v1.0.1 release

dbolya · Aug 31, 2020 · 49a5d2a · 49a5d2a
1 parent ab60912
commit 49a5d2a
Show file tree

Hide file tree

Showing 6 changed files with 29 additions and 17 deletions.
diff --git a/.gitignore b/.gitignore
@@ -6,3 +6,6 @@ dist/
 .ipynb*
 examples/mask_rcnn_mask.json
 examples/mask_rcnn_bbox.json
+examples/mask_rcnn_lvis.json
+minimal_example.py
+figs/
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -0,0 +1,14 @@
+# TIDE Changelog
+
+## v1.0.1
+### Error Calculation
+ - **Important**: Fixed an oversight where detections ignored by AP calculation were not allowed to contribute to fixing errors. This caused a lot of error in datasets with large amounts of ignore regions (e.g., LVIS) to significantly overrepresent Missed Error and underrepresent either Classification or Localization error. This fix will also slightly change errors for other datasets (< .4 dAP on COCO), but conclusions on LVIS change dramatically.
+### Plotting
+ - Fixed issue with pie plots sometimes having negative padding.
+ - Added automatic rescaling to the bar plots. This scalining will persist for the entire TIDE object, so you can compare between runs.
+### Datasets
+ - Fixed bug in the LVIS automatic download script that caused it to crash after downloading.
+
+
+## v1.0.0
+ - Initial release.
diff --git a/README.md b/README.md
@@ -5,14 +5,14 @@
    ██║   ██║██║  ██║█████╗  
    ██║   ██║██║  ██║██╔══╝  
    ██║   ██║██████╔╝███████╗
-   ╚═╝   ╚═╝╚═════╝ ╚══════╝                            
+   ╚═╝   ╚═╝╚═════╝ ╚══════╝
 ```
 
-An easy-to-use, general toolbox to compute and evaluate the effect of object detection and instance segmentation on overall performance. This is the code for our paper: [TIDE: A General Toolbox for Identifying Object Detection Errors](https://dbolya.github.io/tide/paper.pdf).
+An easy-to-use, general toolbox to compute and evaluate the effect of object detection and instance segmentation on overall performance. This is the code for our paper: [TIDE: A General Toolbox for Identifying Object Detection Errors](https://dbolya.github.io/tide/paper.pdf) ([ArXiv](https://arxiv.org/abs/2008.08115)) [ECCV2020 Spotlight].
 
 Check out our ECCV 2020 short video for an explanation of what TIDE can do:
 
-[![TIDE Introduction](https://img.youtube.com/vi/Pz4uggQzJXU/0.jpg)](https://www.youtube.com/watch?v=Pz4uggQzJXU)
+[![TIDE Introduction](https://img.youtube.com/vi/McYFYU3PXcU/0.jpg)](https://youtu.be/McYFYU3PXcU)
 
 # Installation
 
@@ -21,6 +21,8 @@ TIDE is available as a python package for python 3.6+ as [tidecv](https://pypi.o
 pip3 install tidecv
 ```
 
+The current version is v1.0.1 ([changelog](https://github.com/dbolya/tide/blob/master/CHANGELOG.md)).
+
 # Usage
 TIDE is meant as a drop-in replacement for the [COCO Evaluation toolkit](https://github.com/cocodataset/cocoapi), and getting started is easy:
 
@@ -43,7 +45,7 @@ bbox AP @ 50: 61.80
 =============================================================
   Type      Cls      Loc     Both     Dupe      Bkg     Miss
 -------------------------------------------------------------
-   dAP     3.33     6.30     1.18     0.19     3.96     7.99
+   dAP     3.40     6.65     1.18     0.19     3.96     7.53
 =============================================================
 
         Special Error

diff --git a/examples/coco_instance_segmentation.ipynb b/examples/coco_instance_segmentation.ipynb
diff --git a/setup.py b/setup.py
@@ -10,7 +10,7 @@
 # This call to setup() does all the work
 setup(
     name="tidecv",
-    version="1.0.0",
+    version="1.0.1",
     description="A General Toolbox for Identifying ObjectDetection Errors",
     long_description=README,
     long_description_content_type="text/markdown",

diff --git a/tidecv/__init__.py b/tidecv/__init__.py
@@ -1,4 +1,4 @@
-__version__ = '1.0.0'
+__version__ = '1.0.1'
 
 from .quantify import *
 from .errors.qualifiers import *