vit-plant-classifier

Medicinal Plant Classifier (ViT)

This model is a fine-tuned Vision Transformer (ViT) on a dataset of Indian medicinal plant leaves.
It predicts the plant species based on leaf images.

Dataset

• 80 classes of Indian medicinal leaf images
• Public dataset from Kaggle: indian-medicinal-leaves-dataset

Model Details

• Base model: google/vit-base-patch16-224-in21k
• Fine-tuned on a labeled image dataset of 80 medicinal leaf images
• For more visit this repo: GitHub: medicinal-plant-classifier-and-insight-generator

How to Use

from transformers import AutoImageProcessor, AutoModelForImageClassification
from PIL import Image
import torch

image = Image.open("your_plant_leaf.jpg")

processor = AutoImageProcessor.from_pretrained("dhivyasri-d/vit-plant-classifier")
model = AutoModelForImageClassification.from_pretrained("dhivyasri-d/vit-plant-classifier")

inputs = processor(images=image, return_tensors="pt")
with torch.no_grad():
    logits = model(**inputs).logits

predicted_class_idx = logits.argmax(-1).item()
print("Predicted label:", model.config.id2label[predicted_class_idx])