美购精排模型

mlp/train.py

@@ -75,17 +75,21 @@ def getTrainColumns(train_columns,data_vocab):
     # 离散特征
     for feature in train_columns:
         if data_vocab.get(feature):
-            if feature.startswith("userRatedHistory") or feature.count("__") > 0 or feature in embedding_columns:
-                cat_col = tf.feature_column.categorical_column_with_vocabulary_list(key=feature, vocabulary_list=data_vocab[feature])
-                col = tf.feature_column.embedding_column(cat_col, 10)
-                columns.append(col)
-                dataColumns.append(feature)
-
-            elif feature in one_hot_columns or feature.count("Bucket") > 0:
-                cat_col = tf.feature_column.categorical_column_with_vocabulary_list(key=feature, vocabulary_list=data_vocab[feature])
-                col = tf.feature_column.indicator_column(cat_col)
-                columns.append(col)
-                dataColumns.append(feature)
+            cat_col = tf.feature_column.categorical_column_with_vocabulary_list(key=feature,vocabulary_list=data_vocab[feature])
+            col = tf.feature_column.embedding_column(cat_col, 10)
+            columns.append(col)
+            dataColumns.append(feature)
+            # if feature.startswith("userRatedHistory") or feature.count("__") > 0 or feature in embedding_columns:
+            #     cat_col = tf.feature_column.categorical_column_with_vocabulary_list(key=feature, vocabulary_list=data_vocab[feature])
+            #     col = tf.feature_column.embedding_column(cat_col, 10)
+            #     columns.append(col)
+            #     dataColumns.append(feature)
+            #
+            # elif feature in one_hot_columns or feature.count("Bucket") > 0:
+            #     cat_col = tf.feature_column.categorical_column_with_vocabulary_list(key=feature, vocabulary_list=data_vocab[feature])
+            #     col = tf.feature_column.indicator_column(cat_col)
+            #     columns.append(col)
+            #     dataColumns.append(feature)
 
         elif feature in ITEM_NUMBER_COLUMNS:
             col = tf.feature_column.numeric_column(feature)
@@ -129,7 +133,7 @@ def evaluate(model,test_dataset):
 
 def predict(model_path,df):
     print("加载模型中...")
-    model_new = tf.keras.models.load_model("service_fm_v3")
+    model_new = tf.keras.models.load_model(model_path)
     # model_new.summary()
     print("模型加载完成...")
     # model = tf.keras.models.model_from_json(model.to_json)
@@ -180,7 +184,9 @@ if __name__ == '__main__':
     timestmp4 = int(round(time.time()))
     print("读取数据耗时h:{}".format((timestmp4 - timestmp3)/60/60))
 
-    # evaluate(model,test_data)
-    predict(model_file,test_data)
+    evaluate(model,test_data)
+
+
+    predict(model_file,df_test)
     pass