|
201 | 201 | "source": [
|
202 | 202 | "**What happened?**\n",
|
203 | 203 | "\n",
|
204 |
| - "The forward simulation failed because AMICI the solver exceeded the maximum number of steps. Unlike in the previous case of `mxstep steps taken before reaching tout` (see above), here we got several additional warnings that the current step size $h$ is numerically zero.\n", |
| 204 | + "The forward simulation failed because the AMICI solver exceeded the maximum number of steps. Unlike in the previous case of `mxstep steps taken before reaching tout` (see above), here we got several additional warnings that the current step size $h$ is numerically zero.\n", |
205 | 205 | "\n",
|
206 | 206 | "**How to address?**\n",
|
207 | 207 | "\n",
|
|
502 | 502 | "[](https://mermaid.live/edit#pako:eNpdkrFuwyAQhl8FIWVL1MwMndKFtd1wBmJIg2IDwucCivLuxQp2zvaA_P1398Od7kFbpzRl9DdIfyM_p8aS8t3FnZHW2QGkheH8Er3wjHgZZK9Bh1kFAYyA6XXldBTpyIixBozsSHGAJSQSWwlccEa4bN3FSFu1KCIjOvmgh8GUF8y1yoGYDkZU-lBQ5SwyI-4y6PAnL52esl-B3a68pDZDDofPhfyKYEVLacSVERdGXFchzYAu59iBYweOHTh2qBAxvLupI3s9eJ41pndqGdOqvYXThv2G7xuOiBf7jHMzNsr41oyvzNgv0z3tdeilUWXzHlOooXDTvW4oK79KX-XYQUMb-yypcgT3nW1LGYRR7-noVVmhk5FlZ3vKrrIbFvVLGXChis9_j9jNUw)\n",
|
503 | 503 | "\n",
|
504 | 504 | "Always look for the most basic (furthest up) model quantities.\n",
|
505 |
| - "In cases where there non-finite values occur in expressions further down, rerunning the simulation after calling `Model.setAlwaysCheckFinite(True)` may give some further hints on where the issue originates.\n", |
| 505 | + "In cases where non-finite values occur in expressions further down, rerunning the simulation after calling `Model.setAlwaysCheckFinite(True)` may give some further hints on where the issue originates.\n", |
506 | 506 | "\n",
|
507 | 507 | "The `NaN` in $w$ occurred for `flux_v7_v_6` (see error log), i.e., when computing the reaction flux for reaction `v7_v_6`. As $w$ only depends on $(t, p, k, x)$ and no non-finite values have been reported for those, the issue has to be in the respective flux equation.\n",
|
508 | 508 | "\n",
|
|
529 | 529 | "id": "9f49a00a",
|
530 | 530 | "metadata": {},
|
531 | 531 | "source": [
|
532 |
| - "What could go wrong? We can obtain `NaN` from any of these symbols symbols being `NaN`, or through division by zero.\n", |
| 532 | + "What could go wrong? We can obtain `NaN` from any of these symbols being `NaN`, or through division by zero.\n", |
533 | 533 | "\n",
|
534 | 534 | "Let's let's check the denominator first: $$(A\\_state^2 + Kp^2)*(Kd^{n\\_par} + Z\\_state^{n\\_par})$$\n",
|
535 | 535 | "\n",
|
|
670 | 670 | "source": [
|
671 | 671 | "**What happened?**\n",
|
672 | 672 | "\n",
|
673 |
| - "AMICI failed to compute steadystate sensitivities, because it was not able to factorize the Jacobian.\n", |
| 673 | + "AMICI failed to compute steady-state sensitivities, because it was not able to factorize the Jacobian.\n", |
674 | 674 | "\n",
|
675 | 675 | "**How to address?**\n",
|
676 | 676 | "\n",
|
|
815 | 815 | "source": [
|
816 | 816 | "**What happened?**\n",
|
817 | 817 | "\n",
|
818 |
| - "All given experimental conditions require pre-equilibration, i.e., finding a steady state. AMICI first tries find a steady state using the Newton solver, if that fails, it tries simulating until steady state, if that also failes, it tries the Newton solver from the end of the simulation. In this case, all three failed. Neither Newton's method nor simulation yielded a steadystate satisfying the required tolerances.\n", |
| 818 | + "All given experimental conditions require pre-equilibration, i.e., finding a steady state. AMICI first tries to find a steady state using the Newton solver, if that fails, it tries simulating until steady state, if that also failes, it tries the Newton solver from the end of the simulation. In this case, all three failed. Neither Newton's method nor simulation yielded a steady state satisfying the required tolerances.\n", |
819 | 819 | "\n",
|
820 | 820 | "This can also be seen in `ReturnDataView.preeq_status` (the three statuses corresponds to Newton \\#1, Simulation, Newton \\#2):"
|
821 | 821 | ]
|
|
842 | 842 | "\n",
|
843 | 843 | "1. Stricter integration tolerances (preferred if affordable - higher accuracy, but generally slower)\n",
|
844 | 844 | "\n",
|
845 |
| - "2. Looser steadystate tolerances (lower accuracy, generally faster)\n", |
| 845 | + "2. Looser steady-state tolerances (lower accuracy, generally faster)\n", |
846 | 846 | "\n",
|
847 | 847 | "3. Increase the number of allowed steps for Newton's method\n",
|
848 | 848 | "\n",
|
|
0 commit comments