Skip to content

Skip metric regularization if adapt window=0 #3037

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: develop
Choose a base branch
from
Open

Skip metric regularization if adapt window=0 #3037

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions src/stan/mcmc/covar_adaptation.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,10 +23,12 @@ class covar_adaptation : public windowed_adaptation {

estimator_.sample_covariance(covar);

double n = static_cast<double>(estimator_.num_samples());
covar = (n / (n + 5.0)) * covar
+ 1e-3 * (5.0 / (n + 5.0))
* Eigen::MatrixXd::Identity(covar.rows(), covar.cols());
if (estimator_.num_samples() > 1) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Because this behavior probably won't be obvious I think we need to wire a message to the logger. Both here and in var_adaptation can you add a std::string& msg argument which passes the message "Warning: inverse metric is not updated for window sizes less than 2". Then we can pass that to the logger in all of the transition implementations.

double n = static_cast<double>(estimator_.num_samples());
covar = (n / (n + 5.0)) * covar
+ 1e-3 * (5.0 / (n + 5.0))
* Eigen::MatrixXd::Identity(covar.rows(), covar.cols());
}

estimator_.restart();

Expand Down
8 changes: 5 additions & 3 deletions src/stan/mcmc/var_adaptation.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,9 +23,11 @@ class var_adaptation : public windowed_adaptation {

estimator_.sample_variance(var);

double n = static_cast<double>(estimator_.num_samples());
var = (n / (n + 5.0)) * var
+ 1e-3 * (5.0 / (n + 5.0)) * Eigen::VectorXd::Ones(var.size());
if (estimator_.num_samples() > 1) {
double n = static_cast<double>(estimator_.num_samples());
var = (n / (n + 5.0)) * var
+ 1e-3 * (5.0 / (n + 5.0)) * Eigen::VectorXd::Ones(var.size());
}

estimator_.restart();

Expand Down
29 changes: 29 additions & 0 deletions src/test/unit/mcmc/covar_adaptation_test.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,3 +27,32 @@ TEST(McmcCovarAdaptation, learn_covariance) {
}
EXPECT_EQ(0, logger.call_count());
}

TEST(McmcCovarAdaptation, learn_covariance_one_sample) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Corresponding test in var_adaptation_test.cpp?

stan::test::unit::instrumented_logger logger;

const int n = 10;
Eigen::VectorXd q = Eigen::VectorXd::Zero(n);
Eigen::MatrixXd covar(Eigen::MatrixXd::Identity(n, n));

const int n_learn = 1;

Eigen::MatrixXd target_covar(Eigen::MatrixXd::Identity(n, n));

stan::mcmc::covar_adaptation adapter(n);
adapter.set_window_params(50, 0, 0, n_learn, logger);

bool update = false;

for (int i = 0; i < n_learn; ++i)
update = adapter.learn_covariance(covar, q);

EXPECT_TRUE(update);

for (int i = 0; i < n; ++i) {
for (int j = 0; j < n; ++j) {
EXPECT_EQ(target_covar(i, j), covar(i, j));
}
}
EXPECT_EQ(0, logger.call_count());
}