Model card for tf_efficientnetv2_s.ft_plantdoc_384

Overview

A EfficientNet-v2 small image classification model. Trained on PlantDoc

• Dataset size: 8000 images
• Number of classes: 39
• Architecture: EfficientNet-v2 Small (384)

Metrics

• mAP: 0.87
• Accuracy: 0.81

Model

Model Details

• Model Type: Image classification
• Backbone: tf_efficientnetv2_s
• Model Stats:
  • Params (M): 21.5
  • GMACs: 5.4
  • Activations (M): 22.7
  • Image size: 384 x 384
• Papers:
  • EfficientNetV2: Smaller Models and Faster Training: https://arxiv.org/abs/2104.00298
• Original: https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
• Dataset:
  • PlantDoc (8000 images)
• Pretrain Dataset: ImageNet-1k

Model Usage

Built with:

import torch
import timm

# create model
model = timm.create_model(
    "tf_efficientnetv2_s",
    pretrained=False,
    num_classes=39,
)

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