Bugfix: num of augmentations and image meta not match when run TTA on CPU #276
Conversation
|
|
Codecov Report
@@ Coverage Diff @@
## master #276 +/- ##
==========================================
+ Coverage 84.58% 85.36% +0.78%
==========================================
Files 90 90
Lines 4385 4414 +29
Branches 701 706 +5
==========================================
+ Hits 3709 3768 +59
+ Misses 536 501 -35
- Partials 140 145 +5
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
|
Nice PR! |
* FPN deprecated warning * FPN deprecated warning * Support DefaultFormatBundle in image_demo * Fix Multi-Scale error, ref： open-mmlab/mmsegmentation#276 * Rename variable name
* FPN deprecated warning * FPN deprecated warning * Support DefaultFormatBundle in image_demo * Fix Multi-Scale error, ref： open-mmlab/mmsegmentation#276 * Rename variable name
… CPU (open-mmlab#276) * add inference test * fix E501 line too long (81 > 79 characters * fix wrong config path * fix num of augmentations (2) != num of image meta (1) * Update test_inference.py Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com>
|
Hi @youqingxiaozhua！First of all, we want to express our gratitude for your significant PR in the MMSeg project. Your contribution is highly appreciated, and we are grateful for your efforts in helping improve this open-source project during your personal time. We believe that many developers will benefit from your PR. We would also like to invite you to join our Special Interest Group (SIG) private channel on Discord, where you can share your experiences, ideas, and build connections with like-minded peers. To join the SIG channel, simply message moderator— OpenMMLab on Discord or briefly share your open-source contributions in the #introductions channel and we will assist you. Look forward to seeing you there! Join us ：https://discord.gg/raweFPmdzG If you are Chinese or have WeChat，welcome to join our community on WeChat. You can add our assistant ：openmmlabwx. Please add "mmsig + Github ID" as a remark when adding friends：） |
When running test time augmentation(TTA) on CPU, it will raise this Error:
num of augmentations (2) != num of image meta (1). The test log is here: https://github.com/youqingxiaozhua/mmsegmentation/actions/runs/381178359The reason is the
inference_segmentorfunction inmmseg/apis/inference.pyalways returns the first image meta when the model's parameters are located on CPU, which is not suited for augmentation cases.