DCA
DCA copied to clipboard
CompFUSE
DCA++
With the update of the interpolation of G0 in TpEqualTimeAccumulator, I went looking for similar code segments. We have the following places in the code where we might should consider...
I was looking at the class in include/dca/phys/domains/convert.hpp as it gave me problems with changes to variadic template helpers and got quite confused. Why do we have all this machinery,...
The CT-AUX accumulator tries to write two functions with the name "no-name". See https://github.com/CompFUSE/DCA/blob/ed9406d2615dfc7e99e9d488d1c58b0a876b3d52/include/dca/phys/dca_step/cluster_solver/ctaux/ctaux_accumulator.hpp#L316 and https://github.com/CompFUSE/DCA/blob/ed9406d2615dfc7e99e9d488d1c58b0a876b3d52/include/dca/phys/dca_step/cluster_solver/ctaux/accumulator/tp/tp_equal_time_accumulator.hpp#L248 This causes the HDF5 writer to throw an exception and the whole main_dca to...
Can we have mixed self-energy factor as 1 whenever we set finite size Monte Carlo to be true?
hdf5 writes a vector of 2n doubles as [[re1, im1, re2, im2,...], [re(n+1), im(n+1), ..]] instead using a dimension for the real/imaginary choiche, as it is done for the function...
After speaking with the new developers in Oak Ridge I stand more convinced that the test submission script should run with `ctest --output-on-failure` rather than `make test` for the following...
I don't know whether CtauxClusterSolver::computeErrorBars is usually executed. But if it is, we make a redundant copy of G4: https://github.com/CompFUSE/DCA/blob/master/include/dca/phys/dca_step/cluster_solver/ctaux/ctaux_cluster_solver.hpp#L402 @gbalduzz Any thoughts on that?
The jackknife error on G4 is computed even when four-point-type = NONE: https://github.com/CompFUSE/DCA/blob/master/include/dca/phys/dca_step/cluster_solver/ctaux/ctaux_cluster_solver.hpp#L288 Besides the redundant computation, this also causes the memory for G4_err_ in DcaData to be allocated.
With gcc/5.3.0 and cuda/8.0.61 the test fails. This may be simply a case of an overly optimistic epsilon in the test. ```shell start reading function : leading-eigenvalues start reading function...
Using the minimalist printer or the DCA test environment on summit causes the executable to crash after completing the tests. It happens only on Summit.