Star file with particles selected from 2D classification in CS

hi there,
I tried to convert selected particles from a 2D classification job to .star file:

csparc2star.py particles_selected.cs P54_J6_passthrough_particles_selected.cs P54_J6_particles.star

and/or

csparc2star.py --logelevel debug particles_selected.cs P54_J6_passthrough_particles_selected.cs P54_J6_particles_selected2.star

however when I run 3D refine in Relion I get this error:

in: /packages/RELION/relion-3.1.2/src/exp_model.cpp, line 344
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
=== Backtrace  ===
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN11RelionErrorC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x67) [0x562b6e2d9c67]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN10Experiment29divideParticlesInRandomHalvesEib+0x1142) [0x562b6e37de92]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi18initialiseWorkLoadEv+0x261) [0x562b6e2f9b91]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi10initialiseEv+0x9d2) [0x562b6e2fbb92]
/opt/relion/3.1.2/bin/relion_refine_mpi(main+0x4a) [0x562b6e2c490a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7f25f4dc709b]
/opt/relion/3.1.2/bin/relion_refine_mpi(_start+0x2a) [0x562b6e2c7c5a]
==================
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
in: /packages/RELION/relion-3.1.2/src/exp_model.cpp, line 344
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
=== Backtrace  ===
in: /packages/RELION/relion-3.1.2/src/exp_model.cpp, line 344
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
=== Backtrace  ===
in: /packages/RELION/relion-3.1.2/src/exp_model.cpp, line 344
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
=== Backtrace  ===
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN11RelionErrorC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x67) [0x55a5d97a3c67]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN10Experiment29divideParticlesInRandomHalvesEib+0x1142) [0x55a5d9847e92]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi18initialiseWorkLoadEv+0x261) [0x55a5d97c3b91]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi10initialiseEv+0x9d2) [0x55a5d97c5b92]
/opt/relion/3.1.2/bin/relion_refine_mpi(main+0x4a) [0x55a5d978e90a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7f29aebb009b]
/opt/relion/3.1.2/bin/relion_refine_mpi(_start+0x2a) [0x55a5d9791c5a]
==================
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
in: /packages/RELION/relion-3.1.2/src/exp_model.cpp, line 344
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
=== Backtrace  ===
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN11RelionErrorC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x67) [0x55f313b35c67]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN10Experiment29divideParticlesInRandomHalvesEib+0x1142) [0x55f313bd9e92]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi18initialiseWorkLoadEv+0x261) [0x55f313b55b91]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi10initialiseEv+0x9d2) [0x55f313b57b92]
/opt/relion/3.1.2/bin/relion_refine_mpi(main+0x4a) [0x55f313b2090a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7f5967be909b]
/opt/relion/3.1.2/bin/relion_refine_mpi(_start+0x2a) [0x55f313b23c5a]
==================
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
in: /packages/RELION/relion-3.1.2/src/exp_model.cpp, line 344
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
=== Backtrace  ===
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN11RelionErrorC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x67) [0x55b5403ddc67]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN10Experiment29divideParticlesInRandomHalvesEib+0x1142) [0x55b540481e92]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi18initialiseWorkLoadEv+0x261) [0x55b5403fdb91]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi10initialiseEv+0x9d2) [0x55b5403ffb92]
/opt/relion/3.1.2/bin/relion_refine_mpi(main+0x4a) [0x55b5403c890a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7fe681ebd09b]
/opt/relion/3.1.2/bin/relion_refine_mpi(_start+0x2a) [0x55b5403cbc5a]
==================
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN11RelionErrorC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x67) [0x5601b51b2c67]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN10Experiment29divideParticlesInRandomHalvesEib+0x1142) [0x5601b5256e92]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi18initialiseWorkLoadEv+0x261) [0x5601b51d2b91]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi10initialiseEv+0x9d2) [0x5601b51d4b92]
/opt/relion/3.1.2/bin/relion_refine_mpi(main+0x4a) [0x5601b519d90a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7f61cc5bc09b]
/opt/relion/3.1.2/bin/relion_refine_mpi(_start+0x2a) [0x5601b51a0c5a]
==================
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
in: /packages/RELION/relion-3.1.2/src/exp_model.cpp, line 344
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
=== Backtrace  ===
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN11RelionErrorC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x67) [0x55f93d874c67]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN10Experiment29divideParticlesInRandomHalvesEib+0x1142) [0x55f93d918e92]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi18initialiseWorkLoadEv+0x261) [0x55f93d894b91]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi10initialiseEv+0x9d2) [0x55f93d896b92]
/opt/relion/3.1.2/bin/relion_refine_mpi(main+0x4a) [0x55f93d85f90a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7f88a4c3f09b]
/opt/relion/3.1.2/bin/relion_refine_mpi(_start+0x2a) [0x55f93d862c5a]
==================
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN11RelionErrorC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEES7_l+0x67) [0x5627f4b11c67]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN10Experiment29divideParticlesInRandomHalvesEib+0x1142) [0x5627f4bb5e92]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi18initialiseWorkLoadEv+0x261) [0x5627f4b31b91]
/opt/relion/3.1.2/bin/relion_refine_mpi(_ZN14MlOptimiserMpi10initialiseEv+0x9d2) [0x5627f4b33b92]
/opt/relion/3.1.2/bin/relion_refine_mpi(main+0x4a) [0x5627f4afc90a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xeb) [0x7f91e892e09b]
/opt/relion/3.1.2/bin/relion_refine_mpi(_start+0x2a) [0x5627f4affc5a]
==================
ERROR: 
ERROR: one of your half sets has no segments. Is rlnRandomSubset set to 1 or 2 in your particles STAR file? Or in case you're doing helical, half-sets are always per-filament, so provide at least 2 filaments.
[89679] 6 more processes have sent help message help-mpi-api.txt / mpi-abort
[89679] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages

Am I doing something wrong with the conversion csparc2star.py?

thanks in advance for your feedback.

If you could post the the header and first particle line of your star file used for refinement, it would help assess this. It is normal for _rlnRandomSubset to be either 1 or 2, perhaps it is empty and/or you should assign it?