Dataset Versions
Versions
english version-65
v15
· a year ago
english version-48
v14
· a year ago
english version-80-classes
v13
· a year ago
test version with 5 classes
v11
· 2 years ago
Version for YOLACT -700 imgsz-
v9
· 2 years ago
New version with augmentation and 2x volume
v8
· 2 years ago
New version without augmentation
v7
· 2 years ago
Remapping classes
v6
· 2 years ago
2023-03-23 5:25pm
v5
· 2 years ago
2023-03-23 5:14pm
v4
· 2 years ago
2023-03-23 5:10pm
v3
· 2 years ago
2023-02-20 9:36pm
v2
· 2 years ago
planeat-test-v1
v1
· 2 years ago
v9
Version for YOLACT -700 imgsz-
Generated on May 8, 2023
2046 Total Images
View All Images
Dataset Split
Train Set 79%
1620Images
Valid Set 14%
281Images
Test Set 7%
145Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 700x700
Modify Classes: 80 remapped, 60 dropped
Filter Null: Require all images to contain annotations.
Augmentations
Outputs per training example: 2
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Rotation: Between -10° and +10°
Saturation: Between -25% and +25%
Exposure: Between -25% and +25%