swim-224-base-satellite-image-classification

Model Details

• Model Type: Image classification / feature backbone

Model Stats:

• Params (M): 71.1
• GMACs: 13.7
• Activations (M): 48.3
• Image size: 224 x 224

Papers:

• AutoFormerV2: https://arxiv.org/abs/2111.1472
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows: https://arxiv.org/abs/2103.14030
• Original: https://github.com/microsoft/Cream/tree/main/AutoFormerV2

how to load the model

import joblib
from huggingface_hub import hf_hub_download
import safetensors
import torch

REPO_ID = "subhuatharva/swim-224-base-satellite-image-classification"
FILENAME = "model.safetensors"

# Download the model file
model_path = hf_hub_download(repo_id=REPO_ID, filename=FILENAME)
# intialize the model
model = create_model(
    model_name,
    num_classes=17
)
load_model(model, model_path)