RandomForest 기반 2022 MSI 분석

import warnings
warnings.filterwarnings(action='ignore')

import pandas as pd

spring_table = pd.read_csv('2022_Spring_data.csv')
msi_table = pd.read_csv('2022_MSI_Data.csv')

spring_table.shape, msi_table.shape

((201, 25), (62, 25))

spring_line = []
lines = ['TOP','JUNGLE','MID','ADC','SUPPORT']

for i in lines:
    data = spring_table[spring_table['Position'] == i]
    data = data.drop(['Player','Country','Position','Games'], axis=1)
    data = data.astype('float32')
    spring_line.append(data)

len(spring_line)

5

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import GridSearchCV

params = {
    'n_estimators' : [90,100,100],
    'max_depth' : [5,6,7],
    'min_samples_split' : [1,2,3],
    'min_samples_leaf' : [0,1,2]
}

result_table = []

for i in range(len(spring_line)):
    X = spring_line[i].drop(['Win rate'], axis=1).values
    y = (spring_line[i]['Win rate'].values / 0.25).astype('int')
    Forest_model = RandomForestClassifier(random_state=0, n_jobs=-1)
    Grid_model = GridSearchCV(estimator=Forest_model, param_grid=params, verbose=False, n_jobs=-1)
    Grid_model.fit(X, y)
    print(f"{lines[i]} Train Acc : {Grid_model.score(X, y)}")

    model_estimators = Grid_model.best_params_['n_estimators']
    model_max_depth = Grid_model.best_params_['max_depth']
    model_min_samples_split = Grid_model.best_params_['min_samples_split']
    model_min_samples_leaf = Grid_model.best_params_['min_samples_leaf']

    model = RandomForestClassifier(
        n_estimators = model_estimators,
        max_depth = model_max_depth,
        min_samples_split = model_min_samples_split,
        min_samples_leaf = model_min_samples_leaf
    )
    model.fit(X, y)
    print(f"{lines[i]} Model Acc : {model.score(X, y)}")
    result_table.append(model)

TOP Train Acc : 0.9761904761904762
TOP Model Acc : 1.0
JUNGLE Train Acc : 1.0
JUNGLE Model Acc : 1.0
MID Train Acc : 0.9736842105263158
MID Model Acc : 0.9473684210526315
ADC Train Acc : 0.9743589743589743
ADC Model Acc : 1.0
SUPPORT Train Acc : 1.0
SUPPORT Model Acc : 1.0

MinMax_table = pd.DataFrame()

for i in spring_table.columns[5:]:
    MinMax_table[i] = msi_table[i] / max(msi_table[i])

MinMax_table.head(3)

	KDA	Avg kills	Avg deaths	Avg assists	CSM	GPM	KP%	DMG%	DPM	VSPM	Avg WPM	Avg WCPM	Avg VWPM	GD@15	CSD@15	XPD@15	FB %	FB Victim	Penta Kills	Solo Kills
0	0.105882	0.216667	0.626866	0.15625	0.915789	0.743381	0.502793	0.606742	0.577953	0.219512	0.191388	0.367647	0.278689	-0.634301	-0.142857	-0.469976	0.167	0.000	0.0	0.0625
1	0.400000	0.883333	0.417910	0.26875	0.673684	0.800407	0.901676	0.570225	0.565354	0.271003	0.167464	0.294118	0.426230	0.379310	0.357143	0.326661	0.667	0.334	0.0	0.1875
2	0.105882	0.166667	0.552239	0.13750	0.789474	0.665988	0.435754	0.558989	0.478740	0.268293	0.239234	0.294118	0.311475	-0.860254	-1.071429	-0.387510	0.000	0.334	0.0	0.0000

from sklearn.preprocessing import MinMaxScaler

score = []

for i in range(msi_table.shape[0]):
    pos = msi_table['Position'].values[i]
    p = sum([i*j for i,j in zip(MinMax_table.values[i][5:], result_table[lines.index(pos)].feature_importances_)])
    score.append(p)

msi_table['Score'] = score
msi_table['Score'] = (MinMaxScaler().fit_transform([[i] for i in msi_table['Score']])*100).astype('int')
msi_table.head(3)

	Player	Country	Position	Games	Win rate	KDA	Avg kills	Avg deaths	Avg assists	CSM	...	Avg WCPM	Avg VWPM	GD@15	CSD@15	XPD@15	FB %	FB Victim	Penta Kills	Solo Kills	Score
0	5Kid	KR	ADC	6	16.7	0.9	1.3	4.2	2.5	8.7	...	0.25	0.17	-1398	-2	-587	16.7	0.0	0	1	49
1	Aegis	BR	JUNGLE	6	33.3	3.4	5.3	2.8	4.3	6.4	...	0.20	0.26	836	5	408	66.7	16.7	0	3	83
2	Aloned	CL	MID	6	16.7	0.9	1.0	3.7	2.2	7.5	...	0.20	0.19	-1896	-15	-484	0.0	16.7	0	0	30

3 rows × 26 columns

msi_table.to_csv('Result_MSI.csv', index=False)