您好，我在本教程中遇到了有关如何在某些库中使用python以及如何使用体育参考库来预测将来的NCAAB游戏的信息。我将发布代码以及文章。这似乎运作良好，但我认为这只是过去基于游戏的测试。我将如何使用它来预测特定团队的未来比赛？例如，在此日期，A队和B队之间的得分是多少？
我看到的问题是，只有在游戏结束后才能知道所使用的某些数据。该已知数据就是程序中用来预测得分的数据。
第一个实验：我尝试仅填充游戏发生之前我知道的数据，并使用fillna（0）填充其余数据为零。这是csv的外观：
date_team，away_assist_percentage，away_assists，away_block_percentage，away_blocks，away_defensive_rating，away_defensive_rebound_percentage，away_defensive_rebounds，away_effective_field_goal_percentage，away_field_goal_attempts，away_field_goal_percentage，away_field_goals，away_free_throw_attempt_rate，away_free_throw_attempts，away_free_throw_percentage，away_free_throws，away_losses，away_minutes_played，away_offensive_rating，away_offensive_rebound_percentage，away_offensive_rebounds，away_personal_fouls，AWAY_POINTS，away_steal_percentage ，away_steals，away_three_point_attempt_rate，away_three_point_field_goal_attempts，away_three_point_field_goal_percentage，away_three_point_field_goals，away_total_rebound_percentage，away_total_rebounds，away_true_shooting_percentage，away_turnover_percentage，away_turnovers，away_two_point_field_goal_attempts，away_two_point_field_goal_percentage，away_two_point_field_goals，away_win_percentage，away_wins，home_assist_percentage，home_assists，home_block_percentage，home_blocks，H ome_defensive_rating，home_defensive_rebound_percentage，home_defensive_rebounds，home_effective_field_goal_percentage，home_field_goal_attempts，home_field_goal_percentage，home_field_goals，home_free_throw_attempt_rate，home_free_throw_attempts，home_free_throw_percentage，home_free_throws，home_losses，home_minutes_played，home_offensive_rating，home_offensive_rebound_percentage，home_offensive_rebounds，home_personal_fouls，HOME_POINTS，home_steal_percentage，home_steals，home_three_point_attempt_rate，home_three_point_field_goal_attempts，home_three_point_field_goal_percentage，home_three_point_field_goals，home_total_rebound_percentage， home_total_rebounds，home_true_shooting_percentage，home_turnover_percentage，home_turnovers，home_two_point_field_goal_attempts，home_two_point_field_goal_percentage，home_two_point_field_goals，home_win_percentage，home_wins，pace 0,59,8,0， 0,0.7,7,42， 0,0，.1,1,0 代码的最后一行更改为： 打印（model.predict（final_trim）.astype（int），y_test）
“ final_trim”是正在预测的新csv。
结果根本不准确。我想念什么？
这是原始代码：
import pandas as pd
from sportsreference.ncaab.teams import Teams
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import train_test_split
FIELDS_TO_DROP = ['away_points','home_points','date','location','losing_abbr','losing_name','winner','winning_abbr','winning_name','home_ranking','away_ranking']
dataset = pd.DataFrame()
teams = Teams()
for team in teams:
dataset = pd.concat([dataset,team.schedule.dataframe_extended])
X = dataset.drop(FIELDS_TO_DROP,1).dropna().drop_duplicates()
y = dataset[['home_points','away_points']].values
X_train,X_test,y_train,y_test = train_test_split(X,y)
parameters = {'bootstrap': False,'min_samples_leaf': 3,'n_estimators': 50,'min_samples_split': 10,'max_features': 'sqrt','max_depth': 6}
model = RandomForestRegressor(**parameters)
model.fit(X_train,y_train)
print(model.predict(X_test).astype(int),y_test)
这是我从中得到的帖子： https://towardsdatascience.com/predict-college-basketball-scores-in-30-lines-of-python-148f6bd71894
谢谢！