corazzon / kagglestruggle Goto Github PK

Hi corazzon.
Thanks. I'm not familiar with Machine learning. It helps me a lot to learn about Machine Learning.

Let me ask one.
You used RandomForestRegressor in Lesson 3 and Lesson 4 (Youtube video).

////////////////Lesson 3:
from sklearn.ensemble import RandomForestRegressor
max_depth_list = []
model = RandomForestRegressor(n_estimators=100,
n_jobs=-1,
random_state=0) ------------------------------------(1)
%time score = cross_val_score(model, X_train, y_train, cv=k_fold, scoring=rmsle_scorer)
score = score.mean()--------------------------------------------------(4)
model.fit(X_train, y_train) ----------------------------------------------(2)
predictions = model.predict(X_test) -----------------------------------(3)
print(predictions.shape)

//////////////////Lesson 4:
from sklearn.ensemble import RandomForestRegressor
rfModel = RandomForestRegressor(n_estimators=100) -----------------(1)
y_train_log = np.log1p(y_train) ------------------------------------------(*)
rfModel.fit(X_train, y_train_log) -----------------------------------------(2)
preds = rfModel.predict(X_train) ---------------------------------------(3)
score = rmsle(np.exp(y_train_log),np.exp(preds),False) -----------------(4)

In fact, you used the same model and algorithm in this part.
But I don't understand ().
When I remove () and use y_train, the output is "non".

Could you explain about this?
Thanks.
박천.