bug fix

.gitignore

@@ -1,5 +1,6 @@
 .idea
 .DS_Store
+*.py[cod]
 *.ipynb
 *ipynb*
 criteo_sample.txt

examples/ctr_predict.py

@@ -15,6 +15,7 @@
 from models.AFM import AFM
 from models.AutoInt import AutoInt
 from models.CCPM import CCPM
+from models.NFFM import NFFM
 from core.features import FeatureMetas
 
 if __name__ == "__main__":
@@ -24,6 +25,8 @@
 
     # Get columns' names
     sparse_features = list(data.columns)
+    sparse_features.remove('click')
+    sparse_features.remove('id')
     target = ['click']
 
     # Preprocess your data
@@ -37,14 +40,14 @@
     train_0 = train[train.click == 0]
     train_1 = train[train.click == 1]
     train = pd.concat([train_1, train_0[0:len(train_1)]])
-    train_model_input = {name: train[name] for name in sparse_features}
-    test_model_input = {name: test[name] for name in sparse_features}
+    train_model_input = {name: train[name].values for name in sparse_features}
+    test_model_input = {name: test[name].values for name in sparse_features}
 
     # Instantiate a FeatureMetas object, add your features' meta information to it
     feature_metas = FeatureMetas()
     for feat in sparse_features:
         feature_metas.add_sparse_feature(name=feat, one_hot_dim=data[feat].nunique(), embedding_dim=32)
-
+    # a warning need to be fixed see https://stackoverflow.com/questions/35892412/tensorflow-dense-gradient-explanation
     # Instantiate a model and compile it
     model = DeepFM(
         feature_metas=feature_metas,
@@ -60,7 +63,7 @@
     history = model.fit(x=train_model_input,
                         y=train[target].values,
                         batch_size=128,
-                        epochs=1,
+                        epochs=3,
                         verbose=2,
                         validation_split=0.2)
 

models/__init__.py

@@ -0,0 +1,15 @@
+
+from .AFM import AFM
+from .AutoInt import AutoInt
+from .CCPM import CCPM
+from .DCN import DCN
+from .DeepFM import DeepFM
+from .FGCNN import FGCNN
+from .FiBiNet import FiBiNet
+from .FNN import FNN
+from .MLR import MLR
+from .NFFM import NFFM
+from .NFM import NFM
+from .PNN import PNN
+from .WideAndDeep import WideAndDeep
+from .xDeepFM import xDeepFM