[SyneRBI/SIRF-SuperBuild] CCPi regularisation fails test (Issue #832)
Edoardo Pasca
CCPi regularisation toolkit builds but unit tests fail with messages as
#829 (comment)
or
File "/home/runner/work/SIRF-SuperBuild/SIRF-SuperBuild/build/sources/CCPi-Regularisation-Toolkit/test/test_run_test.py", line 5, in <module>
from ccpi.filters.regularisers import ROF_TV, FGP_TV, PD_TV, SB_TV, TGV, LLT_ROF, FGP_dTV, NDF, Diff4th
File "/home/runner/install/python/ccpi/filters/regularisers.py", line 5, in <module>
from ccpi.filters.cpu_regularisers import TV_ROF_CPU, TV_FGP_CPU, TV_PD_CPU, TV_SB_CPU, dTV_FGP_CPU, TNV_CPU, NDF_CPU, Diff4th_CPU, TGV_CPU, LLT_ROF_CPU, PATCHSEL_CPU, NLTV_CPU
ImportError: /home/runner/install/python/ccpi/filters/cpu_regularisers.cpython-310-x86_64-linux-gnu.so: undefined symbol: _Z14SB_TV_CPU_mainPfS_S_fifiiii
See also vais-ral/CCPi-Regularisation-Toolkit#177
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Edoardo Pasca
We should reinstate 90faebc and 6e0b7e3
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Edoardo Pasca
After #829 the ccpi-regularisation is not built as CIL dependency
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Casper da Costa-Luis
@paskino I guess you'll push a new toolkit release?
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Edoardo Pasca
Can we just use a different commit hash
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Kris Thielemans
fine for me
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
Casper da Costa-Luis
Closed #832 as completed via #897.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.