Getting ['AxesSubplot' object is not subscriptable] when trying to use skopt.plots with result object as returned by Optimizer.tell() #740
Description
I am using skopt.optimize.Optimizer, as I need the ask() and tell() functionality, and I would like to use the plotting functions in skopt.plots on the results of my optimization.
As far as I can understand, what the plotting functions need is the scipy.optimize.OptimizeResult object, which is exactly what the tell() method returns, so I'm trying to pass that:
import numpy as np
import matplotlib.pyplot as plt
from skopt import Optimizer
from skopt.plots import plot_objective
# Define objective
def objective(x, noise_level=0.1):
return np.sin(5 * x[0]) * (1 - np.tanh(x[0] ** 2)) + np.random.randn() * noise_level
# Initialize Optimizer
opt = Optimizer([(-2.0, 2.0)], n_initial_points=1)
# Optimize
for i in range(2):
next_x = opt.ask()
f_val = objective(next_x)
res = opt.tell(next_x, f_val)
# Plot results
ax = plot_objective(res)However, this results in the following error:
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-12-d718026f55fc> in <module>
20
21 # Plot results
---> 22 ax = plot_objective(res)
~/miniconda3/envs/py3/lib/python3.6/site-packages/skopt/plots.py in plot_objective(result, levels, n_points, n_samples, size, zscale, dimensions)
425 n_points=n_points)
426
--> 427 ax[i, i].plot(xi, yi)
428 ax[i, i].axvline(result.x[i], linestyle="--", color="r", lw=1)
429
TypeError: 'AxesSubplot' object is not subscriptableIt seems the only difference between what skopt.optimizer.base.base_minimize() and tell() returns is that the result object from tell() has attribute specs set to None.
I can see in the source code for base_minimize() that specs there is set to
{"args": copy.copy(inspect.currentframe().f_locals),
"function": inspect.currentframe().f_code.co_name}Is this difference the reason why the above does not work? If so, what would be a nice way to do this? Also, isn't "function" always just "base_minimize"? Why is this needed?
Thank you.