@@ -56,16 +56,15 @@ def read_data(args):
5656 """
5757
5858 # Load datasets
59- occ = pd .read_csv (r '../data/occupancy .csv' , header = 0 , index_col = 0 )
60- duration = pd .read_csv (r '../data/duration .csv' , header = 0 , index_col = 0 )
61- volume = pd .read_csv (r '../data/volume .csv' , header = 0 , index_col = 0 )
62- e_price = pd .read_csv (r '../data/e_price .csv' , index_col = 0 , header = 0 ). values
63- s_price = pd .read_csv (r '../data/s_price .csv' , index_col = 0 , header = 0 ).values
64- adj = pd .read_csv ('../data/adj .csv' , header = 0 , index_col = 0 )
65- adj . columns = adj . columns . astype ( float ). astype ( int ). astype ( str )
59+ inf = pd .read_csv ('../data/inf .csv' , header = 0 , index_col = None )
60+ occ = pd .read_csv ('../data/occupancy .csv' , header = 0 , index_col = 0 )
61+ duration = pd .read_csv ('../data/duration .csv' , header = 0 , index_col = 0 )
62+ volume = pd .read_csv ('../data/volume .csv' , header = 0 , index_col = 0 )
63+ e_price = pd .read_csv ('../data/e_price .csv' , index_col = 0 , header = 0 ).values
64+ s_price = pd .read_csv ('../data/s_price .csv' , index_col = 0 , header = 0 ). values
65+ adj = pd . read_csv ( '../data/adj.csv' , header = 0 , index_col = None )
6666 adj .index = adj .columns
67- adj = adj .loc [occ .columns ,occ .columns ]
68- adj .to_csv ('../data/adj_filter.csv' )
67+
6968 time = pd .to_datetime (occ .index )
7069
7170 feat = occ
@@ -74,7 +73,12 @@ def read_data(args):
7473 elif args .feat == 'volume' :
7574 feat = volume
7675
77- # Normalize e_price and s_price data
76+ # Normalize
77+ charge_count_dict = dict (zip (inf ['TAZID' ].astype (str ), inf ['charge_count' ]))
78+ for col in occ .columns :
79+ charge_count = charge_count_dict [col ]
80+ occ [col ] = occ [col ] / charge_count
81+
7882 price_scaler = MinMaxScaler (feature_range = (0 , 1 ))
7983 e_price = price_scaler .fit_transform (e_price )
8084 s_price = price_scaler .fit_transform (s_price )
0 commit comments