[RTR] Fixed add_penalty_plot index

[EveCharbie]

Tries to fix issue #641

---

This change is 

[EveCharbie]

@lsechoir I think you had problems with add_plot_penalty too, can you try this fix to see if it helps in your case?

[codecov]

Codecov Report

Patch coverage: 65.00% and project coverage change: -0.01 ⚠️

Comparison is base (2cce426) 81.82% compared to head (d3764bb) 81.81%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #665      +/-   ##
==========================================
- Coverage   81.82%   81.81%   -0.01%     
==========================================
  Files         112      112              
  Lines       12919    12953      +34     
==========================================
+ Hits        10571    10598      +27     
- Misses       2348     2355       +7     

Impacted Files	Coverage Δ
bioptim/gui/plot.py	73.48% <55.71%> (+0.46%)	⬆️
bioptim/dynamics/configure_problem.py	90.92% <76.47%> (-0.35%)	⬇️
bioptim/limits/penalty.py	89.65% <100.00%> (+0.13%)	⬆️
bioptim/limits/penalty_option.py	88.59% <100.00%> (-0.15%)	⬇️
bioptim/optimization/optimal_control_program.py	87.38% <100.00%> (-0.16%)	⬇️

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[EveCharbie]

@pariterre I changed the behavior of the plots. Now it shows all plots even if there is a mapping. I find it easier to understand and it fixes the issue with indexing the plots. What do you think? (If it is ok, I will clean it, ex the legend)

[pariterre]

I understand how it simplifies things... but not sure I like creating useless graphs... @Ipuch ?

[pariterre]

Reviewed 1 of 1 files at r1, 3 of 3 files at r2, all commit messages.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @EveCharbie)

---

bioptim/dynamics/configure_problem.py line 1113 at r2 (raw file):

            as_controls,
            as_states_dot,
            axes_idx=axes_idx  # nlp.variable_mappings[name].to_first,

Remove if not necessary

---

bioptim/gui/plot.py line 99 at r2 (raw file):

            self.phase_mappings = None  # Will be set later
        elif isinstance(axes_idx, (tuple, list)):
            self.phase_mappings = BiMapping(axes_idx) # ?

What is the question here?

[pariterre]

Please have a look why tests fail

[EveCharbie]

Test does not pass because it is not a final version, I just wanted your insights before cleaning the code.

[Ipuch]

I don't care about having extra graphs if it eases the code. Anyways we don't use them for publication but only to check our results.
If it helps to understand which dof are mapped, I'm down for this PR.
The little extra would be to write on which graph to refer to if mapped with another dof, or None if just remove.
Or display the same value as it is mapped ?

[EveCharbie]

Reviewable status: 0 of 5 files reviewed, 2 unresolved discussions (waiting on @pariterre)

---

bioptim/dynamics/configure_problem.py line 1113 at r2 (raw file):

Previously, pariterre (Pariterre) wrote…

Remove if not necessary

Done.

---

bioptim/gui/plot.py line 99 at r2 (raw file):

Previously, pariterre (Pariterre) wrote…

What is the question here?

Done.

[pariterre]

Reviewed 1 of 5 files at r3, 4 of 4 files at r4, all commit messages.
Reviewable status: all files reviewed, 5 unresolved discussions (waiting on @EveCharbie)

---

bioptim/limits/penalty_option.py line 571 at r4 (raw file):

        else:
            target_mapping = BiMapping(
                to_first=list(range(controller.get_nlp.controls[key].cx_start.shape[0])),

congtroller.controls[key].cx.shape

---

bioptim/limits/penalty_option.py line 572 at r4 (raw file):

            target_mapping = BiMapping(
                to_first=list(range(controller.get_nlp.controls[key].cx_start.shape[0])),
                to_second=list(range(controller.get_nlp.controls[key].cx_start.shape[0])),

congtroller.controls[key].cx.shape

---

bioptim/limits/penalty_option.py line 593 at r4 (raw file):

        self.target_plot_name = combine_to
        # if the target is n x ns, we need to add a dimension (n x ns + 1) to make it compatible with the plot
        if self.target[0].shape[1] == controller.get_nlp.ns:

congtroller.ns?

---

bioptim/optimization/optimal_control_program.py line 1034 at r4 (raw file):

                x_scaling = np.vstack([self.nlp[penalty.phase].x_scaling["all"].scaling for _ in range(x_shape)]).T

            x /= x_scaling

check this?

---

bioptim/optimization/optimal_control_program.py line 1035 at r4 (raw file):

            x /= x_scaling
            if u.size != 0:

wait for me?

[EveCharbie]

Reviewable status: 1 of 5 files reviewed, 5 unresolved discussions (waiting on @pariterre)

---

bioptim/limits/penalty_option.py line 571 at r4 (raw file):

Previously, pariterre (Pariterre) wrote…

congtroller.controls[key].cx.shape

Done.

---

bioptim/limits/penalty_option.py line 572 at r4 (raw file):

Previously, pariterre (Pariterre) wrote…

congtroller.controls[key].cx.shape

Done.

---

bioptim/limits/penalty_option.py line 593 at r4 (raw file):

Previously, pariterre (Pariterre) wrote…

congtroller.ns?

Done.

---

bioptim/optimization/optimal_control_program.py line 1034 at r4 (raw file):

Previously, pariterre (Pariterre) wrote…

check this?

did not disapear, dont know why ?

---

bioptim/optimization/optimal_control_program.py line 1035 at r4 (raw file):

Previously, pariterre (Pariterre) wrote…

wait for me?

Merged you PR

[pariterre]

Reviewed 4 of 4 files at r5, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @EveCharbie)