how to load the Catboost model for predicting new data

[HazelSCUT]

I used the Catboost to train the model and saved the trained model by using the code (m.save_model('Catboost.dump')).

However, when I failed to load such a saved model for predicting the new data unless I trained the model again. I used the code (m.load_model('Catboost.dump')); however, the bug is name "m" is not defined.

The question is how to load such a model for predicting new data without training the model each time.

Thanks!

[AutoViML]

Hi @HazelSCUT

Thanks for testing out AutoViML. The instructions to do this correctly are just a click away in Stack Overflow.

# After you train the model using fit(), save like this - 
m.save_model('model_name')    # extension not required.

# And then, later load - 
from catboost import CatBoostClassifier
mx = CatBoostClassifier()      # parameters not required.
mx.load_model('model_name')

# Now, try predict().
mx.predict(test[preds])

Next time please do this 👍 https://stackoverflow.com/questions/51895761/how-to-correctly-load-pretrained-model-in-catboost-in-python Thanks AutoVIML