feat(opendataset): add dataloader for SVHN dataset

tensorbay/opendataset/SVHN/__init__.py

@@ -0,0 +1,11 @@
+#!/usr/bin/env python3
+#
+# Copyright 2021 Graviti. Licensed under MIT License.
+#
+# pylint: disable=invalid-name
+
+"""Dataloader of SVHN."""
+
+from .loader import SVHN
+
+__all__ = ["SVHN"]

tensorbay/opendataset/SVHN/catalog.json

@@ -0,0 +1,36 @@
+{
+    "BOX2D": {
+        "categories": [
+            {
+                "name": "0"
+            },
+            {
+                "name": "1"
+            },
+            {
+                "name": "2"
+            },
+            {
+                "name": "3"
+            },
+            {
+                "name": "4"
+            },
+            {
+                "name": "5"
+            },
+            {
+                "name": "6"
+            },
+            {
+                "name": "7"
+            },
+            {
+                "name": "8"
+            },
+            {
+                "name": "9"
+            }
+        ]
+    }
+}

tensorbay/opendataset/SVHN/loader.py

@@ -0,0 +1,94 @@
+#!/usr/bin/env python3
+#
+# Copyright 2021 Graviti. Licensed under MIT License.
+#
+# pylint: disable=invalid-name, missing-module-docstring
+
+import os
+from typing import Dict, List
+
+from ...dataset import Data, Dataset
+from ...exception import ModuleImportError
+from ...label import LabeledBox2D
+
+_SVHN_SEGMENTS = ("extra", "test", "train")
+
+DATASET_NAME = "SVHN"
+
+
+def SVHN(path: str) -> Dataset:
+    """Dataloader of SVHN open dataset.
+
+    .. SVHN: http://ufldl.stanford.edu/housenumbers/
+
+    The file structure should be like::
+
+        <path>
+            Cropped/
+                extra_32x32.mat
+                test_32x32.mat
+                train_32x32.mat
+            FullNumbers/
+                extra/
+                    116507.png
+                    116508.png
+                    ...
+                    digitStruct.mat
+                    see_bboxes.m
+                test/
+                train/
+
+    Arguments:
+        path: The root directory of the dataset.
+
+    Returns:
+        Loaded :class: `~tensorbay.dataset.dataset.Dataset` instance.
+
+    """
+    try:
+        from h5py import Dataset as h5Dataset
+        from h5py import File  # pylint: disable=import-outside-toplevel
+    except ModuleNotFoundError as error:
+        raise ModuleImportError(error.name) from error  # type: ignore[arg-type]
+
+    root_path = os.path.abspath(os.path.expanduser(path))
+    dataset = Dataset(DATASET_NAME)
+    dataset.load_catalog(os.path.join(os.path.dirname(__file__), "catalog.json"))
+
+    for segment_name in _SVHN_SEGMENTS:
+        segment = dataset.create_segment(f"FullNumbers-{segment_name}")
+        mat_file = File(os.path.join(root_path, "FullNumbers", segment_name, "digitStruct.mat"))
+        bbox = mat_file["digitStruct"]["bbox"]
+        name = mat_file["digitStruct"]["name"]
+        for index in iter(range(bbox.shape[0])):
+            segment.append(_get_fullnumbers_data(mat_file, name, bbox, index))
+    return dataset
+
+
+def _get_fullnumbers_data(
+    mat_file: "File", name: "h5Dataset", bbox: "h5Dataset", index: int
+) -> Data:
+    data = Data("".join([chr(v[0]) for v in mat_file[name[index][0]]]))
+    data.label.box2d = []
+    meta_data: Dict[str, List[float]] = {}
+    mat_bbox = mat_file[bbox[index][0]]
+    length = mat_bbox["label"].shape[0]
+    for _name, _obj in iter(mat_bbox.items()):
+        meta_data[_name] = []
+        if length == 1:
+            meta_data[_name].append(_obj[0][0])
+        else:
+            for i in iter(range(length)):
+                meta_data[_name].append(mat_file[_obj[i][0]][0][0])
+
+    for i in iter(range(length)):
+        data.label.box2d.append(
+            LabeledBox2D.from_xywh(
+                x=meta_data["left"][i],
+                y=meta_data["top"][i],
+                width=meta_data["width"][i],
+                height=meta_data["height"][i],
+                category=str(int(meta_data["label"][i])),
+            )
+        )
+    return data

tensorbay/opendataset/__init__.py

@@ -41,6 +41,7 @@
 from .NightOwls import NightOwls
 from .nuScenes import nuScenes
 from .RP2K import RP2K
+from .SVHN import SVHN
 from .THCHS30 import THCHS30
 from .THUCNews import THUCNews
 from .TLR import TLR
@@ -90,6 +91,7 @@
     "Newsgroups20",
     "NightOwls",
     "RP2K",
+    "SVHN",
     "THCHS30",
     "THUCNews",
     "TLR",