def train(self, ds):
datassc = ds[ds['StoreType_Shopping Center'] == 1]
datasoth = ds[ds['StoreType_Shopping Center'] == 0]
datasc = sb.SetBuilder(target='NumberOfCustomers', autoexclude=True, df=datassc) \
.exclude('NumberOfSales', 'Month') \
.build()
print("SHOPPING CENTERS MODEL TRAINING: ")
for i in range(self.nsc):
print("SC :", i)
x, y = datasc.random_sampling(1.0)
mod = self.model_shopc()
mod.fit(x, y)
pr = mod.predict(datasc.xtr)
self.models_sc.append(mod)
print("SC ", i, " TRAIN R2: ", eval.evaluate(datasc.ytr, pr))
dataoth = sb.SetBuilder(target='NumberOfCustomers', autoexclude=True, df=datasoth) \
.exclude('NumberOfSales', 'Month') \
.build()
print("OTHER SHOPS MODEL TRAINING: ")
for i in range(self.noth):
print("OTH :", i)
x, y = dataoth.random_sampling(1.0)
mod = self.model_others()
mod.fit(x, y)
pr = mod.predict(dataoth.xtr)
self.models_o.append(mod)
print("OTH ", i, " TRAIN R2: ", eval.evaluate(dataoth.ytr, pr))
def test(self, ds):
datassc = ds[ds['StoreType_Shopping Center'] == 1]
datasoth = ds[ds['StoreType_Shopping Center'] == 0]
datasc = sb.SetBuilder(target='NumberOfCustomers', autoexclude=True, df=datassc) \
.exclude('NumberOfSales', 'Month') \
.build()
print("SHOPPING CENTERS MODEL EVALUATION")
preds = self.predict_sc(datasc.xts)
print("SC TEST R2: ", eval.evaluate(datasc.yts, preds))
dataoth = sb.SetBuilder(target='NumberOfCustomers', autoexclude=True, df=datasoth) \
.exclude('NumberOfSales', 'Month') \
.build()
print("OTHERS MODEL EVALUATION")
preds = self.predict_oth(dataoth.xts)
print("OTH TEST R2: ", eval.evaluate(dataoth.yts, preds))
def __init__(self, count):
Bagging.__init__(self, count)
self.description = "customers"
self.training_ds = sb.SetBuilder(
target='NumberOfCustomers',
autoexclude=True,
split=[[(3, 2016, 2, 2018)], []],
dataset="final_for_customer_train.csv").exclude(
'NumberOfSales', 'Month').build()
self.testing_ds = sb.SetBuilder(
target='NumberOfCustomers',
autoexclude=True,
split=[[], [(3, 2018, 4, 2018)]],
dataset="final_for_customer_test_r.csv").exclude(
'NumberOfSales', 'Month').build()
def __init__(self, count, target):
Bagging.__init__(self, count)
print(
"Remember: IS IS MANDATORY THAT THE COLUMN ORDER OF TARGET SET IS THE SAME OF final_for_sales_train.csv"
)
self.description = "sales"
self.target = target
self.training_ds = sb.SetBuilder(
target='NumberOfSales',
autoexclude=True,
split=[[(3, 2016, 2, 2018)], []],
dataset="final_for_sales_train.csv").exclude('Month').build()
self.testing_ds = sb.SetBuilder(target='NumberOfSales',
autoexclude=True,
split=[[], [(3, 2018, 4, 2018)]],
df=target).exclude('Month').build()
from numpy import loadtxt
from xgboost import XGBRegressor
from sklearn import linear_model
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import PolynomialFeatures
import dataset.setbuilder as sb
import models.sklearn.evaluator as eval
data = sb.SetBuilder(target='NumberOfCustomers',
autoexclude=True).exclude('NumberOfSales',
'Month').build()
model = XGBRegressor(n_estimators=100,
learning_rate=0.2,
colsample_bytree=1,
max_depth=4,
silent=False,
n_jobs=8)
model.fit(data.xtr, data.ytr)
pred_tr = model.predict(data.xtr)
pred_ts = model.predict(data.xts)
print('R2 TRAIN = %s' % eval.evaluate(data.ytr, pred_tr))
print('R2 TEST = %s' % eval.evaluate(data.yts, pred_ts))
preds += p
preds[preds < 0] = 0
for i in range(min(len(preds), number_print)):
print("PRED: ", preds[i], " y: ", y[i])
print("R2: ", eva.r2(ds, preds, 'NumberOfCustomers'))
if __name__ == '__main__':
TRAIN = True
LOAD = False
SAVE_DS = True
name = "lin_"
dts = d.read_dataset("best_for_customers.csv")
ds = sb.SetBuilder(target='NumberOfCustomers', autoexclude=True, df=dts)\
.exclude('NumberOfSales', 'Month')\
.build()
models = []
x, y = ds.xtr, ds.ytr
real_y = np.array(y)
x_test, y_test = ds.xts, ds.yts
for i in range(number_of_model):
if not LOAD:
models.append(m.single_relu(x.shape[1], i == 0))
else:
models.append(k.models.load_model("mod" + name + str(i) + ".h5"))
opt = k.optimizers.adam(lr=2e-4)
models[i].compile(optimizer=opt, loss='mean_squared_error', metrics=['mae'])
models[i].summary()
if TRAIN:
models[i].fit(x=x, y=y, batch_size=20000, epochs=80, verbose=2, validation_data=(x_test, y_test))
return ensemble.GradientBoostingRegressor(max_depth=8, n_estimators=5)
if __name__ == '__main__':
datas = ds.read_dataset("mean_var_on_customers_from_tain.csv")
datas['Month'] = datas['Date']
datas['Month'] = datas['Month'].apply(lambda x: x.split("-")[1])
datas = imp.one_hot_numeric(datas, 'Month', 'Month_')
datas = imp.one_hot_numeric(datas, 'Region', 'Region_')
datas = preprocessing_utils.mean_cust_per_month_per_region(
datas, utils.get_frame_in_range(datas, 3, 2016, 12, 2017))
datas = preprocessing_utils.mean_cust_per_month_per_shop(
datas, utils.get_frame_in_range(datas, 3, 2016, 12, 2017))
datas = sb.SetBuilder(target='NumberOfCustomers', autoexclude=True, df=datas)\
.exclude('NumberOfSales', 'Month', 'Max_Humidity', 'Max_Sea_Level_PressurehPa', 'Max_TemperatureC',
'Max_VisibilityKm', 'Max_Wind_SpeedKm_h', 'Mean_Humidity', 'Mean_Sea_Level_PressurehPa',
'Mean_VisibilityKm', 'Mean_Wind_SpeedKm_h', 'Min_Dew_PointC',
'Min_Humidity', 'Min_Sea_Level_PressurehPa', 'Min_TemperatureC', 'Min_VisibilitykM')\
.build()
model = [
ridge, linear_model.LinearRegression, lasso, regtree, gradboostreg
]
final = ridge
n = len(model)
mods = []
modpreds = []
modpreds_t = []
for i in range(n):
print(i + 1)
x, y = datas.xtr, datas.ytr
mod = skc.LinearSklearn(1, model[i])
mod.train(x, y)
def excluded_feats():
return [
"Month", 'Max_Humidity', 'Max_TemperatureC', 'Max_VisibilityKm',
'Max_Wind_SpeedKm_h', 'Min_Dew_PointC', 'Min_Humidity',
'Min_TemperatureC', 'Min_VisibilitykM', 'NumberOfCustomers'
]
datas = d.read_dataset("final_sales_only_train.csv")
regions = d.to_numpy(datas[['Region']]).squeeze()
dates = d.to_numpy(datas[['Date']]).squeeze()
ids = d.to_numpy(datas[['StoreID']]).squeeze()
data = sb.SetBuilder(target='NumberOfSales', autoexclude=True, df=datas.copy(), split=(3, 2016, 2, 2018, 12, 2018, 12, 2018))\
.exclude_list(excluded_feats())\
.build()
it = 1
yy = []
models = []
for i in range(it):
bagx, bagy = data.xtr, data.ytr
dt = nn.MLPRegressor(hidden_layer_sizes=(400, 3),
activation='identity',
solver='adam',
batch_size=50000,
learning_rate='adaptive',
learning_rate_init=0.002,
max_iter=50,
shuffle=True,
das = preu.eliminate_IsOpen_zeros(das)
das = preu.mean_std_cust_per_shop_per_day(das, dfrom)
das = preu.add_avg_cust_per_shop(das, dfrom)
das = preu.add_std_cust_per_shop(das, dfrom)
das = preu.add_max_cust_per_shop(das, dfrom)
das = preu.add_min_cust_per_shop(das, dfrom)
das = preu.mean_cust_per_month_per_shop(das, dfrom)
das = preu.mean_cust_per_month_per_region(das, dfrom)
return das
if __name__ == '__main__':
split = [[(3, 2016, 12, 2016), (3, 2017, 2, 2018)], [(1, 2017, 2, 2017)]]
datas = build_cust_predictor_train_dataset(1, 2017, 2, 2017)
datas = sb.SetBuilder(target='NumberOfCustomers', autoexclude=True, df=datas, split=split)\
.exclude("Month", "NumberOfSales")\
.build()
n = 10
mods = []
for i in range(n):
print(i + 1)
x, y = datas.random_sampling(1.0)
mod = skc.LinearSklearn(1, model)
mod.train(x, y)
mods.append(mod)
p = mod.predict(datas.xtr).squeeze()
print("TRAIN R2: ", eval.r2_score(datas.ytr, p))
print("TEST R2: ", eval.r2_score(datas.yts, mod.predict(datas.xts)))
print("##########################")
preds = []
ds_name = "fully_preprocessed_ds.csv"
NMODELS = 3
SET_OF_MODELS_DIM = 1
models = []
datas = ds.read_dataset(ds_name)
datas = utils.get_frame_in_range(datas, 1, 2018, 2, 2018)
regions = ds.to_numpy(datas[['Region']]).squeeze()
dates = ds.to_numpy(datas[['Date']]).squeeze()
ids = ds.to_numpy(datas[['StoreID']]).squeeze()
def mod():
return skl.linear_model.Ridge(200)
for i in range(SET_OF_MODELS_DIM):
data = sb.SetBuilder(target='NumberOfSales', dataset=ds_name).build()
data.random_sampling(1.0)
model = skc.LinearSklearn(NMODELS, mod)
# Performs simple linear regression
model.train(data.xtr, data.ytr)
models.append(model)
data = sb.SetBuilder(target='NumberOfSales', dataset=ds_name).build()
ypred = np.zeros(shape=data.yts.shape[0])
for i in range(SET_OF_MODELS_DIM):
ypred += models[i].predict(data.xts)
ypred = ypred/SET_OF_MODELS_DIM
print('R2 = %s' % eval.evaluate(data.yts, ypred))
from sklearn import tree
import dataset.setbuilder as sb
import models.sklearn.evaluator as eval
import numpy as np
import models.sklearn.persistence as pr
print("Plain Decision regression tree without bagging")
# Build training & test sets
#
data = sb.SetBuilder(
target='NumberOfCustomers',
autoexclude=True,
dataset='best_for_customers.csv',
).exclude('NumberOfSales', 'Month').build()
# data = sb.SetBuilder(target='NumberOfSales', autoexclude=True, dataset='mean_var_on_cust_from_tain.csv').build()
# Performs simple linear regression
depth = 8
dtree = tree.DecisionTreeRegressor(max_depth=depth)
dtree.fit(data.xtr, data.ytr)
ypred = dtree.predict(data.xts)
pr.save_model(dtree, 'decision_tree_cust')
dtree = pr.load_model('decision_tree_cust')
ypred = dtree.predict(data.xts)
import dataset.setbuilder as sb
import models.sklearn.evaluator as eval
from sklearn import svm
data = sb.SetBuilder(target='NumberOfSales').build().random_sampling(
percentage=0.1)
clf = svm.SVR()
clf.fit(data.xtr, data.ytr.ravel())
ypred = clf.predict(data.xts)
print('R2 = %s' % eval.evaluate(data.yts, ypred))
import dataset.utility as dsutil
import dataset.dataset as ds
import seaborn as sea
import dataset.setbuilder as sb
from sklearn import linear_model
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import models.sklearn.evaluator as eval
from sklearn.preprocessing import PolynomialFeatures
data = sb.SetBuilder(target="NumberOfSales",
dataset="final_for_sales_train.csv",
autoexclude=False,
split=[[(3, 2016, 1, 2018)], [(3, 2016, 2, 2018)]
]).only('NearestCompetitor').build()
poly_degree = 2
# Performs simple linear regression
print("Linear regression started, polynomial degree = %s" % poly_degree)
poly = PolynomialFeatures(degree=poly_degree)
xtr_ = poly.fit_transform(data.xtr)
xts_ = poly.fit_transform(data.xts)
model = linear_model.LinearRegression()
model.fit(data.xtr, data.ytr)
print(eval.evaluate(data.ytr, model.predict(data.xtr)))
print(eval.evaluate(data.yts, model.predict(data.xts)))
