Add `sigma`, `sigma0` to `gaussian_emitter` and `iso_line_emitter`

[itsdawei]

Description

Allow user to configure different standard deviations (sigma, sigma0) for initialization and sampling steps of gaussian_emitter and iso_line_emitter.

TODO / DONE

• [x] Change sigma0 to optional parameter (defaults to sigma). sigma0 is used to generate new solutions when the archive is empty, and added parametersigma, which used to sample solutions from a non-empty archive.
• [x] Add optional parametersigma0 to Iso_Line_Emitter.
• [x] Change iso_line_emitter_test::test_properties_are_correct to test for sigma0.
• [x] Change factory_test to reflect constructor change
• [x] Add additional test iso_line_emitter_test::test_sigma0_is_correct and gaussian_emitter_test::test_sigma0_is_correct

Status

• [x] I have read the guidelines in CONTRIBUTING.md.
• [x] I have formatted my code using yapf
• [x] I have tested my code by running pytest
• [x] I have linted my code with pylint
• [ ] I have added a one-line description of my change to the changelog in HISTORY.md.
• [ ] This PR is ready to go

[itsdawei]

Visualizing using the new sigma0 parameter.

from ribs.archives import GridArchive
from ribs.emitters import GaussianEmitter, IsoLineEmitter
from ribs.optimizers import Optimizer
import numpy as np
import matplotlib.pyplot as plt
from ribs.visualize import grid_archive_heatmap

archive = GridArchive(
    dims=[20, 20],
    ranges=[(-1, 1), (-1, 1)],
)

# emitters = [
#     GaussianEmitter(
#         archive,
#         x0=[0.0] * 10,
#         sigma=1,
#         sigma0=0.1,  # should generate a small circle of solutions around (0,0)
#     )
# ]

emitters = [
    IsoLineEmitter(
        archive,
        x0=[0.0] * 10,
        iso_sigma=1,
        sigma0=0.1,  # should generate a small circle of solutions around (0,0)
    )
]

optimizer = Optimizer(archive, emitters)

for itr in range(1000):
    solutions = optimizer.ask()

    # Optimize the 10D negative Sphere function.
    objectives = -np.sum(np.square(solutions), axis=1)

    # BCs: first 2 coordinates of each 10D solution.
    bcs = solutions[:, :2]
    optimizer.tell(objectives, bcs)

    # plot every 100 iteration
    if itr % 100 == 0:
        plt.figure()
        grid_archive_heatmap(archive)
        plt.savefig(f"img/plot{itr}.png", dpi=300)
        plt.close()

GaussianEmitter (sigma0=0)

GaussianEmitter (sigma0=0.1)

IsoLineEmitter (sigma0=0)

IsoLineEmitter (sigma0=0.1)