from sklearn.model_selection import train_test_split from numpy import ravel import pandas as pd import numpy as np from sklearn.metrics import accuracy_score from sklearn.metrics import cohen_kappa_score from sklearn.metrics import f1_score from sklearn.metrics import roc_auc_score from sklearn.svm import SVC ### Bayes: 0.50316455696202533, 0.49936708860759493 ### Log: .49786628733997157 ### lin svm: 0.49644381223328593 SVC(C = 7.5, kernel = 'linear', gamma = 'auto') ### SVC(C = .005, kernel = 'linear'), acc: 0.520410, roc: 0.519155, kappa: 0.038282, f1: 0.463038 data = classificationdata() xvars = list(data)[:-1] useset = data #useset, holdoutset = train_test_split(data, test_size = .1, random_state = 1108) #kernels = ['poly', 'rbf', 'sigmoid'] #Cs = [.01, .1, 1, 10] #gammas = [.001, .01, .05, .1] Cs = [.005, .01, .05] end = len(Cs) * 9 at = 0 parameterscores = pd.DataFrame()
import numpy as np
import pandas as pd
from classificationdata import classificationdata
from sklearn.linear_model import LogisticRegression
import matplotlib.pyplot as plt
np.random.seed(42)
traindata = pd.read_csv('train_line_data.csv')
testdata = pd.read_csv('test_line_data.csv')
features = list(traindata)[1:]
x_feat = features[:-2]
train_x = traindata[x_feat]
train_y = traindata['y']
train_juice = traindata['juice']
testdata = classificationdata('test')
test_x = testdata[x_feat]
test_y = testdata['y']
test_juice = testdata['juice']
model = LogisticRegression()
model.fit(train_x, train_y)
confidence_threshold = .56
bank = 1000
bankhistory = []
pred = None
inputdata = None
prediction = None
gamejuice = None