*n,0,n,1).cwiseSqrt();
39
+
+ VectorXd eigen_sD=eigen_dpi.segment(bl*n,n).cwiseSqrt();
40
+
LLT<MatrixXd> chol_tmp((eigen_sD*eigen_sD.transpose()).cwiseProduct(eigen_ktrtr*CMath::exp(m_log_scale*2.0))+
41
+
MatrixXd::Identity(m_ktrtr.num_rows, m_ktrtr.num_cols));
42
+
MatrixXd eigen_L_tmp=chol_tmp.matrixU();
43
+
@@ -341,11 +341,11 @@ void CMultiLaplaceInferenceMethod::update_alpha()
44
+
VectorXd tmp2=m_tmp.array().rowwise().sum();
45
+
46
+
for(index_t bl=0; bl<C; bl++)
47
+
- eigen_b.block(bl*n,0,n,1)+=eigen_dpi.block(bl*n,0,n,1).cwiseProduct(eigen_mu.block(bl*n,0,n,1)-eigen_mean_bl-tmp2);
48
+
+ eigen_b.segment(bl*n,n)+=eigen_dpi.segment(bl*n,n).cwiseProduct(eigen_mu.segment(bl*n,n)-eigen_mean_bl-tmp2);
49
+
50
+
Map<VectorXd> &eigen_c=eigen_W;
51
+
for(index_t bl=0; bl<C; bl++)
52
+
- eigen_c.block(bl*n,0,n,1)=eigen_E.block(0,bl*n,n,n)*(eigen_ktrtr*CMath::exp(m_log_scale*2.0)*eigen_b.block(bl*n,0,n,1));
53
+
+ eigen_c.segment(bl*n,n)=eigen_E.block(0,bl*n,n,n)*(eigen_ktrtr*CMath::exp(m_log_scale*2.0)*eigen_b.segment(bl*n,n));
54
+
55
+
Map<MatrixXd> c_tmp(eigen_c.data(),n,C);
56
+
57
+
@@ -409,7 +409,7 @@ float64_t CMultiLaplaceInferenceMethod::get_derivative_helper(SGMatrix<float64_t
58
+
{
59
+
result+=((eigen_E.block(0,bl*n,n,n)-eigen_U.block(0,bl*n,n,n).transpose()*eigen_U.block(0,bl*n,n,n)).array()
60
+
*eigen_dK.array()).sum();
61
+
- result-=(eigen_dK*eigen_alpha.block(bl*n,0,n,1)).dot(eigen_alpha.block(bl*n,0,n,1));
62
+
+ result-=(eigen_dK*eigen_alpha.segment(bl*n,n)).dot(eigen_alpha.segment(bl*n,n));
63
+
}
64
+
65
+
return result/2.0;
66
+
@@ -489,7 +489,7 @@ SGVector<float64_t> CMultiLaplaceInferenceMethod::get_derivative_wrt_mean(
67
+
result[i]=0;
68
+
//currently only compute the explicit term
69
+
for(index_t bl=0; bl<C; bl++)
70
+
- result[i]-=eigen_alpha.block(bl*n,0,n,1).dot(eigen_dmu);
71
+
+ result[i]-=eigen_alpha.segment(bl*n,n).dot(eigen_dmu);
72
+
}
73
+
74
+
return result;