add l2 regularization to reduce the probability of fpe exception

demo/seqToseq/api_train_v2.py

@@ -1,3 +1,4 @@
+import sys
 import paddle.v2 as paddle
 
 
@@ -104,7 +105,9 @@ def main():
     parameters = paddle.parameters.create(cost)
 
     # define optimize method and trainer
-    optimizer = paddle.optimizer.Adam(learning_rate=1e-4)
+    optimizer = paddle.optimizer.Adam(
+        learning_rate=5e-5,
+        regularization=paddle.optimizer.L2Regularization(rate=1e-3))
     trainer = paddle.trainer.SGD(cost=cost,
                                  parameters=parameters,
                                  update_equation=optimizer)
@@ -125,8 +128,11 @@ def main():
     def event_handler(event):
         if isinstance(event, paddle.event.EndIteration):
             if event.batch_id % 10 == 0:
-                print "Pass %d, Batch %d, Cost %f, %s" % (
+                print "\nPass %d, Batch %d, Cost %f, %s" % (
                     event.pass_id, event.batch_id, event.cost, event.metrics)
+            else:
+                sys.stdout.write('.')
+                sys.stdout.flush()
 
     # start to train
     trainer.train(