search

opencobra / cobrapy

COBRApy is a package for constraint-based modeling of metabolic networks.
http://opencobra.github.io/cobrapy/
GNU General Public License v2.0
461 stars 216 forks source link

Enable the hybrid solver interface #1366

Closed cdiener closed 11 months ago

cdiener commented 11 months ago

This enables the hybrid solver from optlang.

Example

In [1]: from cobra.io import load_model

In [2]: mod = load_model("textbook")

In [3]: mod.solver.configuration.lp_method = "auto"

In [4]: mod.solver.configuration.qp_method = "auto"

In [5]: mod.solver = "hybrid"

In [6]: mod.optimize()
Out[6]: <Solution 0.874 at 0x7f27a95d20d0>

In [7]: from cobra.medium import minimal_medium

In [8]: minimal_medium(mod, minimize_components=True, open_exchanges=True)
/home/cdiener/code/cobrapy/src/cobra/medium/minimal_medium.py:112: FutureWarning: The default dtype for empty Series will be 'object' instead of 'float64' in a future version. Specify a dtype explicitly to silence this warning.
  medium = pd.Series()
Out[8]: 
EX_fru_e       7.040736
EX_gln__L_e    0.272640
EX_pi_e        0.367870
dtype: float64

In [9]: from cobra.flux_analysis import moma

In [10]: moma(mod, linear=False)  # kind of a stupid example, just to illustrate it works
Out[10]: <Solution 0.000 at 0x7f27a8b52390>

Things to consider

  1. There is currently an issue with the wheels for highspy. The wheels on PyPI only work with linux at the moment. There are some candiates here in the artifacts. Installing from source will work on all OSs. Broken wheels are handled by optlang and if code is run on a platform with a broken highspy wheel it will not be listed as an available solver.

  2. This is not specific to the hybrid interface but in some case one has to set the configuration for LP/QP methods before switching if the next solver does not support he same methods. Maybe this should be handled more gracefully in optlang by using the default and issuing a warning instead of raising an error.

  3. HIGHS and OSQP only support getting the full set of primals/duals at once, so there is no good way of getting primals for only one variable. Right now, the interface will thus obtain the full solution on every solve, which makes basis recycling a bit slower. So it is not the best solver for FVA or gene deletions.

  4. Is it okay if I also fix the minimal_medium pandas deprecation warning here?

Full test and benchmark results

``` ===================================================== test session starts ====================================================== platform linux -- Python 3.11.6, pytest-7.3.0, pluggy-1.0.0 benchmark: 4.0.0 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000) rootdir: /home/cdiener/code/cobrapy configfile: tox.ini testpaths: tests plugins: cov-4.0.0, mock-3.10.0, anyio-3.6.2, benchmark-4.0.0 collected 670 items tests/test_core/test_configuration.py ...... [ 0%] tests/test_core/test_core_reaction.py ...........s.........................................x....... [ 10%] tests/test_core/test_dictlist.py ...................... [ 13%] tests/test_core/test_formula.py ..... [ 14%] tests/test_core/test_gene.py . [ 14%] tests/test_core/test_gpr.py .....................x..x............................................. [ 24%] tests/test_core/test_group.py .. [ 24%] tests/test_core/test_metabolite.py ....s..s. [ 26%] tests/test_core/test_model.py ...............................................................s. [ 35%] tests/test_core/test_solution.py ..s [ 36%] tests/test_flux_analysis/test_deletion.py ....s..... [ 37%] tests/test_flux_analysis/test_fastcc.py .... [ 38%] tests/test_flux_analysis/test_geometric.py .. [ 38%] tests/test_flux_analysis/test_moma.py . [ 38%] tests/test_flux_analysis/test_parsimonious.py .. [ 39%] tests/test_flux_analysis/test_reaction.py . [ 39%] tests/test_flux_analysis/test_room.py .... [ 40%] tests/test_flux_analysis/test_variability.py ....... [ 41%] tests/test_flux_analysis/test_deletion.py .......... [ 42%] tests/test_flux_analysis/test_fastcc.py .... [ 43%] tests/test_flux_analysis/test_geometric.py .. [ 43%] tests/test_flux_analysis/test_moma.py . [ 43%] tests/test_flux_analysis/test_parsimonious.py .. [ 43%] tests/test_flux_analysis/test_reaction.py . [ 44%] tests/test_flux_analysis/test_room.py .... [ 44%] tests/test_flux_analysis/test_variability.py ....... [ 45%] tests/test_flux_analysis/test_deletion.py .......... [ 47%] tests/test_flux_analysis/test_fastcc.py .... [ 47%] tests/test_flux_analysis/test_geometric.py .. [ 48%] tests/test_flux_analysis/test_moma.py . [ 48%] tests/test_flux_analysis/test_parsimonious.py .. [ 48%] tests/test_flux_analysis/test_reaction.py . [ 48%] tests/test_flux_analysis/test_room.py .... [ 49%] tests/test_flux_analysis/test_variability.py ....... [ 50%] tests/test_flux_analysis/test_deletion.py ... [ 50%] tests/test_flux_analysis/test_moma.py . [ 50%] tests/test_flux_analysis/test_deletion.py ... [ 51%] tests/test_flux_analysis/test_moma.py . [ 51%] tests/test_flux_analysis/test_deletion.py ...... [ 52%] tests/test_flux_analysis/test_gapfilling.py . [ 52%] tests/test_flux_analysis/test_helpers.py ... [ 52%] tests/test_flux_analysis/test_loopless.py ....... [ 54%] tests/test_flux_analysis/test_phenotype_phase_plane.py ...... [ 54%] tests/test_flux_analysis/test_variability.py ....... [ 55%] tests/test_io/test_annotation.py .. [ 56%] tests/test_io/test_annotation_format.py .. [ 56%] tests/test_io/test_io_order.py ........................... [ 60%] tests/test_io/test_json.py .... [ 61%] tests/test_io/test_mat.py ............... [ 63%] tests/test_io/test_notes.py . [ 63%] tests/test_io/test_pickle.py .. [ 63%] tests/test_io/test_sbml.py .ssss............xxxx.x.x..s............. [ 70%] tests/test_io/test_yaml.py .x [ 70%] tests/test_io/test_web/test_load.py ....... [ 71%] tests/test_manipulation/test_annotate.py . [ 71%] tests/test_manipulation/test_delete.py ....... [ 72%] tests/test_manipulation/test_modify.py ... [ 72%] tests/test_manipulation/test_validate.py .. [ 73%] tests/test_medium/test_boundary_types.py ......... [ 74%] tests/test_medium/test_minimal_medium.py .......... [ 76%] tests/test_sampling/test_achr.py ...... [ 77%] tests/test_sampling/test_optgp.py ...... [ 77%] tests/test_sampling/test_sampling.py .......... [ 79%] tests/test_summary/test_metabolite_summary.py ................. [ 81%] tests/test_summary/test_model_summary.py ............... [ 84%] tests/test_summary/test_reaction_summary.py ........... [ 85%] tests/test_summary/test_metabolite_summary.py ................. [ 88%] tests/test_summary/test_model_summary.py ............... [ 90%] tests/test_summary/test_reaction_summary.py .................... [ 93%] tests/test_util/test_array.py .. [ 93%] tests/test_util/test_context.py ... [ 94%] tests/test_util/test_process_pool.py ............... [ 96%] tests/test_util/test_solver.py ..................s [ 99%] tests/test_util/test_util.py .... [100%] ======================================================= warnings summary ======================================================= tests/test_core/test_gpr.py::test_and_gpr[a AND b-2-gpr_genes2-a and b] tests/test_core/test_gpr.py::test_gpr_equality_with_bolean_logic[a_and_b_strs] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists0] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists1] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists2] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists3] /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule 'a AND b'. warn( tests/test_core/test_gpr.py::test_or_gpr[a OR b-2-gpr_genes2-a or b] tests/test_core/test_gpr.py::test_gpr_as_symbolic_boolean[a OR b-symbolic_gpr2] tests/test_core/test_gpr.py::test_gpr_equality_with_bolean_logic[a_or_b_strs] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists0] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists4] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists5] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists6] tests/test_core/test_gpr.py::test_gpr_from_symbolic[a OR b-symbolic_gpr5] /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule 'a OR b'. warn( tests/test_core/test_gpr.py::test_complicated_gpr[(a OR b) AND c] tests/test_core/test_gpr.py::test_gpr_as_symbolic_boolean[(a OR b) AND c-symbolic_gpr6] tests/test_core/test_gpr.py::test_gpr_equality_with_bolean_logic[a_b_c_or_and_strs] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists3] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists6] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists8] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists9] tests/test_core/test_gpr.py::test_gpr_from_symbolic[(a OR b) AND c-symbolic_gpr9] /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule '(a OR b) AND c'. warn( tests/test_core/test_gpr.py::test_gpr_that_needs_two_replacements /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule '(591001.3.peg.1891 AND 591001.3.peg.1892 AND 591001.3.peg.1893)'. warn( tests/test_core/test_gpr.py::test_gpr_equality_with_bolean_logic[a_and_b_strs] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists0] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists1] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists2] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists3] /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule 'b AND a'. warn( tests/test_core/test_gpr.py::test_gpr_equality_with_bolean_logic[a_or_b_strs] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists0] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists4] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists5] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists6] /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule 'b OR a'. warn( tests/test_core/test_gpr.py::test_gpr_equality_with_bolean_logic[a_b_c_or_strs] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists1] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists4] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists7] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists8] /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule 'a OR b or c'. warn( tests/test_core/test_gpr.py::test_gpr_equality_with_bolean_logic[a_b_c_and_strs] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists2] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists5] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists7] tests/test_core/test_gpr.py::test_gpr_inequality_boolean[gpr_lists9] /home/cdiener/code/cobrapy/src/cobra/core/gene.py:358: SyntaxWarning: Uppercase AND/OR found in rule 'a AND b and c'. warn( tests/test_core/test_model.py::test_reaction_delete tests/test_core/test_model.py::test_reaction_delete /home/cdiener/code/cobrapy/src/cobra/core/reaction.py:879: DeprecationWarning: delete is deprecated. Use reaction.remove_from_model instead warn( tests/test_core/test_model.py::test_group_loss_of_elements tests/test_flux_analysis/test_gapfilling.py::test_gapfilling /home/cdiener/code/cobrapy/src/cobra/core/group.py:147: UserWarning: need to pass in a list warn("need to pass in a list") tests/test_core/test_model.py::test_change_objective tests/test_core/test_model.py::test_change_objective tests/test_core/test_model.py::test_change_objective /home/cdiener/code/cobrapy/src/cobra/core/reaction.py:761: DeprecationWarning: Please use reaction.flux instead. warn("Please use reaction.flux instead.", DeprecationWarning) tests/test_flux_analysis/test_geometric.py: 1 warning tests/test_io/test_json.py: 6 warnings tests/test_io/test_mat.py: 14 warnings tests/test_io/test_pickle.py: 2 warnings tests/test_io/test_sbml.py: 18 warnings tests/test_io/test_yaml.py: 2 warnings /home/cdiener/code/cobrapy/src/cobra/util/solver.py:554: UserWarning: Solver status is 'infeasible'. warn(f"Solver status is '{status}'.", UserWarning) tests/test_medium/test_minimal_medium.py: 274 warnings /home/cdiener/code/cobrapy/src/cobra/medium/minimal_medium.py:112: FutureWarning: The default dtype for empty Series will be 'object' instead of 'float64' in a future version. Specify a dtype explicitly to silence this warning. medium = pd.Series() tests/test_util/test_process_pool.py::test_with_context /home/cdiener/code/cobrapy/src/cobra/util/process_pool.py:72: PytestMockWarning: Mocks returned by pytest-mock do not need to be used as context managers. The mocker fixture automatically undoes mocking at the end of a test. This warning can be ignored if it was triggered by mocking a context manager. https://pytest-mock.readthedocs.io/en/latest/remarks.html#usage-as-context-manager self._pool.__enter__() -- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html -------------------------------------------------------------------------------------------------------------------------------- benchmark: 56 tests -------------------------------------------------------------------------------------------------------------------------------- Name (time in us) Min Max Mean StdDev Median IQR Outliers OPS Rounds Iterations ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- test_subtract_metabolite_benchmark[gurobi] 2.2210 (1.0) 42.8750 (1.0) 2.5678 (1.0) 1.6440 (1.0) 2.4480 (1.0) 0.1160 (1.0) 25;378 389,445.0767 (1.0) 4659 1 test_subtract_metabolite_benchmark[cplex] 2.2890 (1.03) 68.3980 (1.60) 4.1441 (1.61) 3.7591 (2.29) 3.0860 (1.26) 2.0360 (17.55) 164;171 241,305.4923 (0.62) 6378 1 test_subtract_metabolite_benchmark[glpk] 2.3340 (1.05) 48.0570 (1.12) 4.9202 (1.92) 2.8143 (1.71) 4.5730 (1.87) 0.9762 (8.42) 29;102 203,242.1688 (0.52) 1943 1 test_change_objective_benchmark[optlang-glpk] 120.6360 (54.32) 522.4520 (12.19) 149.3786 (58.17) 38.1732 (23.22) 132.2330 (54.02) 24.0393 (207.23) 368;368 6,694.3987 (0.02) 2601 1 test_change_objective_benchmark[optlang-cplex] 199.1030 (89.65) 459.3160 (10.71) 223.5738 (87.07) 43.9883 (26.76) 210.1320 (85.84) 21.0570 (181.52) 4;4 4,472.7953 (0.01) 48 1 test_achr_sample_benchmark 240.2170 (108.16) 15,246.6630 (355.61) 383.1738 (149.23) 722.3710 (439.40) 310.4670 (126.82) 168.2637 (>1000.0) 4;7 2,609.7819 (0.01) 967 1 test_optgp_sample_benchmark 310.3160 (139.72) 847.1020 (19.76) 433.5972 (168.86) 118.9395 (72.35) 374.1195 (152.83) 237.7370 (>1000.0) 331;0 2,306.2878 (0.01) 1014 1 test_loopless_benchmark_after 785.5410 (353.69) 2,219.7130 (51.77) 892.6913 (347.65) 142.5749 (86.72) 830.3280 (339.19) 129.0415 (>1000.0) 110;64 1,120.2081 (0.00) 980 1 test_loopless_benchmark_before 942.2160 (424.23) 3,020.4740 (70.45) 1,182.0751 (460.35) 241.3384 (146.80) 1,117.8755 (456.65) 188.4715 (>1000.0) 61;43 845.9700 (0.00) 692 1 test_minimal_medium_mip_benchmark 4,641.6310 (>1000.0) 10,877.9680 (253.71) 5,577.8353 (>1000.0) 1,045.6055 (636.01) 5,150.4305 (>1000.0) 1,408.0545 (>1000.0) 22;3 179.2810 (0.00) 168 1 test_minimal_medium_linear_benchmark 4,746.4850 (>1000.0) 12,780.4260 (298.09) 6,007.7749 (>1000.0) 1,645.3141 (>1000.0) 5,223.0170 (>1000.0) 1,485.7672 (>1000.0) 12;7 166.4510 (0.00) 91 1 test_copy_benchmark[optlang-glpk] 8,639.2220 (>1000.0) 215,628.2820 (>1000.0) 13,219.7811 (>1000.0) 22,812.2820 (>1000.0) 10,561.0590 (>1000.0) 1,718.6073 (>1000.0) 1;3 75.6442 (0.00) 81 1 test_copy_benchmark[optlang-cplex] 8,842.8780 (>1000.0) 14,316.7610 (333.92) 10,919.7518 (>1000.0) 1,222.4106 (743.56) 10,966.1480 (>1000.0) 1,376.3210 (>1000.0) 9;1 91.5772 (0.00) 30 1 test_deepcopy_benchmark 15,694.4280 (>1000.0) 314,824.4660 (>1000.0) 26,095.8490 (>1000.0) 44,629.7774 (>1000.0) 18,967.8615 (>1000.0) 4,623.8495 (>1000.0) 1;1 38.3203 (0.00) 44 1 test_fastcc_benchmark[glpk] 18,808.2300 (>1000.0) 27,621.1050 (644.22) 20,942.8220 (>1000.0) 1,611.5068 (980.24) 20,809.4850 (>1000.0) 1,774.9317 (>1000.0) 11;1 47.7491 (0.00) 43 1 test_gpr_equality_benchmark 20,437.4840 (>1000.0) 25,558.5890 (596.12) 22,338.0347 (>1000.0) 1,337.4611 (813.54) 22,115.0325 (>1000.0) 1,819.1530 (>1000.0) 9;0 44.7667 (0.00) 32 1 test_single_reaction_deletion_benchmark[glpk] 20,840.1330 (>1000.0) 24,988.4670 (582.82) 22,332.7618 (>1000.0) 1,081.6181 (657.92) 22,378.5120 (>1000.0) 1,209.3540 (>1000.0) 12;3 44.7773 (0.00) 37 1 test_single_gene_deletion_linear_moma_benchmark[glpk] 23,055.7340 (>1000.0) 38,615.4990 (900.65) 26,319.2084 (>1000.0) 3,245.1638 (>1000.0) 25,390.9915 (>1000.0) 3,970.6340 (>1000.0) 5;1 37.9951 (0.00) 30 1 test_fastcc_benchmark[cplex] 23,700.8330 (>1000.0) 32,639.7080 (761.28) 25,774.7391 (>1000.0) 2,181.1494 (>1000.0) 25,081.3380 (>1000.0) 2,383.9635 (>1000.0) 6;2 38.7977 (0.00) 36 1 test_single_gene_deletion_linear_room_benchmark[glpk] 25,950.2090 (>1000.0) 47,668.3800 (>1000.0) 31,951.9793 (>1000.0) 5,611.0599 (>1000.0) 31,134.1310 (>1000.0) 6,255.3290 (>1000.0) 6;1 31.2970 (0.00) 19 1 test_single_gene_deletion_moma_benchmark[cplex] 28,314.8480 (>1000.0) 40,975.5620 (955.70) 34,476.9840 (>1000.0) 4,075.9781 (>1000.0) 33,988.3440 (>1000.0) 8,056.1982 (>1000.0) 11;0 29.0049 (0.00) 25 1 test_single_gene_deletion_linear_room_benchmark[cplex] 30,500.0670 (>1000.0) 68,856.1120 (>1000.0) 42,649.5639 (>1000.0) 10,259.0894 (>1000.0) 42,607.3730 (>1000.0) 13,912.1007 (>1000.0) 6;1 23.4469 (0.00) 17 1 test_single_gene_deletion_linear_moma_benchmark[cplex] 33,484.8950 (>1000.0) 46,683.3010 (>1000.0) 37,811.7433 (>1000.0) 4,233.0393 (>1000.0) 36,566.5220 (>1000.0) 5,082.0500 (>1000.0) 5;0 26.4468 (0.00) 22 1 test_single_reaction_deletion_benchmark[cplex] 35,597.8390 (>1000.0) 48,411.2350 (>1000.0) 41,711.3557 (>1000.0) 4,106.7386 (>1000.0) 41,938.5000 (>1000.0) 7,516.9975 (>1000.0) 9;0 23.9743 (0.00) 24 1 test_fastcc_benchmark[hybrid] 43,641.5370 (>1000.0) 80,763.9400 (>1000.0) 55,096.7471 (>1000.0) 8,855.5439 (>1000.0) 53,501.7400 (>1000.0) 3,900.5073 (>1000.0) 4;4 18.1499 (0.00) 15 1 test_add_metabolite_benchmark[gurobi] 47,171.5360 (>1000.0) 51,967.1480 (>1000.0) 48,282.8281 (>1000.0) 1,196.3889 (727.73) 47,844.2310 (>1000.0) 1,031.6945 (>1000.0) 2;2 20.7113 (0.00) 19 1 test_add_metabolite_benchmark[cplex] 47,986.2910 (>1000.0) 60,508.3480 (>1000.0) 49,912.8654 (>1000.0) 3,083.5317 (>1000.0) 49,053.2230 (>1000.0) 1,153.1555 (>1000.0) 2;3 20.0349 (0.00) 20 1 test_add_metabolite_benchmark[glpk] 48,359.1450 (>1000.0) 53,027.6900 (>1000.0) 49,791.9914 (>1000.0) 1,125.7087 (684.74) 49,626.0440 (>1000.0) 934.4312 (>1000.0) 5;2 20.0836 (0.00) 21 1 test_flux_variability_benchmark[cplex] 52,722.9120 (>1000.0) 69,932.2230 (>1000.0) 58,140.0347 (>1000.0) 5,739.9571 (>1000.0) 57,816.9480 (>1000.0) 7,145.2955 (>1000.0) 1;0 17.1999 (0.00) 8 1 test_pfba_benchmark[cplex] 63,524.4740 (>1000.0) 75,204.5270 (>1000.0) 67,568.0630 (>1000.0) 4,581.2748 (>1000.0) 66,729.9390 (>1000.0) 5,462.4390 (>1000.0) 1;0 14.7999 (0.00) 6 1 test_single_gene_deletion_fba_benchmark[glpk] 84,127.1660 (>1000.0) 104,777.2040 (>1000.0) 89,979.0596 (>1000.0) 6,836.7151 (>1000.0) 88,520.1590 (>1000.0) 3,514.5265 (>1000.0) 1;1 11.1137 (0.00) 7 1 test_single_gene_deletion_fba_benchmark[cplex] 84,977.2520 (>1000.0) 115,968.7560 (>1000.0) 95,540.5132 (>1000.0) 9,283.1446 (>1000.0) 92,018.1720 (>1000.0) 12,479.2015 (>1000.0) 3;0 10.4668 (0.00) 11 1 test_single_gene_deletion_linear_moma_benchmark[hybrid] 92,090.5690 (>1000.0) 132,243.2630 (>1000.0) 100,250.8305 (>1000.0) 11,864.3636 (>1000.0) 97,044.2080 (>1000.0) 8,565.2140 (>1000.0) 1;1 9.9750 (0.00) 10 1 test_single_gene_deletion_linear_room_benchmark[hybrid] 98,121.1920 (>1000.0) 136,736.3840 (>1000.0) 111,687.3335 (>1000.0) 11,959.1933 (>1000.0) 108,055.0675 (>1000.0) 14,080.9830 (>1000.0) 3;0 8.9536 (0.00) 10 1 test_flux_variability_loopless_benchmark[cplex] 108,310.3300 (>1000.0) 147,710.6330 (>1000.0) 122,901.2899 (>1000.0) 13,485.2730 (>1000.0) 117,784.3450 (>1000.0) 20,795.1735 (>1000.0) 3;0 8.1366 (0.00) 9 1 test_achr_init_benchmark 109,112.3240 (>1000.0) 330,553.2910 (>1000.0) 157,740.4424 (>1000.0) 74,120.0931 (>1000.0) 120,730.6570 (>1000.0) 78,853.5793 (>1000.0) 1;1 6.3395 (0.00) 9 1 test_single_gene_deletion_fba_benchmark[hybrid] 113,241.4500 (>1000.0) 137,132.8390 (>1000.0) 121,290.4381 (>1000.0) 8,749.0715 (>1000.0) 118,673.8255 (>1000.0) 13,014.4585 (>1000.0) 2;0 8.2447 (0.00) 8 1 test_flux_variability_loopless_benchmark[glpk] 115,074.5190 (>1000.0) 138,949.7130 (>1000.0) 128,862.1222 (>1000.0) 10,297.1189 (>1000.0) 135,734.1680 (>1000.0) 18,519.6488 (>1000.0) 2;0 7.7602 (0.00) 9 1 test_flux_variability_loopless_benchmark[hybrid] 137,314.7830 (>1000.0) 144,399.8830 (>1000.0) 139,559.4689 (>1000.0) 2,447.3373 (>1000.0) 138,523.3675 (>1000.0) 2,867.5130 (>1000.0) 2;0 7.1654 (0.00) 8 1 test_pfba_benchmark[glpk] 149,147.0140 (>1000.0) 171,139.6850 (>1000.0) 156,251.5560 (>1000.0) 9,017.6511 (>1000.0) 151,872.4360 (>1000.0) 11,127.5995 (>1000.0) 1;0 6.3999 (0.00) 5 1 test_geometric_fba_benchmark[cplex] 165,014.1640 (>1000.0) 318,751.5180 (>1000.0) 211,051.4112 (>1000.0) 63,702.0079 (>1000.0) 187,517.5830 (>1000.0) 75,667.7337 (>1000.0) 1;0 4.7382 (0.00) 5 1 test_geometric_fba_benchmark[glpk] 165,591.7880 (>1000.0) 212,340.4690 (>1000.0) 192,479.0373 (>1000.0) 18,253.2109 (>1000.0) 198,895.7325 (>1000.0) 29,966.9620 (>1000.0) 2;0 5.1954 (0.00) 6 1 test_optgp_init_benchmark 169,941.9070 (>1000.0) 209,082.7270 (>1000.0) 185,200.1122 (>1000.0) 14,503.3889 (>1000.0) 182,255.4230 (>1000.0) 13,667.8935 (>1000.0) 2;0 5.3996 (0.00) 5 1 test_double_gene_deletion_benchmark 180,012.9130 (>1000.0) 223,338.0280 (>1000.0) 198,281.6172 (>1000.0) 16,851.7959 (>1000.0) 200,056.9400 (>1000.0) 23,134.4025 (>1000.0) 2;0 5.0433 (0.00) 5 1 test_gpr_symbolism_benchmark 184,335.1930 (>1000.0) 292,671.0500 (>1000.0) 217,380.2356 (>1000.0) 42,954.1847 (>1000.0) 204,945.9880 (>1000.0) 31,897.1125 (>1000.0) 1;1 4.6002 (0.00) 5 1 test_copy_benchmark_large_model[optlang-cplex] 188,534.2380 (>1000.0) 453,784.6940 (>1000.0) 250,888.7276 (>1000.0) 113,911.5680 (>1000.0) 204,521.9650 (>1000.0) 83,553.1767 (>1000.0) 1;1 3.9858 (0.00) 5 1 test_pfba_benchmark[hybrid] 220,623.0740 (>1000.0) 606,065.1290 (>1000.0) 315,945.9808 (>1000.0) 162,754.2904 (>1000.0) 249,365.9490 (>1000.0) 103,682.9632 (>1000.0) 1;1 3.1651 (0.00) 5 1 test_copy_benchmark_large_model[optlang-glpk] 222,081.8320 (>1000.0) 413,136.2280 (>1000.0) 294,932.2254 (>1000.0) 94,491.3229 (>1000.0) 229,418.8770 (>1000.0) 163,661.6210 (>1000.0) 1;0 3.3906 (0.00) 5 1 test_single_reaction_deletion_benchmark[hybrid] 264,429.7450 (>1000.0) 326,895.6420 (>1000.0) 295,573.7994 (>1000.0) 24,217.9928 (>1000.0) 302,236.7100 (>1000.0) 34,803.6612 (>1000.0) 2;0 3.3832 (0.00) 5 1 test_single_gene_deletion_moma_benchmark[hybrid] 308,420.1310 (>1000.0) 364,447.5200 (>1000.0) 331,691.1660 (>1000.0) 22,660.2764 (>1000.0) 333,746.7560 (>1000.0) 34,370.7205 (>1000.0) 2;0 3.0149 (0.00) 5 1 test_flux_variability_benchmark[glpk] 316,016.0120 (>1000.0) 349,599.7290 (>1000.0) 339,069.1798 (>1000.0) 13,435.8760 (>1000.0) 342,282.6260 (>1000.0) 13,813.9683 (>1000.0) 1;0 2.9493 (0.00) 5 1 test_double_reaction_deletion_benchmark 609,252.9860 (>1000.0) 854,443.7430 (>1000.0) 773,973.5060 (>1000.0) 97,167.3241 (>1000.0) 799,363.0170 (>1000.0) 106,056.0932 (>1000.0) 1;0 1.2920 (0.00) 5 1 test_geometric_fba_benchmark[hybrid] 1,023,727.1270 (>1000.0) 1,658,614.1540 (>1000.0) 1,346,921.9078 (>1000.0) 292,585.9094 (>1000.0) 1,267,277.0150 (>1000.0) 541,621.3995 (>1000.0) 3;0 0.7424 (0.00) 5 1 test_flux_variability_benchmark[hybrid] 4,857,905.6960 (>1000.0) 5,153,362.8900 (>1000.0) 5,055,272.9016 (>1000.0) 115,518.2066 (>1000.0) 5,084,840.4870 (>1000.0) 114,921.6588 (>1000.0) 1;0 0.1978 (0.00) 5 1 test_single_gene_deletion_room_benchmark[cplex] 6,066,726.0810 (>1000.0) 7,149,963.7760 (>1000.0) 6,566,601.3488 (>1000.0) 447,866.3932 (>1000.0) 6,612,153.0110 (>1000.0) 747,221.5107 (>1000.0) 2;0 0.1523 (0.00) 5 1 test_single_gene_deletion_room_benchmark[hybrid] 60,739,469.0210 (>1000.0) 170,085,329.6590 (>1000.0) 106,933,118.6170 (>1000.0) 55,809,899.5292 (>1000.0) 77,876,918.9120 (>1000.0) 105,419,248.2898 (>1000.0) 2;0 0.0094 (0.00) 5 1 ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Legend: Outliers: 1 Standard Deviation from Mean; 1.5 IQR (InterQuartile Range) from 1st Quartile and 3rd Quartile. OPS: Operations Per Second, computed as 1 / Mean ============================ 648 passed, 12 skipped, 10 xfailed, 368 warnings in 980.40s (0:16:20) ============================= ```
Midnighter commented 11 months ago

Btw, I saw that both osqp and highs are available as conda packages, so maybe our recommendation for Mac & Windows platforms could be to use conda?

cdiener commented 11 months ago

Btw, I saw that both osqp and highs are available as conda packages, so maybe our recommendation for Mac & Windows platforms could be to use conda?

I wish, but unfortunately the highs conda package is only the compiled (C) library. It does not include highspy (the Python interface).

I did notice in the CI though that the current wheels actually do work on Windows. So it's only MAC that does not work. But I do think this will be fixed in the coming months. I do know the HIGHS team is working on fixing the wheels.