File size: 947 Bytes
3533c69
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch

def fasterrcnn_reshape_transform(x):
    target_size = x['pool'].size()[-2 : ]
    activations = []
    for key, value in x.items():
        activations.append(torch.nn.functional.interpolate(torch.abs(value), target_size, mode='bilinear'))
    activations = torch.cat(activations, axis=1)
    return activations

def swinT_reshape_transform(tensor, height=7, width=7):
    result = tensor.reshape(tensor.size(0),
                            height, width, tensor.size(2))

    # Bring the channels to the first dimension,
    # like in CNNs.
    result = result.transpose(2, 3).transpose(1, 2)
    return result

def vit_reshape_transform(tensor, height=14, width=14):
    result = tensor[:, 1:, :].reshape(tensor.size(0),
                                      height, width, tensor.size(2))

    # Bring the channels to the first dimension,
    # like in CNNs.
    result = result.transpose(2, 3).transpose(1, 2)
    return result