import numpy as np
import pandas as pd
from sklearn.metrics import f1_score
from sklearn.model_selection import KFold
from read_data import connect4
from data_label_num import data_label_num, combination
from decision_tree import decision_tree
if __name__ == '__main__':
dataset = connect4(numerical=True, one_hot=False)
model_data, model_label, train_dataset_num = data_label_num(dataset.train_x, dataset.train_y)
test_data, test_label, _ = data_label_num(dataset.test_x, dataset.test_y)
macro_f1_score = []
kf = KFold(n_splits=5, shuffle=True, random_state=2020)
for train_index, valid_index in kf.split(train_dataset_num):
train_data = pd.DataFrame(train_dataset_num.copy()).drop(valid_index)
valid_data = pd.DataFrame(train_dataset_num.copy()).drop(train_index)
train_data_y = train_data.iloc[:, -1]
valid_data_y = valid_data.iloc[:, -1]
train_data_x = train_data.drop(axis=1, columns=42)
valid_data_x = valid_data.drop(axis=1, columns=42)
train_dataset = combination(train_data_x.values.tolist(), train_data_y.values.tolist())
dt = decision_tree(max_depth=6)
dt.train(train_dataset)
valid_dataset = combination(valid_data_x.values.tolist(), valid_data_y.values.tolist())
result = "win"
elif feature[20] == "o":
if feature[21] == "x":
result = "win"
else:
result = "loss"
else:
if feature[19] == "o":
result = "loss"
else:
result = "win"
return result
if __name__ == "__main__":
dataset = connect4()
data = dataset.get_all_data()
# 计算if-else在数据集上的正确率
right = 0
res_ifelse = []
for i in range(len(data[0])):
feature = data[0][i]
pred = predict(feature)
res_ifelse.append(pred)
if pred == data[1][i]:
right += 1
print("if-else正确率：", right / len(data[0]))
# 计算决策树在数据集上的正确率
right = 0