def test_optimizer_history():
# create the necessary results
result_1 = create_optimization_history()
result_2 = create_optimization_history()
# test a standard call
visualize.optimizer_history(result_1)
# test plotting of lists
visualize.optimizer_history([result_1, result_2])
def test_optimizer_history_with_options():
# create the necessary results
result_1 = create_optimization_history()
result_2 = create_optimization_history()
# alternative figure size and plotting options
(_, _, ref3, _, ref_point) = create_plotting_options()
alt_fig_size = (9.0, 8.0)
# Test with y-limits as vector
with pytest.warns(UserWarning, match="Invalid lower bound"):
visualize.optimizer_history(
result_1,
y_limits=[-0.5, 2.5],
start_indices=[0, 1, 4, 11],
reference=ref_point,
size=alt_fig_size,
trace_x='steps',
trace_y='fval',
offset_y=-10.0,
colors=[1.0, 0.3, 0.3, 0.5],
)
# Test with linear scale
visualize.optimizer_history(
[result_1, result_2],
y_limits=[-0.5, 2.5],
start_indices=[0, 1, 4, 11],
reference=ref_point,
size=alt_fig_size,
scale_y='lin',
)
# Test with y-limits as float
visualize.optimizer_history(
result_1,
y_limits=5.0,
start_indices=3,
reference=ref3,
trace_x='time',
trace_y='gradnorm',
offset_y=10.0,
)
for name in problem.x_names
]
x_sorted = [opt_result['x'][sorted_par_idx[ix]] for ix in
range(len(problem.x_names))]
opt_result['x'] = x_sorted
result.optimize_result.append(opt_result)
result.optimize_result.sort()
prefix = '__'.join([MODEL, DATA, str(N_HIDDEN), OPTIMIZER])
waterfall(result, scale_y='log10', offset_y=0.0)
plot_and_save_fig(prefix + '__waterfall.pdf')
optimizer_history(result, scale_y='log10')
plot_and_save_fig(prefix + '__optimizer_trace.pdf')
parameters(result)
plot_and_save_fig(prefix + '__parameters.pdf')
optimizer_convergence(result)
plot_and_save_fig(prefix + '__optimizer_convergence.pdf')
fig_embedding, axes_embedding = plt.subplots(1, N_STARTS,
figsize=(18.5, 10.5))
embedding_fun = theano.function(
[mae.encoder_pars],
mae.encode(mae.encoder_pars)
)
balance_alpha=False)
visualize.parameters(result1_dogleg,
legends='dogleg',
reference=ref,
size=(15, 10),
start_indices=[0, 1, 2, 3, 4, 5],
balance_alpha=False)
df = result1_tnc.optimize_result.as_dataframe(
['fval', 'n_fval', 'n_grad', 'n_hess', 'n_res', 'n_sres', 'time'])
df.head()
###### Optimizer History
# plot one list of waterfalls
visualize.optimizer_history([result1_bfgs, result1_tnc],
legends=['L-BFGS-B', 'TNC'],
reference=ref)
# plot one list of waterfalls
visualize.optimizer_history(result1_dogleg, reference=ref)
# plot one list of waterfalls
visualize.optimizer_history([result1_bfgs, result1_tnc],
legends=['L-BFGS-B', 'TNC'],
reference=ref,
offset_y=0.)
# plot one list of waterfalls
visualize.optimizer_history([result1_bfgs, result1_tnc],
legends=['L-BFGS-B', 'TNC'],
reference=ref,
scale_y='lin',