], 1)
train_dataset = a.sample(frac=0.85, random_state=16)
test_dataset = a.drop(train_dataset.index)
train_labels = train_dataset.pop('Result')
test_labels = test_dataset.pop('Result')
clf = LogisticRegression(n_jobs=-1)
train_dataset = preprocessing.normalize(train_dataset)
test_dataset = preprocessing.normalize(test_dataset)
clf.fit(train_dataset, train_labels)
joblib.dump(clf, 'Logistic.joblib')
acc = clf.score(test_dataset, test_labels)
preds = clf.predict(test_dataset)
print(f.acc(preds, test_labels))
zeros, ones = 0, 0
for pred in preds:
if round(pred) == 1:
ones += 1
else:
zeros += 1
print('lenght of test:', len(preds))
print('0s:', zeros / len(preds))
print('1s:', ones / len(preds))
if abs(corr[x]) < 0.07:
del2.append(x)
data=data.drop(del2,1)
"""
clf = MLPRegressor(activation='logistic',random_state=1,max_iter=500)
Y=data.pop('Result')
X=data
accs=[]
for rs in range(10):
x_train,x_test,y_train,y_test = train_test_split(X, Y, test_size=0.2, random_state=rs)
clf.fit(x_train,y_train)
preds=clf.predict(x_test)
#print('zeros:',f.get0and1(preds))
accs.append(f.acc(preds,y_test))
print(sum(accs)/len(accs))
#joblib.dump(clf,'regression_linear.joblib')
"""
games=pd.read_csv(path2data+'games.csv')
df2log=pd.DataFrame()
df2log['home']=games['home']
df2log['away']=games['away']
df2log['date']=games['date']
# predict today's games
c2_avg=['PTS', 'FGM', 'FGA','FG%', '3PM', '3PA', '3P%',
'FTM', 'FTA', 'FT%', 'OREB', 'DREB', 'REB',
'AST', 'TOV', 'STL', 'BLK', 'PF', '+/-']
data=pd.read_csv(path2data+'train.csv')
data=data.dropna()
data=data.drop(['Team_home','Match Up_home','Game Date_home','Team_away',
'Match Up_away','Game Date_away','MIN_home','MIN_away',
'W/L_home','W/L_away'],1)
clf = ExtraTreesRegressor(n_estimators=1000, random_state=11,n_jobs=-1)
# split data into train and test sets
Y=data.pop('Result')
X=data
x_train,x_test,y_train,y_test = train_test_split(X, Y, test_size=0.01, random_state=1)
clf.fit(x_train,y_train)
preds=clf.predict(x_test)
print('test:',f.acc(preds,y_test))
#print('zeros:',f.get0and1(preds))
#joblib.dump(clf,'regression_linear.joblib')
games=pd.read_csv(path2data+'games.csv')
df2log=pd.DataFrame()
df2log['home']=games['home']
df2log['away']=games['away']
df2log['date']=games['date']
# predict today's games
c2_avg=['PTS', 'FGM', 'FGA','FG%', '3PM', '3PA', '3P%',
'FTM', 'FTA', 'FT%', 'OREB', 'DREB', 'REB',
'AST', 'TOV', 'STL', 'BLK', 'PF', '+/-']
preds=[]
# split data into train and test sets
X_train, X_test, y_train, y_test = train_test_split(X,
Y,
test_size=0.01,
random_state=7)
# fit model no training data
model = xgb.XGBRegressor()
model.fit(X_train, y_train)
# make predictions for test data
y_pred = model.predict(X_test)
print('zeros:', f.get0and1(y_pred))
# evaluate predictions
accuracy = f.acc(y_test, y_pred)
print("Accuracy: %.2f%%" % (accuracy * 100.0))
preds = model.predict
games = pd.read_csv(path2data + 'games.csv')
df2log = pd.DataFrame()
df2log['home'] = games['home']
df2log['away'] = games['away']
df2log['date'] = games['date']
# predict today's games
c2_avg = [
'PTS', 'FGM', 'FGA', 'FG%', '3PM', '3PA', '3P%', 'FTM', 'FTA', 'FT%',
'OREB', 'DREB', 'REB', 'AST', 'TOV', 'STL', 'BLK', 'PF', '+/-'
]