Image Classification
PyTorch
few-shot-learning
parameter-efficient
File size: 657 Bytes
6edcaab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
from .models import (
    ALPINE_CIFAR,
    ALPINE_Native,
    ALPINE_35k_CIFAR,
    ALPINE_35k_Native,
    IRFEExpF3_CIFAR,
    IRFEExpF3_Native,
    IRFEExpF3_35k_CIFAR,
    IRFEExpF3_35k_Native,
    WideWindowAdaptivePatchLocator,
    get_exp_p1_base_centers,
    extract_patches_grid_sample,
    load_alpine_model,
)

__all__ = [
    "ALPINE_CIFAR",
    "ALPINE_Native",
    "ALPINE_35k_CIFAR",
    "ALPINE_35k_Native",
    "IRFEExpF3_CIFAR",
    "IRFEExpF3_Native",
    "IRFEExpF3_35k_CIFAR",
    "IRFEExpF3_35k_Native",
    "WideWindowAdaptivePatchLocator",
    "get_exp_p1_base_centers",
    "extract_patches_grid_sample",
    "load_alpine_model",
]