Merge remote-tracking branch 'origin/synthetic-data' into feature/heat-map

Author: devhelpr

datasets/__init__.py

@@ -0,0 +1,14 @@
+import pandas as pd
+from pathlib import Path
+import json
+
+
+data_folder = Path(__file__).resolve().parent / "data"
+dtypes_path = (data_folder / "dtypes.json")
+csv_path = str(data_folder / "adult.data")
+
+
+with dtypes_path.open('r') as f:
+    dtypes = json.load(f)
+columns = list(dtypes.keys())
+df = pd.read_csv(csv_path, header=None, names=columns).astype(dtypes)

datasets/adult/__init__.py

@@ -0,0 +1,14 @@
+import pandas as pd
+from pathlib import Path
+import json
+
+
+data_folder = Path(__file__).resolve().parent / "data"
+dtypes_path = (data_folder / "dtypes.json")
+csv_path = str(data_folder / "bar_pass_prediction.csv")
+
+
+with dtypes_path.open('r') as f:
+    dtypes = json.load(f)
+columns = list(dtypes.keys())
+df = pd.read_csv(csv_path, skiprows=1, sep=';', header=None, names=columns).astype(dtypes)

datasets/data/bar_pass_prediction.csv

@@ -0,0 +1,14 @@
+import pandas as pd
+from pathlib import Path
+import json
+
+
+data_folder = Path(__file__).resolve().parent / "data"
+dtypes_path = (data_folder / "dtypes.json")
+csv_path = str(data_folder / "winequality-white.csv")
+
+
+with dtypes_path.open('r') as f:
+    dtypes = json.load(f)
+columns = list(dtypes.keys())
+df = pd.read_csv(csv_path, skiprows=1, sep=';', header=None, names=columns).astype(dtypes)