From b20e38149226dcf28a1c5962fbef82b20c8ffaa3 Mon Sep 17 00:00:00 2001 From: Glenn Jocher Date: Thu, 30 Sep 2021 14:06:00 -0700 Subject: [PATCH] Update README.md (#5015) --- README.md | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/README.md b/README.md index 854c38c..06a2187 100644 --- a/README.md +++ b/README.md @@ -245,17 +245,15 @@ We are super excited about our first-ever Ultralytics YOLOv5 🚀 EXPORT Competi
Table Notes (click to expand) +* All checkpoints are trained to 300 epochs with default settings and hyperparameters. * APtest denotes COCO [test-dev2017](http://cocodataset.org/#upload) server results, all other AP results denote val2017 accuracy. -* AP values are for single-model single-scale unless otherwise noted. **Reproduce mAP** - by `python val.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65` -* SpeedGPU averaged over 5000 COCO val2017 images using a +* **mAP** values are for single-model single-scale unless otherwise noted.
**Reproduce** by `python val.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65` +* **Speed** averaged over 5000 COCO val2017 images using a GCP [n1-standard-16](https://cloud.google.com/compute/docs/machine-types#n1_standard_machine_types) V100 instance, and - includes FP16 inference, postprocessing and NMS. **Reproduce speed** + includes FP16 inference, postprocessing and NMS.
**Reproduce** by `python val.py --data coco.yaml --img 640 --conf 0.25 --iou 0.45 --half` -* All checkpoints are trained to 300 epochs with default settings and hyperparameters (no autoaugmentation). -* Test Time Augmentation ([TTA](https://github.com/ultralytics/yolov5/issues/303)) includes reflection and scale - augmentation. **Reproduce TTA** by `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment` +* **TTA** [Test Time Augmentation](https://github.com/ultralytics/yolov5/issues/303) includes reflection and scale.
**Reproduce** by `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment`