remove irrelevant comments

Author: lazyprogrammer

airline/ann.py

@@ -156,24 +156,15 @@ def predict(self, X):
         X[:,d] = series[d:d+n]
     Y = series[D:D+n]
 
-    # print "X.shape:", X.shape
-    # print "Y.shape:", Y.shape
-
     print "series length:", n
     Xtrain = X[:n/2]
     Ytrain = Y[:n/2]
     Xtest = X[n/2:]
     Ytest = Y[n/2:]
 
-    # print "Xtrain.shape:", Xtrain.shape
-    # print "Ytrain.shape:", Ytrain.shape
-
     model = ANN([200])
     model.fit(Xtrain, Ytrain, activation=T.tanh)
     print "train score:", model.score(Xtrain, Ytrain)
-
-    # print "Xtest.shape:", Xtest.shape
-    # print "Ytest.shape:", Ytest.shape
     print "test score:", model.score(Xtest, Ytest)
 
     # plot the prediction with true values

airline/lr.py

@@ -22,11 +22,6 @@
 # let's try with only the time series itself
 series = df.num_passengers.as_matrix()
 
-# def myr2(T, Y, Ym):
-#     sse = (T - Y).dot(T - Y)
-#     sst = (T - Ym).dot(T - Ym)
-#     return 1 - sse / sst
-
 # let's see if we can use D past values to predict the next value
 N = len(series)
 for D in (2,3,4,5,6,7):
@@ -36,33 +31,17 @@
         X[:,d] = series[d:d+n]
     Y = series[D:D+n]
 
-    # print "X.shape:", X.shape
-    # print "Y.shape:", Y.shape
-
     print "series length:", n
     Xtrain = X[:n/2]
     Ytrain = Y[:n/2]
     Xtest = X[n/2:]
     Ytest = Y[n/2:]
 
-    # print "Xtrain.shape:", Xtrain.shape
-    # print "Ytrain.shape:", Ytrain.shape
-
     model = LinearRegression()
     model.fit(Xtrain, Ytrain)
     print "train score:", model.score(Xtrain, Ytrain)
-
-    # Ytrain_mean = Ytrain.mean()
-    # print "myr2 train:", myr2(Ytrain, model.predict(Xtrain), Ytrain_mean)
-
-    # print "Xtest.shape:", Xtest.shape
-    # print "Ytest.shape:", Ytest.shape
     print "test score:", model.score(Xtest, Ytest)
 
-    # Ytest_mean = Ytest.mean()
-    # print "myr2 test w/ Ytrain mean:", myr2(Ytest, model.predict(Xtest), Ytrain_mean)
-    # print "myr2 test w/ Ytest mean:", myr2(Ytest, model.predict(Xtest), Ytest_mean) # this is the one score uses
-
     # plot the prediction with true values
     plt.plot(series)
 

airline/rnn.py

@@ -86,10 +86,6 @@ def fit(self, X, Y, activation=T.tanh, learning_rate=10e-2, mu=0.5, reg=0, epoch
 
                 c = self.train_op(learning_rate, X[j], Y[j])
                 cost += c
-
-                # if (j+1) % 200 == 0:
-                #     sys.stdout.write("j/N: %d/%d cost so far: %f\r" % (j, N, cost))
-                #     sys.stdout.flush()
             if i % 10 == 0:
                 print "i:", i, "cost:", cost, "time for epoch:", (datetime.now() - t0)
             if (i+1) % 500 == 0:
@@ -145,9 +141,6 @@ def predict(self, X):
         X[:,d] = series[d:d+n]
     Y = series[D:D+n]
 
-    # print "X.shape:", X.shape
-    # print "Y.shape:", Y.shape
-
     print "series length:", n
     Xtrain = X[:n/2]
     Ytrain = Y[:n/2]
@@ -159,15 +152,9 @@ def predict(self, X):
     Ntest = len(Xtest)
     Xtest = Xtest.reshape(Ntest, D, 1)
 
-    # print "Xtrain.shape:", Xtrain.shape
-    # print "Ytrain.shape:", Ytrain.shape
-
     model = RNN([50])
     model.fit(Xtrain, Ytrain, activation=T.tanh)
     print "train score:", model.score(Xtrain, Ytrain)
-
-    # print "Xtest.shape:", Xtest.shape
-    # print "Ytest.shape:", Ytest.shape
     print "test score:", model.score(Xtest, Ytest)
 
     # plot the prediction with true values