Improve the code

convert_col2hist.ipynb

@@ -16,7 +16,7 @@
    "execution_count": 2,
    "outputs": [],
    "source": [
-    "csv = pd.read_csv('111111.csv', usecols=['project full path', 'project url', 'notification count', 'star count'])"
+    "csv = pd.read_csv('weekly_growth.csv', usecols=['project full path', 'project url', 'notification count', 'star count'])"
    ],
    "metadata": {
     "collapsed": false,

convert_col2hist.py

+import pandas as pd
+import matplotlib.pyplot as plt
+import argparse
+
+# Just change the parameters here
+parser = argparse.ArgumentParser()
+parser.add_argument('--sort_values', default='notification count', type=str)
+parser.add_argument('--result_name', default='weekly_growth.csv', type=str)
+parser.add_argument('--use_cols', default=['project full path', 'project url', 'notification count', 'star count'], type=list)
+args = parser.parse_args()
+
+
+csv = pd.read_csv(args.result_name, usecols=args.use_cols)
+
+
+dict_notification = {}
+for count in csv[args.sort_values].unique():
+    num = len(csv[csv[args.sort_values] == count].index)
+    dict_notification[str(count)] = num
+
+
+dict_notification = dict(sorted(dict_notification.items(), key=lambda item: int(item[0])))
+
+
+print(dict_notification)
+
+filter_dict = {}
+for notification_count, num in dict_notification.items():
+    if int(notification_count) < 10:
+        continue
+    filter_dict[notification_count] = num
+
+
+plt.bar(filter_dict.keys(), filter_dict.values(), width=3, color='g')
+plt.show()
+
+plt.plot(filter_dict.keys(), filter_dict.values())
+plt.show()
\ No newline at end of file

updateRepo_dataAnalysis.py

@@ -2,20 +2,22 @@ import pandas as pd
 import numpy as np
 import argparse
 
+# Just change the parameters here
 parser = argparse.ArgumentParser()
 parser.add_argument('--csv1_path', default='0611-projects-star-and-notification.csv', type=str)
 parser.add_argument('--csv2_path', default='0618-projects-star-and-notification.csv', type=str)
 parser.add_argument('--sort_values', default='project full path', type=str)
 parser.add_argument('--result_name', default='weekly_growth.csv', type=str)
+parser.add_argument('--calc_index', default=['notification count', 'star count'], type=list)
 args = parser.parse_args()
 
 csv1 = pd.read_csv(args.csv1_path)
 csv2 = pd.read_csv(args.csv2_path)
-  
+
 
 csv1 = csv1.sort_values(args.sort_values)
 csv2 = csv2.sort_values(args.sort_values)
-  
+
 
 projectNames1 = []
 for i in range(len(csv1)):
@@ -24,32 +26,40 @@ for i in range(len(csv1)):
 projectNames2 = []
 for i in range(len(csv2)):
     projectNames2.append(csv2.iloc[i, 0])
-  
+
 
 in2notin1 = []
 for i in range(len(csv2)):
     if not csv2.iloc[i, 0] in projectNames1:
         in2notin1.append(csv2.iloc[i, 0])
-  
+
 
 for name in in2notin1:
     csv2.drop(csv2[csv2['project full path'] == name].index, inplace=True)
-  
+
+in1notin2 = []
+for i in range(len(csv1)):
+    if not csv1.iloc[i, 0] in projectNames2:
+        in1notin2.append(csv1.iloc[i, 0])
+
+for name in in1notin2:
+    csv1.drop(csv1[csv1['project full path'] == name].index, inplace=True)
+
 
 for i in range(len(csv1)):
     if csv1.iloc[i, 0] == csv2.iloc[i, 0]:
         continue
     print(i, csv1.iloc[i, 0], csv2.iloc[i, 0])
-  
+
 
 csv1.index = np.asarray(list(range(0, len(csv1))))
 csv2.index = np.asarray(list(range(0, len(csv1))))
 
-notification_sub = csv2['notification count'] - csv1['notification count']
-csv1['notification count'] = notification_sub
+# One week data growth calculation
+calculation_list = args.calc_index
+for calc_index in calculation_list:
+    sub = csv2[calc_index] - csv1[calc_index]
+    csv1[calc_index] = sub
 
-star_sub = csv2['star count'] - csv1['star count']
-csv1['star count'] = star_sub
-  
 
 csv1.to_csv(args.result_name, index=False)
\ No newline at end of file

updateRepo_parseYml.py

@@ -2,11 +2,12 @@ import yaml
 import pandas as pd
 import argparse
 
-
+# Just change the parameters here
 parser = argparse.ArgumentParser()
 parser.add_argument('--index_yml', default='/home/mao/Downloads/operation-work/explore/index.yml', type=str)
 parser.add_argument('--sort_values', default='notification count', type=str)
 parser.add_argument('--result_name', default='weekly_growth.csv', type=str)
+parser.add_argument('--use_cols', default=['project full path', 'project url', 'notification count', 'star count'], type=list)
 args = parser.parse_args()
 
 with open(args.index_yml, 'r', encoding='utf-8') as stream:
@@ -15,32 +16,14 @@ with open(args.index_yml, 'r', encoding='utf-8') as stream:
         print(parse_yml)
     except yaml.YAMLError as exec:
         print(exec)
-    
-
-parse_yml['projects']
-    
-
-csv = pd.read_csv(args.result_name, usecols=['project full path', 'project url', 'notification count', 'star count'])
-    
 
-csv.head()
-    
 
+csv = pd.read_csv(args.result_name, usecols=args.use_cols)
 csv = csv.sort_values(args.sort_values, ascending=False)
-    
-
-csv.head()
-    
-
-csv.iloc[0, 0]
-    
 
 for i in range(len(parse_yml['projects'])):
-    parse_yml['projects'][i] = csv.iloc[i+25, 0]
-    
+    parse_yml['projects'][i] = csv.iloc[i, 0]
 
-parse_yml['projects']
-    
 
 with open(args.index_yml, 'w', encoding='utf-8') as stream:
     try: