def _sample_bowl_problems(rng):
"""Sample a bowl problem."""
is_noise = utils.sample_bool(rng, 0.5)
return {
"cond":
utils.sample_log_float(rng, 0.01, 100),
"angle":
rng.choice([0, 0, np.pi / 4., np.pi / 3]),
"noise_stdev":
utils.sample_log_float(rng, 0.01, 10.0) if is_noise else 0.0,
}
def sample_quadratic_family_cfg(seed):
"""Sample a task config for a toy quadratic based problem.
See QuadraticBasedTask for more information.
These configs are nested python structures that provide enough information
to create an instance of the problem.
Args:
seed: int Random seed to generate task from.
Returns:
A nested dictionary containing a configuration.
"""
rng = np.random.RandomState(seed)
cfg = {}
cfg["A_dist"] = _sample_distribution_over_matrix(rng)
cfg["initial_dist"] = _sample_vector_dist(rng)
cfg["output_fn"] = rng.choice(["identity", "log"])
cfg["dims"] = utils.sample_log_int(rng, 2, 3000)
cfg["seed"] = rng.randint(0, 100000)
cfg["loss_scale"] = utils.sample_log_float(rng, 1e-5, 1e3)
if rng.choice([True, False]):
cfg["noise"] = {}
cfg["noise"]["A_noise"] = _sample_matrix_noise_dist(rng)
else:
cfg["noise"] = None
return cfg
def _sample_min_max_well(rng):
"""Sample a min max well problem."""
is_noise = utils.sample_bool(rng, 0.5)
return {
"dim":
utils.sample_log_int(rng, 10, 1000),
"noise_stdev":
utils.sample_log_float(rng, 0.01, 10.0) if is_noise else 0.0,
}
def _sample_optimization_test_problems(rng):
"""Sample an optimization test function problem."""
is_noise = utils.sample_bool(rng, 0.5)
return {
"problem":
rng.choice(sorted(_opt_test_problems.keys())),
"noise_stdev":
utils.sample_log_float(rng, 0.01, 10.0) if is_noise else 0.0,
}
def _sample_quadratic_problem(rng):
"""Sample a quadratic problem."""
is_noise = utils.sample_bool(rng, 0.5)
return {
"dim":
utils.sample_log_int(rng, 10, 1000),
"noise_stdev":
utils.sample_log_float(rng, 0.01, 10.0) if is_noise else 0.0,
}
def _sample_sparse_softmax_regression(rng):
"""Sample a sparse softmax regression problem."""
is_noise = utils.sample_bool(rng, 0.5)
return {
"n_features": utils.sample_log_int(rng, 2, 100),
"n_classes": 2,
"noise_stdev":
utils.sample_log_float(rng, 0.01, 10.0) if is_noise else 0.0,
"bs": utils.sample_log_int(rng, 1, 50),
"n_samples": utils.sample_log_int(rng, 1, 30),
}
def _sample_rescale_problem(rng):
"""Sample a rescale problem.
This problem modifies a sampled base problem by rescaling the parameters.
Args:
rng: Random state
Returns:
The sampled config.
"""
base_config = rng.choice(_to_modify)
return {
"base": (base_config, _problem_sample_get[base_config][0](rng)),
"scale": utils.sample_log_float(rng, 0.001, 1000.0),
}
def _sample_sparse_problem(rng):
"""Sample a sparse problem.
This problem modifies a sampled base problem by setting some gradients to
zero.
Args:
rng: Random state
Returns:
The sampled config.
"""
is_noise = utils.sample_bool(rng, 0.5)
base_config = rng.choice(_to_modify)
return {
"base": (base_config, _problem_sample_get[base_config][0](rng)),
"zero_probability":
rng.uniform(0.9, 0.99),
"noise_stdev":
utils.sample_log_float(rng, 0.01, 10.0) if is_noise else 0.0,
}
