Example powersystem.py doesn't work at all

[dietmarwo]

Steps to reproduce

1. pip install nevergrad
2. git clone [email protected]:facebookresearch/nevergrad.git
3. cd nevergrad/examples
4. python powersystem.py
5. using anaconda python 3.9 on Linux Mint 20.3.

Observed Results

• What happened? This could be a description, log output, etc.

ana39/lib/python3.9/site-packages/nevergrad/optimization/base.py:146: InefficientSettingsWarning: num_workers = 10 > 1 is suboptimal when run sequentially
warnings.warn(msg, e)
Traceback (most recent call last):
File "/home/xxx/sich/examples/powersystem.py", line 43, in
optimizer.minimize(power_system_loss)
File "/home/xxx/ana39/lib/python3.9/site-packages/nevergrad/optimization/base.py", line 641, in minimize
result = job.result()
File "/home/xxx/ana39/lib/python3.9/site-packages/nevergrad/optimization/utils.py", line 137, in result
self._result = self.func(*self.args, **self.kwargs)
File "/home/xxx/ana39/lib/python3.9/site-packages/nevergrad/functions/base.py", line 119, in call
return self.function(*args, **kwargs)
File "/home/xxx/ana39/lib/python3.9/site-packages/nevergrad/functions/powersystems/core.py", line 129, in _simulate_power_system
agent.set_parameters(array)
File "/home/xxx/ana39/lib/python3.9/site-packages/nevergrad/functions/powersystems/core.py", line 37, in set_parameters
raise ValueError(f"length = {weights.size} instead of {self.dimension}: {weights}.")
ValueError: length = 7560 instead of 1260: [-0.73698052 1.41931025 1.34390607 ... -0.79129847 -0.77101758
-0.6903909 ].

Expected Results

• What did you expect to happen?

Expected "optimizer.minimize(power_system_loss)" to perform some optimization and
"power_system_loss.make_plots" to show some plots.

As it is the one and only nevergrad example the observed result is a bit disappointing.

[Elcomiqu321]

`import nevergrad as ng
from nevergrad.functions.powersystems.core import PowerSystem

budget = 3500
width = 6
depth = 6
num_dams = 6
year_to_day_ratio = 0.5
back_to_normal = 0.5
num_thermal_plants = 6
constant_to_year_ratio = 4.0

power_system_loss = PowerSystem(
num_dams=num_dams,
depth=depth,
width=width,
year_to_day_ratio=year_to_day_ratio,
back_to_normal=back_to_normal,
num_thermal_plants=num_thermal_plants,
constant_to_year_ratio=constant_to_year_ratio,
)

added line:

param = int(power_system_loss.dimension/power_system_loss.num_dams)

changed 'parametrization=power_system_loss.dimension' in the optimizer parameters:

optimizer = ng.optimizers.SplitOptimizer(
parametrization=param, budget=budget, num_workers=1
)

optimizer.minimize(power_system_loss)
power_system_loss(optimizer.provide_recommendation().value)
power_system_loss.make_plots(
f"ps_{num_dams}dams_{depth}_{width}_ytdr{year_to_day_ratio}_btn{back_to_normal}"
f"_num_thermal_plants{num_thermal_plants}_ctyr{constant_to_year_ratio}_budget{budget}.png"
)`