from mlxim.model import create_model
from mlxim.io import read_rgb
from mlxim.transform import ImageNetTransform

transform = ImageNetTransform(train=False, img_size=224)
x = transform(read_rgb("cat.png"))
x = mx.expand_dims(x, 0)

model = create_model("swin_tiny_patch4_window7_224")
model.eval()

logits = model(x)

You can also use the embeds from layer before head:

python

from mlxim.model import create_model
from mlxim.io import read_rgb
from mlxim.transform import ImageNetTransform

transform = ImageNetTransform(train=False, img_size=224)
x = transform(read_rgb("cat.png"))
x = mx.expand_dims(x, 0)

# first option
model = create_model("swin_tiny_patch4_window7_224", num_classes=0)
model.eval()

embeds = model(x)

# second option
model = create_model("swin_tiny_patch4_window7_224")
model.eval()

embeds = model.get_features(x)

## Model Comparison

Explore the metrics of this model in mlx-image model results.

mlx-vision/swin_tiny_patch4_window7_224-mlxim

swin_tiny_patch4_window7_224

How to use

mlx-vision/
swin_tiny_patch4_window7_224-mlxim