File size: 1,258 Bytes
e65bf54
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0186575
 
 
e65bf54
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
31
32
33
34
35
import torch
import torchvision

from torch import nn


def create_vit_model(num_classes:int=7, 
                          seed:int=42):
    """Creates an ViT feature extractor model and transforms.

    Args:
        num_classes (int, optional): number of classes in the classifier head. 
            Defaults to 7.
        seed (int, optional): random seed value. Defaults to 42.

    Returns:
        model (torch.nn.Module): ViT feature extractor model. 
        transforms (torchvision.transforms): ViT image transforms.
    """
    # Create ViT pretrained weights, transforms and model
    weights = torchvision.models.ViT_B_16_Weights.DEFAULT
    transforms = weights.transforms()
    model = torchvision.models.vit_b_16(weights=weights)

    # Freeze all layers in base model
    for param in model.parameters():
        param.requires_grad = False

    # Change classifier head with random seed for reproducibility
    torch.manual_seed(seed)
    model.heads = nn.Sequential(nn.Dropout(p=0.3, inplace=True),
                                nn.Linear(in_features=768, # keep this the same as original model
                                          out_features=num_classes)) # update to reflect target number of classes
    return model, transforms