update event_handler

08.recommender_system/README.en.md

@@ -300,10 +300,10 @@ reader=paddle.batch(
         batch_size=256)
 ```
 
-`feeding` is devoted to specifying the correspondence between each yield record and `paddle.layer.data`. For instance, the first column of data generated by `movielens.train` corresponds to `user_id` feature.
+`reader_dict` is devoted to specifying the correspondence between each yield record and `paddle.layer.data`. For instance, the first column of data generated by `movielens.train` corresponds to `user_id` feature.
 
 ```python
-feeding = {
+reader_dict = {
     'user_id': 0,
     'gender_id': 1,
     'age_id': 2,
@@ -315,7 +315,15 @@ feeding = {
 }
 ```
 
-Callback function `event_handler` will be called during training when a pre-defined event happens.
+Callback function `event_handler` and  `event_handler_plot` will be called during training when a pre-defined event happens.
+
+```python
+def event_handler(event):
+    if isinstance(event, paddle.event.EndIteration):
+        if event.batch_id % 100 == 0:
+            print "Pass %d Batch %d Cost %.2f" % (
+                event.pass_id, event.batch_id, event.cost)
+```
 
 ```python
 step=0
@@ -323,7 +331,7 @@ step=0
 train_costs=[],[]
 test_costs=[],[]
 
-def event_handler(event):
+def event_handler_plot(event):
     global step
     global train_costs
     global test_costs
@@ -354,9 +362,9 @@ Finally, we can invoke `trainer.train` to start training:
 ```python
 trainer.train(
     reader=reader,
-    event_handler=event_handler,
-    feeding=feeding,
-    num_passes=200)
+    event_handler=event_handler_plot,
+    feeding=reader_dict,
+    num_passes=2)
 ```
 
 ## Conclusion

08.recommender_system/README.md

@@ -325,17 +325,8 @@ trainer = paddle.trainer.SGD(cost=cost, parameters=parameters,
 
 例如，这里的reader_dict表示的是，对于数据层 `user_id`，使用了reader中每一条数据的第0个元素。`gender_id`数据层使用了第1个元素。以此类推。
 
-训练过程是完全自动的。我们可以使用event_handler来观察训练过程，或进行测试等。这里我们在event_handler里面绘制了训练误差曲线和测试误差曲线。并且保存了模型。
-
-
 ```python
-%matplotlib inline
-
-import matplotlib.pyplot as plt
-from IPython import display
-import cPickle
-
-feeding = {
+reader_dict = {
     'user_id': 0,
     'gender_id': 1,
     'age_id': 2,
@@ -345,13 +336,31 @@ feeding = {
     'movie_title': 6,
     'score': 7
 }
+```
+
+训练过程是完全自动的。我们可以使用event_handler与event_handler_plot来观察训练过程，或进行测试等。这里我们在event_handler_plot里面绘制了训练误差曲线和测试误差曲线。并且保存了模型。
+
+```python
+def event_handler(event):
+    if isinstance(event, paddle.event.EndIteration):
+        if event.batch_id % 100 == 0:
+            print "Pass %d Batch %d Cost %.2f" % (
+                event.pass_id, event.batch_id, event.cost)
+```
+
+```python
+%matplotlib inline
+
+import matplotlib.pyplot as plt
+from IPython import display
+import cPickle
 
 step=0
 
 train_costs=[],[]
 test_costs=[],[]
 
-def event_handler(event):
+def event_handler_plot(event):
     global step
     global train_costs
     global test_costs
@@ -381,8 +390,8 @@ trainer.train(
             paddle.reader.shuffle(
             paddle.dataset.movielens.train(), buf_size=8192),
                             batch_size=256),
-    event_handler=event_handler,
-    feeding=feeding,
+    event_handler=event_handler_plot,
+    feeding=reader_dict,
     num_passes=2)
 ```
 

08.recommender_system/index.en.html

@@ -342,10 +342,10 @@ reader=paddle.batch(
         batch_size=256)
 ```
 
-`feeding` is devoted to specifying the correspondence between each yield record and `paddle.layer.data`. For instance, the first column of data generated by `movielens.train` corresponds to `user_id` feature.
+`reader_dict` is devoted to specifying the correspondence between each yield record and `paddle.layer.data`. For instance, the first column of data generated by `movielens.train` corresponds to `user_id` feature.
 
 ```python
-feeding = {
+reader_dict = {
     'user_id': 0,
     'gender_id': 1,
     'age_id': 2,
@@ -357,7 +357,15 @@ feeding = {
 }
 ```
 
-Callback function `event_handler` will be called during training when a pre-defined event happens.
+Callback function `event_handler` and  `event_handler_plot` will be called during training when a pre-defined event happens.
+
+```python
+def event_handler(event):
+    if isinstance(event, paddle.event.EndIteration):
+        if event.batch_id % 100 == 0:
+            print "Pass %d Batch %d Cost %.2f" % (
+                event.pass_id, event.batch_id, event.cost)
+```
 
 ```python
 step=0
@@ -365,7 +373,7 @@ step=0
 train_costs=[],[]
 test_costs=[],[]
 
-def event_handler(event):
+def event_handler_plot(event):
     global step
     global train_costs
     global test_costs
@@ -396,9 +404,9 @@ Finally, we can invoke `trainer.train` to start training:
 ```python
 trainer.train(
     reader=reader,
-    event_handler=event_handler,
-    feeding=feeding,
-    num_passes=200)
+    event_handler=event_handler_plot,
+    feeding=reader_dict,
+    num_passes=2)
 ```
 
 ## Conclusion

08.recommender_system/index.html

@@ -367,17 +367,8 @@ trainer = paddle.trainer.SGD(cost=cost, parameters=parameters,
 
 例如，这里的reader_dict表示的是，对于数据层 `user_id`，使用了reader中每一条数据的第0个元素。`gender_id`数据层使用了第1个元素。以此类推。
 
-训练过程是完全自动的。我们可以使用event_handler来观察训练过程，或进行测试等。这里我们在event_handler里面绘制了训练误差曲线和测试误差曲线。并且保存了模型。
-
-
 ```python
-%matplotlib inline
-
-import matplotlib.pyplot as plt
-from IPython import display
-import cPickle
-
-feeding = {
+reader_dict = {
     'user_id': 0,
     'gender_id': 1,
     'age_id': 2,
@@ -387,13 +378,31 @@ feeding = {
     'movie_title': 6,
     'score': 7
 }
+```
+
+训练过程是完全自动的。我们可以使用event_handler与event_handler_plot来观察训练过程，或进行测试等。这里我们在event_handler_plot里面绘制了训练误差曲线和测试误差曲线。并且保存了模型。
+
+```python
+def event_handler(event):
+    if isinstance(event, paddle.event.EndIteration):
+        if event.batch_id % 100 == 0:
+            print "Pass %d Batch %d Cost %.2f" % (
+                event.pass_id, event.batch_id, event.cost)
+```
+
+```python
+%matplotlib inline
+
+import matplotlib.pyplot as plt
+from IPython import display
+import cPickle
 
 step=0
 
 train_costs=[],[]
 test_costs=[],[]
 
-def event_handler(event):
+def event_handler_plot(event):
     global step
     global train_costs
     global test_costs
@@ -423,8 +432,8 @@ trainer.train(
             paddle.reader.shuffle(
             paddle.dataset.movielens.train(), buf_size=8192),
                             batch_size=256),
-    event_handler=event_handler,
-    feeding=feeding,
+    event_handler=event_handler_plot,
+    feeding=reader_dict,
     num_passes=2)
 ```