Model card for convnextv2_tiny.ft_plantdoc_384

Overview

This model classifies diseases from plant images.

• Dataset size: 8000 images
• Number of classes: 39
• Architecture: ConvNeXtV2 Tiny (384)

Metrics

• mAP: 0.94
• Accuracy: 0.86

Model

Model Details

• Model Type: Image classification
• Backbone: convnextv2_tiny.fcmae_ft_in22k_in1k_384
• Model Stats:
  • Params (M): 28.6
  • GMACs: 13.1
  • Activations (M): 39.5
  • Image size: 384 x 384
• Papers:
  • ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders: https://arxiv.org/abs/2301.00808
• Original: https://github.com/facebookresearch/ConvNeXt-V2
• Dataset: PlantDoc (8000 images)
• Pretrain Dataset: ImageNet-1k

Model Usage

Built with:

import torch
import timm

# create model
model = timm.create_model(
    "convnextv2_tiny.fcmae_ft_in22k_in1k_384",
    pretrained=False,
    num_classes=39,
    drop_rate=0.2,
    drop_path_rate=0.2,
)

# load weights
state_dict = torch.load("model.bin", map_location="cpu")
model.load_state_dict(state_dict)