|
SOL4Py Sample: XGBRegressor
|
#******************************************************************************
#
# Copyright (c) 2018 Antillia.com TOSHIYUKI ARAI. ALL RIGHTS RESERVED.
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
#******************************************************************************
# 2018/09/10
# XGBRegressor.py
# This is based on the following program:
# http://scikit-learn.org/stable/auto_examples/ensemble/plot_gradient_boosting_regression.html
#
# See also:
# xgboost/demo/guide-python/sklearn_examples.py
# https://github.com/dmlc/xgboost/blob/master/demo/guide-python/sklearn_examples.py
# encodig: utf-8
import sys
import os
import cv2
import time
import traceback
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import pickle
import xgboost as xgb
from sklearn.model_selection import train_test_split
from sklearn.model_selection import GridSearchCV
from sklearn import datasets
from sklearn.utils import shuffle
from sklearn.metrics import mean_squared_error
from PyQt5.QtCore import *
from PyQt5.QtWidgets import *
from PyQt5.QtGui import *
sys.path.append('../')
from SOL4Py.ZMLModel import *
from SOL4Py.ZApplicationView import *
from SOL4Py.ZLabeledComboBox import *
from SOL4Py.ZPushButton import *
from SOL4Py.ZTabbedWindow import *
from SOL4Py.ZVerticalPane import *
from SOL4Py.ZTabbedWindow import *
from SOL4Py.ZScalableScrolledFigureView import *
Boston = 0
Diabetes = 1
############################################################
#
# XGBRegressorModel class
#
class XGBRegressorModel(ZMLModel):
def __init__(self, dataset_id, mainv):
super(XGBRegressorModel, self).__init__(dataset_id, mainv)
def run(self):
self.write("====================================")
self._start(self.run.__name__)
try:
self.load_dataset()
if self.trained():
self.load()
else:
self.build()
self.train()
self.save()
self.predict()
self.visualize()
except:
traceback.print_exc()
self._end(self.run.__name__)
def load_dataset(self):
self._start(self.load_dataset.__name__)
if self.dataset_id == Boston:
self.dataset= datasets.load_boston()
self.write("loaded Boston dataset")
if self.dataset_id == Diabetes:
self.dataset= datasets.load_diabetes()
self.write("loaded Diabetes dataset")
attr = dir(self.dataset)
self.write("dir:" + str(attr))
if "feature_names" in attr:
self.write("feature_names:" + str(self.dataset.feature_names))
if "target_names" in attr:
self.write("target_names:" + str(self.dataset.target_names))
self.set_model_filename()
self.view.description.setText(self.dataset.DESCR)
X, y = self.dataset.data, self.dataset.target
self.X_train, self.X_test, self.y_train, self.y_test = train_test_split(X, y, test_size=0.3, random_state=42)
self._end(self.load_dataset.__name__)
def build(self):
self._start(self.build.__name__)
self.model = xgb.XGBRegressor()
self._end(self.build.__name__)
def train(self):
#We use GridSearchCV
self._start(self.train.__name__)
start = time.time()
params = {'max_depth': [2,4,6], 'n_estimators': [50,100,200]}
grid_search = GridSearchCV(self.model, params, verbose=1)
grid_search.fit(self.X_train, self.y_train)
self.write("GridSearch BestParams " + str(grid_search.best_params_) )
self.write("GridSearch BestScore " + str( grid_search.best_score_) )
# Rebuild a XGBRegressor model by using grid_search.best_params_
self.model = xgb.XGBRegressor(**grid_search.best_params_)
self.model.fit(self.X_train, self.y_train)
elapsed_time = time.time() - start
elapsed = str("Train elapsed_time:{0}".format(elapsed_time) + "[sec]")
self.write(elapsed)
self._end(self.train.__name__)
def predict(self):
self._start(self.predict.__name__)
self.pred_train = self.model.predict(self.X_train)
self.pred_test = self.model.predict(self.X_test)
self.mean_squared_error()
self._end(self.predict.__name__)
def mean_squared_error(self):
self.write("MSE:train " + str(mean_squared_error(self.y_train, self.pred_train)) )
self.write("MSE:test " + str(mean_squared_error(self.y_test, self.pred_test)) )
def visualize(self):
importances = pd.Series(self.model.feature_importances_,
index = self.dataset.feature_names)
importances = importances.sort_values()
self.view.visualize(importances)
############################################################
#
# MainView class, which is a subclass of ZApplicationView
#
class MainView(ZApplicationView):
# MainView Constructor
def __init__(self, title, x, y, width, height):
super(MainView, self).__init__(title, x, y, width, height)
self.font = QFont("Arial", 10)
self.setFont(self.font)
# 1 Add a labeled combobox to top dock area
self.add_datasets_combobox()
# 2 Add a textedit to the left pane of the center area.
self.text_editor = QTextEdit()
self.text_editor.setFont(self.font)
self.text_editor.setLineWrapColumnOrWidth(600)
self.text_editor.setLineWrapMode(QTextEdit.FixedPixelWidth)
self.description = QTextEdit()
self.description.setFont(self.font)
self.description.setLineWrapColumnOrWidth(600)
self.description.setLineWrapMode(QTextEdit.FixedPixelWidth)
# 3 Add a tabbled_window to the right pane of the center area.
self.tabbed_window = ZTabbedWindow(self, 0, 0, width/2, height)
# 4 Add a figure_view to the tabbed_window
self.figure_view = ZScalableScrolledFigureView(self, 0, 0, width/2, height)
self.add(self.text_editor)
self.add(self.tabbed_window)
self.tabbed_window.add("Description", self.description)
self.tabbed_window.add("Importances", self.figure_view)
self.figure_view.hide()
self.show()
def add_datasets_combobox(self):
self.dataset_id = Boston
self.datasets_combobox = ZLabeledComboBox(self, "Datasets", Qt.Horizontal)
self.datasets_combobox.setFont(self.font)
# We use the following datasets of sklearn to test XGBRegressor.
self.datasets = {"Boston": Boston, "Diabetes": Diabetes}
title = self.get_title()
self.setWindowTitle( "Boston" + " - " + title)
self.datasets_combobox.add_items(self.datasets.keys())
self.datasets_combobox.add_activated_callback(self.datasets_activated)
self.datasets_combobox.set_current_text(self.dataset_id)
self.start_button = ZPushButton("Start", self)
self.clear_button = ZPushButton("Clear", self)
self.start_button.setFont(self.font)
self.clear_button.setFont(self.font)
self.start_button.add_activated_callback(self.start_button_activated)
self.clear_button.add_activated_callback(self.clear_button_activated)
self.datasets_combobox.add(self.start_button)
self.datasets_combobox.add(self.clear_button)
self.set_top_dock(self.datasets_combobox)
def write(self, text):
self.text_editor.append(text)
self.text_editor.repaint()
def datasets_activated(self, text):
self.dataset_id = self.datasets[text]
title = self.get_title()
self.setWindowTitle(text + " - " + title)
def start_button_activated(self, text):
self.model = XGBRegressorModel(self.dataset_id, self)
self.start_button.setEnabled(False)
self.clear_button.setEnabled(False)
try:
self.model.run()
except:
pass
self.start_button.setEnabled(True)
self.clear_button.setEnabled(True)
def file_new(self):
self.text_editor.setText("")
self.description.setText("")
self.figure_view.hide()
if plt.gcf() != None:
plt.close()
def clear_button_activated(self, text):
self.file_new()
def visualize(self, importances):
self.figure_view.show()
if plt.gcf() != None:
plt.close()
importances.plot(kind = "barh")
self.figure_view.set_figure( plt )
############################################################
#
# Program main start point
#
if main(__name__):
try:
app_name = os.path.basename(sys.argv[0])
applet = QApplication(sys.argv)
main_view = MainView(app_name, 40, 40, 800, 500)
main_view.show ()
applet.exec_()
except:
traceback.print_exc()
Last modified: 6 May 2018