From b85ee9d64a536937912544c7bbd5b98b635b7e8d Mon Sep 17 00:00:00 2001
From: Christian C <cc@localhost>
Date: Mon, 11 Nov 2024 12:29:32 -0800
Subject: Initial commit

---
 code/sunlab/common/data/shape_dataset.py | 57 ++++++++++++++++++++++++++++++++
 1 file changed, 57 insertions(+)
 create mode 100644 code/sunlab/common/data/shape_dataset.py

(limited to 'code/sunlab/common/data/shape_dataset.py')

diff --git a/code/sunlab/common/data/shape_dataset.py b/code/sunlab/common/data/shape_dataset.py
new file mode 100644
index 0000000..5a68736
--- /dev/null
+++ b/code/sunlab/common/data/shape_dataset.py
@@ -0,0 +1,57 @@
+from .dataset import Dataset
+
+
+class ShapeDataset(Dataset):
+    """# Shape Dataset"""
+
+    def __init__(
+        self,
+        dataset_filename,
+        data_columns=[
+            "Area",
+            "MjrAxisLength",
+            "MnrAxisLength",
+            "Eccentricity",
+            "ConvexArea",
+            "EquivDiameter",
+            "Solidity",
+            "Extent",
+            "Perimeter",
+            "ConvexPerim",
+            "FibLen",
+            "InscribeR",
+            "BlebLen",
+        ],
+        label_columns=["Class"],
+        batch_size=None,
+        shuffle=False,
+        val_split=0.0,
+        scaler=None,
+        sort_columns=None,
+        random_seed=4332,
+        pre_scale=10,
+        **kwargs
+    ):
+        """# Initialize Dataset
+        self.dataset = dataset (N, ...)
+        self.labels = labels (N, ...)
+
+        Optional Arguments:
+        - prescale_function: The function that takes the ratio and transforms
+        the dataset by multiplying the prescale_function output
+        - sort_columns: The columns to sort the data by initially
+        - equal_split: If the classifications should be equally split in
+        training"""
+        super().__init__(
+            dataset_filename,
+            data_columns=data_columns,
+            label_columns=label_columns,
+            batch_size=batch_size,
+            shuffle=shuffle,
+            val_split=val_split,
+            scaler=scaler,
+            sort_columns=sort_columns,
+            random_seed=random_seed,
+            pre_scale=pre_scale,
+            **kwargs
+        )
-- 
cgit v1.2.1