diff --git a/README.md b/README.md
index 78695fd..e20d718 100644
--- a/README.md
+++ b/README.md
@@ -30,7 +30,7 @@ train = [
 ]
 v = DictVectorizer()
 X = v.fit_transform(train)
-print X.toarray()
+print(X.toarray())
 [[ 19.   0.   0.   0.   1.   1.   0.   0.   0.]
  [ 33.   0.   0.   1.   0.   0.   1.   0.   0.]
  [ 55.   0.   1.   0.   0.   0.   0.   1.   0.]
@@ -102,7 +102,7 @@ Training RMSE: 0.43527
 # Evaluate
 preds = fm.predict(X_test)
 from sklearn.metrics import mean_squared_error
-print "FM RMSE: %.4f" % mean_squared_error(y_test,preds)
+print("FM RMSE: %.4f" % mean_squared_error(y_test,preds))
 FM RMSE: 0.9253
 
 ```
\ No newline at end of file
diff --git a/pylibfm.py b/pylibfm.py
index 48d6f41..ef22277 100644
--- a/pylibfm.py
+++ b/pylibfm.py
@@ -1,7 +1,9 @@
+import random
+
 import numpy as np
 from sklearn import cross_validation
-import random
-from pyfm_fast import FM_fast, CSRDataset
+
+from pyfm_fast import CSRDataset, FM_fast
 
 LEARNING_RATE_TYPES = {"optimal": 0, "invscaling": 1, "constant": 2}
 TASKS = {"regression": 0, "classification" : 1}
@@ -172,7 +174,7 @@ def fit(self, X, y):
 
         # use sklearn to create a validation dataset for lambda updates
         if self.verbose == True:
-            print "Creating validation dataset of %.2f of training for adaptive regularization" % self.validation_size
+            print("Creating validation dataset of %.2f of training for adaptive regularization" % self.validation_size)
         X_train, validation, train_labels, validation_labels = cross_validation.train_test_split(
             X, y, test_size=self.validation_size)
         self.num_attribute = X_train.shape[1]
@@ -212,7 +214,7 @@ def fit(self, X, y):
         # report epoch information
         if self.verbose == True:
             print("-- Epoch %d" % (epoch + 1))
-            print "Train RMSE: %.5f" % (self.sumloss / self.count)
+            print("Train RMSE: %.5f" % (self.sumloss / self.count))
 
     def predict(self, X):
         """Predict using the factorization machine
@@ -238,4 +240,3 @@ def _make_dataset(X, y_i):
     sample_weight = np.ones(X.shape[0], dtype=np.float64, order='C') # ignore sample weight for the moment
     dataset = CSRDataset(X.data, X.indptr, X.indices, y_i, sample_weight)
     return dataset
-