import sys
sys.path.insert(0, "../myTools")
import joblib
import os
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.ensemble import RandomForestRegressor
from sklearn.neighbors import KNeighborsRegressor
from loadDataSet import loadMainDataSet, loadTesteDataSet, loadCompletDataSet, loadMainDataSetWithElevation
from sklearn.model_selection import GridSearchCV
from sklearn.model_selection import train_test_split
dataSet, _, _ = loadMainDataSetWithElevation()
# Melhores modelos
# Modelo Magnésio
#6, 2, 'log2', True, 6, 33
def MELHOR_RESULTADO_MG():
X = dataSet[:, :4]
y = dataSet[:, 4]
params = {
'random_state': 0,
'learning_rate': 0.05,
'loss': 'lad',
'max_depth': None,
'max_features': 'log2',
'min_samples_leaf': 5,
]
reg =GridSearchCV(model, cv=3,param_grid=param_grid,verbose=0,n_jobs=-1,scoring=scores,refit='r2',iid=True)
reg.fit(X_train,y_train)
return reg.best_estimator_,reg.best_params_,reg.best_score_
args = sys.argv[1:]
y_column,random_state = verifyArgs(args)
# Load data set
dataSet = loadMainDataSetWithElevation()
#dataSet = loadTesteDataSet()
#dataSet = loadMainDataSet()
#Set features and target
y_column = 2
X = dataSet[:,0:2]
y = dataSet[:,2]
best_model , best_params, best_score,best_seed = findBalancedDataSet(range(1,10),X,y,GridSearchCVKNeighborsRegressor)
print("#Best Params",best_params)
print("#Best Score",best_score)
print("#Best Seed",best_seed)
import sys
sys.path.insert(0, "../myTools")
import numpy as np
from dataSetPreProcessing import train_validation_test_split
from sklearn.model_selection import train_test_split
from loadDataSet import loadMainDataSet, loadTesteDataSet, loadCompletDataSet, loadMainDataSetWithElevation
from tools import verifyArgs, plotLeanrningCurve, findBalancedDataSet, pltResults, pltCorrelation, pltLossGraph, pltShow, plotXY, getMetrics, getBalancedDataSetIndexRandomState
from sklearn.metrics import r2_score, mean_squared_error
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.model_selection import GridSearchCV
# Load data set
dataSet, features_names, target_names = loadMainDataSetWithElevation()
def getParamGrid():
'''Função que retorna os parâmetro utilizados para encontrar
o conjunto de treino balanceado e também na tunagem.
'''
param_grid_half = {
'n_estimators': [100],
'learning_rate': [0.05],
'max_depth': [None],
'min_samples_leaf': [3, 8],
'max_features': ['auto', 'log2'],
'loss': ['ls']
}
param_grid_full = {