HongChuang/ultralytics-ascend
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 3

Add dota8.yaml and O tests (#7394)

Browse source 
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
This commit is contained in:
Laughing 2024-01-09 02:54:09 +08:00 • committed by GitHub
parent d0562d7a2f
commit a6a2c256d4
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
13 changed files with 176 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

29
docs/en/datasets/obb/dota-v2.md
View file

@ -66,13 +66,40 @@ Typically, datasets incorporate a YAML (Yet Another Markup Language) file detail
--8<-- "ultralytics/cfg/datasets/DOTAv1.yaml"
```
## Split DOTA images
To train DOTA dataset, We split original DOTA images with high-resolution into images with 1024x1024 resolution in multi-scale way.
!!! Example "Split images"
=== "Python"
```python
from ultralytics.data.split_dota import split_trainval, split_test
# split train and val set, with labels.
split_trainval(
data_root='path/to/DOTAv1.0/',
save_dir='path/to/DOTAv1.0-split/',
rates=[0.5, 1.0, 1.5], # multi-scale
gap=500
)
# split test set, without labels.
split_test(
data_root='path/to/DOTAv1.0/',
save_dir='path/to/DOTAv1.0-split/',
rates=[0.5, 1.0, 1.5], # multi-scale
gap=500
)
```
## Usage
To train a model on the DOTA v1 dataset, you can utilize the following code snippets. Always refer to your model's documentation for a thorough list of available arguments.
!!! Warning
Please note that all images and associated annotations in the DOTAv2 dataset can be used for academic purposes, but commercial use is prohibited. Your understanding and respect for the dataset creators' wishes are greatly appreciated!
Please note that all images and associated annotations in the DOTAv1 dataset can be used for academic purposes, but commercial use is prohibited. Your understanding and respect for the dataset creators' wishes are greatly appreciated!
!!! Example "Train Example"