initial commit

README.md

@@ -0,0 +1,265 @@
+基于 AI 的集成电路版图识别：RoRD 模型
+描述
+本项目实现了 RoRD（Rotation-Robust Descriptors）模型，用于集成电路（IC）版图识别。RoRD 是一种先进的局部特征匹配方法，具有旋转鲁棒性，特别适合于 IC 版图，因为它们可能以各种方向出现（0°、90°、180°、270°及其镜像）。项目通过自监督学习和随机旋转增强，解决了数据稀缺性、几何多变性、动态扩展性和结构复杂性等挑战。
+项目包括：
+
+模型实现：适用于 IC 版图的 RoRD 模型，使用 PyTorch，基于 D2-Net 架构。
+数据加载：自定义数据集类 ICLayoutDataset，用于加载光栅化的 IC 版图图像。
+训练脚本：通过随机旋转进行自监督训练，确保模型对旋转鲁棒。
+评估脚本：在验证集上评估模型性能，计算精确率、召回率和 F1 分数。
+匹配工具：使用训练好的模型进行模板匹配，支持多实例匹配和可视化。
+
+安装
+环境要求
+
+Python 3.8 或更高版本
+CUDA（可选，用于 GPU 加速）
+
+依赖安装
+使用 uv 安装依赖库：
+uv add torch torchvision opencv-python numpy Pillow
+uv lock
+uv sync
+
+或者使用 pip：
+pip install torch torchvision opencv-python numpy Pillow
+
+使用方法
+项目结构
+ic_layout_recognition/
+├── data/
+│   ├── ic_dataset.py
+├── utils/
+│   ├── transforms.py
+├── models/
+│   ├── rord.py
+├── train.py
+├── evaluate.py
+├── match.py
+├── requirements.txt
+└── README.md
+
+训练
+运行以下命令训练模型：
+python train.py --data_dir path/to/layouts --save_dir path/to/save
+
+
+--data_dir：包含 PNG 格式 IC 版图图像的目录。
+--save_dir：模型权重保存目录。训练过程使用自监督学习，通过随机旋转生成训练对，优化关键点检测和描述子生成。
+
+评估
+运行以下命令评估模型性能：
+python evaluate.py --model_path path/to/model.pth --val_dir path/to/val/images --annotations_dir path/to/val/annotations --templates path/to/templates
+
+
+--model_path：训练好的模型权重路径。
+--val_dir：验证集图像目录。
+--annotations_dir：JSON 格式的真实标注目录。
+--templates：模板图像路径列表。评估结果包括精确率、召回率和 F1 分数，基于 IoU（Intersection over Union）阈值。
+
+模板匹配
+运行以下命令进行模板匹配：
+python match.py --model_path path/to/model.pth --layout_path path/to/layout.png --template_path path/to/template.png --output_path path/to/output.png
+
+
+--layout_path：版图图像路径。
+--template_path：模板图像路径。
+--output_path：可视化结果保存路径（可选）。匹配过程使用 RoRD 模型提取关键点和描述子，通过互最近邻（MNN）匹配和 RANSAC 几何验证，生成边界框并支持多实例匹配。
+
+数据准备
+训练数据
+
+格式：PNG 格式的 IC 版图图像，从 GDSII 或 OASIS 文件光栅化。
+要求：数据集应包含多个版图图像，建议分辨率适中（如 1024x1024）。
+路径：存储在 path/to/layouts 目录中。
+
+验证数据
+
+图像：PNG 格式的验证集图像，存储在 path/to/val/images。
+注释：JSON 格式的真实标注，存储在 path/to/val/annotations，示例：{
+    "boxes": [
+        {"template": "template1.png", "x": 100, "y": 200, "width": 50, "height": 50},
+        {"template": "template2.png", "x": 300, "y": 400, "width": 60, "height": 60}
+    ]
+}
+
+
+模板：模板图像存储在 path/to/templates，文件名需与注释中的 template 字段一致。
+
+模型
+RoRD 模型基于 D2-Net 架构，使用 VGG-16 作为骨干网络。它包括：
+
+检测头：用于关键点检测，输出概率图。
+描述子头：生成旋转鲁棒的 128 维描述子，适配 IC 版图的 8 个离散旋转方向。模型通过自监督学习训练，使用随机旋转（0°~360°）生成训练对，优化检测重复性和描述子相似性。gi
+
+
+结果
+[待补充：如果有预训练模型或基准测试结果，请在此列出。例如：]
+
+预训练模型：[链接](待补充)
+验证集评估指标：精确率：X，召回率：Y，F1 分数：Z
+
+贡献
+欢迎贡献代码或提出建议！请遵循以下步骤：
+
+Fork 本仓库。
+创建新分支（git checkout -b feature/your-feature）。
+提交更改（git commit -m "Add your feature"）。
+推送到分支（git push origin feature/your-feature）。
+提交 Pull Request。
+
+许可证
+本项目采用 MIT 许可证。
+联系
+如有问题或建议，请通过 [您的电子邮件] 联系或在 GitHub 上提交 issue。
+
+AI-based Integrated Circuit Layout Recognition with RoRD
+Description
+This project implements the RoRD (Rotation-Robust Descriptors) model for integrated circuit (IC) layout recognition. RoRD is a state-of-the-art method for local feature matching that is robust to rotations, making it particularly suitable for IC layouts which can be oriented in various directions (0°, 90°, 180°, 270°, and their mirrors). The project addresses challenges such as data scarcity, geometric variability, dynamic scalability, and structural complexity through self-supervised learning and random rotation augmentation.
+The project includes:
+
+Model Implementation: The RoRD model adapted for IC layouts, using PyTorch, based on the D2-Net architecture.
+Data Loading: Custom dataset class ICLayoutDataset for loading rasterized IC layout images.
+Training Script: Self-supervised training with random rotations to ensure rotation robustness.
+Evaluation Script: Evaluates model performance on a validation set, computing precision, recall, and F1 score.
+Matching Utility: Performs template matching with the trained model, supporting multi-instance matching and visualization.
+
+Installation
+Requirements
+
+Python 3.8 or higher
+CUDA (optional, for GPU acceleration)
+
+Dependency Installation
+Install dependencies using uv:
+uv add torch torchvision opencv-python numpy Pillow
+uv lock
+uv sync
+
+Alternatively, use pip:
+pip install torch torchvision opencv-python numpy Pillow
+
+Usage
+Project Structure
+ic_layout_recognition/
+├── data/
+│   ├── ic_dataset.py
+├── utils/
+│   ├── transforms.py
+├── models/
+│   ├── rord.py
+├── train.py
+├── evaluate.py
+├── match.py
+├── requirements.txt
+└── README.md
+
+Training
+Run the following command to train the model:
+python train.py --data_dir path/to/layouts --save_dir path/to/save
+
+
+--data_dir: Directory containing PNG format IC layout images.
+--save_dir: Directory to save model weights.The training process uses self-supervised learning, generating training pairs with random rotations to optimize keypoint detection and descriptor generation.
+
+Evaluation
+Run the following command to evaluate model performance:
+python evaluate.py --model_path path/to/model.pth --val_dir path/to/val/images --annotations_dir path/to/val/annotations --templates path/to/templates
+
+
+--model_path: Path to the trained model weights.
+--val_dir: Directory containing validation images.
+--annotations_dir: Directory containing JSON format ground truth annotations.
+--templates: List of template image paths.Evaluation metrics include precision, recall, and F1 score, based on IoU (Intersection over Union) thresholds.
+
+Template Matching
+Run the following command to perform template matching:
+python match.py --model_path path/to/model.pth --layout_path path/to/layout.png --template_path path/to/template.png --output_path path/to/output.png
+
+
+--layout_path: Path to the layout image.
+--template_path: Path to the template image.
+--output_path: Path to save visualization results (optional).The matching process extracts keypoints and descriptors using the RoRD model, performs mutual nearest neighbor (MNN) matching, and applies RANSAC for geometric verification, generating bounding boxes for multiple instances.
+
+Data Preparation
+Training Data
+
+Format: PNG format IC layout images, rasterized from GDSII or OASIS files.
+Requirements: The dataset should include multiple layout images, preferably with moderate resolution (e.g., 1024x1024).
+Path: Stored in path/to/layouts.
+
+Validation Data
+
+Images: PNG format validation images, stored in path/to/val/images.
+Annotations: JSON format ground truth annotations, stored in path/to/val/annotations, example:{
+    "boxes": [
+        {"template": "template1.png", "x": 100, "y": 200, "width": 50, "height": 50},
+        {"template": "template2.png", "x": 300, "y": 400, "width": 60, "height": 60}
+    ]
+}
+
+
+Templates: Template images stored in path/to/templates, with filenames matching the template field in annotations.
+
+Model
+The RoRD model is based on the D2-Net architecture, using VGG-16 as the backbone. It includes:
+
+Detection Head: Outputs a probability map for keypoint detection.
+Descriptor Head: Generates 128-dimensional rotation-robust descriptors, tailored for the 8 discrete rotation directions in IC layouts.The model is trained using self-supervised learning with random rotations (0°~360°), optimizing for detection repeatability and descriptor similarity.
+
+Technical Comparison
+
+
+
+Feature
+U-Net
+YOLO
+Transformer (ViT)
+SuperPoint
+RoRD
+
+
+
+Core Principle
+Semantic Segmentation
+Object Detection
+Global Self-Attention
+Self-Supervised Features
+Rotation-Robust Features
+
+
+Data Requirement
+Large Pixel-Level Labels
+Large Bounding Box Labels
+Massive Pretraining Data
+Synthetic Data
+Synthetic Rotation Data
+
+
+Rotation Robustness
+Low
+Low-Medium
+Medium
+Medium-High
+Very High
+
+
+Results
+[To be added: If pre-trained models or benchmarks are available, list them here. For example:]
+
+Pre-trained model: [link]
+Validation set metrics: Precision: X, Recall: Y, F1 Score: Z
+
+Contributing
+Contributions are welcome! Please follow these steps:
+
+Fork the repository.
+Create a new branch (git checkout -b feature/your-feature).
+Commit your changes (git commit -m "Add your feature").
+Push to the branch (git push origin feature/your-feature).
+Submit a Pull Request.
+
+License
+This project is licensed under the MIT License.
+Contact
+For questions or issues, please contact [your email] or open an issue on GitHub.

data/ic_dataset.py

@@ -0,0 +1,56 @@
+import os
+from PIL import Image
+from torch.utils.data import Dataset
+import json
+
+class ICLayoutDataset(Dataset):
+    def __init__(self, image_dir, annotation_dir=None, transform=None):
+        """
+        初始化 IC 版图数据集。
+
+        参数：
+            image_dir (str): 存储 PNG 格式 IC 版图图像的目录路径。
+            annotation_dir (str, optional): 存储 JSON 格式注释文件的目录路径。
+            transform (callable, optional): 应用于图像的可选变换（如 Sobel 边缘检测）。
+        """
+        self.image_dir = image_dir
+        self.annotation_dir = annotation_dir
+        self.transform = transform
+        self.images = [f for f in os.listdir(image_dir) if f.endswith('.png')]
+        if annotation_dir:
+            self.annotations = [f.replace('.png', '.json') for f in self.images]
+        else:
+            self.annotations = [None] * len(self.images)
+
+    def __len__(self):
+        """
+        返回数据集中的图像数量。
+
+        返回：
+            int: 数据集大小。
+        """
+        return len(self.images)
+
+    def __getitem__(self, idx):
+        """
+        获取指定索引的图像和注释。
+
+        参数：
+            idx (int): 图像索引。
+
+        返回：
+            tuple: (image, annotation)，image 为处理后的图像，annotation 为注释字典或空字典。
+        """
+        img_path = os.path.join(self.image_dir, self.images[idx])
+        image = Image.open(img_path).convert('L')  # 转换为灰度图
+        if self.transform:
+            image = self.transform(image)
+        
+        annotation = {}
+        if self.annotation_dir and self.annotations[idx]:
+            ann_path = os.path.join(self.annotation_dir, self.annotations[idx])
+            if os.path.exists(ann_path):
+                with open(ann_path, 'r') as f:
+                    annotation = json.load(f)
+        
+        return image, annotation

evaluate.py

@@ -0,0 +1,123 @@
+from models.rord import RoRD
+from data.ic_dataset import ICLayoutDataset
+from utils.transforms import SobelTransform
+from match import match_template_to_layout
+import torch
+from torchvision import transforms
+import json
+import os
+from PIL import Image
+
+def compute_iou(box1, box2):
+    x1, y1, w1, h1 = box1['x'], box1['y'], box1['width'], box1['height']
+    x2, y2, w2, h2 = box2['x'], box2['y'], box2['width'], box2['height']
+    
+    inter_x1 = max(x1, x2)
+    inter_y1 = max(y1, y2)
+    inter_x2 = min(x1 + w1, x2 + w2)
+    inter_y2 = min(y1 + h1, y2 + h2)
+    
+    inter_area = max(0, inter_x2 - inter_x1) * max(0, inter_y2 - inter_y1)
+    
+    box1_area = w1 * h1
+    box2_area = w2 * h2
+    union_area = box1_area + box2_area - inter_area
+    
+    iou = inter_area / union_area if union_area > 0 else 0
+    return iou
+
+def evaluate(model, val_dataset, templates, iou_threshold=0.5):
+    model.eval()
+    all_true_positives = 0
+    all_false_positives = 0
+    all_false_negatives = 0
+
+    for layout_idx in range(len(val_dataset)):
+        layout_image, annotation = val_dataset[layout_idx]
+        # layout_image is [3, H, W]
+        layout_tensor = layout_image.unsqueeze(0).cuda()  # [1, 3, H, W]
+        
+        # 假设 annotation 是 {"boxes": [{"template": "template1.png", "x": x, "y": y, "width": w, "height": h}, ...]}
+        gt_boxes_by_template = {}
+        for box in annotation.get('boxes', []):
+            template_name = box['template']
+            if template_name not in gt_boxes_by_template:
+                gt_boxes_by_template[template_name] = []
+            gt_boxes_by_template[template_name].append(box)
+
+        for template_path in templates:
+            template_name = os.path.basename(template_path)
+            template_image = Image.open(template_path).convert('L')
+            template_tensor = transform(template_image).unsqueeze(0).cuda()  # [1, 3, H, W]
+
+            # 执行匹配
+            detected_bboxes = match_template_to_layout(model, layout_tensor, template_tensor)
+
+            # 获取当前模板的 gt_boxes
+            gt_boxes = gt_boxes_by_template.get(template_name, [])
+
+            # 初始化已分配的 gt_box 索引
+            assigned_gt = set()
+
+            for det_box in detected_bboxes:
+                best_iou = 0
+                best_gt_idx = -1
+                for idx, gt_box in enumerate(gt_boxes):
+                    if idx in assigned_gt:
+                        continue
+                    iou = compute_iou(det_box, gt_box)
+                    if iou > best_iou:
+                        best_iou = iou
+                        best_gt_idx = idx
+                if best_iou > iou_threshold and best_gt_idx != -1:
+                    all_true_positives += 1
+                    assigned_gt.add(best_gt_idx)
+                else:
+                    all_false_positives += 1
+
+            # 计算 FN：未分配的 gt_box
+            for idx in range(len(gt_boxes)):
+                if idx not in assigned_gt:
+                    all_false_negatives += 1
+
+    # 计算评估指标
+    precision = all_true_positives / (all_true_positives + all_false_positives) if (all_true_positives + all_false_positives) > 0 else 0
+    recall = all_true_positives / (all_true_positives + all_false_negatives) if (all_true_positives + all_false_negatives) > 0 else 0
+    f1 = 2 * (precision * recall) / (precision + recall) if (precision + recall) > 0 else 0
+
+    return {
+        'precision': precision,
+        'recall': recall,
+        'f1': f1
+    }
+
+if __name__ == "__main__":
+    # 设置变换
+    transform = transforms.Compose([
+        SobelTransform(),
+        transforms.ToTensor(),
+        transforms.Lambda(lambda x: x.repeat(3, 1, 1)),  # [1, H, W] -> [3, H, W]
+        transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])
+    ])
+
+    # 加载模型
+    model = RoRD().cuda()
+    model.load_state_dict(torch.load('path/to/weights.pth'))
+    model.eval()
+
+    # 定义验证数据集
+    val_dataset = ICLayoutDataset(
+        image_dir='path/to/val/images',
+        annotation_dir='path/to/val/annotations',
+        transform=transform
+    )
+
+    # 定义模板列表
+    templates = ['path/to/templates/template1.png', 'path/to/templates/template2.png']  # 替换为实际模板路径
+
+    # 评估模型
+    results = evaluate(model, val_dataset, templates)
+    print("评估结果：")
+    print(f"精确率: {results['precision']:.4f}")
+    print(f"召回率: {results['recall']:.4f}")
+    print(f"F1 分数: {results['f1']:.4f}")

main.py

@@ -0,0 +1,6 @@
+def main():
+    print("Hello from rord-layout-recognation!")
+
+
+if __name__ == "__main__":
+    main()

match.py

@@ -0,0 +1,199 @@
+import torch
+import torch.nn.functional as F
+from models.rord import RoRD
+from torchvision import transforms
+from utils.transforms import SobelTransform
+import numpy as np
+import cv2
+from PIL import Image
+
+def extract_keypoints_and_descriptors(model, image):
+    """
+    从 RoRD 模型中提取关键点和描述子。
+
+    参数：
+        model (RoRD): RoRD 模型。
+        image (torch.Tensor): 输入图像张量，形状为 [1, 1, H, W]。
+
+    返回：
+        tuple: (keypoints_input, descriptors)
+            - keypoints_input: [N, 2] float tensor，关键点在输入图像中的坐标。
+            - descriptors: [N, 128] float tensor，L2 归一化的描述子。
+    """
+    with torch.no_grad():
+        detection_map, _, desc_rord = model(image)
+        desc = desc_rord  # 使用 RoRD 描述子头
+
+        # 从检测图中提取关键点
+        thresh = 0.5
+        binary_map = (detection_map > thresh).float()
+        coords = torch.nonzero(binary_map[0, 0] > thresh).float()  # [N, 2]，每个行是 (i_d, j_d)
+        keypoints_input = coords * 16.0  # 将特征图坐标映射到输入图像坐标（stride=16）
+
+        # 从描述子图中提取描述子
+        # detection_map 的形状为 [1, 1, H/16, W/16]，desc 的形状为 [1, 128, H/8, W/8]
+        # 将 detection_map 的坐标映射到 desc 的坐标：(i_d * 2, j_d * 2)
+        keypoints_desc = (coords * 2).long()  # [N, 2]，整数坐标
+        H_desc, W_desc = desc.shape[2], desc.shape[3]
+        mask = (keypoints_desc[:, 0] < H_desc) & (keypoints_desc[:, 1] < W_desc)
+        keypoints_desc = keypoints_desc[mask]
+        keypoints_input = keypoints_input[mask]
+
+        # 提取描述子
+        descriptors = desc[0, :, keypoints_desc[:, 0], keypoints_desc[:, 1]].T  # [N, 128]
+
+        # L2 归一化描述子
+        descriptors = F.normalize(descriptors, p=2, dim=1)
+
+        return keypoints_input, descriptors
+
+def mutual_nearest_neighbor(template_descs, layout_descs):
+    """
+    使用互最近邻（MNN）找到模板和版图之间的匹配。
+
+    参数：
+        template_descs (torch.Tensor): 模板描述子，形状为 [M, 128]。
+        layout_descs (torch.Tensor): 版图描述子，形状为 [N, 128]。
+
+    返回：
+        list: [(i_template, i_layout)]，互最近邻匹配对的列表。
+    """
+    M, N = template_descs.size(0), layout_descs.size(0)
+    if M == 0 or N == 0:
+        return []
+    similarity_matrix = template_descs @ layout_descs.T  # [M, N]，点积矩阵
+
+    # 找到每个模板描述子的最近邻
+    nn_template_to_layout = torch.argmax(similarity_matrix, dim=1)  # [M]
+
+    # 找到每个版图描述子的最近邻
+    nn_layout_to_template = torch.argmax(similarity_matrix, dim=0)  # [N]
+
+    # 找到互最近邻
+    mutual_matches = []
+    for i in range(M):
+        j = nn_template_to_layout[i]
+        if nn_layout_to_template[j] == i:
+            mutual_matches.append((i.item(), j.item()))
+
+    return mutual_matches
+
+def ransac_filter(matches, template_kps, layout_kps):
+    """
+    使用 RANSAC 对匹配进行几何验证，并返回内点。
+
+    参数：
+        matches (list): [(i_template, i_layout)]，匹配对列表。
+        template_kps (torch.Tensor): 模板关键点，形状为 [M, 2]。
+        layout_kps (torch.Tensor): 版图关键点，形状为 [N, 2]。
+
+    返回：
+        tuple: (inlier_matches, num_inliers)
+            - inlier_matches: [(i_template, i_layout)]，内点匹配对。
+            - num_inliers: int，内点数量。
+    """
+    src_pts = np.array([template_kps[i].cpu().numpy() for i, _ in matches])
+    dst_pts = np.array([layout_kps[j].cpu().numpy() for _, j in matches])
+
+    if len(src_pts) < 4:
+        return [], 0
+
+    try:
+        H, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC, ransacReprojThreshold=5.0)
+        if H is None:
+            return [], 0
+        inliers = mask.ravel() > 0
+        num_inliers = np.sum(inliers)
+        inlier_matches = [matches[k] for k in range(len(matches)) if inliers[k]]
+        return inlier_matches, num_inliers
+    except cv2.error:
+        return [], 0
+
+def match_template_to_layout(model, layout_image, template_image):
+    """
+    使用 RoRD 模型执行模板匹配，迭代找到所有匹配并屏蔽已匹配区域。
+
+    参数：
+        model (RoRD): RoRD 模型。
+        layout_image (torch.Tensor): 版图图像张量，形状为 [1, 1, H_layout, W_layout]。
+        template_image (torch.Tensor): 模板图像张量，形状为 [1, 1, H_template, W_template]。
+
+    返回：
+        list: [{'x': x_min, 'y': y_min, 'width': w, 'height': h}]，所有检测到的边框。
+    """
+    # 提取版图和模板的关键点和描述子
+    layout_kps, layout_descs = extract_keypoints_and_descriptors(model, layout_image)
+    template_kps, template_descs = extract_keypoints_and_descriptors(model, template_image)
+
+    # 初始化活动版图关键点掩码
+    active_layout = torch.ones(len(layout_kps), dtype=bool)
+
+    bboxes = []
+    while True:
+        # 获取当前活动的版图关键点和描述子
+        current_layout_kps = layout_kps[active_layout]
+        current_layout_descs = layout_descs[active_layout]
+
+        if len(current_layout_descs) == 0:
+            break
+
+        # MNN 匹配
+        matches = mutual_nearest_neighbor(template_descs, current_layout_descs)
+
+        if len(matches) == 0:
+            break
+
+        # 将当前版图索引映射回原始版图索引
+        active_indices = torch.nonzero(active_layout).squeeze(1)
+        matches_original = [(i_template, active_indices[i_layout].item()) for i_template, i_layout in matches]
+
+        # RANSAC 过滤
+        inlier_matches, num_inliers = ransac_filter(matches_original, template_kps, layout_kps)
+
+        if num_inliers > 10:  # 设置内点阈值
+            # 获取内点在版图中的关键点
+            inlier_layout_kps = [layout_kps[j].cpu().numpy() for _, j in inlier_matches]
+            inlier_layout_kps = np.array(inlier_layout_kps)
+
+            # 计算边框
+            x_min = int(inlier_layout_kps[:, 0].min())
+            y_min = int(inlier_layout_kps[:, 1].min())
+            x_max = int(inlier_layout_kps[:, 0].max())
+            y_max = int(inlier_layout_kps[:, 1].max())
+            bboxes.append({'x': x_min, 'y': y_min, 'width': x_max - x_min, 'height': y_max - y_min})
+
+            # 屏蔽内点
+            for _, j in inlier_matches:
+                active_layout[j] = False
+        else:
+            break
+
+    return bboxes
+
+if __name__ == "__main__":
+    # 设置变换
+    transform = transforms.Compose([
+        SobelTransform(),
+        transforms.ToTensor(),
+        transforms.Normalize(mean=[0.5], std=[0.5])
+    ])
+
+    # 加载模型
+    model = RoRD().cuda()
+    model.load_state_dict(torch.load('path/to/weights.pth'))
+    model.eval()
+
+    # 加载版图和模板图像
+    layout_image = Image.open('path/to/layout.png').convert('L')
+    layout_tensor = transform(layout_image).unsqueeze(0).cuda()
+
+    template_image = Image.open('path/to/template.png').convert('L')
+    template_tensor = transform(template_image).unsqueeze(0).cuda()
+
+    # 执行匹配
+    detected_bboxes = match_template_to_layout(model, layout_tensor, template_tensor)
+
+    # 打印检测到的边框
+    print("检测到的边框：")
+    for bbox in detected_bboxes:
+        print(bbox)

models/rord.py

@@ -0,0 +1,47 @@
+import torch
+import torch.nn as nn
+from torchvision import models
+
+class RoRD(nn.Module):
+    def __init__(self):
+        super(RoRD, self).__init__()
+        # 检测骨干网络：VGG-16 直到 relu5_3（层 0 到 29）
+        self.backbone_det = models.vgg16(pretrained=True).features[:30]
+        # 描述骨干网络：VGG-16 直到 relu4_3（层 0 到 22）
+        self.backbone_desc = models.vgg16(pretrained=True).features[:23]
+        
+        # 检测头：输出关键点概率图
+        self.detection_head = nn.Sequential(
+            nn.Conv2d(512, 256, kernel_size=3, padding=1),
+            nn.ReLU(inplace=True),
+            nn.Conv2d(256, 1, kernel_size=1),
+            nn.Sigmoid()
+        )
+        
+        # 普通描述子头（D2-Net 风格）
+        self.descriptor_head_vanilla = nn.Sequential(
+            nn.Conv2d(512, 256, kernel_size=3, padding=1),
+            nn.ReLU(inplace=True),
+            nn.Conv2d(256, 128, kernel_size=1),
+            nn.InstanceNorm2d(128)
+        )
+        
+        # RoRD 描述子头（旋转鲁棒）
+        self.descriptor_head_rord = nn.Sequential(
+            nn.Conv2d(512, 256, kernel_size=3, padding=1),
+            nn.ReLU(inplace=True),
+            nn.Conv2d(256, 128, kernel_size=1),
+            nn.InstanceNorm2d(128)
+        )
+
+    def forward(self, x):
+        # 检测分支
+        features_det = self.backbone_det(x)
+        detection = self.detection_head(features_det)
+        
+        # 描述分支
+        features_desc = self.backbone_desc(x)
+        desc_vanilla = self.descriptor_head_vanilla(features_desc)
+        desc_rord = self.descriptor_head_rord(features_desc)
+        
+        return detection, desc_vanilla, desc_rord

pyproject.toml

@@ -0,0 +1,17 @@
+[project]
+name = "rord-layout-recognation"
+version = "0.1.0"
+description = "Add your description here"
+readme = "README.md"
+requires-python = ">=3.12"
+dependencies = [
+    "numpy>=2.3.0",
+    "opencv-python>=4.11.0.86",
+    "pillow>=11.2.1",
+    "torch>=2.7.1",
+    "torchvision>=0.22.1",
+]
+
+[[tool.uv.index]]
+url = "https://pypi.tuna.tsinghua.edu.cn/simple"
+default = true

train.py

@@ -0,0 +1,236 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from torch.utils.data import Dataset, DataLoader
+from torchvision import transforms
+from PIL import Image
+import numpy as np
+import cv2
+import os
+from models.rord import RoRD
+
+# 数据集类：生成随机旋转的训练对
+class ICLayoutTrainingDataset(Dataset):
+    def __init__(self, image_dir, patch_size=256, transform=None):
+        """
+        初始化 IC 版图训练数据集。
+
+        参数：
+            image_dir (str): 存储 PNG 格式 IC 版图图像的目录路径。
+            patch_size (int): 裁剪的 patch 大小（默认 256x256）。
+            transform (callable, optional): 应用于图像的变换。
+        """
+        self.image_dir = image_dir
+        self.image_paths = [os.path.join(image_dir, f) for f in os.listdir(image_dir) if f.endswith('.png')]
+        self.patch_size = patch_size
+        self.transform = transform
+
+    def __len__(self):
+        """
+        返回数据集中的图像数量。
+
+        返回：
+            int: 数据集大小。
+        """
+        return len(self.image_paths)
+
+    def __getitem__(self, index):
+        """
+        获取指定索引的训练对（原始 patch、旋转 patch、Homography 矩阵）。
+
+        参数：
+            index (int): 图像索引。
+
+        返回：
+            tuple: (patch, rotated_patch, H_tensor)
+                - patch: 原始 patch 张量。
+                - rotated_patch: 旋转后的 patch 张量。
+                - H_tensor: Homography 矩阵张量。
+        """
+        img_path = self.image_paths[index]
+        image = Image.open(img_path).convert('L')  # 灰度图像
+
+        # 获取图像大小
+        W, H = image.size
+
+        # 随机选择裁剪的左上角坐标
+        x = np.random.randint(0, W - self.patch_size + 1)
+        y = np.random.randint(0, H - self.patch_size + 1)
+        patch = image.crop((x, y, x + self.patch_size, y + self.patch_size))
+
+        # 转换为 NumPy 数组
+        patch_np = np.array(patch)
+
+        # 随机旋转角度（0°~360°）
+        theta = np.random.uniform(0, 360)
+        theta_rad = np.deg2rad(theta)
+        cos_theta = np.cos(theta_rad)
+        sin_theta = np.sin(theta_rad)
+
+        # 计算旋转中心（patch 的中心）
+        cx = self.patch_size / 2.0
+        cy = self.patch_size / 2.0
+
+        # 计算旋转的齐次矩阵（Homography）
+        H = np.array([
+            [cos_theta, -sin_theta, cx * (1 - cos_theta) + cy * sin_theta],
+            [sin_theta,  cos_theta, cy * (1 - cos_theta) - cx * sin_theta],
+            [0,          0,          1]
+        ], dtype=np.float32)
+
+        # 应用旋转到 patch
+        rotated_patch_np = cv2.warpPerspective(patch_np, H, (self.patch_size, self.patch_size))
+
+        # 转换回 PIL Image
+        rotated_patch = Image.fromarray(rotated_patch_np)
+
+        # 应用变换
+        if self.transform:
+            patch = self.transform(patch)
+            rotated_patch = self.transform(rotated_patch)
+
+        # 转换 H 为张量
+        H_tensor = torch.from_numpy(H).float()
+
+        return patch, rotated_patch, H_tensor
+
+# 特征图变换函数
+def warp_feature_map(feature_map, H_inv):
+    """
+    使用逆 Homography 矩阵变换特征图。
+
+    参数：
+        feature_map (torch.Tensor): 输入特征图，形状为 [B, C, H, W]。
+        H_inv (torch.Tensor): 逆 Homography 矩阵，形状为 [B, 3, 3]。
+
+    返回：
+        torch.Tensor: 变换后的特征图，形状为 [B, C, H, W]。
+    """
+    B, C, H, W = feature_map.size()
+    # 生成网格
+    grid_y, grid_x = torch.meshgrid(
+        torch.linspace(-1, 1, H, device=feature_map.device),
+        torch.linspace(-1, 1, W, device=feature_map.device),
+        indexing='ij'
+    )
+    grid = torch.stack((grid_x, grid_y, torch.ones_like(grid_x)), dim=-1)  # [H, W, 3]
+    grid = grid.unsqueeze(0).expand(B, H, W, 3)  # [B, H, W, 3]
+
+    # 将网格转换为齐次坐标并应用 H_inv
+    grid_flat = grid.view(B, -1, 3)  # [B, H*W, 3]
+    grid_transformed = torch.bmm(grid_flat, H_inv.transpose(1, 2))  # [B, H*W, 3]
+    grid_transformed = grid_transformed.view(B, H, W, 3)  # [B, H, W, 3]
+    grid_transformed = grid_transformed[..., :2] / (grid_transformed[..., 2:3] + 1e-8)  # [B, H, W, 2]
+
+    # 使用 grid_sample 进行变换
+    warped_feature = F.grid_sample(feature_map, grid_transformed, align_corners=True)
+    return warped_feature
+
+# 检测损失函数
+def compute_detection_loss(det_original, det_rotated, H):
+    """
+    计算检测损失（MSE），比较原始检测图与旋转检测图（逆变换后）。
+
+    参数：
+        det_original (torch.Tensor): 原始图像的检测图，形状为 [B, 1, H, W]。
+        det_rotated (torch.Tensor): 旋转图像的检测图，形状为 [B, 1, H, W]。
+        H (torch.Tensor): Homography 矩阵，形状为 [B, 3, 3]。
+
+    返回：
+        torch.Tensor: 检测损失。
+    """
+    H_inv = torch.inverse(H)  # 计算逆 Homography
+    warped_det_rotated = warp_feature_map(det_rotated, H_inv)
+    return F.mse_loss(det_original, warped_det_rotated)
+
+# 描述子损失函数
+def compute_description_loss(desc_original, desc_rotated, H, margin=1.0):
+    """
+    计算描述子损失（三元组损失），基于对应点的描述子。
+
+    参数：
+        desc_original (torch.Tensor): 原始图像的描述子图，形状为 [B, 128, H, W]。
+        desc_rotated (torch.Tensor): 旋转图像的描述子图，形状为 [B, 128, H, W]。
+        H (torch.Tensor): Homography 矩阵，形状为 [B, 3, 3]。
+        margin (float): 三元组损失的边距。
+
+    返回：
+        torch.Tensor: 描述子损失。
+    """
+    B, C, H, W = desc_original.size()
+    # 随机选择锚点（anchor）
+    num_samples = min(100, H * W)  # 每张图像采样 100 个点
+    idx = torch.randint(0, H * W, (B, num_samples), device=desc_original.device)
+    idx_y = idx // W
+    idx_x = idx % W
+    coords = torch.stack((idx_x.float(), idx_y.float()), dim=-1)  # [B, num_samples, 2]
+
+    # 转换为齐次坐标
+    coords_hom = torch.cat((coords, torch.ones(B, num_samples, 1, device=coords.device)), dim=-1)  # [B, num_samples, 3]
+    coords_transformed = torch.bmm(coords_hom, H.transpose(1, 2))  # [B, num_samples, 3]
+    coords_transformed = coords_transformed[..., :2] / (coords_transformed[..., 2:3] + 1e-8)  # [B, num_samples, 2]
+
+    # 归一化到 [-1, 1] 用于 grid_sample
+    coords_transformed = coords_transformed / torch.tensor([W/2, H/2], device=coords.device) - 1
+
+    # 提取锚点和正样本描述子
+    anchor = desc_original.view(B, C, -1)[:, :, idx.view(-1)]  # [B, 128, num_samples]
+    positive = F.grid_sample(desc_rotated, coords_transformed.unsqueeze(2), align_corners=True).squeeze(3)  # [B, 128, num_samples]
+
+    # 随机选择负样本
+    neg_idx = torch.randint(0, H * W, (B, num_samples), device=desc_original.device)
+    negative = desc_rotated.view(B, C, -1)[:, :, neg_idx.view(-1)]  # [B, 128, num_samples]
+
+    # 三元组损失
+    triplet_loss = nn.TripletMarginLoss(margin=margin, p=2)
+    loss = triplet_loss(anchor.transpose(1, 2), positive.transpose(1, 2), negative.transpose(1, 2))
+    return loss
+
+# 定义变换
+transform = transforms.Compose([
+    transforms.ToTensor(),  # (1, 256, 256)
+    transforms.Lambda(lambda x: x.repeat(3, 1, 1)),  # (3, 256, 256)
+    transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])
+])
+
+# 创建数据集和 DataLoader
+dataset = ICLayoutTrainingDataset('path/to/layouts', patch_size=256, transform=transform)
+dataloader = DataLoader(dataset, batch_size=4, shuffle=True, num_workers=4)
+
+# 定义模型
+model = RoRD().cuda()
+
+# 定义优化器
+optimizer = torch.optim.Adam(model.parameters(), lr=1e-4)
+
+# 训练循环
+num_epochs = 10
+for epoch in range(num_epochs):
+    model.train()
+    total_loss = 0
+    for batch in dataloader:
+        original, rotated, H = batch
+        original = original.cuda()
+        rotated = rotated.cuda()
+        H = H.cuda()
+
+        # 前向传播
+        det_original, _, desc_rord_original = model(original)
+        det_rotated, _, desc_rord_rotated = model(rotated)
+
+        # 计算损失
+        detection_loss = compute_detection_loss(det_original, det_rotated, H)
+        description_loss = compute_description_loss(desc_rord_original, desc_rord_rotated, H)
+        total_loss_batch = detection_loss + description_loss
+
+        # 反向传播
+        optimizer.zero_grad()
+        total_loss_batch.backward()
+        optimizer.step()
+
+        total_loss += total_loss_batch.item()
+
+    print(f"Epoch {epoch+1}/{num_epochs}, Loss: {total_loss / len(dataloader):.4f}")
+
+# 保存模型
+torch.save(model.state_dict(), 'path/to/save/model.pth')

utils/transforms.py

@@ -0,0 +1,21 @@
+import cv2
+import numpy as np
+from PIL import Image
+
+class SobelTransform:
+    def __call__(self, image):
+        """
+        应用 Sobel 边缘检测，增强 IC 版图的几何边界。
+
+        参数：
+            image (PIL.Image): 输入图像（灰度图）。
+
+        返回：
+            PIL.Image: 边缘增强后的图像。
+        """
+        img_np = np.array(image)
+        sobelx = cv2.Sobel(img_np, cv2.CV_64F, 1, 0, ksize=3)
+        sobely = cv2.Sobel(img_np, cv2.CV_64F, 0, 1, ksize=3)
+        sobel = np.hypot(sobelx, sobely)
+        sobel = np.clip(sobel, 0, 255).astype(np.uint8)
+        return Image.fromarray(sobel)

uv.lock

@@ -0,0 +1,457 @@
+version = 1
+revision = 2
+requires-python = ">=3.12"
+resolution-markers = [
+    "sys_platform == 'darwin'",
+    "platform_machine == 'aarch64' and sys_platform == 'linux'",
+    "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')",
+]
+
+[[package]]
+name = "filelock"
+version = "3.18.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/0a/10/c23352565a6544bdc5353e0b15fc1c563352101f30e24bf500207a54df9a/filelock-3.18.0.tar.gz", hash = "sha256:adbc88eabb99d2fec8c9c1b229b171f18afa655400173ddc653d5d01501fb9f2", size = 18075, upload-time = "2025-03-14T07:11:40.47Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/4d/36/2a115987e2d8c300a974597416d9de88f2444426de9571f4b59b2cca3acc/filelock-3.18.0-py3-none-any.whl", hash = "sha256:c401f4f8377c4464e6db25fff06205fd89bdd83b65eb0488ed1b160f780e21de", size = 16215, upload-time = "2025-03-14T07:11:39.145Z" },
+]
+
+[[package]]
+name = "fsspec"
+version = "2025.5.1"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/00/f7/27f15d41f0ed38e8fcc488584b57e902b331da7f7c6dcda53721b15838fc/fsspec-2025.5.1.tar.gz", hash = "sha256:2e55e47a540b91843b755e83ded97c6e897fa0942b11490113f09e9c443c2475", size = 303033, upload-time = "2025-05-24T12:03:23.792Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/bb/61/78c7b3851add1481b048b5fdc29067397a1784e2910592bc81bb3f608635/fsspec-2025.5.1-py3-none-any.whl", hash = "sha256:24d3a2e663d5fc735ab256263c4075f374a174c3410c0b25e5bd1970bceaa462", size = 199052, upload-time = "2025-05-24T12:03:21.66Z" },
+]
+
+[[package]]
+name = "jinja2"
+version = "3.1.6"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "markupsafe" },
+]
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/df/bf/f7da0350254c0ed7c72f3e33cef02e048281fec7ecec5f032d4aac52226b/jinja2-3.1.6.tar.gz", hash = "sha256:0137fb05990d35f1275a587e9aee6d56da821fc83491a0fb838183be43f66d6d", size = 245115, upload-time = "2025-03-05T20:05:02.478Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/62/a1/3d680cbfd5f4b8f15abc1d571870c5fc3e594bb582bc3b64ea099db13e56/jinja2-3.1.6-py3-none-any.whl", hash = "sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67", size = 134899, upload-time = "2025-03-05T20:05:00.369Z" },
+]
+
+[[package]]
+name = "markupsafe"
+version = "3.0.2"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b2/97/5d42485e71dfc078108a86d6de8fa46db44a1a9295e89c5d6d4a06e23a62/markupsafe-3.0.2.tar.gz", hash = "sha256:ee55d3edf80167e48ea11a923c7386f4669df67d7994554387f84e7d8b0a2bf0", size = 20537, upload-time = "2024-10-18T15:21:54.129Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/22/09/d1f21434c97fc42f09d290cbb6350d44eb12f09cc62c9476effdb33a18aa/MarkupSafe-3.0.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:9778bd8ab0a994ebf6f84c2b949e65736d5575320a17ae8984a77fab08db94cf", size = 14274, upload-time = "2024-10-18T15:21:13.777Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/6b/b0/18f76bba336fa5aecf79d45dcd6c806c280ec44538b3c13671d49099fdd0/MarkupSafe-3.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:846ade7b71e3536c4e56b386c2a47adf5741d2d8b94ec9dc3e92e5e1ee1e2225", size = 12348, upload-time = "2024-10-18T15:21:14.822Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/e0/25/dd5c0f6ac1311e9b40f4af06c78efde0f3b5cbf02502f8ef9501294c425b/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1c99d261bd2d5f6b59325c92c73df481e05e57f19837bdca8413b9eac4bd8028", size = 24149, upload-time = "2024-10-18T15:21:15.642Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f3/f0/89e7aadfb3749d0f52234a0c8c7867877876e0a20b60e2188e9850794c17/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e17c96c14e19278594aa4841ec148115f9c7615a47382ecb6b82bd8fea3ab0c8", size = 23118, upload-time = "2024-10-18T15:21:17.133Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/d5/da/f2eeb64c723f5e3777bc081da884b414671982008c47dcc1873d81f625b6/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:88416bd1e65dcea10bc7569faacb2c20ce071dd1f87539ca2ab364bf6231393c", size = 22993, upload-time = "2024-10-18T15:21:18.064Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/da/0e/1f32af846df486dce7c227fe0f2398dc7e2e51d4a370508281f3c1c5cddc/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2181e67807fc2fa785d0592dc2d6206c019b9502410671cc905d132a92866557", size = 24178, upload-time = "2024-10-18T15:21:18.859Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/c4/f6/bb3ca0532de8086cbff5f06d137064c8410d10779c4c127e0e47d17c0b71/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:52305740fe773d09cffb16f8ed0427942901f00adedac82ec8b67752f58a1b22", size = 23319, upload-time = "2024-10-18T15:21:19.671Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a2/82/8be4c96ffee03c5b4a034e60a31294daf481e12c7c43ab8e34a1453ee48b/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ad10d3ded218f1039f11a75f8091880239651b52e9bb592ca27de44eed242a48", size = 23352, upload-time = "2024-10-18T15:21:20.971Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/51/ae/97827349d3fcffee7e184bdf7f41cd6b88d9919c80f0263ba7acd1bbcb18/MarkupSafe-3.0.2-cp312-cp312-win32.whl", hash = "sha256:0f4ca02bea9a23221c0182836703cbf8930c5e9454bacce27e767509fa286a30", size = 15097, upload-time = "2024-10-18T15:21:22.646Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/c1/80/a61f99dc3a936413c3ee4e1eecac96c0da5ed07ad56fd975f1a9da5bc630/MarkupSafe-3.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:8e06879fc22a25ca47312fbe7c8264eb0b662f6db27cb2d3bbbc74b1df4b9b87", size = 15601, upload-time = "2024-10-18T15:21:23.499Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/83/0e/67eb10a7ecc77a0c2bbe2b0235765b98d164d81600746914bebada795e97/MarkupSafe-3.0.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ba9527cdd4c926ed0760bc301f6728ef34d841f405abf9d4f959c478421e4efd", size = 14274, upload-time = "2024-10-18T15:21:24.577Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/2b/6d/9409f3684d3335375d04e5f05744dfe7e9f120062c9857df4ab490a1031a/MarkupSafe-3.0.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f8b3d067f2e40fe93e1ccdd6b2e1d16c43140e76f02fb1319a05cf2b79d99430", size = 12352, upload-time = "2024-10-18T15:21:25.382Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/d2/f5/6eadfcd3885ea85fe2a7c128315cc1bb7241e1987443d78c8fe712d03091/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:569511d3b58c8791ab4c2e1285575265991e6d8f8700c7be0e88f86cb0672094", size = 24122, upload-time = "2024-10-18T15:21:26.199Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/0c/91/96cf928db8236f1bfab6ce15ad070dfdd02ed88261c2afafd4b43575e9e9/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:15ab75ef81add55874e7ab7055e9c397312385bd9ced94920f2802310c930396", size = 23085, upload-time = "2024-10-18T15:21:27.029Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/c2/cf/c9d56af24d56ea04daae7ac0940232d31d5a8354f2b457c6d856b2057d69/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f3818cb119498c0678015754eba762e0d61e5b52d34c8b13d770f0719f7b1d79", size = 22978, upload-time = "2024-10-18T15:21:27.846Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/2a/9f/8619835cd6a711d6272d62abb78c033bda638fdc54c4e7f4272cf1c0962b/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:cdb82a876c47801bb54a690c5ae105a46b392ac6099881cdfb9f6e95e4014c6a", size = 24208, upload-time = "2024-10-18T15:21:28.744Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f9/bf/176950a1792b2cd2102b8ffeb5133e1ed984547b75db47c25a67d3359f77/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:cabc348d87e913db6ab4aa100f01b08f481097838bdddf7c7a84b7575b7309ca", size = 23357, upload-time = "2024-10-18T15:21:29.545Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/ce/4f/9a02c1d335caabe5c4efb90e1b6e8ee944aa245c1aaaab8e8a618987d816/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:444dcda765c8a838eaae23112db52f1efaf750daddb2d9ca300bcae1039adc5c", size = 23344, upload-time = "2024-10-18T15:21:30.366Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/ee/55/c271b57db36f748f0e04a759ace9f8f759ccf22b4960c270c78a394f58be/MarkupSafe-3.0.2-cp313-cp313-win32.whl", hash = "sha256:bcf3e58998965654fdaff38e58584d8937aa3096ab5354d493c77d1fdd66d7a1", size = 15101, upload-time = "2024-10-18T15:21:31.207Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/29/88/07df22d2dd4df40aba9f3e402e6dc1b8ee86297dddbad4872bd5e7b0094f/MarkupSafe-3.0.2-cp313-cp313-win_amd64.whl", hash = "sha256:e6a2a455bd412959b57a172ce6328d2dd1f01cb2135efda2e4576e8a23fa3b0f", size = 15603, upload-time = "2024-10-18T15:21:32.032Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/62/6a/8b89d24db2d32d433dffcd6a8779159da109842434f1dd2f6e71f32f738c/MarkupSafe-3.0.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:b5a6b3ada725cea8a5e634536b1b01c30bcdcd7f9c6fff4151548d5bf6b3a36c", size = 14510, upload-time = "2024-10-18T15:21:33.625Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/7a/06/a10f955f70a2e5a9bf78d11a161029d278eeacbd35ef806c3fd17b13060d/MarkupSafe-3.0.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:a904af0a6162c73e3edcb969eeeb53a63ceeb5d8cf642fade7d39e7963a22ddb", size = 12486, upload-time = "2024-10-18T15:21:34.611Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/34/cf/65d4a571869a1a9078198ca28f39fba5fbb910f952f9dbc5220afff9f5e6/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4aa4e5faecf353ed117801a068ebab7b7e09ffb6e1d5e412dc852e0da018126c", size = 25480, upload-time = "2024-10-18T15:21:35.398Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/0c/e3/90e9651924c430b885468b56b3d597cabf6d72be4b24a0acd1fa0e12af67/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0ef13eaeee5b615fb07c9a7dadb38eac06a0608b41570d8ade51c56539e509d", size = 23914, upload-time = "2024-10-18T15:21:36.231Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/66/8c/6c7cf61f95d63bb866db39085150df1f2a5bd3335298f14a66b48e92659c/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d16a81a06776313e817c951135cf7340a3e91e8c1ff2fac444cfd75fffa04afe", size = 23796, upload-time = "2024-10-18T15:21:37.073Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/bb/35/cbe9238ec3f47ac9a7c8b3df7a808e7cb50fe149dc7039f5f454b3fba218/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:6381026f158fdb7c72a168278597a5e3a5222e83ea18f543112b2662a9b699c5", size = 25473, upload-time = "2024-10-18T15:21:37.932Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/e6/32/7621a4382488aa283cc05e8984a9c219abad3bca087be9ec77e89939ded9/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:3d79d162e7be8f996986c064d1c7c817f6df3a77fe3d6859f6f9e7be4b8c213a", size = 24114, upload-time = "2024-10-18T15:21:39.799Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/0d/80/0985960e4b89922cb5a0bac0ed39c5b96cbc1a536a99f30e8c220a996ed9/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:131a3c7689c85f5ad20f9f6fb1b866f402c445b220c19fe4308c0b147ccd2ad9", size = 24098, upload-time = "2024-10-18T15:21:40.813Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/82/78/fedb03c7d5380df2427038ec8d973587e90561b2d90cd472ce9254cf348b/MarkupSafe-3.0.2-cp313-cp313t-win32.whl", hash = "sha256:ba8062ed2cf21c07a9e295d5b8a2a5ce678b913b45fdf68c32d95d6c1291e0b6", size = 15208, upload-time = "2024-10-18T15:21:41.814Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/4f/65/6079a46068dfceaeabb5dcad6d674f5f5c61a6fa5673746f42a9f4c233b3/MarkupSafe-3.0.2-cp313-cp313t-win_amd64.whl", hash = "sha256:e444a31f8db13eb18ada366ab3cf45fd4b31e4db1236a4448f68778c1d1a5a2f", size = 15739, upload-time = "2024-10-18T15:21:42.784Z" },
+]
+
+[[package]]
+name = "mpmath"
+version = "1.3.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f", size = 508106, upload-time = "2023-03-07T16:47:11.061Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", size = 536198, upload-time = "2023-03-07T16:47:09.197Z" },
+]
+
+[[package]]
+name = "networkx"
+version = "3.5"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/6c/4f/ccdb8ad3a38e583f214547fd2f7ff1fc160c43a75af88e6aec213404b96a/networkx-3.5.tar.gz", hash = "sha256:d4c6f9cf81f52d69230866796b82afbccdec3db7ae4fbd1b65ea750feed50037", size = 2471065, upload-time = "2025-05-29T11:35:07.804Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/eb/8d/776adee7bbf76365fdd7f2552710282c79a4ead5d2a46408c9043a2b70ba/networkx-3.5-py3-none-any.whl", hash = "sha256:0030d386a9a06dee3565298b4a734b68589749a544acbb6c412dc9e2489ec6ec", size = 2034406, upload-time = "2025-05-29T11:35:04.961Z" },
+]
+
+[[package]]
+name = "numpy"
+version = "2.3.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f3/db/8e12381333aea300890829a0a36bfa738cac95475d88982d538725143fd9/numpy-2.3.0.tar.gz", hash = "sha256:581f87f9e9e9db2cba2141400e160e9dd644ee248788d6f90636eeb8fd9260a6", size = 20382813, upload-time = "2025-06-07T14:54:32.608Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/89/59/9df493df81ac6f76e9f05cdbe013cdb0c9a37b434f6e594f5bd25e278908/numpy-2.3.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:389b85335838155a9076e9ad7f8fdba0827496ec2d2dc32ce69ce7898bde03ba", size = 20897025, upload-time = "2025-06-07T14:40:33.558Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/2f/86/4ff04335901d6cf3a6bb9c748b0097546ae5af35e455ae9b962ebff4ecd7/numpy-2.3.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:9498f60cd6bb8238d8eaf468a3d5bb031d34cd12556af53510f05fcf581c1b7e", size = 14129882, upload-time = "2025-06-07T14:40:55.034Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/71/8d/a942cd4f959de7f08a79ab0c7e6cecb7431d5403dce78959a726f0f57aa1/numpy-2.3.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:622a65d40d8eb427d8e722fd410ac3ad4958002f109230bc714fa551044ebae2", size = 5110181, upload-time = "2025-06-07T14:41:04.4Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/86/5d/45850982efc7b2c839c5626fb67fbbc520d5b0d7c1ba1ae3651f2f74c296/numpy-2.3.0-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:b9446d9d8505aadadb686d51d838f2b6688c9e85636a0c3abaeb55ed54756459", size = 6647581, upload-time = "2025-06-07T14:41:14.695Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/1a/c0/c871d4a83f93b00373d3eebe4b01525eee8ef10b623a335ec262b58f4dc1/numpy-2.3.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:50080245365d75137a2bf46151e975de63146ae6d79f7e6bd5c0e85c9931d06a", size = 14262317, upload-time = "2025-06-07T14:41:35.862Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b7/f6/bc47f5fa666d5ff4145254f9e618d56e6a4ef9b874654ca74c19113bb538/numpy-2.3.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:c24bb4113c66936eeaa0dc1e47c74770453d34f46ee07ae4efd853a2ed1ad10a", size = 16633919, upload-time = "2025-06-07T14:42:00.622Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f5/b4/65f48009ca0c9b76df5f404fccdea5a985a1bb2e34e97f21a17d9ad1a4ba/numpy-2.3.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:4d8d294287fdf685281e671886c6dcdf0291a7c19db3e5cb4178d07ccf6ecc67", size = 15567651, upload-time = "2025-06-07T14:42:24.429Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f1/62/5367855a2018578e9334ed08252ef67cc302e53edc869666f71641cad40b/numpy-2.3.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:6295f81f093b7f5769d1728a6bd8bf7466de2adfa771ede944ce6711382b89dc", size = 18361723, upload-time = "2025-06-07T14:42:51.167Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/d4/75/5baed8cd867eabee8aad1e74d7197d73971d6a3d40c821f1848b8fab8b84/numpy-2.3.0-cp312-cp312-win32.whl", hash = "sha256:e6648078bdd974ef5d15cecc31b0c410e2e24178a6e10bf511e0557eed0f2570", size = 6318285, upload-time = "2025-06-07T14:43:02.052Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/bc/49/d5781eaa1a15acb3b3a3f49dc9e2ff18d92d0ce5c2976f4ab5c0a7360250/numpy-2.3.0-cp312-cp312-win_amd64.whl", hash = "sha256:0898c67a58cdaaf29994bc0e2c65230fd4de0ac40afaf1584ed0b02cd74c6fdd", size = 12732594, upload-time = "2025-06-07T14:43:21.071Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/c2/1c/6d343e030815c7c97a1f9fbad00211b47717c7fe446834c224bd5311e6f1/numpy-2.3.0-cp312-cp312-win_arm64.whl", hash = "sha256:bd8df082b6c4695753ad6193018c05aac465d634834dca47a3ae06d4bb22d9ea", size = 9891498, upload-time = "2025-06-07T14:43:36.332Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/73/fc/1d67f751fd4dbafc5780244fe699bc4084268bad44b7c5deb0492473127b/numpy-2.3.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:5754ab5595bfa2c2387d241296e0381c21f44a4b90a776c3c1d39eede13a746a", size = 20889633, upload-time = "2025-06-07T14:44:06.839Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/e8/95/73ffdb69e5c3f19ec4530f8924c4386e7ba097efc94b9c0aff607178ad94/numpy-2.3.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d11fa02f77752d8099573d64e5fe33de3229b6632036ec08f7080f46b6649959", size = 14151683, upload-time = "2025-06-07T14:44:28.847Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/64/d5/06d4bb31bb65a1d9c419eb5676173a2f90fd8da3c59f816cc54c640ce265/numpy-2.3.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:aba48d17e87688a765ab1cd557882052f238e2f36545dfa8e29e6a91aef77afe", size = 5102683, upload-time = "2025-06-07T14:44:38.417Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/12/8b/6c2cef44f8ccdc231f6b56013dff1d71138c48124334aded36b1a1b30c5a/numpy-2.3.0-cp313-cp313-macosx_14_0_x86_64.whl", hash = "sha256:4dc58865623023b63b10d52f18abaac3729346a7a46a778381e0e3af4b7f3beb", size = 6640253, upload-time = "2025-06-07T14:44:49.359Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/62/aa/fca4bf8de3396ddb59544df9b75ffe5b73096174de97a9492d426f5cd4aa/numpy-2.3.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:df470d376f54e052c76517393fa443758fefcdd634645bc9c1f84eafc67087f0", size = 14258658, upload-time = "2025-06-07T14:45:10.156Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/1c/12/734dce1087eed1875f2297f687e671cfe53a091b6f2f55f0c7241aad041b/numpy-2.3.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:87717eb24d4a8a64683b7a4e91ace04e2f5c7c77872f823f02a94feee186168f", size = 16628765, upload-time = "2025-06-07T14:45:35.076Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/48/03/ffa41ade0e825cbcd5606a5669962419528212a16082763fc051a7247d76/numpy-2.3.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:d8fa264d56882b59dcb5ea4d6ab6f31d0c58a57b41aec605848b6eb2ef4a43e8", size = 15564335, upload-time = "2025-06-07T14:45:58.797Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/07/58/869398a11863310aee0ff85a3e13b4c12f20d032b90c4b3ee93c3b728393/numpy-2.3.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:e651756066a0eaf900916497e20e02fe1ae544187cb0fe88de981671ee7f6270", size = 18360608, upload-time = "2025-06-07T14:46:25.687Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/2f/8a/5756935752ad278c17e8a061eb2127c9a3edf4ba2c31779548b336f23c8d/numpy-2.3.0-cp313-cp313-win32.whl", hash = "sha256:e43c3cce3b6ae5f94696669ff2a6eafd9a6b9332008bafa4117af70f4b88be6f", size = 6310005, upload-time = "2025-06-07T14:50:13.138Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/08/60/61d60cf0dfc0bf15381eaef46366ebc0c1a787856d1db0c80b006092af84/numpy-2.3.0-cp313-cp313-win_amd64.whl", hash = "sha256:81ae0bf2564cf475f94be4a27ef7bcf8af0c3e28da46770fc904da9abd5279b5", size = 12729093, upload-time = "2025-06-07T14:50:31.82Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/66/31/2f2f2d2b3e3c32d5753d01437240feaa32220b73258c9eef2e42a0832866/numpy-2.3.0-cp313-cp313-win_arm64.whl", hash = "sha256:c8738baa52505fa6e82778580b23f945e3578412554d937093eac9205e845e6e", size = 9885689, upload-time = "2025-06-07T14:50:47.888Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f1/89/c7828f23cc50f607ceb912774bb4cff225ccae7131c431398ad8400e2c98/numpy-2.3.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:39b27d8b38942a647f048b675f134dd5a567f95bfff481f9109ec308515c51d8", size = 20986612, upload-time = "2025-06-07T14:46:56.077Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/dd/46/79ecf47da34c4c50eedec7511e53d57ffdfd31c742c00be7dc1d5ffdb917/numpy-2.3.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:0eba4a1ea88f9a6f30f56fdafdeb8da3774349eacddab9581a21234b8535d3d3", size = 14298953, upload-time = "2025-06-07T14:47:18.053Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/59/44/f6caf50713d6ff4480640bccb2a534ce1d8e6e0960c8f864947439f0ee95/numpy-2.3.0-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:b0f1f11d0a1da54927436505a5a7670b154eac27f5672afc389661013dfe3d4f", size = 5225806, upload-time = "2025-06-07T14:47:27.524Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a6/43/e1fd1aca7c97e234dd05e66de4ab7a5be54548257efcdd1bc33637e72102/numpy-2.3.0-cp313-cp313t-macosx_14_0_x86_64.whl", hash = "sha256:690d0a5b60a47e1f9dcec7b77750a4854c0d690e9058b7bef3106e3ae9117808", size = 6735169, upload-time = "2025-06-07T14:47:38.057Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/84/89/f76f93b06a03177c0faa7ca94d0856c4e5c4bcaf3c5f77640c9ed0303e1c/numpy-2.3.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:8b51ead2b258284458e570942137155978583e407babc22e3d0ed7af33ce06f8", size = 14330701, upload-time = "2025-06-07T14:47:59.113Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/aa/f5/4858c3e9ff7a7d64561b20580cf7cc5d085794bd465a19604945d6501f6c/numpy-2.3.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:aaf81c7b82c73bd9b45e79cfb9476cb9c29e937494bfe9092c26aece812818ad", size = 16692983, upload-time = "2025-06-07T14:48:24.196Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/08/17/0e3b4182e691a10e9483bcc62b4bb8693dbf9ea5dc9ba0b77a60435074bb/numpy-2.3.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:f420033a20b4f6a2a11f585f93c843ac40686a7c3fa514060a97d9de93e5e72b", size = 15641435, upload-time = "2025-06-07T14:48:47.712Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/4e/d5/463279fda028d3c1efa74e7e8d507605ae87f33dbd0543cf4c4527c8b882/numpy-2.3.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:d344ca32ab482bcf8735d8f95091ad081f97120546f3d250240868430ce52555", size = 18433798, upload-time = "2025-06-07T14:49:14.866Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/0e/1e/7a9d98c886d4c39a2b4d3a7c026bffcf8fbcaf518782132d12a301cfc47a/numpy-2.3.0-cp313-cp313t-win32.whl", hash = "sha256:48a2e8eaf76364c32a1feaa60d6925eaf32ed7a040183b807e02674305beef61", size = 6438632, upload-time = "2025-06-07T14:49:25.67Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/fe/ab/66fc909931d5eb230107d016861824f335ae2c0533f422e654e5ff556784/numpy-2.3.0-cp313-cp313t-win_amd64.whl", hash = "sha256:ba17f93a94e503551f154de210e4d50c5e3ee20f7e7a1b5f6ce3f22d419b93bb", size = 12868491, upload-time = "2025-06-07T14:49:44.898Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/ee/e8/2c8a1c9e34d6f6d600c83d5ce5b71646c32a13f34ca5c518cc060639841c/numpy-2.3.0-cp313-cp313t-win_arm64.whl", hash = "sha256:f14e016d9409680959691c109be98c436c6249eaf7f118b424679793607b5944", size = 9935345, upload-time = "2025-06-07T14:50:02.311Z" },
+]
+
+[[package]]
+name = "nvidia-cublas-cu12"
+version = "12.6.4.1"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/af/eb/ff4b8c503fa1f1796679dce648854d58751982426e4e4b37d6fce49d259c/nvidia_cublas_cu12-12.6.4.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:08ed2686e9875d01b58e3cb379c6896df8e76c75e0d4a7f7dace3d7b6d9ef8eb", size = 393138322, upload-time = "2024-11-20T17:40:25.65Z" },
+]
+
+[[package]]
+name = "nvidia-cuda-cupti-cu12"
+version = "12.6.80"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/49/60/7b6497946d74bcf1de852a21824d63baad12cd417db4195fc1bfe59db953/nvidia_cuda_cupti_cu12-12.6.80-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:6768bad6cab4f19e8292125e5f1ac8aa7d1718704012a0e3272a6f61c4bce132", size = 8917980, upload-time = "2024-11-20T17:36:04.019Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a5/24/120ee57b218d9952c379d1e026c4479c9ece9997a4fb46303611ee48f038/nvidia_cuda_cupti_cu12-12.6.80-py3-none-manylinux2014_x86_64.whl", hash = "sha256:a3eff6cdfcc6a4c35db968a06fcadb061cbc7d6dde548609a941ff8701b98b73", size = 8917972, upload-time = "2024-10-01T16:58:06.036Z" },
+]
+
+[[package]]
+name = "nvidia-cuda-nvrtc-cu12"
+version = "12.6.77"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/75/2e/46030320b5a80661e88039f59060d1790298b4718944a65a7f2aeda3d9e9/nvidia_cuda_nvrtc_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl", hash = "sha256:35b0cc6ee3a9636d5409133e79273ce1f3fd087abb0532d2d2e8fff1fe9efc53", size = 23650380, upload-time = "2024-10-01T17:00:14.643Z" },
+]
+
+[[package]]
+name = "nvidia-cuda-runtime-cu12"
+version = "12.6.77"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/e1/23/e717c5ac26d26cf39a27fbc076240fad2e3b817e5889d671b67f4f9f49c5/nvidia_cuda_runtime_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ba3b56a4f896141e25e19ab287cd71e52a6a0f4b29d0d31609f60e3b4d5219b7", size = 897690, upload-time = "2024-11-20T17:35:30.697Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f0/62/65c05e161eeddbafeca24dc461f47de550d9fa8a7e04eb213e32b55cfd99/nvidia_cuda_runtime_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl", hash = "sha256:a84d15d5e1da416dd4774cb42edf5e954a3e60cc945698dc1d5be02321c44dc8", size = 897678, upload-time = "2024-10-01T16:57:33.821Z" },
+]
+
+[[package]]
+name = "nvidia-cudnn-cu12"
+version = "9.5.1.17"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "nvidia-cublas-cu12", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+]
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/2a/78/4535c9c7f859a64781e43c969a3a7e84c54634e319a996d43ef32ce46f83/nvidia_cudnn_cu12-9.5.1.17-py3-none-manylinux_2_28_x86_64.whl", hash = "sha256:30ac3869f6db17d170e0e556dd6cc5eee02647abc31ca856634d5a40f82c15b2", size = 570988386, upload-time = "2024-10-25T19:54:26.39Z" },
+]
+
+[[package]]
+name = "nvidia-cufft-cu12"
+version = "11.3.0.4"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "nvidia-nvjitlink-cu12", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+]
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/8f/16/73727675941ab8e6ffd86ca3a4b7b47065edcca7a997920b831f8147c99d/nvidia_cufft_cu12-11.3.0.4-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ccba62eb9cef5559abd5e0d54ceed2d9934030f51163df018532142a8ec533e5", size = 200221632, upload-time = "2024-11-20T17:41:32.357Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/60/de/99ec247a07ea40c969d904fc14f3a356b3e2a704121675b75c366b694ee1/nvidia_cufft_cu12-11.3.0.4-py3-none-manylinux2014_x86_64.whl", hash = "sha256:768160ac89f6f7b459bee747e8d175dbf53619cfe74b2a5636264163138013ca", size = 200221622, upload-time = "2024-10-01T17:03:58.79Z" },
+]
+
+[[package]]
+name = "nvidia-cufile-cu12"
+version = "1.11.1.6"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b2/66/cc9876340ac68ae71b15c743ddb13f8b30d5244af344ec8322b449e35426/nvidia_cufile_cu12-1.11.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:cc23469d1c7e52ce6c1d55253273d32c565dd22068647f3aa59b3c6b005bf159", size = 1142103, upload-time = "2024-11-20T17:42:11.83Z" },
+]
+
+[[package]]
+name = "nvidia-curand-cu12"
+version = "10.3.7.77"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/73/1b/44a01c4e70933637c93e6e1a8063d1e998b50213a6b65ac5a9169c47e98e/nvidia_curand_cu12-10.3.7.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:a42cd1344297f70b9e39a1e4f467a4e1c10f1da54ff7a85c12197f6c652c8bdf", size = 56279010, upload-time = "2024-11-20T17:42:50.958Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/4a/aa/2c7ff0b5ee02eaef890c0ce7d4f74bc30901871c5e45dee1ae6d0083cd80/nvidia_curand_cu12-10.3.7.77-py3-none-manylinux2014_x86_64.whl", hash = "sha256:99f1a32f1ac2bd134897fc7a203f779303261268a65762a623bf30cc9fe79117", size = 56279000, upload-time = "2024-10-01T17:04:45.274Z" },
+]
+
+[[package]]
+name = "nvidia-cusolver-cu12"
+version = "11.7.1.2"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "nvidia-cublas-cu12", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "nvidia-cusparse-cu12", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "nvidia-nvjitlink-cu12", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+]
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f0/6e/c2cf12c9ff8b872e92b4a5740701e51ff17689c4d726fca91875b07f655d/nvidia_cusolver_cu12-11.7.1.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e9e49843a7707e42022babb9bcfa33c29857a93b88020c4e4434656a655b698c", size = 158229790, upload-time = "2024-11-20T17:43:43.211Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/9f/81/baba53585da791d043c10084cf9553e074548408e04ae884cfe9193bd484/nvidia_cusolver_cu12-11.7.1.2-py3-none-manylinux2014_x86_64.whl", hash = "sha256:6cf28f17f64107a0c4d7802be5ff5537b2130bfc112f25d5a30df227058ca0e6", size = 158229780, upload-time = "2024-10-01T17:05:39.875Z" },
+]
+
+[[package]]
+name = "nvidia-cusparse-cu12"
+version = "12.5.4.2"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "nvidia-nvjitlink-cu12", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+]
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/06/1e/b8b7c2f4099a37b96af5c9bb158632ea9e5d9d27d7391d7eb8fc45236674/nvidia_cusparse_cu12-12.5.4.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:7556d9eca156e18184b94947ade0fba5bb47d69cec46bf8660fd2c71a4b48b73", size = 216561367, upload-time = "2024-11-20T17:44:54.824Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/43/ac/64c4316ba163e8217a99680c7605f779accffc6a4bcd0c778c12948d3707/nvidia_cusparse_cu12-12.5.4.2-py3-none-manylinux2014_x86_64.whl", hash = "sha256:23749a6571191a215cb74d1cdbff4a86e7b19f1200c071b3fcf844a5bea23a2f", size = 216561357, upload-time = "2024-10-01T17:06:29.861Z" },
+]
+
+[[package]]
+name = "nvidia-cusparselt-cu12"
+version = "0.6.3"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/3b/9a/72ef35b399b0e183bc2e8f6f558036922d453c4d8237dab26c666a04244b/nvidia_cusparselt_cu12-0.6.3-py3-none-manylinux2014_x86_64.whl", hash = "sha256:e5c8a26c36445dd2e6812f1177978a24e2d37cacce7e090f297a688d1ec44f46", size = 156785796, upload-time = "2024-10-15T21:29:17.709Z" },
+]
+
+[[package]]
+name = "nvidia-nccl-cu12"
+version = "2.26.2"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/67/ca/f42388aed0fddd64ade7493dbba36e1f534d4e6fdbdd355c6a90030ae028/nvidia_nccl_cu12-2.26.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:694cf3879a206553cc9d7dbda76b13efaf610fdb70a50cba303de1b0d1530ac6", size = 201319755, upload-time = "2025-03-13T00:29:55.296Z" },
+]
+
+[[package]]
+name = "nvidia-nvjitlink-cu12"
+version = "12.6.85"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/9d/d7/c5383e47c7e9bf1c99d5bd2a8c935af2b6d705ad831a7ec5c97db4d82f4f/nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:eedc36df9e88b682efe4309aa16b5b4e78c2407eac59e8c10a6a47535164369a", size = 19744971, upload-time = "2024-11-20T17:46:53.366Z" },
+]
+
+[[package]]
+name = "nvidia-nvtx-cu12"
+version = "12.6.77"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/56/9a/fff8376f8e3d084cd1530e1ef7b879bb7d6d265620c95c1b322725c694f4/nvidia_nvtx_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:b90bed3df379fa79afbd21be8e04a0314336b8ae16768b58f2d34cb1d04cd7d2", size = 89276, upload-time = "2024-11-20T17:38:27.621Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/9e/4e/0d0c945463719429b7bd21dece907ad0bde437a2ff12b9b12fee94722ab0/nvidia_nvtx_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl", hash = "sha256:6574241a3ec5fdc9334353ab8c479fe75841dbe8f4532a8fc97ce63503330ba1", size = 89265, upload-time = "2024-10-01T17:00:38.172Z" },
+]
+
+[[package]]
+name = "opencv-python"
+version = "4.11.0.86"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "numpy" },
+]
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/17/06/68c27a523103dad5837dc5b87e71285280c4f098c60e4fe8a8db6486ab09/opencv-python-4.11.0.86.tar.gz", hash = "sha256:03d60ccae62304860d232272e4a4fda93c39d595780cb40b161b310244b736a4", size = 95171956, upload-time = "2025-01-16T13:52:24.737Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/05/4d/53b30a2a3ac1f75f65a59eb29cf2ee7207ce64867db47036ad61743d5a23/opencv_python-4.11.0.86-cp37-abi3-macosx_13_0_arm64.whl", hash = "sha256:432f67c223f1dc2824f5e73cdfcd9db0efc8710647d4e813012195dc9122a52a", size = 37326322, upload-time = "2025-01-16T13:52:25.887Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/3b/84/0a67490741867eacdfa37bc18df96e08a9d579583b419010d7f3da8ff503/opencv_python-4.11.0.86-cp37-abi3-macosx_13_0_x86_64.whl", hash = "sha256:9d05ef13d23fe97f575153558653e2d6e87103995d54e6a35db3f282fe1f9c66", size = 56723197, upload-time = "2025-01-16T13:55:21.222Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f3/bd/29c126788da65c1fb2b5fb621b7fed0ed5f9122aa22a0868c5e2c15c6d23/opencv_python-4.11.0.86-cp37-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1b92ae2c8852208817e6776ba1ea0d6b1e0a1b5431e971a2a0ddd2a8cc398202", size = 42230439, upload-time = "2025-01-16T13:51:35.822Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/2c/8b/90eb44a40476fa0e71e05a0283947cfd74a5d36121a11d926ad6f3193cc4/opencv_python-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b02611523803495003bd87362db3e1d2a0454a6a63025dc6658a9830570aa0d", size = 62986597, upload-time = "2025-01-16T13:52:08.836Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/fb/d7/1d5941a9dde095468b288d989ff6539dd69cd429dbf1b9e839013d21b6f0/opencv_python-4.11.0.86-cp37-abi3-win32.whl", hash = "sha256:810549cb2a4aedaa84ad9a1c92fbfdfc14090e2749cedf2c1589ad8359aa169b", size = 29384337, upload-time = "2025-01-16T13:52:13.549Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a4/7d/f1c30a92854540bf789e9cd5dde7ef49bbe63f855b85a2e6b3db8135c591/opencv_python-4.11.0.86-cp37-abi3-win_amd64.whl", hash = "sha256:085ad9b77c18853ea66283e98affefe2de8cc4c1f43eda4c100cf9b2721142ec", size = 39488044, upload-time = "2025-01-16T13:52:21.928Z" },
+]
+
+[[package]]
+name = "pillow"
+version = "11.2.1"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/af/cb/bb5c01fcd2a69335b86c22142b2bccfc3464087efb7fd382eee5ffc7fdf7/pillow-11.2.1.tar.gz", hash = "sha256:a64dd61998416367b7ef979b73d3a85853ba9bec4c2925f74e588879a58716b6", size = 47026707, upload-time = "2025-04-12T17:50:03.289Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/c7/40/052610b15a1b8961f52537cc8326ca6a881408bc2bdad0d852edeb6ed33b/pillow-11.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:78afba22027b4accef10dbd5eed84425930ba41b3ea0a86fa8d20baaf19d807f", size = 3190185, upload-time = "2025-04-12T17:48:00.417Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/e5/7e/b86dbd35a5f938632093dc40d1682874c33dcfe832558fc80ca56bfcb774/pillow-11.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:78092232a4ab376a35d68c4e6d5e00dfd73454bd12b230420025fbe178ee3b0b", size = 3030306, upload-time = "2025-04-12T17:48:02.391Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a4/5c/467a161f9ed53e5eab51a42923c33051bf8d1a2af4626ac04f5166e58e0c/pillow-11.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:25a5f306095c6780c52e6bbb6109624b95c5b18e40aab1c3041da3e9e0cd3e2d", size = 4416121, upload-time = "2025-04-12T17:48:04.554Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/62/73/972b7742e38ae0e2ac76ab137ca6005dcf877480da0d9d61d93b613065b4/pillow-11.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0c7b29dbd4281923a2bfe562acb734cee96bbb129e96e6972d315ed9f232bef4", size = 4501707, upload-time = "2025-04-12T17:48:06.831Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/e4/3a/427e4cb0b9e177efbc1a84798ed20498c4f233abde003c06d2650a6d60cb/pillow-11.2.1-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:3e645b020f3209a0181a418bffe7b4a93171eef6c4ef6cc20980b30bebf17b7d", size = 4522921, upload-time = "2025-04-12T17:48:09.229Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/fe/7c/d8b1330458e4d2f3f45d9508796d7caf0c0d3764c00c823d10f6f1a3b76d/pillow-11.2.1-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:b2dbea1012ccb784a65349f57bbc93730b96e85b42e9bf7b01ef40443db720b4", size = 4612523, upload-time = "2025-04-12T17:48:11.631Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b3/2f/65738384e0b1acf451de5a573d8153fe84103772d139e1e0bdf1596be2ea/pillow-11.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:da3104c57bbd72948d75f6a9389e6727d2ab6333c3617f0a89d72d4940aa0443", size = 4587836, upload-time = "2025-04-12T17:48:13.592Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/6a/c5/e795c9f2ddf3debb2dedd0df889f2fe4b053308bb59a3cc02a0cd144d641/pillow-11.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:598174aef4589af795f66f9caab87ba4ff860ce08cd5bb447c6fc553ffee603c", size = 4669390, upload-time = "2025-04-12T17:48:15.938Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/96/ae/ca0099a3995976a9fce2f423166f7bff9b12244afdc7520f6ed38911539a/pillow-11.2.1-cp312-cp312-win32.whl", hash = "sha256:1d535df14716e7f8776b9e7fee118576d65572b4aad3ed639be9e4fa88a1cad3", size = 2332309, upload-time = "2025-04-12T17:48:17.885Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/7c/18/24bff2ad716257fc03da964c5e8f05d9790a779a8895d6566e493ccf0189/pillow-11.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:14e33b28bf17c7a38eede290f77db7c664e4eb01f7869e37fa98a5aa95978941", size = 2676768, upload-time = "2025-04-12T17:48:19.655Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/da/bb/e8d656c9543276517ee40184aaa39dcb41e683bca121022f9323ae11b39d/pillow-11.2.1-cp312-cp312-win_arm64.whl", hash = "sha256:21e1470ac9e5739ff880c211fc3af01e3ae505859392bf65458c224d0bf283eb", size = 2415087, upload-time = "2025-04-12T17:48:21.991Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/36/9c/447528ee3776e7ab8897fe33697a7ff3f0475bb490c5ac1456a03dc57956/pillow-11.2.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:fdec757fea0b793056419bca3e9932eb2b0ceec90ef4813ea4c1e072c389eb28", size = 3190098, upload-time = "2025-04-12T17:48:23.915Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b5/09/29d5cd052f7566a63e5b506fac9c60526e9ecc553825551333e1e18a4858/pillow-11.2.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:b0e130705d568e2f43a17bcbe74d90958e8a16263868a12c3e0d9c8162690830", size = 3030166, upload-time = "2025-04-12T17:48:25.738Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/71/5d/446ee132ad35e7600652133f9c2840b4799bbd8e4adba881284860da0a36/pillow-11.2.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7bdb5e09068332578214cadd9c05e3d64d99e0e87591be22a324bdbc18925be0", size = 4408674, upload-time = "2025-04-12T17:48:27.908Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/69/5f/cbe509c0ddf91cc3a03bbacf40e5c2339c4912d16458fcb797bb47bcb269/pillow-11.2.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d189ba1bebfbc0c0e529159631ec72bb9e9bc041f01ec6d3233d6d82eb823bc1", size = 4496005, upload-time = "2025-04-12T17:48:29.888Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f9/b3/dd4338d8fb8a5f312021f2977fb8198a1184893f9b00b02b75d565c33b51/pillow-11.2.1-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:191955c55d8a712fab8934a42bfefbf99dd0b5875078240943f913bb66d46d9f", size = 4518707, upload-time = "2025-04-12T17:48:31.874Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/13/eb/2552ecebc0b887f539111c2cd241f538b8ff5891b8903dfe672e997529be/pillow-11.2.1-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:ad275964d52e2243430472fc5d2c2334b4fc3ff9c16cb0a19254e25efa03a155", size = 4610008, upload-time = "2025-04-12T17:48:34.422Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/72/d1/924ce51bea494cb6e7959522d69d7b1c7e74f6821d84c63c3dc430cbbf3b/pillow-11.2.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:750f96efe0597382660d8b53e90dd1dd44568a8edb51cb7f9d5d918b80d4de14", size = 4585420, upload-time = "2025-04-12T17:48:37.641Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/43/ab/8f81312d255d713b99ca37479a4cb4b0f48195e530cdc1611990eb8fd04b/pillow-11.2.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:fe15238d3798788d00716637b3d4e7bb6bde18b26e5d08335a96e88564a36b6b", size = 4667655, upload-time = "2025-04-12T17:48:39.652Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/94/86/8f2e9d2dc3d308dfd137a07fe1cc478df0a23d42a6c4093b087e738e4827/pillow-11.2.1-cp313-cp313-win32.whl", hash = "sha256:3fe735ced9a607fee4f481423a9c36701a39719252a9bb251679635f99d0f7d2", size = 2332329, upload-time = "2025-04-12T17:48:41.765Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/6d/ec/1179083b8d6067a613e4d595359b5fdea65d0a3b7ad623fee906e1b3c4d2/pillow-11.2.1-cp313-cp313-win_amd64.whl", hash = "sha256:74ee3d7ecb3f3c05459ba95eed5efa28d6092d751ce9bf20e3e253a4e497e691", size = 2676388, upload-time = "2025-04-12T17:48:43.625Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/23/f1/2fc1e1e294de897df39fa8622d829b8828ddad938b0eaea256d65b84dd72/pillow-11.2.1-cp313-cp313-win_arm64.whl", hash = "sha256:5119225c622403afb4b44bad4c1ca6c1f98eed79db8d3bc6e4e160fc6339d66c", size = 2414950, upload-time = "2025-04-12T17:48:45.475Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/c4/3e/c328c48b3f0ead7bab765a84b4977acb29f101d10e4ef57a5e3400447c03/pillow-11.2.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:8ce2e8411c7aaef53e6bb29fe98f28cd4fbd9a1d9be2eeea434331aac0536b22", size = 3192759, upload-time = "2025-04-12T17:48:47.866Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/18/0e/1c68532d833fc8b9f404d3a642991441d9058eccd5606eab31617f29b6d4/pillow-11.2.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:9ee66787e095127116d91dea2143db65c7bb1e232f617aa5957c0d9d2a3f23a7", size = 3033284, upload-time = "2025-04-12T17:48:50.189Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b7/cb/6faf3fb1e7705fd2db74e070f3bf6f88693601b0ed8e81049a8266de4754/pillow-11.2.1-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9622e3b6c1d8b551b6e6f21873bdcc55762b4b2126633014cea1803368a9aa16", size = 4445826, upload-time = "2025-04-12T17:48:52.346Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/07/94/8be03d50b70ca47fb434a358919d6a8d6580f282bbb7af7e4aa40103461d/pillow-11.2.1-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:63b5dff3a68f371ea06025a1a6966c9a1e1ee452fc8020c2cd0ea41b83e9037b", size = 4527329, upload-time = "2025-04-12T17:48:54.403Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/fd/a4/bfe78777076dc405e3bd2080bc32da5ab3945b5a25dc5d8acaa9de64a162/pillow-11.2.1-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:31df6e2d3d8fc99f993fd253e97fae451a8db2e7207acf97859732273e108406", size = 4549049, upload-time = "2025-04-12T17:48:56.383Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/65/4d/eaf9068dc687c24979e977ce5677e253624bd8b616b286f543f0c1b91662/pillow-11.2.1-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:062b7a42d672c45a70fa1f8b43d1d38ff76b63421cbbe7f88146b39e8a558d91", size = 4635408, upload-time = "2025-04-12T17:48:58.782Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/1d/26/0fd443365d9c63bc79feb219f97d935cd4b93af28353cba78d8e77b61719/pillow-11.2.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:4eb92eca2711ef8be42fd3f67533765d9fd043b8c80db204f16c8ea62ee1a751", size = 4614863, upload-time = "2025-04-12T17:49:00.709Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/49/65/dca4d2506be482c2c6641cacdba5c602bc76d8ceb618fd37de855653a419/pillow-11.2.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:f91ebf30830a48c825590aede79376cb40f110b387c17ee9bd59932c961044f9", size = 4692938, upload-time = "2025-04-12T17:49:02.946Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b3/92/1ca0c3f09233bd7decf8f7105a1c4e3162fb9142128c74adad0fb361b7eb/pillow-11.2.1-cp313-cp313t-win32.whl", hash = "sha256:e0b55f27f584ed623221cfe995c912c61606be8513bfa0e07d2c674b4516d9dd", size = 2335774, upload-time = "2025-04-12T17:49:04.889Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a5/ac/77525347cb43b83ae905ffe257bbe2cc6fd23acb9796639a1f56aa59d191/pillow-11.2.1-cp313-cp313t-win_amd64.whl", hash = "sha256:36d6b82164c39ce5482f649b437382c0fb2395eabc1e2b1702a6deb8ad647d6e", size = 2681895, upload-time = "2025-04-12T17:49:06.635Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/67/32/32dc030cfa91ca0fc52baebbba2e009bb001122a1daa8b6a79ad830b38d3/pillow-11.2.1-cp313-cp313t-win_arm64.whl", hash = "sha256:225c832a13326e34f212d2072982bb1adb210e0cc0b153e688743018c94a2681", size = 2417234, upload-time = "2025-04-12T17:49:08.399Z" },
+]
+
+[[package]]
+name = "rord-layout-recognation"
+version = "0.1.0"
+source = { virtual = "." }
+dependencies = [
+    { name = "numpy" },
+    { name = "opencv-python" },
+    { name = "pillow" },
+    { name = "torch" },
+    { name = "torchvision" },
+]
+
+[package.metadata]
+requires-dist = [
+    { name = "numpy", specifier = ">=2.3.0" },
+    { name = "opencv-python", specifier = ">=4.11.0.86" },
+    { name = "pillow", specifier = ">=11.2.1" },
+    { name = "torch", specifier = ">=2.7.1" },
+    { name = "torchvision", specifier = ">=0.22.1" },
+]
+
+[[package]]
+name = "setuptools"
+version = "80.9.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/18/5d/3bf57dcd21979b887f014ea83c24ae194cfcd12b9e0fda66b957c69d1fca/setuptools-80.9.0.tar.gz", hash = "sha256:f36b47402ecde768dbfafc46e8e4207b4360c654f1f3bb84475f0a28628fb19c", size = 1319958, upload-time = "2025-05-27T00:56:51.443Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a3/dc/17031897dae0efacfea57dfd3a82fdd2a2aeb58e0ff71b77b87e44edc772/setuptools-80.9.0-py3-none-any.whl", hash = "sha256:062d34222ad13e0cc312a4c02d73f059e86a4acbfbdea8f8f76b28c99f306922", size = 1201486, upload-time = "2025-05-27T00:56:49.664Z" },
+]
+
+[[package]]
+name = "sympy"
+version = "1.14.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "mpmath" },
+]
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/83/d3/803453b36afefb7c2bb238361cd4ae6125a569b4db67cd9e79846ba2d68c/sympy-1.14.0.tar.gz", hash = "sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517", size = 7793921, upload-time = "2025-04-27T18:05:01.611Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a2/09/77d55d46fd61b4a135c444fc97158ef34a095e5681d0a6c10b75bf356191/sympy-1.14.0-py3-none-any.whl", hash = "sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5", size = 6299353, upload-time = "2025-04-27T18:04:59.103Z" },
+]
+
+[[package]]
+name = "torch"
+version = "2.7.1"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "filelock" },
+    { name = "fsspec" },
+    { name = "jinja2" },
+    { name = "networkx" },
+    { name = "nvidia-cublas-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-cupti-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-nvrtc-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-runtime-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cudnn-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufft-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufile-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-curand-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusolver-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparse-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparselt-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nccl-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvjitlink-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvtx-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "setuptools" },
+    { name = "sympy" },
+    { name = "triton", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "typing-extensions" },
+]
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/87/93/fb505a5022a2e908d81fe9a5e0aa84c86c0d5f408173be71c6018836f34e/torch-2.7.1-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:27ea1e518df4c9de73af7e8a720770f3628e7f667280bce2be7a16292697e3fa", size = 98948276, upload-time = "2025-06-04T17:39:12.852Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/56/7e/67c3fe2b8c33f40af06326a3d6ae7776b3e3a01daa8f71d125d78594d874/torch-2.7.1-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:c33360cfc2edd976c2633b3b66c769bdcbbf0e0b6550606d188431c81e7dd1fc", size = 821025792, upload-time = "2025-06-04T17:34:58.747Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/a1/37/a37495502bc7a23bf34f89584fa5a78e25bae7b8da513bc1b8f97afb7009/torch-2.7.1-cp312-cp312-win_amd64.whl", hash = "sha256:d8bf6e1856ddd1807e79dc57e54d3335f2b62e6f316ed13ed3ecfe1fc1df3d8b", size = 216050349, upload-time = "2025-06-04T17:38:59.709Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/3a/60/04b77281c730bb13460628e518c52721257814ac6c298acd25757f6a175c/torch-2.7.1-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:787687087412c4bd68d315e39bc1223f08aae1d16a9e9771d95eabbb04ae98fb", size = 68645146, upload-time = "2025-06-04T17:38:52.97Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/66/81/e48c9edb655ee8eb8c2a6026abdb6f8d2146abd1f150979ede807bb75dcb/torch-2.7.1-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:03563603d931e70722dce0e11999d53aa80a375a3d78e6b39b9f6805ea0a8d28", size = 98946649, upload-time = "2025-06-04T17:38:43.031Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/3a/24/efe2f520d75274fc06b695c616415a1e8a1021d87a13c68ff9dce733d088/torch-2.7.1-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:d632f5417b6980f61404a125b999ca6ebd0b8b4bbdbb5fbbba44374ab619a412", size = 821033192, upload-time = "2025-06-04T17:38:09.146Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/dd/d9/9c24d230333ff4e9b6807274f6f8d52a864210b52ec794c5def7925f4495/torch-2.7.1-cp313-cp313-win_amd64.whl", hash = "sha256:23660443e13995ee93e3d844786701ea4ca69f337027b05182f5ba053ce43b38", size = 216055668, upload-time = "2025-06-04T17:38:36.253Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/95/bf/e086ee36ddcef9299f6e708d3b6c8487c1651787bb9ee2939eb2a7f74911/torch-2.7.1-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:0da4f4dba9f65d0d203794e619fe7ca3247a55ffdcbd17ae8fb83c8b2dc9b585", size = 68925988, upload-time = "2025-06-04T17:38:29.273Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/69/6a/67090dcfe1cf9048448b31555af6efb149f7afa0a310a366adbdada32105/torch-2.7.1-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:e08d7e6f21a617fe38eeb46dd2213ded43f27c072e9165dc27300c9ef9570934", size = 99028857, upload-time = "2025-06-04T17:37:50.956Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/90/1c/48b988870823d1cc381f15ec4e70ed3d65e043f43f919329b0045ae83529/torch-2.7.1-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:30207f672328a42df4f2174b8f426f354b2baa0b7cca3a0adb3d6ab5daf00dc8", size = 821098066, upload-time = "2025-06-04T17:37:33.939Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/7b/eb/10050d61c9d5140c5dc04a89ed3257ef1a6b93e49dd91b95363d757071e0/torch-2.7.1-cp313-cp313t-win_amd64.whl", hash = "sha256:79042feca1c634aaf6603fe6feea8c6b30dfa140a6bbc0b973e2260c7e79a22e", size = 216336310, upload-time = "2025-06-04T17:36:09.862Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/b1/29/beb45cdf5c4fc3ebe282bf5eafc8dfd925ead7299b3c97491900fe5ed844/torch-2.7.1-cp313-none-macosx_11_0_arm64.whl", hash = "sha256:988b0cbc4333618a1056d2ebad9eb10089637b659eb645434d0809d8d937b946", size = 68645708, upload-time = "2025-06-04T17:34:39.852Z" },
+]
+
+[[package]]
+name = "torchvision"
+version = "0.22.1"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "numpy" },
+    { name = "pillow" },
+    { name = "torch" },
+]
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/02/90/f4e99a5112dc221cf68a485e853cc3d9f3f1787cb950b895f3ea26d1ea98/torchvision-0.22.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:153f1790e505bd6da123e21eee6e83e2e155df05c0fe7d56347303067d8543c5", size = 1947827, upload-time = "2025-06-04T17:43:11.945Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/25/f6/53e65384cdbbe732cc2106bb04f7fb908487e4fb02ae4a1613ce6904a122/torchvision-0.22.1-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:964414eef19459d55a10e886e2fca50677550e243586d1678f65e3f6f6bac47a", size = 2514576, upload-time = "2025-06-04T17:43:02.707Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/17/8b/155f99042f9319bd7759536779b2a5b67cbd4f89c380854670850f89a2f4/torchvision-0.22.1-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:699c2d70d33951187f6ed910ea05720b9b4aaac1dcc1135f53162ce7d42481d3", size = 7485962, upload-time = "2025-06-04T17:42:43.606Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/05/17/e45d5cd3627efdb47587a0634179a3533593436219de3f20c743672d2a79/torchvision-0.22.1-cp312-cp312-win_amd64.whl", hash = "sha256:75e0897da7a8e43d78632f66f2bdc4f6e26da8d3f021a7c0fa83746073c2597b", size = 1707992, upload-time = "2025-06-04T17:42:53.207Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/7a/30/fecdd09fb973e963da68207fe9f3d03ec6f39a935516dc2a98397bf495c6/torchvision-0.22.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:9c3ae3319624c43cc8127020f46c14aa878406781f0899bb6283ae474afeafbf", size = 1947818, upload-time = "2025-06-04T17:42:51.954Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/55/f4/b45f6cd92fa0acfac5e31b8e9258232f25bcdb0709a604e8b8a39d76e411/torchvision-0.22.1-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:4a614a6a408d2ed74208d0ea6c28a2fbb68290e9a7df206c5fef3f0b6865d307", size = 2471597, upload-time = "2025-06-04T17:42:48.838Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/8d/b0/3cffd6a285b5ffee3fe4a31caff49e350c98c5963854474d1c4f7a51dea5/torchvision-0.22.1-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:7ee682be589bb1a002b7704f06b8ec0b89e4b9068f48e79307d2c6e937a9fdf4", size = 7485894, upload-time = "2025-06-04T17:43:01.371Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/fd/1d/0ede596fedc2080d18108149921278b59f220fbb398f29619495337b0f86/torchvision-0.22.1-cp313-cp313-win_amd64.whl", hash = "sha256:2566cafcfa47ecfdbeed04bab8cef1307c8d4ef75046f7624b9e55f384880dfe", size = 1708020, upload-time = "2025-06-04T17:43:06.085Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/0f/ca/e9a06bd61ee8e04fb4962a3fb524fe6ee4051662db07840b702a9f339b24/torchvision-0.22.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:043d9e35ed69c2e586aff6eb9e2887382e7863707115668ac9d140da58f42cba", size = 2137623, upload-time = "2025-06-04T17:43:05.028Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/ab/c8/2ebe90f18e7ffa2120f5c3eab62aa86923185f78d2d051a455ea91461608/torchvision-0.22.1-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:27142bcc8a984227a6dcf560985e83f52b82a7d3f5fe9051af586a2ccc46ef26", size = 2476561, upload-time = "2025-06-04T17:42:59.691Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/94/8b/04c6b15f8c29b39f0679589753091cec8b192ab296d4fdaf9055544c4ec9/torchvision-0.22.1-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:ef46e065502f7300ad6abc98554131c35dc4c837b978d91306658f1a65c00baa", size = 7658543, upload-time = "2025-06-04T17:42:46.064Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/ab/c0/131628e6d42682b0502c63fd7f647b8b5ca4bd94088f6c85ca7225db8ac4/torchvision-0.22.1-cp313-cp313t-win_amd64.whl", hash = "sha256:7414eeacfb941fa21acddcd725f1617da5630ec822e498660a4b864d7d998075", size = 1629892, upload-time = "2025-06-04T17:42:57.156Z" },
+]
+
+[[package]]
+name = "triton"
+version = "3.3.1"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "setuptools", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+]
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/24/5f/950fb373bf9c01ad4eb5a8cd5eaf32cdf9e238c02f9293557a2129b9c4ac/triton-3.3.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9999e83aba21e1a78c1f36f21bce621b77bcaa530277a50484a7cb4a822f6e43", size = 155669138, upload-time = "2025-05-29T23:39:51.771Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/74/1f/dfb531f90a2d367d914adfee771babbd3f1a5b26c3f5fbc458dee21daa78/triton-3.3.1-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b89d846b5a4198317fec27a5d3a609ea96b6d557ff44b56c23176546023c4240", size = 155673035, upload-time = "2025-05-29T23:40:02.468Z" },
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/28/71/bd20ffcb7a64c753dc2463489a61bf69d531f308e390ad06390268c4ea04/triton-3.3.1-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a3198adb9d78b77818a5388bff89fa72ff36f9da0bc689db2f0a651a67ce6a42", size = 155735832, upload-time = "2025-05-29T23:40:10.522Z" },
+]
+
+[[package]]
+name = "typing-extensions"
+version = "4.14.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/d1/bc/51647cd02527e87d05cb083ccc402f93e441606ff1f01739a62c8ad09ba5/typing_extensions-4.14.0.tar.gz", hash = "sha256:8676b788e32f02ab42d9e7c61324048ae4c6d844a399eebace3d4979d75ceef4", size = 107423, upload-time = "2025-06-02T14:52:11.399Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/69/e0/552843e0d356fbb5256d21449fa957fa4eff3bbc135a74a691ee70c7c5da/typing_extensions-4.14.0-py3-none-any.whl", hash = "sha256:a1514509136dd0b477638fc68d6a91497af5076466ad0fa6c338e44e359944af", size = 43839, upload-time = "2025-06-02T14:52:10.026Z" },
+]