gemma3n-plant-disease-classifier

Gemma3n Plant Disease Classifier

This model is a fine-tuned version of Gemma3n-E2B-it for plant disease classification.

Model Description

• Base Model: unsloth/gemma-3n-E2B-it
• Task: Plant Disease Classification
• Training Data: New Plant Diseases Dataset (Augmented)
• Classes: 38 plant disease categories

Usage

from huggingface_hub import snapshot_download
import sys

# Download model from HuggingFace Hub
model_path = snapshot_download("hiro785/gemma3n-plant-disease-classifier")

# Add the model directory to Python path
sys.path.append(model_path)

# Load the model using the provided loader
from load_model import load_vision_model
from PIL import Image

# Load model, tokenizer, and class names
model, tokenizer, class_names = load_vision_model(model_path)

# Predict plant disease
image = Image.open("plant_image.jpg")
# Use your prediction pipeline here

Training Details

• Training Framework: Unsloth + Transformers
• Optimization: 4-bit quantization
• GPU Memory: Optimized for efficient inference
• Fine-tuning: Vision tower with frozen language model

Performance

The model has been trained on a comprehensive dataset of plant diseases with data augmentation for improved generalization.

Citation

@misc{gemma3n-plant-disease-classifier,
  author = {Plant Disease Diagnosis Team},
  title = {Gemma3n Plant Disease Classifier},
  year = {2025},
  publisher = {HuggingFace},
  journal = {HuggingFace repository},
  howpublished = {\url{hiro785/gemma3n-plant-disease-classifier}}
}