LCPQ / quantum_package

Set of quantum chemistry programs and libraries
https://quantumpackage.github.io/qp2/
GNU General Public License v2.0
44 stars 35 forks source link

Error: Cannot locate deriver sexp #232

Closed kgasperich closed 6 years ago

kgasperich commented 6 years ago

Written by Thomas

make
./create_git_sha1.sh
rm -f -- qp_print
ocamlbuild -j 0 -cflags "-g" -lflags "-g"  qp_print.native -use-ocamlfind 
+ ocamlfind ocamldep -package core,cryptokit,zmq,str,ppx_sexp_conv,ppx_deriving -modules Charge.mli > Charge.mli.depends
File "Charge.mli", line 1, characters 27-31:
Error: Cannot locate deriver sexp
Command exited with code 2.
Compilation unsuccessful after building 2 targets (1 cached) in 00:00:00.
make: *** [qp_print.native] Error 10
>opam --version
1.2.2

>ocaml --version
The OCaml toplevel, version 4.06.0

>opam install ppx_sexp_conv
[NOTE] Package ppx_sexp_conv is already installed (current version is v0.11.0).

Potential relevant links: https://github.com/janestreet-deprecated/ppx_core/issues/1 https://github.com/janestreet/ppx_sexp_conv/issues/1

💇‍♂️

TApplencourt commented 6 years ago

Side note: In the new version of opam, ZMQ seem to be renamed into zmq. More precissely you can opam install ZMQ or opam install zmq, but the ocamlfind will only work with zmq (no capital).

You may need to update the .merlin file.

TApplencourt commented 6 years ago

I rebuild the last Travis after removing the Ocaml/Opam cache and it's working (https://travis-ci.org/LCPQ/quantum_package/builds/377046178)

ocam 4.06.0
ppx_sexp_conv  v0.10.0 

They use the version 0.10 and not 0.11 of ppx_sexp_conv

Maybe this is the issue. Can you try to check @kgasperich?

anbenali commented 6 years ago

Quantum Package has been reinstalled properly on theta and cooley and can be found on /soft/applications/quantum_package

By default the network is selected to be "ib0". In previous version, I was still losing packets in the zmq_communication for unknown reasons...

f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, ntask, 4, ZMQ_SNDMORE) f77_zmq_send( zmq_socket_push, task_id(1), ntask*4, 0) push_selection_results: error in receiving ok

This job was run with the following submission script:

!/bin/bash

COBALT -q default

COBALT -A QMCPACK_AESP

COBALT -n 61

COBALT -t 180

COBALT --attrs mcdram=cache:numa=quad

COBALT -O diamondC_2x1x1-CIPSI-4

echo $COBALT_JOBID echo $COBALT_JOBSIZE echo $COBALT_PARTNAME

QPTYPE0="fci_zmq" QPTYPE1="-slave selection_davidson_slave"

NCORES=64 HT=1 NTHREADS=$((NCORES * HT))

source /soft/applications/quantum_package/quantum_package.rc

Input=diamondC_2x1x1-3.37316115

aprun -n 1 -N 1 -cc none -j $HT -d $NTHREADS qp_run ${QPTYPE0} ${Input}.ezfio > ${Input}_FCI4.out & sleep 55

for i in {1..6} do sleep 5 aprun -n 10 -N 1 -cc depth -j $HT -d $NTHREADS qp_run ${QPTYPE1} ${Input}.ezfio >> ${Input}_Slave-FCI4.out & done

wait

I will let you know if the behavior has improved with the new zmq

Anouar

On Wed, May 9, 2018 at 5:57 PM, Thomas Applencourt <notifications@github.com

wrote:

I rebuild the last Travis after remove the Ocaml/Opam cache and it's working (https://travis-ci.org/LCPQ/quantum_package/builds/377046178)

ocam 4.06.0 ppx_sexp_conv v0.10.0

They use the version 0.10 and not 0.11 of ppx_sexp_conv

May this is the issue. Can you try to check @kgasperich https://github.com/kgasperich?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/LCPQ/quantum_package/issues/232#issuecomment-387899723, or mute the thread https://github.com/notifications/unsubscribe-auth/AXl6w28c-yAFyipahvuM6vB9EvKf4WBIks5tw3RNgaJpZM4T3YPy .

--

Anouar Benali, PhD Leadership Computing Facility Argonne National Laboratory Building 240 Office - 2127 9700 S Cass Av., Argonne Il, 60439 (630) 252-0058

kgasperich commented 6 years ago

@TApplencourt I am now able to build without errors, but the two tests in the tests directory fail


 >./run_tests.sh 

-~-~-~-~-~-~

Running tests for convert

✗ gamess convert HBO.out
   (in test file bats/convert.bats, line 11)
     `qp_run SCF HBO.out.ezfio' failed
   2018-05-10 09:47:19.512396-04:00
   ===============
   Quantum Package
   ===============

   Git Commit: Change the url for GMP. The officiel website is down sometime
   Git Date  : Wed May 9 17:17:12 2018 -0500
   Git SHA1  : 850c438956a0320a46b9002400ffb3544b5f6606

   Task server running : tcp://136.142.28.137:41279

   .. >>>>> [ WALL TIME:        0.000000 s ] [ CPU  TIME:        0.007288 s ] <<<<< ..

    Read  ao_num
   * mo_tot_num                                                    20
    Read  mo_coef

   .. >>>>> [ WALL TIME:        0.028400 s ] [ CPU  TIME:        0.008528 s ] <<<<< ..

    Read  ao_prim_num

   .. >>>>> [ WALL TIME:        0.039400 s ] [ CPU  TIME:        0.008933 s ] <<<<< ..

    Read  ao_expo

   .. >>>>> [ WALL TIME:        0.051000 s ] [ CPU  TIME:        0.009907 s ] <<<<< ..

    Read  ao_coef

   .. >>>>> [ WALL TIME:        0.062800 s ] [ CPU  TIME:        0.010806 s ] <<<<< ..

    Read  ao_power

   .. >>>>> [ WALL TIME:        0.074900 s ] [ CPU  TIME:        0.010806 s ] <<<<< ..

    Read  ao_nucl

   .. >>>>> [ WALL TIME:        0.075600 s ] [ CPU  TIME:        0.010806 s ] <<<<< ..

    Read  nucl_num

   .. >>>>> [ WALL TIME:        0.091100 s ] [ CPU  TIME:        0.011856 s ] <<<<< ..

    Read  nucl_label

   .. >>>>> [ WALL TIME:        0.108600 s ] [ CPU  TIME:        0.011856 s ] <<<<< ..

    Read  nucl_charge

   .. >>>>> [ WALL TIME:        0.121700 s ] [ CPU  TIME:        0.011856 s ] <<<<< ..

   Nuclear Coordinates (Angstroms)
   ===============================

   ================ ============ ============ ============ ============
         Atom          Charge          X            Y            Z 
   ================ ============ ============ ============ ============
   H                    1.000000     0.000000     0.000000     0.000000
   B                    5.000000     1.166000     0.000000     0.000000
   B                    8.000000     2.366000     0.000000     0.000000
   ================ ============ ============ ============ ============

    * mo_label          Orthonormalized

   .. >>>>> [ WALL TIME:        0.150300 s ] [ CPU  TIME:        0.285588 s ] <<<<< ..

    Read  elec_beta_num     

   .. >>>>> [ WALL TIME:        0.150600 s ] [ CPU  TIME:        0.289294 s ] <<<<< ..

    Read  elec_alpha_num
    Read mo_occ

   .. >>>>> [ WALL TIME:        0.164600 s ] [ CPU  TIME:        0.442287 s ] <<<<< ..

    Read  ao_md5
   Wall time : 2.11924s

   /ihome/kjordan/keg56/anl/qp/quantum_package/tests/work/HBO.out recognized as gamessFile

   Possible choices:
   * H_CORE_guess
   * check_orthonormality
   * davidson_slave
   * diagonalize_restart_and_save_all_states
   * diagonalize_restart_and_save_lowest_state
   * diagonalize_restart_and_save_one_state
   * diagonalize_restart_and_save_two_states
   * four_idx_transform
   * guess_doublet
   * guess_lowest_state
   * guess_singlet
   * guess_triplet
   * mo_permutation
   * print_H_matrix_restart
   * print_aos
   * print_bitmask
   * print_energy
   * print_holes_particles
   * print_mo_in_space
   * print_wf
   * qp_ao_ints
   * save_HF_determinant
   * save_natorb
   * save_only_singles
   * save_ortho_mos
   * swap_mos
   * test_integrals
   * truncate_wf
   Uncaught exception:

     (Failure "Executable SCF not found")

   Raised at file "pervasives.ml", line 32, characters 17-33
   Called from file "qp_run.ml", line 61, characters 8-49
   Called from file "src/command.ml", line 2734, characters 6-97
   Called from file "src/exn.ml", line 108, characters 6-10
 ✗ g09 convert H2O.log
   (in test file bats/convert.bats, line 22)
     `qp_run SCF h2o.log.ezfio' failed
   2018-05-10 09:47:23.900480-04:00
   ===============
   Quantum Package
   ===============

   Git Commit: Change the url for GMP. The officiel website is down sometime
   Git Date  : Wed May 9 17:17:12 2018 -0500
   Git SHA1  : 850c438956a0320a46b9002400ffb3544b5f6606

   Task server running : tcp://136.142.28.137:41279

   .. >>>>> [ WALL TIME:        0.000000 s ] [ CPU  TIME:        0.006590 s ] <<<<< ..

    Read  ao_num
   * mo_tot_num                                                    24
    Read  mo_coef

   .. >>>>> [ WALL TIME:        0.028100 s ] [ CPU  TIME:        0.008028 s ] <<<<< ..

    Read  ao_prim_num

   .. >>>>> [ WALL TIME:        0.040300 s ] [ CPU  TIME:        0.008580 s ] <<<<< ..

    Read  ao_expo

   .. >>>>> [ WALL TIME:        0.052700 s ] [ CPU  TIME:        0.008886 s ] <<<<< ..

    Read  ao_coef

   .. >>>>> [ WALL TIME:        0.064700 s ] [ CPU  TIME:        0.009261 s ] <<<<< ..

    Read  ao_power

   .. >>>>> [ WALL TIME:        0.077600 s ] [ CPU  TIME:        0.009821 s ] <<<<< ..

    Read  ao_nucl

   .. >>>>> [ WALL TIME:        0.078200 s ] [ CPU  TIME:        0.009960 s ] <<<<< ..

    Read  nucl_num

   .. >>>>> [ WALL TIME:        0.090900 s ] [ CPU  TIME:        0.010547 s ] <<<<< ..

    Read  nucl_label

   .. >>>>> [ WALL TIME:        0.104500 s ] [ CPU  TIME:        0.011103 s ] <<<<< ..

    Read  nucl_charge

   .. >>>>> [ WALL TIME:        0.117300 s ] [ CPU  TIME:        0.011497 s ] <<<<< ..

   Nuclear Coordinates (Angstroms)
   ===============================

   ================ ============ ============ ============ ============
         Atom          Charge          X            Y            Z 
   ================ ============ ============ ============ ============
   H                    1.000000     0.000000     0.751000    -0.465600
   O                    8.000000     0.000000     0.000000     0.116400
   H                    1.000000     0.000000    -0.751000    -0.465600
   ================ ============ ============ ============ ============

    * mo_label          Orthonormalized

   .. >>>>> [ WALL TIME:        0.145500 s ] [ CPU  TIME:        0.280484 s ] <<<<< ..

    Read  elec_beta_num

   .. >>>>> [ WALL TIME:        0.145700 s ] [ CPU  TIME:        0.282618 s ] <<<<< ..

    Read  elec_alpha_num
    Read mo_occ

   .. >>>>> [ WALL TIME:        0.159600 s ] [ CPU  TIME:        0.433763 s ] <<<<< ..

    Read  ao_md5
   Wall time : 2.13075s

   /ihome/kjordan/keg56/anl/qp/quantum_package/tests/work/h2o.log recognized as gaussianFile

   Possible choices:
   * H_CORE_guess
   * check_orthonormality
   * davidson_slave
   * diagonalize_restart_and_save_all_states
   * diagonalize_restart_and_save_lowest_state
   * diagonalize_restart_and_save_one_state
   * diagonalize_restart_and_save_two_states
   * four_idx_transform
   * guess_doublet
   * guess_lowest_state
   * guess_singlet
   * guess_triplet
   * mo_permutation
   * print_H_matrix_restart
   * print_aos
   * print_bitmask
   * print_energy
   * print_holes_particles
   * print_mo_in_space
   * print_wf
   * qp_ao_ints
   * save_HF_determinant
   * save_natorb
   * save_only_singles
   * save_ortho_mos
   * swap_mos
   * test_integrals
   * truncate_wf
   Uncaught exception:

     (Failure "Executable SCF not found")

   Raised at file "pervasives.ml", line 32, characters 17-33
   Called from file "qp_run.ml", line 61, characters 8-49
   Called from file "src/command.ml", line 2734, characters 6-97
   Called from file "src/exn.ml", line 108, characters 6-10

2 tests, 2 failures
TApplencourt commented 6 years ago

@kgasperich
Nice! This one is not a bug! :) (or a least I know how to fix it).

You need to install the plugins who contain the required executable:

qp_module.py install Full_CI_ZMQ Hartree_Fock

for example

kgasperich commented 6 years ago

@TApplencourt Thanks, that makes sense; I looked for SCF and RHF in the list of available plugins, but somehow I managed to look past Hartree_Fock

tests pass now:

>./run_tests.sh 

-~-~-~-~-~-~

Running tests for convert

 ✓ gamess convert HBO.out
 ✓ g09 convert H2O.log

2 tests, 0 failures

-~-~-~-~-~-~

Running tests for hf

 ✓ init DHNO chipman-dzp
 ✓ SCF DHNO chipman-dzp
 ✓ init HBO STO-3G
 ✓ SCF HBO STO-3G
 ✓ init H2O cc-pVDZ
 ✓ SCF H2O cc-pVDZ

6 tests, 0 failures

-~-~-~-~-~-~

Running tests for fci

 ✓ qp_set_mo_class H2O cc-pVDZ
 - FCI H2O cc-pVDZ (skipped)
 ✓ FCI-ZMQ H2O cc-pVDZ

3 tests, 0 failures, 1 skipped

-~-~-~-~-~-~

Running tests for cassd

 - CAS_SD H2O cc-pVDZ (skipped)

1 test, 0 failures, 1 skipped

-~-~-~-~-~-~

Running tests for mrcepa0

 - MRCC-lambda H2O cc-pVDZ (skipped)
 - MRCC H2O cc-pVDZ (skipped)
 - MRSC2 H2O cc-pVDZ (skipped)
 - MRCEPA0 H2O cc-pVDZ (skipped)

4 tests, 0 failures, 4 skipped

-~-~-~-~-~-~

Running tests for pseudo

 ✓ init H2O VDZ pseudo
 ✓ SCF H2O VDZ pseudo
 ✓ FCI H2O VDZ pseudo

3 tests, 0 failures