read tidb

eda/smart_rank/gm_tag_cf.py

@@ -4,7 +4,7 @@ import numpy as np
 import pandas as pd
 from sklearn.metrics import pairwise_distances
 from scipy.spatial.distance import cosine, correlation
-from tool import get_all_tag_parent_tag, get_all_tags_name, get_all_word_synonym_words
+from tool import get_all_tag_parent_tag, get_all_tags_name, get_all_word_synonym_words, get_data_by_mysql
 
 
 # Recommender Engine
@@ -14,12 +14,22 @@ def get_similary_tags(tag_id, tags, ratings_matrix):
     tags['similarity'] = ratings_matrix.iloc[tag_id_index]
     return tags.sort_values(["similarity"], ascending=False)[:-1].tag_id.tolist()  # rm self
 
+
+def get_user_log():
+    sql = "select userId, tagId, rating from item_cf_log"
+    mysql_results = get_data_by_mysql('172.16.40.158', 4000, 'root', '3SYz54LS9#^9sBvC', 'jerry_test', sql)
+    df = pd.DataFrame(mysql_results)
+    return df
+
 if __name__ == '__main__':
     # User Log
     # index: sample id
     # columns: userId,tagId,rating
-    ratings = pd.read_csv("/home/gmuser/gyz/log/cf/user_item_log.csv", sep="\t")
-    ratings.drop("timestamp", inplace=True, axis=1)
+    # ratings = pd.read_csv("/home/gmuser/gyz/log/cf/user_item_log.csv", sep="\t")
+    # ratings.drop("timestamp", inplace=True, axis=1)
+    # sorted_tag_ids = sorted(ratings.tagId.unique())
+    # cf_tags = pd.DataFrame({'tag_id': sorted_tag_ids})
+    ratings = get_user_log()
     sorted_tag_ids = sorted(ratings.tagId.unique())
     cf_tags = pd.DataFrame({'tag_id': sorted_tag_ids})
     # ratings.head()