I have been conducting a learning algorithm using Decision Tree Classifier in Python.
from sklearn.tree import DecisionTreeClassifier
clf = DecisionTreeClassifier()
clf.fit(train, train_label)
predicted_label = clf.predict(test)
The Decision Tree Classifier accepts training labels from a large text file. I want to run the program without performing again the training process. How will I do it in Python? How will I include a precompiled learning model and used it for testing in another program? Is precompiled python files does have anything to do with it?
After training your model, you could save your model for future use for avoiding the process of training.
import pickle
model.fit(X,y)
saved_model = pickle.dump(model,open('saved_model.sav', 'wb'))#save your model
.
.
.
model = pickle.loads(open('saved_model.sav', 'rb'))#get your model from saved model file
model.predict(X[0:1])#use without training