Doctorado-ML/stree_datasets
1
0
Fork 0
mirror of https://github.com/Doctorado-ML/Stree_datasets.git synced 2025-08-17 08:26:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add nodes, leaves, depth to mysql

Browse Source 
Add nodes, leaves, depth, samples, features and classes to analysis
This commit is contained in:
Ricardo Montañana Gómez
2021-03-24 12:38:11 +01:00
parent ae6e2ba374
commit 66bceff179
5 changed files with 124 additions and 74 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
analysis_mysql.py
View File

@@ -1,5 +1,6 @@
import argparse import argparse
from typing import Tuple from typing import Tuple
import numpy as np
from experimentation.Sets import Datasets from experimentation.Sets import Datasets
from experimentation.Utils import TextColor from experimentation.Utils import TextColor
from experimentation.Database import MySQL from experimentation.Database import MySQL
@@ -14,8 +15,10 @@ models_tree = [
"baseRaF", "baseRaF",
] ]
models_ensemble = ["odte", "adaBoost", "bagging", "TBRaF", "TBRoF", "TBRRoF"] models_ensemble = ["odte", "adaBoost", "bagging", "TBRaF", "TBRoF", "TBRRoF"]
description = ["samp", "var", "cls"]
complexity = ["nodes", "leaves", "depth"]
title = "Best model results" title = "Best model results"
lengths = (30, 12, 12, 12, 12, 12, 12) lengths = (30, 4, 3, 3, 3, 3, 3, 12, 12, 12, 12, 12, 12)
def parse_arguments() -> Tuple[str, str, str, bool, bool]: def parse_arguments() -> Tuple[str, str, str, bool, bool]:
@@ -79,9 +82,11 @@ def report_header(title, experiment, model_type):
def report_line(line): def report_line(line):
output = f"{line['dataset']:{lengths[0] + 5}s} " output = f"{line['dataset']:{lengths[0] + 5}s} "
for key, item in enumerate(description + complexity):
output += f"{line[item]:{lengths[key + 1]}d} "
data = models.copy() data = models.copy()
for key, model in enumerate(data): for key, model in enumerate(data):
output += f"{line[model]:{lengths[key + 1]}s} " output += f"{line[model]:{lengths[key + 7]}s} "
return output return output
@@ -101,7 +106,15 @@ def report_footer(agg):
dbh = MySQL() dbh = MySQL()
database = dbh.get_connection() database = dbh.get_connection()
dt = Datasets(False, False, "tanveer") dt = Datasets(False, False, "tanveer")
fields = ("Dataset",) fields = (
"Dataset",
"Samp",
"Var",
"Cls",
"Nod",
"Lea",
"Dep",
)
models = models_tree if model_type == "tree" else models_ensemble models = models_tree if model_type == "tree" else models_ensemble
for item in models: for item in models:
fields += (f"{item}",) fields += (f"{item}",)
@@ -121,13 +134,23 @@ for dataset in dt:
find_one = False find_one = False
# Look for max accuracy for any given dataset # Look for max accuracy for any given dataset
line = {"dataset": color + dataset[0]} line = {"dataset": color + dataset[0]}
X, y = dt.load(dataset[0]) # type: ignore
line["samp"], line["var"] = X.shape
line["cls"] = len(np.unique(y))
record = dbh.find_best(dataset[0], models, experiment) record = dbh.find_best(dataset[0], models, experiment)
max_accuracy = 0.0 if record is None else record[5] max_accuracy = 0.0 if record is None else record[5]
line["nodes"] = 0
line["leaves"] = 0
line["depth"] = 0
for model in models: for model in models:
record = dbh.find_best(dataset[0], model, experiment) record = dbh.find_best(dataset[0], model, experiment)
if record is None: if record is None:
line[model] = color + "-" * 12 line[model] = color + "-" * 12
else: else:
if model == "stree":
line["nodes"] = record[12]
line["leaves"] = record[13]
line["depth"] = record[14]
reference = record[13] reference = record[13]
accuracy = record[5] accuracy = record[5]
acc_std = record[11] acc_std = record[11]

13
experimentation/Database.py
View File

@@ -176,6 +176,7 @@ class BD(ABC):
accuracy, accuracy,
time_spent, time_spent,
parameters, parameters,
complexity,
) -> None: ) -> None:
"""Create a record in MySQL database """Create a record in MySQL database
@@ -187,8 +188,8 @@ class BD(ABC):
command_insert = ( command_insert = (
"replace into results (date, time, type, accuracy, " "replace into results (date, time, type, accuracy, "
"dataset, classifier, norm, stand, parameters, accuracy_std, " "dataset, classifier, norm, stand, parameters, accuracy_std, "
"time_spent, time_spent_std) values (%s, %s, " "time_spent, time_spent_std, nodes, leaves, depth) values (%s, %s,"
"%s, %s, %s, %s, %s, %s, %s, %s, %s, %s)" " %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s)"
) )
now = datetime.now() now = datetime.now()
date = now.strftime("%Y-%m-%d") date = now.strftime("%Y-%m-%d")
@@ -206,6 +207,9 @@ class BD(ABC):
accuracy[1], accuracy[1],
time_spent[0], time_spent[0],
time_spent[1], time_spent[1],
complexity["nodes"],
complexity["leaves"],
complexity["depth"],
) )
cursor = database.cursor() cursor = database.cursor()
cursor.execute(command_insert, values) cursor.execute(command_insert, values)
@@ -319,7 +323,9 @@ class Outcomes(BD):
self._table = "outcomes" self._table = "outcomes"
super().__init__(host=host, model=model) super().__init__(host=host, model=model)
def store(self, dataset, normalize, standardize, parameters, results): def store(
self, dataset, normalize, standardize, parameters, results, complexity
):
outcomes = ["fit_time", "score_time", "train_score", "test_score"] outcomes = ["fit_time", "score_time", "train_score", "test_score"]
data = "" data = ""
for index in outcomes: for index in outcomes:
@@ -350,6 +356,7 @@ class Outcomes(BD):
float(np.std(results["fit_time"])), float(np.std(results["fit_time"])),
], ],
parameters, parameters,
complexity,
) )
def report(self, dataset, exclude_params): def report(self, dataset, exclude_params):

15
experimentation/Experiments.py
View File

@@ -2,9 +2,8 @@ import json
import os import os
import time import time
import warnings import warnings
import numpy as np
from sklearn.model_selection import GridSearchCV, cross_validate from sklearn.model_selection import GridSearchCV, cross_validate
from . import Models from . import Models
from .Database import Hyperparameters, MySQL, Outcomes from .Database import Hyperparameters, MySQL, Outcomes
from .Sets import Datasets from .Sets import Datasets
@@ -94,15 +93,25 @@ class Experiment:
X, X,
y, y,
return_train_score=True, return_train_score=True,
return_estimator=True,
n_jobs=self._threads, n_jobs=self._threads,
cv=kfold, cv=kfold,
) )
for item in outcomes: for item in outcomes:
total[item].append(results[item]) total[item].append(results[item])
print("end") print("end")
if type(model).__name__ == "Stree":
best_model = results["estimator"][np.argmax(results["test_score"])]
nodes, leaves = best_model.nodes_leaves()
depth = best_model.depth_
else:
nodes = leaves = depth = 0
complexity = dict(nodes=nodes, leaves=leaves, depth=depth)
outcomes = Outcomes(host=self._host, model=self._model_name) outcomes = Outcomes(host=self._host, model=self._model_name)
parameters = json.dumps(parameters, sort_keys=True) parameters = json.dumps(parameters, sort_keys=True)
outcomes.store(dataset, normalize, standardize, parameters, total) outcomes.store(
dataset, normalize, standardize, parameters, total, complexity
)
if self._num_warnings > 0: if self._num_warnings > 0:
print(f"{self._num_warnings} warnings have happend") print(f"{self._num_warnings} warnings have happend")

94
report.csv
View File

@@ -1,107 +1,107 @@
dataset, classifier, accuracy dataset, classifier, accuracy
balance-scale, stree, 0.91184 balance-scale, stree, 0.97056
balance-scale, wodt, 0.912 balance-scale, wodt, 0.912
balance-scale, j48svm, 0.94 balance-scale, j48svm, 0.94
balance-scale, oc1, 0.9192 balance-scale, oc1, 0.9192
balance-scale, cart, 0.78816 balance-scale, cart, 0.78816
balance-scale, baseRaF, 0.706738 balance-scale, baseRaF, 0.706738
balloons, stree, 0.653333 balloons, stree, 0.86
balloons, wodt, 0.688333 balloons, wodt, 0.688333
balloons, j48svm, 0.595 balloons, j48svm, 0.595
balloons, oc1, 0.62 balloons, oc1, 0.62
balloons, cart, 0.671667 balloons, cart, 0.671667
balloons, baseRaF, 0.605 balloons, baseRaF, 0.605
breast-cancer-wisc-diag, stree, 0.968898 breast-cancer-wisc-diag, stree, 0.972764
breast-cancer-wisc-diag, wodt, 0.967317 breast-cancer-wisc-diag, wodt, 0.967317
breast-cancer-wisc-diag, j48svm, 0.952878 breast-cancer-wisc-diag, j48svm, 0.952878
breast-cancer-wisc-diag, oc1, 0.933477 breast-cancer-wisc-diag, oc1, 0.933477
breast-cancer-wisc-diag, cart, 0.93953 breast-cancer-wisc-diag, cart, 0.93953
breast-cancer-wisc-diag, baseRaF, 0.965694 breast-cancer-wisc-diag, baseRaF, 0.965694
breast-cancer-wisc-prog, stree, 0.802051 breast-cancer-wisc-prog, stree, 0.811128
breast-cancer-wisc-prog, wodt, 0.710141 breast-cancer-wisc-prog, wodt, 0.710141
breast-cancer-wisc-prog, j48svm, 0.724038 breast-cancer-wisc-prog, j48svm, 0.724038
breast-cancer-wisc-prog, oc1, 0.71 breast-cancer-wisc-prog, oc1, 0.71
breast-cancer-wisc-prog, cart, 0.699833 breast-cancer-wisc-prog, cart, 0.699833
breast-cancer-wisc-prog, baseRaF, 0.74485 breast-cancer-wisc-prog, baseRaF, 0.74485
breast-cancer-wisc, stree, 0.966661 breast-cancer-wisc, stree, 0.965802
breast-cancer-wisc, wodt, 0.946208 breast-cancer-wisc, wodt, 0.946208
breast-cancer-wisc, j48svm, 0.967674 breast-cancer-wisc, j48svm, 0.967674
breast-cancer-wisc, oc1, 0.940194 breast-cancer-wisc, oc1, 0.940194
breast-cancer-wisc, cart, 0.940629 breast-cancer-wisc, cart, 0.940629
breast-cancer-wisc, baseRaF, 0.942857 breast-cancer-wisc, baseRaF, 0.942857
breast-cancer, stree, 0.734211 breast-cancer, stree, 0.733158
breast-cancer, wodt, 0.650236 breast-cancer, wodt, 0.650236
breast-cancer, j48svm, 0.707719 breast-cancer, j48svm, 0.707719
breast-cancer, oc1, 0.649728 breast-cancer, oc1, 0.649728
breast-cancer, cart, 0.65444 breast-cancer, cart, 0.65444
breast-cancer, baseRaF, 0.656438 breast-cancer, baseRaF, 0.656438
cardiotocography-10clases, stree, 0.552558 cardiotocography-10clases, stree, 0.712009
cardiotocography-10clases, wodt, 0.773706 cardiotocography-10clases, wodt, 0.773706
cardiotocography-10clases, j48svm, 0.830812 cardiotocography-10clases, j48svm, 0.830812
cardiotocography-10clases, oc1, 0.795528 cardiotocography-10clases, oc1, 0.795528
cardiotocography-10clases, cart, 0.818864 cardiotocography-10clases, cart, 0.818864
cardiotocography-10clases, baseRaF, 0.774788 cardiotocography-10clases, baseRaF, 0.774788
cardiotocography-3clases, stree, 0.35207 cardiotocography-3clases, stree, 0.891956
cardiotocography-3clases, wodt, 0.897509 cardiotocography-3clases, wodt, 0.897509
cardiotocography-3clases, j48svm, 0.927327 cardiotocography-3clases, j48svm, 0.927327
cardiotocography-3clases, oc1, 0.899811 cardiotocography-3clases, oc1, 0.899811
cardiotocography-3clases, cart, 0.929258 cardiotocography-3clases, cart, 0.929258
cardiotocography-3clases, baseRaF, 0.896715 cardiotocography-3clases, baseRaF, 0.896715
conn-bench-sonar-mines-rocks, stree, 0.755528 conn-bench-sonar-mines-rocks, stree, 0.71439
conn-bench-sonar-mines-rocks, wodt, 0.824959 conn-bench-sonar-mines-rocks, wodt, 0.824959
conn-bench-sonar-mines-rocks, j48svm, 0.73892 conn-bench-sonar-mines-rocks, j48svm, 0.73892
conn-bench-sonar-mines-rocks, oc1, 0.710798 conn-bench-sonar-mines-rocks, oc1, 0.710798
conn-bench-sonar-mines-rocks, cart, 0.728711 conn-bench-sonar-mines-rocks, cart, 0.728711
conn-bench-sonar-mines-rocks, baseRaF, 0.772981 conn-bench-sonar-mines-rocks, baseRaF, 0.772981
cylinder-bands, stree, 0.715049 cylinder-bands, stree, 0.687101
cylinder-bands, wodt, 0.704074 cylinder-bands, wodt, 0.704074
cylinder-bands, j48svm, 0.726351 cylinder-bands, j48svm, 0.726351
cylinder-bands, oc1, 0.67106 cylinder-bands, oc1, 0.67106
cylinder-bands, cart, 0.712703 cylinder-bands, cart, 0.712703
cylinder-bands, baseRaF, 0.675117 cylinder-bands, baseRaF, 0.675117
dermatology, stree, 0.966087 dermatology, stree, 0.971833
dermatology, wodt, 0.965557 dermatology, wodt, 0.965557
dermatology, j48svm, 0.955735 dermatology, j48svm, 0.955735
dermatology, oc1, 0.916087 dermatology, oc1, 0.916087
dermatology, cart, 0.932766 dermatology, cart, 0.932766
dermatology, baseRaF, 0.970723 dermatology, baseRaF, 0.970723
echocardiogram, stree, 0.808832 echocardiogram, stree, 0.814758
echocardiogram, wodt, 0.733875 echocardiogram, wodt, 0.733875
echocardiogram, j48svm, 0.805527 echocardiogram, j48svm, 0.805527
echocardiogram, oc1, 0.748291 echocardiogram, oc1, 0.748291
echocardiogram, cart, 0.745043 echocardiogram, cart, 0.745043
echocardiogram, baseRaF, 0.753522 echocardiogram, baseRaF, 0.753522
fertility, stree, 0.866 fertility, stree, 0.88
fertility, wodt, 0.785 fertility, wodt, 0.785
fertility, j48svm, 0.857 fertility, j48svm, 0.857
fertility, oc1, 0.793 fertility, oc1, 0.793
fertility, cart, 0.8 fertility, cart, 0.8
fertility, baseRaF, 0.798 fertility, baseRaF, 0.798
haberman-survival, stree, 0.735637 haberman-survival, stree, 0.727795
haberman-survival, wodt, 0.664707 haberman-survival, wodt, 0.664707
haberman-survival, j48svm, 0.714056 haberman-survival, j48svm, 0.714056
haberman-survival, oc1, 0.651634 haberman-survival, oc1, 0.651634
haberman-survival, cart, 0.65 haberman-survival, cart, 0.65
haberman-survival, baseRaF, 0.720133 haberman-survival, baseRaF, 0.720133
heart-hungarian, stree, 0.817674 heart-hungarian, stree, 0.827522
heart-hungarian, wodt, 0.764909 heart-hungarian, wodt, 0.764909
heart-hungarian, j48svm, 0.785026 heart-hungarian, j48svm, 0.785026
heart-hungarian, oc1, 0.758298 heart-hungarian, oc1, 0.758298
heart-hungarian, cart, 0.760508 heart-hungarian, cart, 0.760508
heart-hungarian, baseRaF, 0.779804 heart-hungarian, baseRaF, 0.779804
hepatitis, stree, 0.796129 hepatitis, stree, 0.824516
hepatitis, wodt, 0.785806 hepatitis, wodt, 0.785806
hepatitis, j48svm, 0.761935 hepatitis, j48svm, 0.761935
hepatitis, oc1, 0.756774 hepatitis, oc1, 0.756774
hepatitis, cart, 0.765161 hepatitis, cart, 0.765161
hepatitis, baseRaF, 0.773671 hepatitis, baseRaF, 0.773671
ilpd-indian-liver, stree, 0.723498 ilpd-indian-liver, stree, 0.719207
ilpd-indian-liver, wodt, 0.676176 ilpd-indian-liver, wodt, 0.676176
ilpd-indian-liver, j48svm, 0.690339 ilpd-indian-liver, j48svm, 0.690339
ilpd-indian-liver, oc1, 0.660139 ilpd-indian-liver, oc1, 0.660139
ilpd-indian-liver, cart, 0.663423 ilpd-indian-liver, cart, 0.663423
ilpd-indian-liver, baseRaF, 0.696685 ilpd-indian-liver, baseRaF, 0.696685
ionosphere, stree, 0.866056 ionosphere, stree, 0.953276
ionosphere, wodt, 0.88008 ionosphere, wodt, 0.88008
ionosphere, j48svm, 0.891984 ionosphere, j48svm, 0.891984
ionosphere, oc1, 0.879742 ionosphere, oc1, 0.879742
@@ -113,49 +113,49 @@ iris, j48svm, 0.947333
iris, oc1, 0.948 iris, oc1, 0.948
iris, cart, 0.938667 iris, cart, 0.938667
iris, baseRaF, 0.953413 iris, baseRaF, 0.953413
led-display, stree, 0.7007 led-display, stree, 0.703
led-display, wodt, 0.7049 led-display, wodt, 0.7049
led-display, j48svm, 0.7204 led-display, j48svm, 0.7204
led-display, oc1, 0.6993 led-display, oc1, 0.6993
led-display, cart, 0.7037 led-display, cart, 0.7037
led-display, baseRaF, 0.70178 led-display, baseRaF, 0.70178
libras, stree, 0.747778 libras, stree, 0.788333
libras, wodt, 0.764167 libras, wodt, 0.764167
libras, j48svm, 0.66 libras, j48svm, 0.66
libras, oc1, 0.645 libras, oc1, 0.645
libras, cart, 0.655 libras, cart, 0.655
libras, baseRaF, 0.726722 libras, baseRaF, 0.726722
low-res-spect, stree, 0.853102 low-res-spect, stree, 0.865713
low-res-spect, wodt, 0.856459 low-res-spect, wodt, 0.856459
low-res-spect, j48svm, 0.83358 low-res-spect, j48svm, 0.83358
low-res-spect, oc1, 0.824671 low-res-spect, oc1, 0.824671
low-res-spect, cart, 0.829206 low-res-spect, cart, 0.829206
low-res-spect, baseRaF, 0.790875 low-res-spect, baseRaF, 0.790875
lymphography, stree, 0.77046 lymphography, stree, 0.823425
lymphography, wodt, 0.808782 lymphography, wodt, 0.808782
lymphography, j48svm, 0.778552 lymphography, j48svm, 0.778552
lymphography, oc1, 0.734634 lymphography, oc1, 0.734634
lymphography, cart, 0.766276 lymphography, cart, 0.766276
lymphography, baseRaF, 0.761622 lymphography, baseRaF, 0.761622
mammographic, stree, 0.81915 mammographic, stree, 0.817068
mammographic, wodt, 0.759839 mammographic, wodt, 0.759839
mammographic, j48svm, 0.821435 mammographic, j48svm, 0.821435
mammographic, oc1, 0.768805 mammographic, oc1, 0.768805
mammographic, cart, 0.757131 mammographic, cart, 0.757131
mammographic, baseRaF, 0.780206 mammographic, baseRaF, 0.780206
molec-biol-promoter, stree, 0.764416 molec-biol-promoter, stree, 0.767056
molec-biol-promoter, wodt, 0.798528 molec-biol-promoter, wodt, 0.798528
molec-biol-promoter, j48svm, 0.744935 molec-biol-promoter, j48svm, 0.744935
molec-biol-promoter, oc1, 0.734805 molec-biol-promoter, oc1, 0.734805
molec-biol-promoter, cart, 0.748701 molec-biol-promoter, cart, 0.748701
molec-biol-promoter, baseRaF, 0.667239 molec-biol-promoter, baseRaF, 0.667239
musk-1, stree, 0.843463 musk-1, stree, 0.916388
musk-1, wodt, 0.838914 musk-1, wodt, 0.838914
musk-1, j48svm, 0.82693 musk-1, j48svm, 0.82693
musk-1, oc1, 0.776401 musk-1, oc1, 0.776401
musk-1, cart, 0.780215 musk-1, cart, 0.780215
musk-1, baseRaF, 0.834034 musk-1, baseRaF, 0.834034
oocytes_merluccius_nucleus_4d, stree, 0.810657 oocytes_merluccius_nucleus_4d, stree, 0.835125
oocytes_merluccius_nucleus_4d, wodt, 0.737673 oocytes_merluccius_nucleus_4d, wodt, 0.737673
oocytes_merluccius_nucleus_4d, j48svm, 0.741766 oocytes_merluccius_nucleus_4d, j48svm, 0.741766
oocytes_merluccius_nucleus_4d, oc1, 0.743199 oocytes_merluccius_nucleus_4d, oc1, 0.743199
@@ -167,127 +167,127 @@ oocytes_merluccius_states_2f, j48svm, 0.901374
oocytes_merluccius_states_2f, oc1, 0.889223 oocytes_merluccius_states_2f, oc1, 0.889223
oocytes_merluccius_states_2f, cart, 0.891193 oocytes_merluccius_states_2f, cart, 0.891193
oocytes_merluccius_states_2f, baseRaF, 0.910551 oocytes_merluccius_states_2f, baseRaF, 0.910551
oocytes_trisopterus_nucleus_2f, stree, 0.800986 oocytes_trisopterus_nucleus_2f, stree, 0.799995
oocytes_trisopterus_nucleus_2f, wodt, 0.751431 oocytes_trisopterus_nucleus_2f, wodt, 0.751431
oocytes_trisopterus_nucleus_2f, j48svm, 0.756587 oocytes_trisopterus_nucleus_2f, j48svm, 0.756587
oocytes_trisopterus_nucleus_2f, oc1, 0.747697 oocytes_trisopterus_nucleus_2f, oc1, 0.747697
oocytes_trisopterus_nucleus_2f, cart, 0.734313 oocytes_trisopterus_nucleus_2f, cart, 0.734313
oocytes_trisopterus_nucleus_2f, baseRaF, 0.76193 oocytes_trisopterus_nucleus_2f, baseRaF, 0.76193
oocytes_trisopterus_states_5b, stree, 0.9023 oocytes_trisopterus_states_5b, stree, 0.924441
oocytes_trisopterus_states_5b, wodt, 0.89165 oocytes_trisopterus_states_5b, wodt, 0.89165
oocytes_trisopterus_states_5b, j48svm, 0.887943 oocytes_trisopterus_states_5b, j48svm, 0.887943
oocytes_trisopterus_states_5b, oc1, 0.86393 oocytes_trisopterus_states_5b, oc1, 0.86393
oocytes_trisopterus_states_5b, cart, 0.870263 oocytes_trisopterus_states_5b, cart, 0.870263
oocytes_trisopterus_states_5b, baseRaF, 0.922149 oocytes_trisopterus_states_5b, baseRaF, 0.922149
parkinsons, stree, 0.882051 parkinsons, stree, 0.865641
parkinsons, wodt, 0.901538 parkinsons, wodt, 0.901538
parkinsons, j48svm, 0.844615 parkinsons, j48svm, 0.844615
parkinsons, oc1, 0.865641 parkinsons, oc1, 0.865641
parkinsons, cart, 0.855897 parkinsons, cart, 0.855897
parkinsons, baseRaF, 0.87924 parkinsons, baseRaF, 0.87924
pima, stree, 0.766651 pima, stree, 0.764053
pima, wodt, 0.681591 pima, wodt, 0.681591
pima, j48svm, 0.749876 pima, j48svm, 0.749876
pima, oc1, 0.693027 pima, oc1, 0.693027
pima, cart, 0.701172 pima, cart, 0.701172
pima, baseRaF, 0.697005 pima, baseRaF, 0.697005
pittsburg-bridges-MATERIAL, stree, 0.787446 pittsburg-bridges-MATERIAL, stree, 0.867749
pittsburg-bridges-MATERIAL, wodt, 0.79961 pittsburg-bridges-MATERIAL, wodt, 0.79961
pittsburg-bridges-MATERIAL, j48svm, 0.855844 pittsburg-bridges-MATERIAL, j48svm, 0.855844
pittsburg-bridges-MATERIAL, oc1, 0.81026 pittsburg-bridges-MATERIAL, oc1, 0.81026
pittsburg-bridges-MATERIAL, cart, 0.783593 pittsburg-bridges-MATERIAL, cart, 0.783593
pittsburg-bridges-MATERIAL, baseRaF, 0.81136 pittsburg-bridges-MATERIAL, baseRaF, 0.81136
pittsburg-bridges-REL-L, stree, 0.62519 pittsburg-bridges-REL-L, stree, 0.564048
pittsburg-bridges-REL-L, wodt, 0.617143 pittsburg-bridges-REL-L, wodt, 0.617143
pittsburg-bridges-REL-L, j48svm, 0.645048 pittsburg-bridges-REL-L, j48svm, 0.645048
pittsburg-bridges-REL-L, oc1, 0.604957 pittsburg-bridges-REL-L, oc1, 0.604957
pittsburg-bridges-REL-L, cart, 0.625333 pittsburg-bridges-REL-L, cart, 0.625333
pittsburg-bridges-REL-L, baseRaF, 0.622107 pittsburg-bridges-REL-L, baseRaF, 0.622107
pittsburg-bridges-SPAN, stree, 0.630234 pittsburg-bridges-SPAN, stree, 0.658713
pittsburg-bridges-SPAN, wodt, 0.606959 pittsburg-bridges-SPAN, wodt, 0.606959
pittsburg-bridges-SPAN, j48svm, 0.621579 pittsburg-bridges-SPAN, j48svm, 0.621579
pittsburg-bridges-SPAN, oc1, 0.579333 pittsburg-bridges-SPAN, oc1, 0.579333
pittsburg-bridges-SPAN, cart, 0.557544 pittsburg-bridges-SPAN, cart, 0.557544
pittsburg-bridges-SPAN, baseRaF, 0.630217 pittsburg-bridges-SPAN, baseRaF, 0.630217
pittsburg-bridges-T-OR-D, stree, 0.861619 pittsburg-bridges-T-OR-D, stree, 0.849952
pittsburg-bridges-T-OR-D, wodt, 0.818429 pittsburg-bridges-T-OR-D, wodt, 0.818429
pittsburg-bridges-T-OR-D, j48svm, 0.838333 pittsburg-bridges-T-OR-D, j48svm, 0.838333
pittsburg-bridges-T-OR-D, oc1, 0.831545 pittsburg-bridges-T-OR-D, oc1, 0.831545
pittsburg-bridges-T-OR-D, cart, 0.821619 pittsburg-bridges-T-OR-D, cart, 0.821619
pittsburg-bridges-T-OR-D, baseRaF, 0.821007 pittsburg-bridges-T-OR-D, baseRaF, 0.821007
planning, stree, 0.70455 planning, stree, 0.73527
planning, wodt, 0.576847 planning, wodt, 0.576847
planning, j48svm, 0.711381 planning, j48svm, 0.711381
planning, oc1, 0.566988 planning, oc1, 0.566988
planning, cart, 0.586712 planning, cart, 0.586712
planning, baseRaF, 0.590586 planning, baseRaF, 0.590586
post-operative, stree, 0.573333 post-operative, stree, 0.703333
post-operative, wodt, 0.535556 post-operative, wodt, 0.535556
post-operative, j48svm, 0.701111 post-operative, j48svm, 0.701111
post-operative, oc1, 0.542222 post-operative, oc1, 0.542222
post-operative, cart, 0.567778 post-operative, cart, 0.567778
post-operative, baseRaF, 0.539375 post-operative, baseRaF, 0.539375
seeds, stree, 0.949048 seeds, stree, 0.952857
seeds, wodt, 0.940476 seeds, wodt, 0.940476
seeds, j48svm, 0.909524 seeds, j48svm, 0.909524
seeds, oc1, 0.932381 seeds, oc1, 0.932381
seeds, cart, 0.900476 seeds, cart, 0.900476
seeds, baseRaF, 0.942518 seeds, baseRaF, 0.942518
statlog-australian-credit, stree, 0.667246 statlog-australian-credit, stree, 0.678261
statlog-australian-credit, wodt, 0.561594 statlog-australian-credit, wodt, 0.561594
statlog-australian-credit, j48svm, 0.66029 statlog-australian-credit, j48svm, 0.66029
statlog-australian-credit, oc1, 0.573913 statlog-australian-credit, oc1, 0.573913
statlog-australian-credit, cart, 0.595507 statlog-australian-credit, cart, 0.595507
statlog-australian-credit, baseRaF, 0.678261 statlog-australian-credit, baseRaF, 0.678261
statlog-german-credit, stree, 0.7625 statlog-german-credit, stree, 0.7569
statlog-german-credit, wodt, 0.6929 statlog-german-credit, wodt, 0.6929
statlog-german-credit, j48svm, 0.7244 statlog-german-credit, j48svm, 0.7244
statlog-german-credit, oc1, 0.6874 statlog-german-credit, oc1, 0.6874
statlog-german-credit, cart, 0.6738 statlog-german-credit, cart, 0.6738
statlog-german-credit, baseRaF, 0.68762 statlog-german-credit, baseRaF, 0.68762
statlog-heart, stree, 0.822963 statlog-heart, stree, 0.822222
statlog-heart, wodt, 0.777778 statlog-heart, wodt, 0.777778
statlog-heart, j48svm, 0.795926 statlog-heart, j48svm, 0.795926
statlog-heart, oc1, 0.749259 statlog-heart, oc1, 0.749259
statlog-heart, cart, 0.762222 statlog-heart, cart, 0.762222
statlog-heart, baseRaF, 0.747605 statlog-heart, baseRaF, 0.747605
statlog-image, stree, 0.850649 statlog-image, stree, 0.956623
statlog-image, wodt, 0.954632 statlog-image, wodt, 0.954632
statlog-image, j48svm, 0.967403 statlog-image, j48svm, 0.967403
statlog-image, oc1, 0.95013 statlog-image, oc1, 0.95013
statlog-image, cart, 0.964892 statlog-image, cart, 0.964892
statlog-image, baseRaF, 0.953604 statlog-image, baseRaF, 0.953604
statlog-vehicle, stree, 0.695151 statlog-vehicle, stree, 0.788537
statlog-vehicle, wodt, 0.726492 statlog-vehicle, wodt, 0.726492
statlog-vehicle, j48svm, 0.729651 statlog-vehicle, j48svm, 0.729651
statlog-vehicle, oc1, 0.708496 statlog-vehicle, oc1, 0.708496
statlog-vehicle, cart, 0.728367 statlog-vehicle, cart, 0.728367
statlog-vehicle, baseRaF, 0.789572 statlog-vehicle, baseRaF, 0.789572
synthetic-control, stree, 0.938833 synthetic-control, stree, 0.95
synthetic-control, wodt, 0.973167 synthetic-control, wodt, 0.973167
synthetic-control, j48svm, 0.922333 synthetic-control, j48svm, 0.922333
synthetic-control, oc1, 0.863167 synthetic-control, oc1, 0.863167
synthetic-control, cart, 0.908333 synthetic-control, cart, 0.908333
synthetic-control, baseRaF, 0.971567 synthetic-control, baseRaF, 0.971567
tic-tac-toe, stree, 0.983296 tic-tac-toe, stree, 0.984444
tic-tac-toe, wodt, 0.93905 tic-tac-toe, wodt, 0.93905
tic-tac-toe, j48svm, 0.983295 tic-tac-toe, j48svm, 0.983295
tic-tac-toe, oc1, 0.91849 tic-tac-toe, oc1, 0.91849
tic-tac-toe, cart, 0.951558 tic-tac-toe, cart, 0.951558
tic-tac-toe, baseRaF, 0.974906 tic-tac-toe, baseRaF, 0.974906
vertebral-column-2clases, stree, 0.852903 vertebral-column-2clases, stree, 0.851936
vertebral-column-2clases, wodt, 0.801935 vertebral-column-2clases, wodt, 0.801935
vertebral-column-2clases, j48svm, 0.84871 vertebral-column-2clases, j48svm, 0.84871
vertebral-column-2clases, oc1, 0.815161 vertebral-column-2clases, oc1, 0.815161
vertebral-column-2clases, cart, 0.784839 vertebral-column-2clases, cart, 0.784839
vertebral-column-2clases, baseRaF, 0.822601 vertebral-column-2clases, baseRaF, 0.822601
wine, stree, 0.97581 wine, stree, 0.949333
wine, wodt, 0.973048 wine, wodt, 0.973048
wine, j48svm, 0.979143 wine, j48svm, 0.979143
wine, oc1, 0.916165 wine, oc1, 0.916165
wine, cart, 0.921937 wine, cart, 0.921937
wine, baseRaF, 0.97748 wine, baseRaF, 0.97748
zoo, stree, 0.947619 zoo, stree, 0.955524
zoo, wodt, 0.954429 zoo, wodt, 0.954429
zoo, j48svm, 0.92381 zoo, j48svm, 0.92381
zoo, oc1, 0.890952 zoo, oc1, 0.890952
1 dataset classifier accuracy
2 balance-scale stree 0.91184 0.97056
3 balance-scale wodt 0.912
4 balance-scale j48svm 0.94
5 balance-scale oc1 0.9192
6 balance-scale cart 0.78816
7 balance-scale baseRaF 0.706738
8 balloons stree 0.653333 0.86
9 balloons wodt 0.688333
10 balloons j48svm 0.595
11 balloons oc1 0.62
12 balloons cart 0.671667
13 balloons baseRaF 0.605
14 breast-cancer-wisc-diag stree 0.968898 0.972764
15 breast-cancer-wisc-diag wodt 0.967317
16 breast-cancer-wisc-diag j48svm 0.952878
17 breast-cancer-wisc-diag oc1 0.933477
18 breast-cancer-wisc-diag cart 0.93953
19 breast-cancer-wisc-diag baseRaF 0.965694
20 breast-cancer-wisc-prog stree 0.802051 0.811128
21 breast-cancer-wisc-prog wodt 0.710141
22 breast-cancer-wisc-prog j48svm 0.724038
23 breast-cancer-wisc-prog oc1 0.71
24 breast-cancer-wisc-prog cart 0.699833
25 breast-cancer-wisc-prog baseRaF 0.74485
26 breast-cancer-wisc stree 0.966661 0.965802
27 breast-cancer-wisc wodt 0.946208
28 breast-cancer-wisc j48svm 0.967674
29 breast-cancer-wisc oc1 0.940194
30 breast-cancer-wisc cart 0.940629
31 breast-cancer-wisc baseRaF 0.942857
32 breast-cancer stree 0.734211 0.733158
33 breast-cancer wodt 0.650236
34 breast-cancer j48svm 0.707719
35 breast-cancer oc1 0.649728
36 breast-cancer cart 0.65444
37 breast-cancer baseRaF 0.656438
38 cardiotocography-10clases stree 0.552558 0.712009
39 cardiotocography-10clases wodt 0.773706
40 cardiotocography-10clases j48svm 0.830812
41 cardiotocography-10clases oc1 0.795528
42 cardiotocography-10clases cart 0.818864
43 cardiotocography-10clases baseRaF 0.774788
44 cardiotocography-3clases stree 0.35207 0.891956
45 cardiotocography-3clases wodt 0.897509
46 cardiotocography-3clases j48svm 0.927327
47 cardiotocography-3clases oc1 0.899811
48 cardiotocography-3clases cart 0.929258
49 cardiotocography-3clases baseRaF 0.896715
50 conn-bench-sonar-mines-rocks stree 0.755528 0.71439
51 conn-bench-sonar-mines-rocks wodt 0.824959
52 conn-bench-sonar-mines-rocks j48svm 0.73892
53 conn-bench-sonar-mines-rocks oc1 0.710798
54 conn-bench-sonar-mines-rocks cart 0.728711
55 conn-bench-sonar-mines-rocks baseRaF 0.772981
56 cylinder-bands stree 0.715049 0.687101
57 cylinder-bands wodt 0.704074
58 cylinder-bands j48svm 0.726351
59 cylinder-bands oc1 0.67106
60 cylinder-bands cart 0.712703
61 cylinder-bands baseRaF 0.675117
62 dermatology stree 0.966087 0.971833
63 dermatology wodt 0.965557
64 dermatology j48svm 0.955735
65 dermatology oc1 0.916087
66 dermatology cart 0.932766
67 dermatology baseRaF 0.970723
68 echocardiogram stree 0.808832 0.814758
69 echocardiogram wodt 0.733875
70 echocardiogram j48svm 0.805527
71 echocardiogram oc1 0.748291
72 echocardiogram cart 0.745043
73 echocardiogram baseRaF 0.753522
74 fertility stree 0.866 0.88
75 fertility wodt 0.785
76 fertility j48svm 0.857
77 fertility oc1 0.793
78 fertility cart 0.8
79 fertility baseRaF 0.798
80 haberman-survival stree 0.735637 0.727795
81 haberman-survival wodt 0.664707
82 haberman-survival j48svm 0.714056
83 haberman-survival oc1 0.651634
84 haberman-survival cart 0.65
85 haberman-survival baseRaF 0.720133
86 heart-hungarian stree 0.817674 0.827522
87 heart-hungarian wodt 0.764909
88 heart-hungarian j48svm 0.785026
89 heart-hungarian oc1 0.758298
90 heart-hungarian cart 0.760508
91 heart-hungarian baseRaF 0.779804
92 hepatitis stree 0.796129 0.824516
93 hepatitis wodt 0.785806
94 hepatitis j48svm 0.761935
95 hepatitis oc1 0.756774
96 hepatitis cart 0.765161
97 hepatitis baseRaF 0.773671
98 ilpd-indian-liver stree 0.723498 0.719207
99 ilpd-indian-liver wodt 0.676176
100 ilpd-indian-liver j48svm 0.690339
101 ilpd-indian-liver oc1 0.660139
102 ilpd-indian-liver cart 0.663423
103 ilpd-indian-liver baseRaF 0.696685
104 ionosphere stree 0.866056 0.953276
105 ionosphere wodt 0.88008
106 ionosphere j48svm 0.891984
107 ionosphere oc1 0.879742
113 iris oc1 0.948
114 iris cart 0.938667
115 iris baseRaF 0.953413
116 led-display stree 0.7007 0.703
117 led-display wodt 0.7049
118 led-display j48svm 0.7204
119 led-display oc1 0.6993
120 led-display cart 0.7037
121 led-display baseRaF 0.70178
122 libras stree 0.747778 0.788333
123 libras wodt 0.764167
124 libras j48svm 0.66
125 libras oc1 0.645
126 libras cart 0.655
127 libras baseRaF 0.726722
128 low-res-spect stree 0.853102 0.865713
129 low-res-spect wodt 0.856459
130 low-res-spect j48svm 0.83358
131 low-res-spect oc1 0.824671
132 low-res-spect cart 0.829206
133 low-res-spect baseRaF 0.790875
134 lymphography stree 0.77046 0.823425
135 lymphography wodt 0.808782
136 lymphography j48svm 0.778552
137 lymphography oc1 0.734634
138 lymphography cart 0.766276
139 lymphography baseRaF 0.761622
140 mammographic stree 0.81915 0.817068
141 mammographic wodt 0.759839
142 mammographic j48svm 0.821435
143 mammographic oc1 0.768805
144 mammographic cart 0.757131
145 mammographic baseRaF 0.780206
146 molec-biol-promoter stree 0.764416 0.767056
147 molec-biol-promoter wodt 0.798528
148 molec-biol-promoter j48svm 0.744935
149 molec-biol-promoter oc1 0.734805
150 molec-biol-promoter cart 0.748701
151 molec-biol-promoter baseRaF 0.667239
152 musk-1 stree 0.843463 0.916388
153 musk-1 wodt 0.838914
154 musk-1 j48svm 0.82693
155 musk-1 oc1 0.776401
156 musk-1 cart 0.780215
157 musk-1 baseRaF 0.834034
158 oocytes_merluccius_nucleus_4d stree 0.810657 0.835125
159 oocytes_merluccius_nucleus_4d wodt 0.737673
160 oocytes_merluccius_nucleus_4d j48svm 0.741766
161 oocytes_merluccius_nucleus_4d oc1 0.743199
167 oocytes_merluccius_states_2f oc1 0.889223
168 oocytes_merluccius_states_2f cart 0.891193
169 oocytes_merluccius_states_2f baseRaF 0.910551
170 oocytes_trisopterus_nucleus_2f stree 0.800986 0.799995
171 oocytes_trisopterus_nucleus_2f wodt 0.751431
172 oocytes_trisopterus_nucleus_2f j48svm 0.756587
173 oocytes_trisopterus_nucleus_2f oc1 0.747697
174 oocytes_trisopterus_nucleus_2f cart 0.734313
175 oocytes_trisopterus_nucleus_2f baseRaF 0.76193
176 oocytes_trisopterus_states_5b stree 0.9023 0.924441
177 oocytes_trisopterus_states_5b wodt 0.89165
178 oocytes_trisopterus_states_5b j48svm 0.887943
179 oocytes_trisopterus_states_5b oc1 0.86393
180 oocytes_trisopterus_states_5b cart 0.870263
181 oocytes_trisopterus_states_5b baseRaF 0.922149
182 parkinsons stree 0.882051 0.865641
183 parkinsons wodt 0.901538
184 parkinsons j48svm 0.844615
185 parkinsons oc1 0.865641
186 parkinsons cart 0.855897
187 parkinsons baseRaF 0.87924
188 pima stree 0.766651 0.764053
189 pima wodt 0.681591
190 pima j48svm 0.749876
191 pima oc1 0.693027
192 pima cart 0.701172
193 pima baseRaF 0.697005
194 pittsburg-bridges-MATERIAL stree 0.787446 0.867749
195 pittsburg-bridges-MATERIAL wodt 0.79961
196 pittsburg-bridges-MATERIAL j48svm 0.855844
197 pittsburg-bridges-MATERIAL oc1 0.81026
198 pittsburg-bridges-MATERIAL cart 0.783593
199 pittsburg-bridges-MATERIAL baseRaF 0.81136
200 pittsburg-bridges-REL-L stree 0.62519 0.564048
201 pittsburg-bridges-REL-L wodt 0.617143
202 pittsburg-bridges-REL-L j48svm 0.645048
203 pittsburg-bridges-REL-L oc1 0.604957
204 pittsburg-bridges-REL-L cart 0.625333
205 pittsburg-bridges-REL-L baseRaF 0.622107
206 pittsburg-bridges-SPAN stree 0.630234 0.658713
207 pittsburg-bridges-SPAN wodt 0.606959
208 pittsburg-bridges-SPAN j48svm 0.621579
209 pittsburg-bridges-SPAN oc1 0.579333
210 pittsburg-bridges-SPAN cart 0.557544
211 pittsburg-bridges-SPAN baseRaF 0.630217
212 pittsburg-bridges-T-OR-D stree 0.861619 0.849952
213 pittsburg-bridges-T-OR-D wodt 0.818429
214 pittsburg-bridges-T-OR-D j48svm 0.838333
215 pittsburg-bridges-T-OR-D oc1 0.831545
216 pittsburg-bridges-T-OR-D cart 0.821619
217 pittsburg-bridges-T-OR-D baseRaF 0.821007
218 planning stree 0.70455 0.73527
219 planning wodt 0.576847
220 planning j48svm 0.711381
221 planning oc1 0.566988
222 planning cart 0.586712
223 planning baseRaF 0.590586
224 post-operative stree 0.573333 0.703333
225 post-operative wodt 0.535556
226 post-operative j48svm 0.701111
227 post-operative oc1 0.542222
228 post-operative cart 0.567778
229 post-operative baseRaF 0.539375
230 seeds stree 0.949048 0.952857
231 seeds wodt 0.940476
232 seeds j48svm 0.909524
233 seeds oc1 0.932381
234 seeds cart 0.900476
235 seeds baseRaF 0.942518
236 statlog-australian-credit stree 0.667246 0.678261
237 statlog-australian-credit wodt 0.561594
238 statlog-australian-credit j48svm 0.66029
239 statlog-australian-credit oc1 0.573913
240 statlog-australian-credit cart 0.595507
241 statlog-australian-credit baseRaF 0.678261
242 statlog-german-credit stree 0.7625 0.7569
243 statlog-german-credit wodt 0.6929
244 statlog-german-credit j48svm 0.7244
245 statlog-german-credit oc1 0.6874
246 statlog-german-credit cart 0.6738
247 statlog-german-credit baseRaF 0.68762
248 statlog-heart stree 0.822963 0.822222
249 statlog-heart wodt 0.777778
250 statlog-heart j48svm 0.795926
251 statlog-heart oc1 0.749259
252 statlog-heart cart 0.762222
253 statlog-heart baseRaF 0.747605
254 statlog-image stree 0.850649 0.956623
255 statlog-image wodt 0.954632
256 statlog-image j48svm 0.967403
257 statlog-image oc1 0.95013
258 statlog-image cart 0.964892
259 statlog-image baseRaF 0.953604
260 statlog-vehicle stree 0.695151 0.788537
261 statlog-vehicle wodt 0.726492
262 statlog-vehicle j48svm 0.729651
263 statlog-vehicle oc1 0.708496
264 statlog-vehicle cart 0.728367
265 statlog-vehicle baseRaF 0.789572
266 synthetic-control stree 0.938833 0.95
267 synthetic-control wodt 0.973167
268 synthetic-control j48svm 0.922333
269 synthetic-control oc1 0.863167
270 synthetic-control cart 0.908333
271 synthetic-control baseRaF 0.971567
272 tic-tac-toe stree 0.983296 0.984444
273 tic-tac-toe wodt 0.93905
274 tic-tac-toe j48svm 0.983295
275 tic-tac-toe oc1 0.91849
276 tic-tac-toe cart 0.951558
277 tic-tac-toe baseRaF 0.974906
278 vertebral-column-2clases stree 0.852903 0.851936
279 vertebral-column-2clases wodt 0.801935
280 vertebral-column-2clases j48svm 0.84871
281 vertebral-column-2clases oc1 0.815161
282 vertebral-column-2clases cart 0.784839
283 vertebral-column-2clases baseRaF 0.822601
284 wine stree 0.97581 0.949333
285 wine wodt 0.973048
286 wine j48svm 0.979143
287 wine oc1 0.916165
288 wine cart 0.921937
289 wine baseRaF 0.97748
290 zoo stree 0.947619 0.955524
291 zoo wodt 0.954429
292 zoo j48svm 0.92381
293 zoo oc1 0.890952

47
report_score.py
View File

@@ -55,22 +55,30 @@ def parse_arguments():
return (args.set_of_files, args.model, args.dataset, args.sql, args.param) return (args.set_of_files, args.model, args.dataset, args.sql, args.param)
def nodes_leaves(clf):
nodes = 0
leaves = 0
for node in clf:
if node.is_leaf():
leaves += 1
else:
nodes += 1
return nodes, leaves
def compute_auto_hyperparams(X, y): def compute_auto_hyperparams(X, y):
params = {"max_iter": 1e4, "C": 0.1} """Propuesta de auto configuración de hiperparámetros
classes = len(np.unique(y)) max_it = 10e4
if classes > 2: (1 valor)
params["split_criteria"] = "max_samples" split = impurity si clases==2 y split=max_samples si clases > 2
(1 valor)
kernel=linear o polinómico
(2 valores)
C = 0.1, 0.5 y 1.0
(3 valores)
Caso 1: C=1, max_iter=1e4 + condicional split_max kernel lineal
Caso 2: C=0.5, max_iter=1e4 + condicional split_max kernel lineal
Caso 3: C=0.1, max_iter=1e4 + condicional split_max kernel lineal
Caso 4: C=1, max_iter=1e4 + condicional split_max kernel poly
Caso 5: C=0.5, max_iter=1e4 + condicional split_max kernel poly
Caso 6: C=0.1, max_iter=1e4 + condicional split_max kernel poly
Caso 7: C=1, max_iter=1e4 + condicional + kernel rbf
Caso 8: kernel rbf
"""
# params = {"max_iter": 1e4, "kernel": "rbf"}
# classes = len(np.unique(y))
# if classes > 2:
# params["split_criteria"] = "max_samples"
params = {"kernel": "rbf"}
return params return params
@@ -97,7 +105,7 @@ def process_dataset(dataset, verbose, model, auto_params):
clf = Stree(random_state=random_state) clf = Stree(random_state=random_state)
clf.set_params(**hyperparameters) clf.set_params(**hyperparameters)
res = cross_validate(clf, X, y, cv=kfold, return_estimator=True) res = cross_validate(clf, X, y, cv=kfold, return_estimator=True)
nodes, leaves = nodes_leaves(res["estimator"][0]) nodes, leaves = res["estimator"][0].nodes_leaves()
depth = res["estimator"][0].depth_ depth = res["estimator"][0].depth_
scores.append(res["test_score"]) scores.append(res["test_score"])
times.append(res["fit_time"]) times.append(res["fit_time"])
@@ -222,6 +230,9 @@ if dataset == "all":
parameters = json.loads("{}") parameters = json.loads("{}")
accuracy_best = 0.0 accuracy_best = 0.0
acc_best_std = 0.0 acc_best_std = 0.0
if auto_params:
# show parameters computed
parameters = json.loads(hyperparameters)
accuracy_computed = np.mean(scores) accuracy_computed = np.mean(scores)
diff = accuracy_best - accuracy_computed diff = accuracy_best - accuracy_computed
print( print(
@@ -243,12 +254,12 @@ else:
accuracy_best = record[5] if record is not None else 0.0 accuracy_best = record[5] if record is not None else 0.0
acc_best_std = record[11] if record is not None else 0.0 acc_best_std = record[11] if record is not None else 0.0
print( print(
f"* Accuracy Computed : {accuracy:6.4f}±{np.std(scores):6.4f} " f"* Accuracy Computed .: {accuracy:6.4f}±{np.std(scores):6.4f} "
f"{np.mean(times):5.3f}s" f"{np.mean(times):5.3f}s"
) )
print(f"* Accuracy Best .....: {accuracy_best:6.4f}±{acc_best_std:6.4f}") print(f"* Accuracy Best .....: {accuracy_best:6.4f}±{acc_best_std:6.4f}")
print(f"* Difference ........: {accuracy_best - accuracy:6.4f}") print(f"* Difference ........: {accuracy_best - accuracy:6.4f}")
print(f"* Nodes/Leaves/Depth :{nodes:2d} {leaves:2d} " f"{depth:2d} ") print(f"* Nodes/Leaves/Depth : {nodes:2d} {leaves:2d} " f"{depth:2d} ")
stop = time.time() stop = time.time()
print(f"- Auto Hyperparams ..: {hyperparameters}") print(f"- Auto Hyperparams ..: {hyperparameters}")
hours, rem = divmod(stop - start, 3600) hours, rem = divmod(stop - start, 3600)