first run out the result

Author: moneyDboat

models.py

@@ -60,10 +60,11 @@ def ConvMF(res_dir, train_user, train_item, valid_user, test_user,
 
     endure_count = 5
     count = 0
+    #max_iter is 50
     for iteration in range(max_iter):
         loss = 0
         tic = time.time()
-        print ("%d iteration\t(patience: %d)" % (iteration, count))
+        print("%d iteration\t(patience: %d)" % (iteration, count))
 
         VV = b * (V.T.dot(V)) + lambda_u * np.eye(dimension)
         sub_loss = np.zeros(num_user)
@@ -130,6 +131,7 @@ def ConvMF(res_dir, train_user, train_item, valid_user, test_user,
         f1.write("Loss: %.5f Elpased: %.4fs Converge: %.6f Tr: %.5f Val: %.5f Te: %.5f\n" % (
             loss, elapsed, converge, tr_eval, val_eval, te_eval))
 
+        # endure_count = 5
         if count == endure_count:
             break
 

test/ml-1m/result/1_100_200/CNN_weights.hdf5

@@ -0,0 +1,23 @@
+Loss: -2436327.67233 Elpased: 94.5476s Converge: 243632767233256604059592609928863639373774189679626682368.000000 Tr: 2.09378 Val: 2.09782 Te: 2.10208
+Loss: -1341523.05926 Elpased: 90.4797s Converge: 0.449367 Tr: 1.11262 Val: 1.12680 Te: 1.13408
+Loss: -848435.39393 Elpased: 90.7476s Converge: 0.367558 Tr: 0.91739 Val: 0.94267 Te: 0.94872
+Loss: -680144.23749 Elpased: 91.0255s Converge: 0.198355 Tr: 0.87135 Val: 0.90376 Te: 0.90901
+Loss: -598651.87307 Elpased: 91.3713s Converge: 0.119816 Tr: 0.85217 Val: 0.89071 Te: 0.89528
+Loss: -554200.64811 Elpased: 92.1919s Converge: 0.074252 Tr: 0.83663 Val: 0.88113 Te: 0.88541
+Loss: -519567.88498 Elpased: 100.5679s Converge: 0.062491 Tr: 0.82112 Val: 0.87114 Te: 0.87512
+Loss: -490528.84064 Elpased: 96.1168s Converge: 0.055891 Tr: 0.81409 Val: 0.86824 Te: 0.87230
+Loss: -468264.29304 Elpased: 94.4288s Converge: 0.045389 Tr: 0.80505 Val: 0.86364 Te: 0.86770
+Loss: -448083.33829 Elpased: 93.9075s Converge: 0.043097 Tr: 0.79767 Val: 0.86036 Te: 0.86398
+Loss: -432335.05202 Elpased: 92.3625s Converge: 0.035146 Tr: 0.79214 Val: 0.85844 Te: 0.86197
+Loss: -418138.53040 Elpased: 92.9699s Converge: 0.032837 Tr: 0.78939 Val: 0.85878 Te: 0.86190
+Loss: -406712.88303 Elpased: 91.3156s Converge: 0.027325 Tr: 0.78511 Val: 0.85773 Te: 0.86124
+Loss: -395646.63541 Elpased: 92.2882s Converge: 0.027209 Tr: 0.77927 Val: 0.85483 Te: 0.85797
+Loss: -385836.61482 Elpased: 92.1456s Converge: 0.024795 Tr: 0.77652 Val: 0.85481 Te: 0.85875
+Loss: -377371.94193 Elpased: 96.3971s Converge: 0.021938 Tr: 0.77206 Val: 0.85292 Te: 0.85693
+Loss: -370309.21071 Elpased: 90.6924s Converge: 0.018716 Tr: 0.76959 Val: 0.85311 Te: 0.85713
+Loss: -363733.96550 Elpased: 90.8471s Converge: 0.017756 Tr: 0.76531 Val: 0.85148 Te: 0.85529
+Loss: -358078.32508 Elpased: 90.3065s Converge: 0.015549 Tr: 0.76462 Val: 0.85251 Te: 0.85634
+Loss: -351743.24889 Elpased: 90.2835s Converge: 0.017692 Tr: 0.76068 Val: 0.85078 Te: 0.85441
+Loss: -345406.45162 Elpased: 90.3411s Converge: 0.018015 Tr: 0.75794 Val: 0.84983 Te: 0.85338
+Loss: -339322.08225 Elpased: 95.5237s Converge: 0.017615 Tr: 0.75560 Val: 0.84986 Te: 0.85308
+Loss: -334450.03947 Elpased: 96.1664s Converge: 0.014358 Tr: 0.75467 Val: 0.85177 Te: 0.85479

test/ml-1m/result/1_100_200/U.dat

@@ -5,7 +5,7 @@
 '''
 import numpy as np
 
-
+# calculate RMSE
 def eval_RMSE(R, U, V, TS):
     num_user = U.shape[0]
     sub_rmse = np.zeros(num_user)