Update README.md (#5015)
Browse files
README.md
CHANGED
@@ -245,17 +245,15 @@ We are super excited about our first-ever Ultralytics YOLOv5 🚀 EXPORT Competi
|
|
245 |
<details>
|
246 |
<summary>Table Notes (click to expand)</summary>
|
247 |
|
|
|
248 |
* AP<sup>test</sup> denotes COCO [test-dev2017](http://cocodataset.org/#upload) server results, all other AP results
|
249 |
denote val2017 accuracy.
|
250 |
-
*
|
251 |
-
|
252 |
-
* Speed<sub>GPU</sub> averaged over 5000 COCO val2017 images using a
|
253 |
GCP [n1-standard-16](https://cloud.google.com/compute/docs/machine-types#n1_standard_machine_types) V100 instance, and
|
254 |
-
includes FP16 inference, postprocessing and NMS
|
255 |
by `python val.py --data coco.yaml --img 640 --conf 0.25 --iou 0.45 --half`
|
256 |
-
*
|
257 |
-
* Test Time Augmentation ([TTA](https://github.com/ultralytics/yolov5/issues/303)) includes reflection and scale
|
258 |
-
augmentation. **Reproduce TTA** by `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment`
|
259 |
|
260 |
</details>
|
261 |
|
|
|
245 |
<details>
|
246 |
<summary>Table Notes (click to expand)</summary>
|
247 |
|
248 |
+
* All checkpoints are trained to 300 epochs with default settings and hyperparameters.
|
249 |
* AP<sup>test</sup> denotes COCO [test-dev2017](http://cocodataset.org/#upload) server results, all other AP results
|
250 |
denote val2017 accuracy.
|
251 |
+
* **mAP** values are for single-model single-scale unless otherwise noted.<br>**Reproduce** by `python val.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65`
|
252 |
+
* **Speed** averaged over 5000 COCO val2017 images using a
|
|
|
253 |
GCP [n1-standard-16](https://cloud.google.com/compute/docs/machine-types#n1_standard_machine_types) V100 instance, and
|
254 |
+
includes FP16 inference, postprocessing and NMS.<br>**Reproduce**
|
255 |
by `python val.py --data coco.yaml --img 640 --conf 0.25 --iou 0.45 --half`
|
256 |
+
* **TTA** [Test Time Augmentation](https://github.com/ultralytics/yolov5/issues/303) includes reflection and scale.<br>**Reproduce** by `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment`
|
|
|
|
|
257 |
|
258 |
</details>
|
259 |
|