[CodeCamp2023-552] Complete Chinese and English documents advanced_gu…

…ides/structures.md (#1982) * Add files via upload [CodeCamp2023-552] Complete Chinese and English documents advanced_guides/structures.md * Add files via upload 1 * Add files via upload test * fix lint error --------- Co-authored-by: LeoXing1996 <[email protected]>
open-mmlab · Aug 21, 2023 · a06fb3b · a06fb3b
1 parent 309ad75
commit a06fb3b
Show file tree

Hide file tree

Showing 2 changed files with 206 additions and 4 deletions.
diff --git a/docs/en/advanced_guides/structures.md b/docs/en/advanced_guides/structures.md
@@ -1,5 +1,106 @@
-# Data structure in MMagic \[Coming Soon!\]
+# Data Structure
 
-We're improving this documentation. Don't hesitate to join us!
+`DataSample` , the data structure interface of MMagic, inherits from [` BaseDataElement`](https://mmengine.readthedocs.io/en/latest/advanced_tutorials/data_element.html). The base class has implemented basic add/delete/update/check functions and supports data migration between different devices, as well as dictionary-like and tensor-like operations, which also allows the interfaces of different algorithms to be unified.
 
-[Make a pull request](https://github.com/open-mmlab/mmagic/compare) or [discuss with us](https://github.com/open-mmlab/mmagic/discussions/1429)!
+Specifically, an instance of BaseDataElement consists of two components:
+
+- `metainfo`, which contains some meta information,
+  e.g., `img_shape`, `img_id`, `color_order`, etc.
+- `data`, which contains the data used in the loop.
+
+Thanks to ` DataSample` , the data flow between each module in the algorithm libraries, such as [`visualizer`](https://mmagic.readthedocs.io/en/latest/user_guides/visualization.html), [`evaluator`](https://mmagic.readthedocs.io/en/latest/advanced_guides/evaluator.html), [`model`](https://mmagic.readthedocs.io/en/latest/howto/models.html), is greatly simplified.
+
+The attributes in `DataSample` are divided into several parts:
+
+```python
+- ``gt_img``: Ground truth image(s).
+- ``pred_img``: Image(s) of model predictions.
+- ``ref_img``: Reference image(s).
+- ``mask``: Mask in Inpainting.
+- ``trimap``: Trimap in Matting.
+- ``gt_alpha``: Ground truth alpha image in Matting.
+- ``pred_alpha``: Predicted alpha image in Matting.
+- ``gt_fg``: Ground truth foreground image in Matting.
+- ``pred_fg``: Predicted foreground image in Matting.
+- ``gt_bg``: Ground truth background image in Matting.
+- ``pred_bg``: Predicted background image in Matting.
+- ``gt_merged``: Ground truth merged image in Matting.
+```
+
+The following sample code demonstrates the components of `DataSample`:
+
+```python
+     >>> import torch
+     >>> import numpy as np
+     >>> from mmagic.structures import DataSample
+     >>> img_meta = dict(img_shape=(800, 1196, 3))
+     >>> img = torch.rand((3, 800, 1196))
+     >>> data_sample = DataSample(gt_img=img, metainfo=img_meta)
+     >>> assert 'img_shape' in data_sample.metainfo_keys()
+     >>> data_sample
+	 >>># metainfo and data of DataSample
+    <DataSample(
+
+        META INFORMATION
+        img_shape: (800, 1196, 3)
+
+        DATA FIELDS
+        gt_img: tensor(3, 800, 1196)
+    ) at 0x1f6a5a99a00>
+```
+
+We also support `stack` and `split` operation to handle a batch of data samples.
+
+1. Stack
+
+Stack a list of data samples to one. All tensor fields will be stacked at first dimension. Otherwise the values will be saved in a list.
+
+```
+    Args:
+        data_samples (Sequence['DataSample']): A sequence of `DataSample` to stack.
+
+    Returns:
+        DataSample: The stacked data sample.
+```
+
+2. Split
+
+Split a sequence of data sample in the first dimension.
+
+```
+	Args:
+         allow_nonseq_value (bool): Whether allow non-sequential data in
+         split operation. If True, non-sequential data will be copied
+         for all split data samples. Otherwise, an error will be
+         raised. Defaults to False.
+
+    Returns:
+         Sequence[DataSample]: The list of data samples after splitting.
+```
+
+The following sample code demonstrates the use of `stack` and ` split`:
+
+```py
+import torch
+import numpy as np
+from mmagic.structures import DataSample
+img_meta1 = img_meta2 = dict(img_shape=(800, 1196, 3))
+img1 = torch.rand((3, 800, 1196))
+img2 = torch.rand((3, 800, 1196))
+data_sample1 = DataSample(gt_img=img1, metainfo=img_meta1)
+data_sample2 = DataSample(gt_img=img2, metainfo=img_meta1)
+```
+
+```py
+# stack them and then use as batched-tensor!
+data_sample = DataSample.stack([data_sample1, data_sample2])
+print(data_sample.gt_img.shape)
+    torch.Size([2, 3, 800, 1196])
+print(data_sample.metainfo)
+    {'img_shape': [(800, 1196, 3), (800, 1196, 3)]}
+
+# split them if you want
+data_sample1_, data_sample2_ = data_sample.split()
+assert (data_sample1_.gt_img == img1).all()
+assert (data_sample2_.gt_img == img2).all()
+```
diff --git a/docs/zh_cn/advanced_guides/structures.md b/docs/zh_cn/advanced_guides/structures.md
@@ -1 +1,102 @@
-# 数据结构（待更新）
+# Data Structure
+
+MMaigc的数据结构接口`DataSample` 继承自 MMEngine 的 [` BaseDataElement`](https://mmengine.readthedocs.io/zh_CN/latest/advanced_tutorials/data_element.html).MMEngine 的抽象数据接口实现了基础的增/删/改/查功能，且支持不同设备间的数据迁移，也支持了类字典和张量的操作，充分满足了数据的日常使用需求，这也使得不同算法的数据接口可以得到统一。
+
+特别的，`BaseDataElement` 中存在两种类型的数据:
+
+- `metainfo` 类型，包含数据的元信息以确保数据的完整性，如 `img_shape`, `img_id` 等数据所在图片的一些基本信息，方便可视化等情况下对数据进行恢复和使用。
+- `data` 类型，如标注框、框的标签、和实例掩码等。
+
+得益于统一的数据封装，算法库内的 [`visualizer`](https://mmagic.readthedocs.io/zh_CN/latest/user_guides/visualization.html), [`evaluator`](https://mmagic.readthedocs.io/zh_CN/latest/advanced_guides/evaluator.html), [`model`](https://mmagic.readthedocs.io/zh_CN/latest/howto/models.html) 等各个模块间的数据流通都得到了极大的简化。
+
+`DataSample`中的数据分为以下几个属性：
+
+```python
+- ``gt_img``: 原始图像
+- ``pred_img``: 模型预测图像
+- ``ref_img``:参考图像
+- ``mask``: 图像修复中的遮挡区域
+- ``trimap``: 图像抠图中的三通道图
+- ``gt_alpha``: 图像抠图中原始Alpha图
+- ``pred_alpha``: 图像抠图中模型预测Alpha图
+- ``gt_fg``: 图像抠图中原始前景图
+- ``pred_fg``: 图像抠图中模型预测前景图
+- ``gt_bg``:  图像抠图中原始背景图
+- ``pred_bg``: 图像抠图中模型预测背景图
+- ``gt_merged``:  图像抠图中原始合并图
+```
+
+以下示例代码展示了 `DataSample` 的组成元素类型：
+
+```python
+import torch
+import numpy as np
+from mmagic.structures import DataSample
+img_meta = dict(img_shape=(800, 1196, 3))
+img = torch.rand((3, 800, 1196))
+data_sample = DataSample(gt_img=img, metainfo=img_meta)
+assert 'img_shape' in data_sample.metainfo_keys()
+data_sample
+# `DataSample` 的组成元素类型
+<DataSample(
+
+        META INFORMATION
+        img_shape: (800, 1196, 3)
+
+        DATA FIELDS
+        gt_img: tensor(3, 800, 1196)
+    ) at 0x1f6a5a99a00>
+```
+
+`DataSample`同样支持`stack`和`split`操作对数据进行批处理:
+
+1. Stack
+
+该函数用于将数据样本列表堆叠成一个。当数据样本堆叠时，所有张量字段都将堆叠在第一维度。如果数据样本中有非张量字段，例如列表或字典，则这些字段的值将保存在列表中。
+
+```
+    Args:
+        data_samples (Sequence['DataSample']): 待堆叠的数据样本序列
+
+    Returns:
+        DataSample: 堆叠的数据样本
+```
+
+2. Split
+
+该函数将在第一维度拆分数据样本序列。
+
+```
+	Args:
+         allow_nonseq_value (bool): 是否允许在拆分操作中使用非顺序数据。如果为 "True"，			将为所有拆分数据样本复制非序列数据；否则，将引发错误。默认为 "False"。
+
+    Returns:
+         Sequence[DataSample]: 拆分后的数据样本列表。
+```
+
+以下示例代码展示了 `stack`和`split` 的使用方法：
+
+```py
+import torch
+import numpy as np
+from mmagic.structures import DataSample
+img_meta1 = img_meta2 = dict(img_shape=(800, 1196, 3))
+img1 = torch.rand((3, 800, 1196))
+img2 = torch.rand((3, 800, 1196))
+data_sample1 = DataSample(gt_img=img1, metainfo=img_meta1)
+data_sample2 = DataSample(gt_img=img2, metainfo=img_meta1)
+```
+
+```py
+# 堆叠stack
+data_sample = DataSample.stack([data_sample1, data_sample2])
+print(data_sample.gt_img.shape)
+    torch.Size([2, 3, 800, 1196])
+print(data_sample.metainfo)
+    {'img_shape': [(800, 1196, 3), (800, 1196, 3)]}
+
+# 拆分split
+data_sample1_, data_sample2_ = data_sample.split()
+assert (data_sample1_.gt_img == img1).all()
+assert (data_sample2_.gt_img == img2).all()
+```