# Load model directly
from transformers import AutoImageProcessor, AutoModelForImageClassification
processor = AutoImageProcessor.from_pretrained("dima806/14_flower_types_image_detection")
model = AutoModelForImageClassification.from_pretrained("dima806/14_flower_types_image_detection")Quick Links
Returns flower type given an image with about 99% accuracy.
See https://www.kaggle.com/code/dima806/14-flowers-image-detection-vit for more details.
Classification report:
precision recall f1-score support
rose 0.9951 0.9737 0.9843 419
astilbe 0.9952 0.9905 0.9928 419
carnation 0.9627 0.9857 0.9741 419
tulip 0.9929 1.0000 0.9964 420
water_lily 1.0000 0.9905 0.9952 419
bellflower 0.9811 0.9905 0.9857 419
coreopsis 0.9881 0.9881 0.9881 419
common_daisy 0.9858 0.9928 0.9893 419
iris 0.9953 1.0000 0.9976 420
dandelion 0.9905 1.0000 0.9952 419
sunflower 0.9976 0.9976 0.9976 419
california_poppy 0.9951 0.9761 0.9855 419
black_eyed_susan 0.9882 1.0000 0.9941 419
calendula 0.9854 0.9667 0.9760 420
accuracy 0.9894 5869
macro avg 0.9895 0.9894 0.9894 5869
weighted avg 0.9895 0.9894 0.9894 5869
- Downloads last month
- 9
Model tree for dima806/14_flower_types_image_detection
Base model
google/vit-base-patch16-224-in21k
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="dima806/14_flower_types_image_detection") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")