0.91484382948417953, 0.91484382948417953, 0.91509144783897012,
0.91665816586759163, 0.91717666258669917, 0.92136210246446459,
0.92136210246446459, 0.92405452495644158, 0.92405452495644158,
0.92696399345335523
])
fscore_metadata_no_svd_random.append([
0.0, 0.0, 0.30198446937014672, 0.58478038815117472, 0.58478038815117472,
0.58532227185705177, 0.72376963350785328, 0.72376963350785328,
0.74567417564479266, 0.74679417517930879, 0.75023582433707159,
0.8216451857467777, 0.82221379833206987, 0.82596739343940284,
0.84447900466562986, 0.8568035625927759, 0.85702978133966556,
0.85960396039603959, 0.86617174959871579, 0.86617174959871579,
0.88902077151335313, 0.88902077151335313, 0.88902077151335313,
0.88902077151335313, 0.89243813467415956, 0.88971742543171106,
0.88971742543171106, 0.89410372707552754, 0.89410372707552754,
0.89410372707552754, 0.8890005022601708, 0.88732677244426594,
0.89332401955892626, 0.89365951073389926, 0.89779280174985099
])
average_metadata_no_svd_random = list(
np.mean(np.matrix(fscore_metadata_no_svd_random), axis=0).A1)
ranges = [label_random, label_0, label_potential]
list = [
average_metadata_no_svd_random, average_metadata_no_svd,
average_metadata_no_svd_absolute_potential
]
names = ["Random", "Round-robin", "Potential prediction"]
plot_list_latex(ranges, list, names, "Flights", x_max=200)
0.68916625923902108, 0.68810233990545633, 0.72854729551752351,
0.72753034591360399, 0.72806657330679936, 0.73634726526509764,
0.697587070339176, 0.69853880268305202, 0.6986324792055707,
0.69867698435582615, 0.70774194643979149, 0.72170204783073311,
0.73280706993918332, 0.73005324986240139, 0.75145071397031471,
0.7333233275572717, 0.73322578554789686, 0.74274468129732296,
0.75949473360175468, 0.75832009080590246, 0.75838365915592332,
0.83748186108532807, 0.83750375861212434, 0.84416974984268389,
0.85507194303352985, 0.86163133748359166, 0.86692910296483272,
0.86109760130731661, 0.86186238811976301, 0.8636445445055646,
0.89151365017749062, 0.89473968923722158, 0.89686158204594757,
0.89606242004011805, 0.89770986568090039
])
average_metadata_no_svd_absolute_potential_20error = list(
np.mean(np.matrix(fscore_metadata_no_svd_absolute_potential_20error),
axis=0).A1)
ranges = [label_potential, label_potential, label_potential, label_potential]
list = [
average_metadata_no_svd_absolute_potential,
average_metadata_no_svd_absolute_potential_05error,
average_metadata_no_svd_absolute_potential_10error,
average_metadata_no_svd_absolute_potential_20error
]
names = [
"No User Error", "5\% User Error", "10\% User Error", "20\% User Error"
]
plot_list_latex(ranges, list, names, "BlackOak", x_max=350)
0.67453625632377734, 0.67453625632377734, 0.70829909613804432,
0.70829909613804432, 0.70829909613804432, 0.70829909613804432,
0.70829909613804432, 0.70829909613804432, 0.72918350848827806,
0.72918350848827806, 0.72918350848827806, 0.72918350848827806,
0.72918350848827806, 0.8954372623574145, 0.90273843248347496,
0.90273843248347496, 0.90273843248347496, 0.90273843248347496,
0.90273843248347496, 0.90273843248347496, 0.90273843248347496,
0.90273843248347496, 0.90273843248347496, 0.90273843248347496,
0.91134413727359387, 0.91134413727359387, 0.91134413727359387,
0.91134413727359387, 0.91134413727359387, 0.93271028037383186,
0.93271028037383186, 0.93271028037383186, 0.93271028037383186,
0.93271028037383186, 0.93271028037383186, 0.93271028037383186,
0.93271028037383186, 0.93271028037383186, 0.99007936507936511,
0.99007936507936511, 0.99007936507936511, 0.99007936507936511,
0.99007936507936511, 0.99007936507936511, 0.99007936507936511,
0.99007936507936511, 0.99007936507936511, 0.99007936507936511,
0.99007936507936511
])
average_metadata_no_svd_random = list(
np.mean(np.matrix(fscore_metadata_no_svd_random), axis=0).A1)
ranges = [label_random, label_0, label_potential]
list = [
average_metadata_no_svd_random, average_metadata_no_svd,
average_metadata_no_svd_absolute_potential
]
names = ["Random", "Round-robin", "Potential prediction"]
plot_list_latex(ranges, list, names, "Hospital", x_max=800)
average_001_02_opt = [
0.0, 0.0, 0.0, 0.16054503410666085, 0.1742153678163337, 0.175188264462044,
0.17800861736869286, 0.18723517157321715, 0.18712116127485837,
0.18942693706942779, 0.1887192639499303, 0.18984482431443722,
0.19068199462259514, 0.19083405040995338
]
average_001_02_round = [
0.0, 0.0, 0.0, 0.16054503410666085, 0.16123376011436252,
0.11867152703903208, 0.12126007977224942, 0.083813880826394183,
0.083854838905432549, 0.080477359248977101, 0.080845752592025491,
0.10537650533256877, 0.10453070277460338, 0.10527981465983556
]
ranges = [labels_optimum, labels_optimum]
list = [average_001_02_opt, average_001_02_round]
names = ["optimum", "round robin"]
plot_list(ranges, list, names, "Synthetic", x_max=200, end_of_round=28)
plot_list_latex(ranges, list, names, "Synthetic", x_max=200)
plot_integral(ranges,
list,
names,
"Synthetic",
x_max=200,
x_min=28,
sorted=True)
#plot_end(ranges, list, names, "Synthetic", x_max=200,x_min=28, sorted=True)
#plot_outperform(ranges, list, names, "Flights", 0.7366, x_max=200)
#plot_outperform(ranges, list, names, "Flights", 0.9, x_max=200)
average_metadata_no_svd_softcertainty_squared = list(np.mean(np.matrix(fscore_metadata_no_svd_softcertainty_squared), axis=0).A1)
ranges = [label_random,
label_0,
label_potential,
label_predictionchange,
label_mincertainty,
label_crossval,
label_softcertainty_squared]
list = [average_metadata_no_svd_random,
average_metadata_with_extr_number,
average_metadata_no_svd_absolute_potential,
average_metadata_no_svd_predictionchange,
average_metadata_no_svd_mincertainty,
average_metadata_no_svd_crossval,
average_metadata_no_svd_softcertainty_squared]
names = ["Random",
"Round-robin",
"Potential prediction",
"by max prediction change",
"by max uncertainty",
"by min crossval",
"by normalized squared uncertainty"]
plot_list_latex(ranges, list, names, "Address", x_max=200)
plot_list(ranges, list, names, "Address", x_max=200)
#plot_integral(ranges, list, names, "Address", x_max=350)
#plot_integral_latex(ranges, list, names, "Address", x_max=350)
#plot_outperform_latex(ranges, list, names, "Address",0.904, x_max=350)