使用 sk-learn 根据字符串特征预测数值特征

Question

我正在尝试预测“ Full_Time_Home_Goals”列（功能）。我遵循了Kaggle 的例子。该代码适用于不同的维度，如我的示例（测试数据中的 419 行和训练数据中的 892 行）我收到的错误是：代码如下：回溯（最近一次调用最后一次）：文件“C:/Users/harsh/PycharmProjects/Kaggle-Machine Learning from Start to Finish with Scikit-Learn/EPL Predicting.py”，第 98 行，预测 = clf.predict(data_test. drop('Id', axis=1)) 文件“C:\Users\harsh\PycharmProjects\GitHub\venv\lib\site-packages\sklearn\ensemble_forest.py”，第 629 行，预测 ValueError：特征数量模型必须与输入匹配。模型 n_features 为 4，输入 n_features 为 2将代码更改为 后predictions = clf.predict(data_test.drop('Id', axis=1)) to predictions = clf.predict(X_test)，错误为： raise ValueError(msg) ValueError: array length 37921 does not match index length 380我该如何解决这个问题？我使用的数据集可以在这里找到

Answer 1

以下是您经过测试且完全有效的代码：

data_train = pd.read_csv(r"train.csv")

data_test = pd.read_csv(r"test.csv")

columns = ['Id', 'HomeTeam', 'AwayTeam', 'Full_Time_Home_Goals']

col = ['Id', 'HomeTeam', 'AwayTeam']

data_test = data_test[col]

data_train = data_train[columns]

data_train = data_train.dropna()

data_test = data_test.dropna()

data_train['Full_Time_Home_Goals'] = data_train['Full_Time_Home_Goals'].astype(int)

from sklearn import preprocessing

def encode_features(df_train, df_test):

    features = ['HomeTeam', 'AwayTeam']

    df_combined = pd.concat([df_train[features], df_test[features]])

    for feature in features:

        le = preprocessing.LabelEncoder()

        le = le.fit(df_combined[feature])

        df_train[feature] = le.transform(df_train[feature])

        df_test[feature] = le.transform(df_test[feature])

    return df_train, df_test

data_train, data_test = encode_features(data_train, data_test)

print(data_train.head())

print(data_test.head())

# X_all would contain all columns required for prediction and y_all would have that one columns we want to predict

y_all = data_train['Full_Time_Home_Goals']

X_all = data_train.drop(['Full_Time_Home_Goals'], axis=1)

from sklearn.model_selection import train_test_split

num_test = 0.20  # 80-20 split

X_train, X_test, y_train, y_test = train_test_split(X_all, y_all, test_size=num_test, random_state=23)

from sklearn.ensemble import RandomForestClassifier

from sklearn.metrics import make_scorer, accuracy_score

from sklearn.model_selection import GridSearchCV

# Using Random Forest and using parameters that we defined

clf = RandomForestClassifier()

parameters = {'n_estimators': [4, 6, 9],

              'max_features': ['log2', 'sqrt', 'auto'],

              'criterion': ['entropy', 'gini'],

              'max_depth': [2, 3, 5, 10],

              'min_samples_split': [2, 3, 5],

              'min_samples_leaf': [1, 5, 8]

              }

acc_scorer = make_scorer(accuracy_score)

grid_obj = GridSearchCV(clf, parameters, scoring=acc_scorer)

grid_obj = grid_obj.fit(X_train, y_train)

clf = grid_obj.best_estimator_

clf.fit(X_train, y_train)

predictions = clf.predict(X_test)

print(accuracy_score(y_test, predictions))

ids = data_test['Id']

predictions = clf.predict(data_test)

df_preds = pd.DataFrame({"id":ids, "predictions":predictions})

df_preds

   Id  HomeTeam  AwayTeam  Full_Time_Home_Goals

0   1        55       440                     3

1   2       158       493                     2

2   3       178       745                     1

3   4       185       410                     1

4   5       249        57                     2

       Id  HomeTeam  AwayTeam

0  190748       284        54

1  190749       124       441

2  190750       446        57

3  190751       185       637

4  190752       749       482

0.33213786556261704

id  predictions

0   190748  1

1   190749  1

2   190750  1

3   190751  1

4   190752  1

... ... ...

375 191123  1

376 191124  1

377 191125  1

378 191126  1

379 191127  1

380 rows × 2 columns