TobiTob commited on
Commit
dfb8e6c
1 Parent(s): 4e2e391

Update CityLearn.py

Browse files
Files changed (1) hide show
  1. CityLearn.py +20 -0
CityLearn.py CHANGED
@@ -17,6 +17,10 @@ _URLS = {
17
  "f_50": f"{_BASE_URL}/f_50x5x1750.pkl",
18
  "fr_24": f"{_BASE_URL}/fr_24x5x364.pkl",
19
  "fn_24": f"{_BASE_URL}/fn_24x5x3649.pkl",
 
 
 
 
20
  }
21
 
22
 
@@ -65,6 +69,22 @@ class DecisionTransformerCityLearnDataset(datasets.GeneratorBasedBuilder):
65
  name="fn_24",
66
  description="Data sampled from an expert policy in CityLearn environment. Used the new reward function and changed some interactions with noise. Sequence length = 24, Buildings = 5, Episodes = 10 ",
67
  ),
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
68
  ]
69
 
70
  def _info(self):
 
17
  "f_50": f"{_BASE_URL}/f_50x5x1750.pkl",
18
  "fr_24": f"{_BASE_URL}/fr_24x5x364.pkl",
19
  "fn_24": f"{_BASE_URL}/fn_24x5x3649.pkl",
20
+ "rb_24": f"{_BASE_URL}/rb_24x5x364.pkl",
21
+ "rb_230": f"{_BASE_URL}/rb_230x5x38.pkl",
22
+ "rb_2189": f"{_BASE_URL}/rb_2189x5x4.pkl",
23
+ "rbn_24": f"{_BASE_URL}/rb_24x5x18247.pkl",
24
  }
25
 
26
 
 
69
  name="fn_24",
70
  description="Data sampled from an expert policy in CityLearn environment. Used the new reward function and changed some interactions with noise. Sequence length = 24, Buildings = 5, Episodes = 10 ",
71
  ),
72
+ datasets.BuilderConfig(
73
+ name="rb_24",
74
+ description="Data sampled from a simple rule based policy. Used the new reward function. Sequence length = 24, Buildings = 5, Episodes = 1 ",
75
+ ),
76
+ datasets.BuilderConfig(
77
+ name="rb_230",
78
+ description="Data sampled from a simple rule based policy. Used the new reward function. Sequence length = 230, Buildings = 5, Episodes = 1 ",
79
+ ),
80
+ datasets.BuilderConfig(
81
+ name="rb_2189",
82
+ description="Data sampled from a simple rule based policy. Used the new reward function. Sequence length = 2189, Buildings = 5, Episodes = 1 ",
83
+ ),
84
+ datasets.BuilderConfig(
85
+ name="rbn_24",
86
+ description="Data sampled from a simple rule based policy. Used the new reward function and changed some interactions with noise. Sequence length = 24, Buildings = 5, Episodes = 50 ",
87
+ ),
88
  ]
89
 
90
  def _info(self):