From ded2280772c0eca26ba81a147ab3af54d4d5f752 Mon Sep 17 00:00:00 2001
From: tangwei <tangwei12@baidu.com>
Date: Mon, 18 May 2020 10:46:38 +0800
Subject: [PATCH] rebase user

---
 .../classification/config.yaml                |  2 +-
 .../classification/model.py                   | 22 +++++++++++--------
 models/contentunderstanding/readme.md         | 11 +++++++++-
 3 files changed, 24 insertions(+), 11 deletions(-)

diff --git a/models/contentunderstanding/classification/config.yaml b/models/contentunderstanding/classification/config.yaml
index 9bdfffe1..4cf4d1bb 100644
--- a/models/contentunderstanding/classification/config.yaml
+++ b/models/contentunderstanding/classification/config.yaml
@@ -18,7 +18,7 @@ train:
     strategy: "async"
 
   epochs: 10
-  workspace: "paddlerec.models.rank.text_classification"
+  workspace: "paddlerec.models.contentunderstandin.classification"
 
   reader:
     batch_size: 5
diff --git a/models/contentunderstanding/classification/model.py b/models/contentunderstanding/classification/model.py
index 833e9445..6254199c 100644
--- a/models/contentunderstanding/classification/model.py
+++ b/models/contentunderstanding/classification/model.py
@@ -26,29 +26,35 @@ import paddle.fluid.layers.control_flow as cf
 class Model(ModelBase):
     def __init__(self, config):
         ModelBase.__init__(self, config)
+        self.dict_dim = 100
+        self.max_len = 10
+        self.cnn_dim = 32
+        self.cnn_filter_size = 128
+        self.emb_dim = 8
+        self.hid_dim = 128
+        self.class_dim = 2
 
     def train_net(self):
         """ network definition """
        
-        data = fluid.data(name="input", shape=[None, max_len], dtype='int64')
+        data = fluid.data(name="input", shape=[None, self.max_len], dtype='int64')
         label = fluid.data(name="label", shape=[None, 1], dtype='int64')
         seq_len = fluid.data(name="seq_len", shape=[None], dtype='int64')
         # embedding layer
-        emb = fluid.embedding(input=data, size=[dict_dim, emb_dim])
-        emb = fluid.layers.sequence_unpad(emb, length=seq_len)
+        emb = fluid.embedding(input=data, size=[self.dict_dim, self.emb_dim])
+        emb = fluid.layers.sequence_unpad(emb, length=self.seq_len)
         # convolution layer
         conv = fluid.nets.sequence_conv_pool(
             input=emb,
-            num_filters=cnn_dim,
-            filter_size=cnn_filter_size,
+            num_filters=self.cnn_dim,
+            filter_size=self.cnn_filter_size,
             act="tanh",
             pool_type="max")
 
         # full connect layer
         fc_1 = fluid.layers.fc(input=[conv], size=hid_dim)
         # softmax layer
-        prediction = fluid.layers.fc(input=[fc_1], size=class_dim, act="softmax")
-
+        prediction = fluid.layers.fc(input=[fc_1], size=self.class_dim, act="softmax")
         cost = fluid.layers.cross_entropy(input=prediction, label=label)
         avg_cost = fluid.layers.mean(x=cost)
         acc = fluid.layers.accuracy(input=prediction, label=label) 
@@ -65,9 +71,7 @@ class Model(ModelBase):
     def optimizer(self):
         learning_rate = 0.01
         sgd_optimizer = fluid.optimizer.Adagrad(learning_rate=learning_rate)
-
         return sgd_optimizer
 
-
     def infer_net(self, parameter_list):
         self.train_net()
diff --git a/models/contentunderstanding/readme.md b/models/contentunderstanding/readme.md
index ef3fd883..417fdde3 100644
--- a/models/contentunderstanding/readme.md
+++ b/models/contentunderstanding/readme.md
@@ -69,14 +69,23 @@ python text2paddle.py raw_big_train_data/ raw_big_test_data/ train_big_data test
 无
 
 ### 训练
+
+```
+python -m paddlerec.run -m paddlerec.models.rank.dnn -d cpu -e single
+```
+
 ### 预测
 
+```
+python -m paddlerec.run -m paddlerec.models.rank.dnn -d cpu -e single
+```
+
 ## 效果对比
 ### 模型效果 (测试)
 
 |       数据集        |       模型       |       loss        |       auc          |       acc         |       mae          |
 | :------------------: | :--------------------: | :---------: |:---------: | :---------: |:---------: |
-|       --        |       TagSpace       |       --        |       --          |       --          |       --          |
+|       ag news dataset        |       TagSpace       |       --        |       --          |       --          |       --          |
 |       --        |       Classification       |       --        |       --          |       --          |       --          |
 
 
-- 
GitLab