AttributeError: 'DecisionTreeRegressor' object has no attribute 'tree_'

Question

everyone. I'm a noob in data science and I'm building my first ML model. But I keep getting this error,

'DecisionTreeRegressor' object has no attribute 'tree_'

This is my code below.

df = pd.read_csv('/Users/mog/Downloads/train_data.csv')

from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 0.3, random_state = 1)

df.model = DecisionTreeRegressor()
df.model.fit(x_train, y_train)

from sklearn.metrics import accuracy_score

print('Training score...', accuracy_score(y_train, df_model.predict(x_train)))
print('Test score...', accuracy_score(y_test, df_model.predict(x_test)))

Please help.

Answer 1

The problem is that you are trying to estimate a classification metric for a regression model.

So, depending on your target you have two options:

1. If it is categorical you need to use DecisionTreeClassifier instead of DecisionTreeRegressor

2. If it is continuous, you need to change the metric accuracy_score for example for r2_score

I also noticed you named your model df.model but when predicting this is named df_model. So I recommend to change both to df_model

Hope it helps!