Update balance_scale.py
Browse files- balance_scale.py +15 -4
balance_scale.py
CHANGED
|
@@ -39,7 +39,14 @@ features_types_per_config = {
|
|
| 39 |
"right_weight": datasets.Value("int64"),
|
| 40 |
"right_distance": datasets.Value("int64"),
|
| 41 |
"balance": datasets.ClassLabel(num_classes=3, names=("tips_left", "balanced", "tips_right"))
|
| 42 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
}
|
| 44 |
features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}
|
| 45 |
|
|
@@ -54,8 +61,8 @@ class Balance_Scale(datasets.GeneratorBasedBuilder):
|
|
| 54 |
# dataset versions
|
| 55 |
DEFAULT_CONFIG = "balance"
|
| 56 |
BUILDER_CONFIGS = [
|
| 57 |
-
Balance_ScaleConfig(name="balance",
|
| 58 |
-
|
| 59 |
]
|
| 60 |
|
| 61 |
|
|
@@ -75,8 +82,12 @@ class Balance_Scale(datasets.GeneratorBasedBuilder):
|
|
| 75 |
def _generate_examples(self, filepath: str):
|
| 76 |
data = pandas.read_csv(filepath, header=None)
|
| 77 |
data.columns = _BASE_FEATURE_NAMES
|
|
|
|
| 78 |
data = data[["left_weight", "left_distance", "right_weight", "right_distance", "balance"]]
|
| 79 |
-
data
|
|
|
|
|
|
|
|
|
|
| 80 |
|
| 81 |
for row_id, row in data.iterrows():
|
| 82 |
data_row = dict(row)
|
|
|
|
| 39 |
"right_weight": datasets.Value("int64"),
|
| 40 |
"right_distance": datasets.Value("int64"),
|
| 41 |
"balance": datasets.ClassLabel(num_classes=3, names=("tips_left", "balanced", "tips_right"))
|
| 42 |
+
},
|
| 43 |
+
"is_balanced": {
|
| 44 |
+
"left_weight": datasets.Value("int64"),
|
| 45 |
+
"left_distance": datasets.Value("int64"),
|
| 46 |
+
"right_weight": datasets.Value("int64"),
|
| 47 |
+
"right_distance": datasets.Value("int64"),
|
| 48 |
+
"is_balanced": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
|
| 49 |
+
},
|
| 50 |
}
|
| 51 |
features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}
|
| 52 |
|
|
|
|
| 61 |
# dataset versions
|
| 62 |
DEFAULT_CONFIG = "balance"
|
| 63 |
BUILDER_CONFIGS = [
|
| 64 |
+
Balance_ScaleConfig(name="balance", description="Multiclass classification of the scale balance."),
|
| 65 |
+
Balance_ScaleConfig(name="is_balanced", description="Binary classification of the scale balance."),
|
| 66 |
]
|
| 67 |
|
| 68 |
|
|
|
|
| 82 |
def _generate_examples(self, filepath: str):
|
| 83 |
data = pandas.read_csv(filepath, header=None)
|
| 84 |
data.columns = _BASE_FEATURE_NAMES
|
| 85 |
+
|
| 86 |
data = data[["left_weight", "left_distance", "right_weight", "right_distance", "balance"]]
|
| 87 |
+
data["balance"] = data.balance.apply(lambda x: 0 if x == "L" else 1 if x == "B" else 2)
|
| 88 |
+
if self.config.name == "is_balanced":
|
| 89 |
+
data["balance"] = data.balance.apply(lambda x: 1 if x == 1 else 0)
|
| 90 |
+
data = data.rename(columns={"balance": "is_balanced"})
|
| 91 |
|
| 92 |
for row_id, row in data.iterrows():
|
| 93 |
data_row = dict(row)
|