From 189f986f072af798c14da7342bab4c4260cb6e69 Mon Sep 17 00:00:00 2001
From: yaoxuefeng <yaoxuefeng@baidu.com>
Date: Tue, 2 Jun 2020 12:16:17 +0800
Subject: [PATCH] add lr fix codes

---
 models/rank/deepfm/data/get_slot_data.py              | 6 ++++--
 models/rank/logistic_regression/data/get_slot_data.py | 9 +++++++--
 models/rank/logistic_regression/data/preprocess.py    | 1 -
 3 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/models/rank/deepfm/data/get_slot_data.py b/models/rank/deepfm/data/get_slot_data.py
index 6177c990..c03b2924 100755
--- a/models/rank/deepfm/data/get_slot_data.py
+++ b/models/rank/deepfm/data/get_slot_data.py
@@ -12,9 +12,11 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import yaml
+import yaml, os
+
 from paddlerec.core.reader import Reader
 from paddlerec.core.utils import envs
+import paddle.fluid.incubate.data_generator as dg
 try:
     import cPickle as pickle
 except ImportError:
@@ -44,7 +46,7 @@ class TrainReader(dg.MultiSlotDataGenerator):
         self.continuous_range_ = range(1, 14)
         self.categorical_range_ = range(14, 40)
         # load preprocessed feature dict 
-        self.feat_dict_name = "aid_data/feat_dict_10.pkl2"
+        self.feat_dict_name = "sample_data/feat_dict_10.pkl2"
         self.feat_dict_ = pickle.load(open(self.feat_dict_name, 'rb'))
 
     def _process_line(self, line):
diff --git a/models/rank/logistic_regression/data/get_slot_data.py b/models/rank/logistic_regression/data/get_slot_data.py
index 6177c990..799762ce 100755
--- a/models/rank/logistic_regression/data/get_slot_data.py
+++ b/models/rank/logistic_regression/data/get_slot_data.py
@@ -13,8 +13,12 @@
 # limitations under the License.
 
 import yaml
+import os
+
 from paddlerec.core.reader import Reader
 from paddlerec.core.utils import envs
+import paddle.fluid.incubate.data_generator as dg
+
 try:
     import cPickle as pickle
 except ImportError:
@@ -44,7 +48,7 @@ class TrainReader(dg.MultiSlotDataGenerator):
         self.continuous_range_ = range(1, 14)
         self.categorical_range_ = range(14, 40)
         # load preprocessed feature dict 
-        self.feat_dict_name = "aid_data/feat_dict_10.pkl2"
+        self.feat_dict_name = "sample_data/feat_dict_10.pkl2"
         self.feat_dict_ = pickle.load(open(self.feat_dict_name, 'rb'))
 
     def _process_line(self, line):
@@ -90,6 +94,7 @@ class TrainReader(dg.MultiSlotDataGenerator):
         return data_iter
 
 
-reader = TrainReader("../config.yaml")
+reader = TrainReader(
+    "../config.yaml")  # run this file in original folder to find config.yaml
 reader.init()
 reader.run_from_stdin()
diff --git a/models/rank/logistic_regression/data/preprocess.py b/models/rank/logistic_regression/data/preprocess.py
index 9da3bdc3..0eaed5eb 100755
--- a/models/rank/logistic_regression/data/preprocess.py
+++ b/models/rank/logistic_regression/data/preprocess.py
@@ -66,7 +66,6 @@ def get_feat_dict():
     categorical_range_ = range(14, 40)
 
     if not os.path.exists(dir_feat_dict_):
-        # print('generate a feature dict')
         # Count the number of occurrences of discrete features
         feat_cnt = Counter()
         with open('train.txt', 'r') as fin:
-- 
GitLab