Commits · c89156babf6d19e2744df6c57032491134500a92 · OpenDAS / vision

09 Apr, 2020 1 commit
- Make norm layer as parameters in models.detection.backbone_utils.py (#2081) · c89156ba
  Luan Pham authored Apr 09, 2020
  
  c89156ba
25 Nov, 2019 1 commit

Make maskrcnn scriptable (#1407) · d88d8961

eellison authored Nov 25, 2019

* almost working...

* respond to comments

* add empty tensor op, handle different output types in generalized rcnn

* clean ups

* address comments

* more changes

* it's working!

* torchscript bugs

* add script/ eager test

* eval script model

* fix flake

* division import

* py2 compat

* update test, fix arange bug

* import division statement

* fix linter

* fixes

* changes needed for JIT master

* cleanups

* remove imagelist_to

* requested changes

* Make FPN backwards-compatible and torchscript compatible

We remove support for feature channels=0, but support for it was already a bit limited

* Fix ONNX regression

d88d8961

23 Jul, 2019 1 commit
- Fix resnet fpn backbone for resnet18 and resnet34 (#1147) · 737966a3
  Kyryl Truskovskyi authored Jul 23, 2019
```
* in_channels_stage2 from backbone.inplanes

* remove type for backward compatible
```
  737966a3
20 May, 2019 1 commit

Improve documentation for detection models (#928) · f76e598d

Francisco Massa authored May 21, 2019

* Add more documentation for the ops

* Add documentation for Faster R-CNN

* Add documentation for Mask R-CNN and Keypoint R-CNN

* Improve doc for RPN

* Add basic doc for GeneralizedRCNNTransform

* Lint fixes

f76e598d

19 May, 2019 1 commit
- Split mask_rcnn.py into several files (#921) · cf401a70
  Francisco Massa authored May 20, 2019
```
* Split mask_rcnn.py into several files

* Lint
```
  cf401a70