deploy: 70e91c39c93ea64c5ddd81aee7881cf044abb76f

README.md

@@ -1,12 +0,0 @@
----
-title: Layout Maximum Iou
-emoji: 🌍
-colorFrom: gray
-colorTo: yellow
-sdk: gradio
-sdk_version: 4.18.0
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

layout_maximum_iou.py

@@ -0,0 +1,193 @@
+from collections import defaultdict
+from itertools import chain
+from typing import Dict, List, Tuple, TypedDict
+
+import datasets as ds
+import evaluate
+import numpy as np
+import numpy.typing as npt
+from scipy.optimize import linear_sum_assignment
+
+
+class Layout(TypedDict):
+    bboxes: npt.NDArray[np.float64]
+    categories: npt.NDArray[np.int64]
+
+
+_DESCRIPTION = """\
+
+"""
+
+_CITATION = """\
+
+"""
+
+
+def convert_xywh_to_ltrb(
+    batch_bbox: npt.NDArray[np.float64],
+) -> Tuple[
+    npt.NDArray[np.float64],
+    npt.NDArray[np.float64],
+    npt.NDArray[np.float64],
+    npt.NDArray[np.float64],
+]:
+    xc, yc, w, h = batch_bbox
+    x1 = xc - w / 2
+    y1 = yc - h / 2
+    x2 = xc + w / 2
+    y2 = yc + h / 2
+    return (x1, y1, x2, y2)
+
+
+def _compute_iou(
+    bbox1: npt.NDArray[np.float64],
+    bbox2: npt.NDArray[np.float64],
+    generalized: bool = False,
+):
+    # shape: bbox1 (N, 4), bbox2 (N, 4)
+    assert bbox1.shape[0] == bbox2.shape[0]
+    assert bbox1.shape[1] == bbox1.shape[1] == 4
+
+    l1, t1, r1, b1 = convert_xywh_to_ltrb(bbox1.T)
+    l2, t2, r2, b2 = convert_xywh_to_ltrb(bbox2.T)
+    a1, a2 = (r1 - l1) * (b1 - t1), (r2 - l2) * (b2 - t2)
+
+    # intersection
+    l_max = np.maximum(l1, l2)
+    r_min = np.minimum(r1, r2)
+    t_max = np.maximum(t1, t2)
+    b_min = np.minimum(b1, b2)
+    cond = (l_max < r_min) & (t_max < b_min)
+    ai = np.where(cond, (r_min - l_max) * (b_min - t_max), np.zeros_like(a1[0]))
+
+    au = a1 + a2 - ai
+    iou = ai / au
+
+    if not generalized:
+        return iou
+
+    # outer region
+    l_min = np.minimum(l1, l2)
+    r_max = np.maximum(r1, r2)
+    t_min = np.minimum(t1, t2)
+    b_max = np.maximum(b1, b2)
+    ac = (r_max - l_min) * (b_max - t_min)
+
+    giou = iou - (ac - au) / ac
+
+    return giou
+
+
+def _compute_maximum_iou_for_layout(layout1: Layout, layout2: Layout):
+    score = 0.0
+    bi, ci = layout1["bboxes"], layout1["categories"]
+    bj, cj = layout2["bboxes"], layout2["categories"]
+    N = len(bi)
+
+    for c in list(set(ci.tolist())):
+        _bi = bi[np.where(ci == c)]
+        _bj = bj[np.where(cj == c)]
+        n = len(_bi)
+        ii, jj = np.meshgrid(range(n), range(n))
+        ii, jj = ii.flatten(), jj.flatten()
+        iou = _compute_iou(_bi[ii], _bj[jj]).reshape(n, n)
+        # Note: maximize is supported only when scipy >= 1.4
+        ii, jj = linear_sum_assignment(iou, maximize=True)
+        score += iou[ii, jj].sum().item()
+    return score / N
+
+
+def _compute_maximum_iou(
+    layouts_1_and_2: Tuple[List[Layout], List[Layout]]
+) -> npt.NDArray[np.float64]:
+    assert len(layouts_1_and_2) == 2
+    layouts1, layouts2 = layouts_1_and_2
+
+    N, M = len(layouts1), len(layouts2)
+    ii, jj = np.meshgrid(range(N), range(M))
+    ii, jj = ii.flatten(), jj.flatten()
+    scores = np.asarray(
+        [
+            _compute_maximum_iou_for_layout(layouts1[i], layouts2[j])
+            for i, j in zip(ii, jj)
+        ]
+    )
+    scores = scores.reshape(N, M)
+    ii, jj = linear_sum_assignment(scores, maximize=True)
+    return scores[ii, jj]
+
+
+def _get_cond_to_layouts(layouts: List[Layout]) -> Dict[str, List[Layout]]:
+    out = defaultdict(list)
+    for layout in layouts:
+        bboxes_list = layout["bboxes"]
+        categories_list = layout["categories"]
+        assert len(bboxes_list) == len(categories_list)
+
+        for bboxes, categories in zip(bboxes_list, categories_list):
+            bboxes = np.array(bboxes)
+            cond_key = str(sorted(categories))
+            categories = np.array(categories)
+            layout_dict: Layout = {"bboxes": bboxes, "categories": categories}
+            out[cond_key].append(layout_dict)
+    return out
+
+
+def compute_maximum_iou(args):
+    return [_compute_maximum_iou(a) for a in args]
+
+
+class LayoutMaximumIoU(evaluate.Metric):
+
+    def _info(self) -> evaluate.EvaluationModuleInfo:
+        return evaluate.EvaluationModuleInfo(
+            description=_DESCRIPTION,
+            citation=_CITATION,
+            features=ds.Features(
+                {
+                    "layouts1": ds.Sequence(
+                        {
+                            "bboxes": ds.Sequence(ds.Sequence((ds.Value("float64")))),
+                            "categories": ds.Sequence(ds.Value("int64")),
+                        }
+                    ),
+                    "layouts2": ds.Sequence(
+                        {
+                            "bboxes": ds.Sequence(ds.Sequence((ds.Value("float64")))),
+                            "categories": ds.Sequence(ds.Value("int64")),
+                        }
+                    ),
+                }
+            ),
+            codebase_urls=[
+                "https://github.com/CyberAgentAILab/layout-dm/blob/main/src/trainer/trainer/helpers/metric.py#L206-L247",
+                "https://github.com/CyberAgentAILab/layout-dm/blob/main/src/trainer/trainer/helpers/metric.py#L250-L297",
+                "https://github.com/CyberAgentAILab/layout-dm/blob/main/src/trainer/trainer/helpers/metric.py#L300-L314",
+                "https://github.com/CyberAgentAILab/layout-dm/blob/main/src/trainer/trainer/helpers/metric.py#L317-L329",
+                "https://github.com/CyberAgentAILab/layout-dm/blob/main/src/trainer/trainer/helpers/metric.py#L332-L340",
+                "https://github.com/CyberAgentAILab/layout-dm/blob/main/src/trainer/trainer/helpers/metric.py#L343-L371",
+            ],
+        )
+
+    def _compute(
+        self,
+        *,
+        layouts1: List[Layout],
+        layouts2: List[Layout],
+    ) -> float:
+        c2bl_1 = _get_cond_to_layouts(layouts1)
+        keys_1 = set(c2bl_1.keys())
+        c2bl_2 = _get_cond_to_layouts(layouts2)
+        keys_2 = set(c2bl_2.keys())
+        keys = list(keys_1.intersection(keys_2))
+        args = [(c2bl_1[key], c2bl_2[key]) for key in keys]
+
+        # to check actual number of layouts for evaluation
+        # ans = 0
+        # for x in args:
+        #     ans += len(x[0])
+
+        scores = compute_maximum_iou(args)
+        scores = np.asarray(list(chain.from_iterable(scores)))
+
+        return scores.mean().item() if len(scores) != 0 else 0.0

requirements.txt

@@ -0,0 +1,90 @@
+aiofiles==23.2.1 ; python_version >= "3.9" and python_version < "4.0"
+aiohttp==3.9.3 ; python_version >= "3.9" and python_version < "4.0"
+aiosignal==1.3.1 ; python_version >= "3.9" and python_version < "4.0"
+altair==5.2.0 ; python_version >= "3.9" and python_version < "4.0"
+annotated-types==0.6.0 ; python_version >= "3.9" and python_version < "4.0"
+anyio==4.2.0 ; python_version >= "3.9" and python_version < "4.0"
+arrow==1.3.0 ; python_version >= "3.9" and python_version < "4.0"
+async-timeout==4.0.3 ; python_version >= "3.9" and python_version < "3.11"
+attrs==23.2.0 ; python_version >= "3.9" and python_version < "4.0"
+binaryornot==0.4.4 ; python_version >= "3.9" and python_version < "4.0"
+certifi==2024.2.2 ; python_version >= "3.9" and python_version < "4.0"
+chardet==5.2.0 ; python_version >= "3.9" and python_version < "4.0"
+charset-normalizer==3.3.2 ; python_version >= "3.9" and python_version < "4.0"
+click==8.1.7 ; python_version >= "3.9" and python_version < "4.0"
+colorama==0.4.6 ; python_version >= "3.9" and python_version < "4.0"
+contourpy==1.2.0 ; python_version >= "3.9" and python_version < "4.0"
+cookiecutter==2.5.0 ; python_version >= "3.9" and python_version < "4.0"
+cycler==0.12.1 ; python_version >= "3.9" and python_version < "4.0"
+datasets==2.17.0 ; python_version >= "3.9" and python_version < "4.0"
+dill==0.3.8 ; python_version >= "3.9" and python_version < "4.0"
+evaluate[template]==0.4.1 ; python_version >= "3.9" and python_version < "4.0"
+exceptiongroup==1.2.0 ; python_version >= "3.9" and python_version < "3.11"
+fastapi==0.109.2 ; python_version >= "3.9" and python_version < "4.0"
+ffmpy==0.3.1 ; python_version >= "3.9" and python_version < "4.0"
+filelock==3.13.1 ; python_version >= "3.9" and python_version < "4.0"
+fonttools==4.48.1 ; python_version >= "3.9" and python_version < "4.0"
+frozenlist==1.4.1 ; python_version >= "3.9" and python_version < "4.0"
+fsspec==2023.10.0 ; python_version >= "3.9" and python_version < "4.0"
+fsspec[http]==2023.10.0 ; python_version >= "3.9" and python_version < "4.0"
+gradio-client==0.10.0 ; python_version >= "3.9" and python_version < "4.0"
+gradio==4.18.0 ; python_version >= "3.9" and python_version < "4.0"
+h11==0.14.0 ; python_version >= "3.9" and python_version < "4.0"
+httpcore==1.0.2 ; python_version >= "3.9" and python_version < "4.0"
+httpx==0.26.0 ; python_version >= "3.9" and python_version < "4.0"
+huggingface-hub==0.20.3 ; python_version >= "3.9" and python_version < "4.0"
+idna==3.6 ; python_version >= "3.9" and python_version < "4.0"
+importlib-resources==6.1.1 ; python_version >= "3.9" and python_version < "4.0"
+jinja2==3.1.3 ; python_version >= "3.9" and python_version < "4.0"
+jsonschema-specifications==2023.12.1 ; python_version >= "3.9" and python_version < "4.0"
+jsonschema==4.21.1 ; python_version >= "3.9" and python_version < "4.0"
+kiwisolver==1.4.5 ; python_version >= "3.9" and python_version < "4.0"
+markdown-it-py==3.0.0 ; python_version >= "3.9" and python_version < "4.0"
+markupsafe==2.1.5 ; python_version >= "3.9" and python_version < "4.0"
+matplotlib==3.8.2 ; python_version >= "3.9" and python_version < "4.0"
+mdurl==0.1.2 ; python_version >= "3.9" and python_version < "4.0"
+multidict==6.0.5 ; python_version >= "3.9" and python_version < "4.0"
+multiprocess==0.70.16 ; python_version >= "3.9" and python_version < "4.0"
+numpy==1.26.4 ; python_version >= "3.9" and python_version < "4.0"
+orjson==3.9.13 ; python_version >= "3.9" and python_version < "4.0"
+packaging==23.2 ; python_version >= "3.9" and python_version < "4.0"
+pandas==2.2.0 ; python_version >= "3.9" and python_version < "4.0"
+pillow==10.2.0 ; python_version >= "3.9" and python_version < "4.0"
+pyarrow-hotfix==0.6 ; python_version >= "3.9" and python_version < "4.0"
+pyarrow==15.0.0 ; python_version >= "3.9" and python_version < "4.0"
+pydantic-core==2.16.2 ; python_version >= "3.9" and python_version < "4.0"
+pydantic==2.6.1 ; python_version >= "3.9" and python_version < "4.0"
+pydub==0.25.1 ; python_version >= "3.9" and python_version < "4.0"
+pygments==2.17.2 ; python_version >= "3.9" and python_version < "4.0"
+pyparsing==3.1.1 ; python_version >= "3.9" and python_version < "4.0"
+python-dateutil==2.8.2 ; python_version >= "3.9" and python_version < "4.0"
+python-multipart==0.0.9 ; python_version >= "3.9" and python_version < "4.0"
+python-slugify==8.0.4 ; python_version >= "3.9" and python_version < "4.0"
+pytz==2024.1 ; python_version >= "3.9" and python_version < "4.0"
+pyyaml==6.0.1 ; python_version >= "3.9" and python_version < "4.0"
+referencing==0.33.0 ; python_version >= "3.9" and python_version < "4.0"
+requests==2.31.0 ; python_version >= "3.9" and python_version < "4.0"
+responses==0.18.0 ; python_version >= "3.9" and python_version < "4.0"
+rich==13.7.0 ; python_version >= "3.9" and python_version < "4.0"
+rpds-py==0.17.1 ; python_version >= "3.9" and python_version < "4.0"
+ruff==0.2.1 ; python_version >= "3.9" and python_version < "4.0"
+scipy==1.12.0 ; python_version >= "3.9" and python_version < "4.0"
+semantic-version==2.10.0 ; python_version >= "3.9" and python_version < "4.0"
+shellingham==1.5.4 ; python_version >= "3.9" and python_version < "4.0"
+six==1.16.0 ; python_version >= "3.9" and python_version < "4.0"
+sniffio==1.3.0 ; python_version >= "3.9" and python_version < "4.0"
+starlette==0.36.3 ; python_version >= "3.9" and python_version < "4.0"
+text-unidecode==1.3 ; python_version >= "3.9" and python_version < "4.0"
+tomlkit==0.12.0 ; python_version >= "3.9" and python_version < "4.0"
+toolz==0.12.1 ; python_version >= "3.9" and python_version < "4.0"
+tqdm==4.66.2 ; python_version >= "3.9" and python_version < "4.0"
+typer[all]==0.9.0 ; python_version >= "3.9" and python_version < "4.0"
+types-python-dateutil==2.8.19.20240106 ; python_version >= "3.9" and python_version < "4.0"
+typing-extensions==4.9.0 ; python_version >= "3.9" and python_version < "4.0"
+tzdata==2024.1 ; python_version >= "3.9" and python_version < "4.0"
+urllib3==2.2.0 ; python_version >= "3.9" and python_version < "4.0"
+uvicorn==0.27.1 ; python_version >= "3.9" and python_version < "4.0"
+websockets==11.0.3 ; python_version >= "3.9" and python_version < "4.0"
+xxhash==3.4.1 ; python_version >= "3.9" and python_version < "4.0"
+yarl==1.9.4 ; python_version >= "3.9" and python_version < "4.0"
+zipp==3.17.0 ; python_version >= "3.9" and python_version < "3.10"