ITTR - Pytorch
Implementation of the Hybrid Perception Block (HPB) and Dual-Pruned Self-Attention (DPSA) block from the ITTR paper for Image to Image Translation using Transformers.
Install
$ pip install ITTR-pytorch
Usage
They had 9 blocks of Hybrid Perception Block (HPB) in the paper
import torch
from ITTR_pytorch import HPB
block = HPB(
dim = 512, # dimension
dim_head = 32, # dimension per attention head
heads = 8, # number of attention heads
attn_height_top_k = 16, # number of top indices to select along height, for the attention pruning
attn_width_top_k = 16, # number of top indices to select along width, for the attention pruning
attn_dropout = 0., # attn dropout
ff_mult = 4, # expansion factor of feedforward
ff_dropout = 0. # feedforward dropout
)
fmap = torch.randn(1, 512, 32, 32)
out = block(fmap) # (1, 512, 32, 32)
You can also use the dual-pruned self-attention as so
import torch
from ITTR_pytorch import DPSA
attn = DPSA(
dim = 512, # dimension
dim_head = 32, # dimension per attention head
heads = 8, # number of attention heads
height_top_k = 48, # number of top indices to select along height, for the attention pruning
width_top_k = 48, # number of top indices to select along width, for the attention pruning
dropout = 0. # attn dropout
)
fmap = torch.randn(1, 512, 32, 32)
out = attn(fmap) # (1, 512, 32, 32)
Citations
@inproceedings{Zheng2022ITTRUI,
title = {ITTR: Unpaired Image-to-Image Translation with Transformers},
author = {Wanfeng Zheng and Qiang Li and Guoxin Zhang and Pengfei Wan and Zhongyuan Wang},
year = {2022}
}
5 Sep 15, 2022
43 Dec 28, 2022
18 Nov 17, 2022
23 Nov 30, 2022
20 Dec 12, 2022
138 Dec 24, 2022
22 Dec 14, 2022
211 Dec 28, 2022
12 Sep 28, 2022
1.8k Dec 28, 2022
819 Dec 30, 2022
919 Jan 03, 2023
1.9k Jan 08, 2023
481 Dec 30, 2022
13 Jan 06, 2023
65 Sep 21, 2022
9 Mar 02, 2022
151 Dec 14, 2022
75 Dec 27, 2022
7.8k Jan 09, 2023
2 Sep 27, 2022
22 Dec 06, 2022
230 Nov 16, 2022
2 Aug 16, 2022
106 Jan 01, 2023
28 Oct 30, 2022
9 Oct 11, 2022