Dataset Versions
Versions
armv8-cpu-deploy-test
v8
· 2 years ago
TEST-Resize640_augmented15x
v7
· 2 years ago
Resize640_augmented15x
v6
· 2 years ago
test2-newClassificationVisualization
v5
· 2 years ago
test
v4
· 3 years ago
2022-01-29 1:34pm
v3
· 3 years ago
raw-images_originalUploads
v2
· 3 years ago
augmented-v1
v1
· 3 years ago
v1
augmented-v1
Generated on Jan 29, 2022
1080 Total Images
View All Images
Dataset Split
Train Set 97%
1050Images
Valid Set 2%
20Images
Test Set 1%
10Images
Preprocessing
Auto-Orient: Applied
Resize: Fill (with center crop) in 416x416
Augmentations
Outputs per training example: 15
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -5° and +5°
Grayscale: Apply to 5% of images
Hue: Between -10° and +10°
Exposure: Between -10% and +10%
Blur: Up to 1px