beit_large_patch16_224.in22k_ft_in22k

Clasificación de Imágenes

from urllib.request import urlopen
from PIL import Image
import timm

img = Image.open(urlopen('https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'))

model = timm.create_model('beit_large_patch16_224.in22k_ft_in22k', pretrained=True)
model = model.eval()

# obtener transformaciones específicas del modelo (normalización, cambio de tamaño)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)

output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1

top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)

Embeddings de Imágenes

from urllib.request import urlopen
from PIL import Image
import timm

img = Image.open(urlopen('https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'))

model = timm.create_model('beit_large_patch16_224.in22k_ft_in22k', pretrained=True, num_classes=0) # remover clasificador nn.Linear
model = model.eval()

# obtener transformaciones específicas del modelo (normalización, cambio de tamaño)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)

output = model(transforms(img).unsqueeze(0)) # salida es un tensor con forma (batch_size, num_features)

# o equivalentemente (sin necesidad de poner num_classes=0)

output = model.forward_features(transforms(img).unsqueeze(0))
# salida es sin agrupar, un tensor con forma (1, 197, 1024)

output = model.forward_head(output, pre_logits=True)
# salida es un tensor con forma (1, num_features)