darjja/oxford-17-flowers-cnn
Image Classification • Updated • 48
image imagewidth (px) 499 1.06k | label class label 17
classes |
|---|---|
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
0Bluebell | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup | |
1Buttercup |
17-class flower image dataset with train/test splits (80/20 per class).
| Split | Images |
|---|---|
| train | 1,088 |
| test | 272 |
Classes: Bluebell, Buttercup, Coltsfoot, Cowslip, Crocus, Daffodil, Daisy, Dandelion, Fritillary, Iris, Lilyvalley, Pansy, Snowdrop, Sunflower, Tigerlily, Tulip, Windflower
from datasets import load_dataset
dataset = load_dataset("darjja/oxford-17-flowers")
print(dataset["train"][0]) # {'image': <PIL.Image>, 'label': 0}
Splits are defined in train.txt and test.txt inside the source folder. Regenerate them with:
python create_split.py