[Datumaro] Update documentation

datumaro/CONTRIBUTING.md

@@ -129,7 +129,7 @@ Plugins reside in plugin directories:
 - `<project_dir>/.datumaro/plugins` for project-specific components
 
 A plugin is a python file or package with any name, which exports some symbols.
-To export a symbol put it to `exports` list of the module like this:
+To export a symbol, put it to `exports` list of the module like this:
 
 ``` python
 class MyComponent1: ...

datumaro/README.md

@@ -1,13 +1,13 @@
-# Dataset Framework (Datumaro)
+# Dataset Management Framework (Datumaro)
 
 A framework to build, transform, and analyze datasets.
 
 <!--lint disable fenced-code-flag-->
 ```
 CVAT annotations  --                              ---> Annotation tool
-...                  \                          /
+                     \                          /
 COCO-like dataset -----> Datumaro ---> dataset ------> Model training
-...                  /                          \
+                     /                          \
 VOC-like dataset  --                              ---> Publication etc.
 ```
 <!--lint enable fenced-code-flag-->
@@ -55,12 +55,12 @@ VOC-like dataset  --                              ---> Publication etc.
 - Dataset building operations:
   - Merging multiple datasets into one
   - Dataset filtering with custom conditions, for instance:
-    - remove all annotations except polygons of a certain class
+    - remove polygons of a certain class
     - remove images without a specific class
-    - remove occluded annotations from images
+    - remove `occluded` annotations from images
     - keep only vertically-oriented images
     - remove small area bounding boxes from annotations
-  - Annotation conversions, for instance
+  - Annotation conversions, for instance:
     - polygons to instance masks and vise-versa
     - apply a custom colormap for mask annotations
     - rename or remove dataset labels

datumaro/datumaro/cli/contexts/project/__init__.py

@@ -579,7 +579,7 @@ def build_transform_parser(parser_ctor=argparse.ArgumentParser):
             |n
             Examples:|n
             - Convert instance polygons to masks:|n
-            |s|stransform -n polygons_to_masks
+            |s|stransform -t polygons_to_masks
         """ % ', '.join(builtins),
         formatter_class=MultilineFormatter)
 

datumaro/datumaro/plugins/transforms.py

@@ -409,6 +409,13 @@ def transform_item(self, item):
             .format(item=item))
 
 class RemapLabels(Transform, CliPlugin):
+    """
+    Changes labels in the dataset.|n
+    Examples:|n
+    - Rename 'person' to 'car' and 'cat' to 'dog', keep 'bus', remove others:|n
+    |s|sremap_labels -l person:car -l bus:bus -l cat:dog --default delete
+    """
+
     DefaultAction = Enum('DefaultAction', ['keep', 'delete'])
 
     @staticmethod
@@ -428,7 +435,7 @@ def build_cmdline_parser(cls, **kwargs):
         parser.add_argument('--default',
             choices=[a.name for a in cls.DefaultAction],
             default=cls.DefaultAction.keep.name,
-            help="Action for unspecified labels")
+            help="Action for unspecified labels (default: %(default)s)")
         return parser
 
     def __init__(self, extractor, mapping, default=None):

datumaro/docs/design.md

@@ -49,6 +49,8 @@ Datumaro is:
 - Versioning (for images, annotations, subsets, sources etc., comparison)
 - Documentation generation
 - Provision of iterators for user code
+- Dataset downloading
+- Dataset generation
 - Dataset building (export in a specific format, indexation, statistics, documentation)
 - Dataset exporting to other formats
 - Dataset debugging (run inference, generate dataset slices, compute statistics)
@@ -111,25 +113,25 @@ can be downloaded by user to be operated on with Datumaro CLI.
     - [ ] with TensorBoard
 
 - Calculation of statistics for datasets
-  - [ ] Pixel mean, std
-  - [ ] Object counts (detection scenario)
-  - [ ] Image-Class distribution (classification scenario)
-  - [ ] Pixel-Class distribution (segmentation scenario)
-  - [ ] Image clusters
+  - [x] Pixel mean, std
+  - [x] Object counts (detection scenario)
+  - [x] Image-Class distribution (classification scenario)
+  - [x] Pixel-Class distribution (segmentation scenario)
+  - [ ] Image similarity clusters
   - [ ] Custom statistics
 
 - Dataset building
   - [x] Composite dataset building
-  - [ ] Annotation remapping
-  - [ ] Subset splitting
+  - [x] Class remapping
+  - [x] Subset splitting
   - [x] Dataset filtering (`extract`)
   - [x] Dataset merging (`merge`)
   - [ ] Dataset item editing (`edit`)
 
 - Dataset comparison (`diff`)
   - [x] Annotation-annotation comparison
   - [x] Annotation-inference comparison
-  - [ ] Annotation quality estimation (for CVAT)
+  - [x] Annotation quality estimation (for CVAT)
     - Provide a simple method to check
       annotation quality with a model and generate summary
 
@@ -142,9 +144,9 @@ can be downloaded by user to be operated on with Datumaro CLI.
   - [x] Task export
     - [x] Datumaro project export
     - [x] Dataset export
-    - [ ] Original raw data (images, a video file) can be downloaded (exported)
+    - [x] Original raw data (images, a video file) can be downloaded (exported)
       together with annotations or just have links
-      on CVAT server (in the future support S3, etc)
+      on CVAT server (in future, support S3, etc)
       - [x] Be able to use local files instead of remote links
         - [ ] Specify cache directory
   - [x] Use case "annotate for model training"
@@ -154,7 +156,7 @@ can be downloaded by user to be operated on with Datumaro CLI.
     - convert to a training format
     - train a DL model
   - [x] Use case "annotate - reannotate problematic images - merge"
-  - [ ] Use case "annotate and estimate quality"
+  - [x] Use case "annotate and estimate quality"
     - create a task
     - annotate
     - estimate quality of annotations