Hugging Face's logo

Datasets:
mstz
/
glass

Tasks:
Tabular Classification
Languages:
English
Tags:
glass
tabular_classification
binary_classification
UCI
License:
Dataset card Files Community
mstz commited on
Commit
58df4ea
•
1 Parent(s): 558be89

Upload glass.py

Browse files
Files changed (1) hide show
  1. glass.py +105 -105
glass.py CHANGED
@@ -23,94 +23,94 @@ _CITATION = """
23
 
24
  # Dataset info
25
  _BASE_FEATURE_NAMES = [
26
- "relative_compactness",
27
- "surface_area",
28
- "wall_area",
29
- "roof_area",
30
- "overall_height",
31
- "orientation",
32
- "glazing_area",
33
- "glazing_area_distribution",
34
- "heating_load",
35
- "cooling_load"
36
  ]
37
  urls_per_split = {
38
  "train": "https://huggingface.co/datasets/mstz/glass/raw/main/glass.data"
39
  }
40
  features_types_per_config = {
41
  "glass": {
42
- "refractive_index": datasets.Value("float64"),
43
- "sodium": datasets.Value("float64"),
44
- "magnesium": datasets.Value("float64"),
45
- "aluminum": datasets.Value("float64"),
46
- "silicon": datasets.Value("float64"),
47
- "potassium": datasets.Value("float64"),
48
- "calcium": datasets.Value("float64"),
49
- "barium": datasets.Value("int8"),
50
- "iron": datasets.Value("float64"),
51
- "glass_type": datasets.ClassLabel(num_classes=6, names=("windows_1", "windows_2",
52
- "vehicle_windows_1", "vehicle_windows_2",
53
- "containers", "tableware", "headlamps"))
54
  },
55
- "windows": {
56
- "refractive_index": datasets.Value("float64"),
57
- "sodium": datasets.Value("float64"),
58
- "magnesium": datasets.Value("float64"),
59
- "aluminum": datasets.Value("float64"),
60
- "silicon": datasets.Value("float64"),
61
- "potassium": datasets.Value("float64"),
62
- "calcium": datasets.Value("float64"),
63
- "barium": datasets.Value("int8"),
64
- "iron": datasets.Value("float64"),
65
- "is_windows_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
66
  },
67
- "vehicles": {
68
- "refractive_index": datasets.Value("float64"),
69
- "sodium": datasets.Value("float64"),
70
- "magnesium": datasets.Value("float64"),
71
- "aluminum": datasets.Value("float64"),
72
- "silicon": datasets.Value("float64"),
73
- "potassium": datasets.Value("float64"),
74
- "calcium": datasets.Value("float64"),
75
- "barium": datasets.Value("int8"),
76
- "iron": datasets.Value("float64"),
77
- "is_vehicle_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
78
  },
79
- "containers": {
80
- "refractive_index": datasets.Value("float64"),
81
- "sodium": datasets.Value("float64"),
82
- "magnesium": datasets.Value("float64"),
83
- "aluminum": datasets.Value("float64"),
84
- "silicon": datasets.Value("float64"),
85
- "potassium": datasets.Value("float64"),
86
- "calcium": datasets.Value("float64"),
87
- "barium": datasets.Value("int8"),
88
- "iron": datasets.Value("float64"),
89
- "is_container_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
90
  },
91
- "tableware": {
92
- "refractive_index": datasets.Value("float64"),
93
- "sodium": datasets.Value("float64"),
94
- "magnesium": datasets.Value("float64"),
95
- "aluminum": datasets.Value("float64"),
96
- "silicon": datasets.Value("float64"),
97
- "potassium": datasets.Value("float64"),
98
- "calcium": datasets.Value("float64"),
99
- "barium": datasets.Value("int8"),
100
- "iron": datasets.Value("float64"),
101
- "is_tableware_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
102
  },
103
- "headlamps": {
104
- "refractive_index": datasets.Value("float64"),
105
- "sodium": datasets.Value("float64"),
106
- "magnesium": datasets.Value("float64"),
107
- "aluminum": datasets.Value("float64"),
108
- "silicon": datasets.Value("float64"),
109
- "potassium": datasets.Value("float64"),
110
- "calcium": datasets.Value("float64"),
111
- "barium": datasets.Value("int8"),
112
- "iron": datasets.Value("float64"),
113
- "is_headlamp_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
114
  },
115
  }
116
  features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}
@@ -127,12 +127,12 @@ class Glass(datasets.GeneratorBasedBuilder):
127
  DEFAULT_CONFIG = "glass"
128
  BUILDER_CONFIGS = [
129
  GlassConfig(name="glass", description="Glass dataset."),
130
- GlassConfig(name="windows", description="Is this windows glass?"),
131
- GlassConfig(name="vehicles", description="Is this vehicles glass?"),
132
- GlassConfig(name="containers", description="Is this containers glass?"),
133
- GlassConfig(name="tableware", description="Is this tableware glass?"),
134
- GlassConfig(name="headlamps", description="Is this headlamps glass?")
135
- ]
136
 
137
 
138
  def _info(self):
@@ -160,27 +160,27 @@ class Glass(datasets.GeneratorBasedBuilder):
160
  def preprocess(self, data: pandas.DataFrame) -> pandas.DataFrame:
161
  data.columns = _BASE_FEATURE_NAMES
162
 
163
- if self.config.name == "windows":
164
- data = data.rename(columns={"glass_type", "is_windows_glass"})
165
- data.loc[:, "is_windows_glass"] = data.is_windows_glass.apply(lambda x: 0 if data.is_windows_glass in {1, 2} else 0)
166
-
167
- elif self.config.name == "vehicles":
168
- data = data.rename(columns={"glass_type", "is_vehicles_glass"})
169
- data.loc[:, "is_vehicles_glass"] = data.is_vehicles_glass.apply(lambda x: 0 if data.is_vehicles_glass in {3, 4} else 0)
170
-
171
- elif self.config.name == "containers":
172
- data = data.rename(columns={"glass_type", "is_containers_glass"})
173
- data.loc[:, "is_containers_glass"] = data.is_containers_glass.apply(lambda x: 0 if data.is_containers_glass == 5 else 0)
174
-
175
- elif self.config.name == "tableware":
176
- data = data.rename(columns={"glass_type", "is_tableware_glass"})
177
- data.loc[:, "is_tableware_glass"] = data.is_tableware_glass.apply(lambda x: 0 if data.is_tableware_glass == 6 else 0)
178
-
179
- elif self.config.name == "headlamps":
180
- data = data.rename(columns={"glass_type", "is_headlamps_glass"})
181
- data.loc[:, "is_headlamps_glass"] = data.is_headlamps_glass.apply(lambda x: 0 if data.is_headlamps_glass == 7 else 0)
182
-
183
- else:
184
- data.loc[:, "glass_type"] = data.glass_type.apply(lambda x: x - 1)
185
 
186
  return data
 
23
 
24
  # Dataset info
25
  _BASE_FEATURE_NAMES = [
26
+ "relative_compactness",
27
+ "surface_area",
28
+ "wall_area",
29
+ "roof_area",
30
+ "overall_height",
31
+ "orientation",
32
+ "glazing_area",
33
+ "glazing_area_distribution",
34
+ "heating_load",
35
+ "cooling_load"
36
  ]
37
  urls_per_split = {
38
  "train": "https://huggingface.co/datasets/mstz/glass/raw/main/glass.data"
39
  }
40
  features_types_per_config = {
41
  "glass": {
42
+ "refractive_index": datasets.Value("float64"),
43
+ "sodium": datasets.Value("float64"),
44
+ "magnesium": datasets.Value("float64"),
45
+ "aluminum": datasets.Value("float64"),
46
+ "silicon": datasets.Value("float64"),
47
+ "potassium": datasets.Value("float64"),
48
+ "calcium": datasets.Value("float64"),
49
+ "barium": datasets.Value("int8"),
50
+ "iron": datasets.Value("float64"),
51
+ "glass_type": datasets.ClassLabel(num_classes=6, names=("windows_1", "windows_2",
52
+ "vehicle_windows_1", "vehicle_windows_2",
53
+ "containers", "tableware", "headlamps"))
54
  },
55
+ "windows": {
56
+ "refractive_index": datasets.Value("float64"),
57
+ "sodium": datasets.Value("float64"),
58
+ "magnesium": datasets.Value("float64"),
59
+ "aluminum": datasets.Value("float64"),
60
+ "silicon": datasets.Value("float64"),
61
+ "potassium": datasets.Value("float64"),
62
+ "calcium": datasets.Value("float64"),
63
+ "barium": datasets.Value("int8"),
64
+ "iron": datasets.Value("float64"),
65
+ "is_windows_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
66
  },
67
+ "vehicles": {
68
+ "refractive_index": datasets.Value("float64"),
69
+ "sodium": datasets.Value("float64"),
70
+ "magnesium": datasets.Value("float64"),
71
+ "aluminum": datasets.Value("float64"),
72
+ "silicon": datasets.Value("float64"),
73
+ "potassium": datasets.Value("float64"),
74
+ "calcium": datasets.Value("float64"),
75
+ "barium": datasets.Value("int8"),
76
+ "iron": datasets.Value("float64"),
77
+ "is_vehicle_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
78
  },
79
+ "containers": {
80
+ "refractive_index": datasets.Value("float64"),
81
+ "sodium": datasets.Value("float64"),
82
+ "magnesium": datasets.Value("float64"),
83
+ "aluminum": datasets.Value("float64"),
84
+ "silicon": datasets.Value("float64"),
85
+ "potassium": datasets.Value("float64"),
86
+ "calcium": datasets.Value("float64"),
87
+ "barium": datasets.Value("int8"),
88
+ "iron": datasets.Value("float64"),
89
+ "is_container_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
90
  },
91
+ "tableware": {
92
+ "refractive_index": datasets.Value("float64"),
93
+ "sodium": datasets.Value("float64"),
94
+ "magnesium": datasets.Value("float64"),
95
+ "aluminum": datasets.Value("float64"),
96
+ "silicon": datasets.Value("float64"),
97
+ "potassium": datasets.Value("float64"),
98
+ "calcium": datasets.Value("float64"),
99
+ "barium": datasets.Value("int8"),
100
+ "iron": datasets.Value("float64"),
101
+ "is_tableware_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
102
  },
103
+ "headlamps": {
104
+ "refractive_index": datasets.Value("float64"),
105
+ "sodium": datasets.Value("float64"),
106
+ "magnesium": datasets.Value("float64"),
107
+ "aluminum": datasets.Value("float64"),
108
+ "silicon": datasets.Value("float64"),
109
+ "potassium": datasets.Value("float64"),
110
+ "calcium": datasets.Value("float64"),
111
+ "barium": datasets.Value("int8"),
112
+ "iron": datasets.Value("float64"),
113
+ "is_headlamp_glass": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
114
  },
115
  }
116
  features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}
 
127
  DEFAULT_CONFIG = "glass"
128
  BUILDER_CONFIGS = [
129
  GlassConfig(name="glass", description="Glass dataset."),
130
+ GlassConfig(name="windows", description="Is this windows glass?"),
131
+ GlassConfig(name="vehicles", description="Is this vehicles glass?"),
132
+ GlassConfig(name="containers", description="Is this containers glass?"),
133
+ GlassConfig(name="tableware", description="Is this tableware glass?"),
134
+ GlassConfig(name="headlamps", description="Is this headlamps glass?")
135
+ ]
136
 
137
 
138
  def _info(self):
 
160
  def preprocess(self, data: pandas.DataFrame) -> pandas.DataFrame:
161
  data.columns = _BASE_FEATURE_NAMES
162
 
163
+ if self.config.name == "windows":
164
+ data = data.rename(columns={"glass_type", "is_windows_glass"})
165
+ data.loc[:, "is_windows_glass"] = data.is_windows_glass.apply(lambda x: 0 if data.is_windows_glass in {1, 2} else 0)
166
+
167
+ elif self.config.name == "vehicles":
168
+ data = data.rename(columns={"glass_type", "is_vehicles_glass"})
169
+ data.loc[:, "is_vehicles_glass"] = data.is_vehicles_glass.apply(lambda x: 0 if data.is_vehicles_glass in {3, 4} else 0)
170
+
171
+ elif self.config.name == "containers":
172
+ data = data.rename(columns={"glass_type", "is_containers_glass"})
173
+ data.loc[:, "is_containers_glass"] = data.is_containers_glass.apply(lambda x: 0 if data.is_containers_glass == 5 else 0)
174
+
175
+ elif self.config.name == "tableware":
176
+ data = data.rename(columns={"glass_type", "is_tableware_glass"})
177
+ data.loc[:, "is_tableware_glass"] = data.is_tableware_glass.apply(lambda x: 0 if data.is_tableware_glass == 6 else 0)
178
+
179
+ elif self.config.name == "headlamps":
180
+ data = data.rename(columns={"glass_type", "is_headlamps_glass"})
181
+ data.loc[:, "is_headlamps_glass"] = data.is_headlamps_glass.apply(lambda x: 0 if data.is_headlamps_glass == 7 else 0)
182
+
183
+ else:
184
+ data.loc[:, "glass_type"] = data.glass_type.apply(lambda x: x - 1)
185
 
186
  return data