Spaces:
Sleeping
Sleeping
| import math | |
| import numpy as np | |
| from torch.nn import functional as F | |
| def get_abs_pos(abs_pos, tgt_size): | |
| # abs_pos: L, C | |
| # tgt_size: M | |
| # return: M, C | |
| src_size = int(math.sqrt(abs_pos.size(0))) | |
| tgt_size = int(math.sqrt(tgt_size)) | |
| dtype = abs_pos.dtype | |
| if src_size != tgt_size: | |
| return F.interpolate( | |
| abs_pos.float().reshape(1, src_size, src_size, -1).permute(0, 3, 1, 2), | |
| size=(tgt_size, tgt_size), | |
| mode="bicubic", | |
| align_corners=False, | |
| ).permute(0, 2, 3, 1).flatten(0, 2).to(dtype=dtype) | |
| else: | |
| return abs_pos | |
| # https://github.com/facebookresearch/mae/blob/efb2a8062c206524e35e47d04501ed4f544c0ae8/util/pos_embed.py#L20 | |
| def get_2d_sincos_pos_embed(embed_dim, grid_size, cls_token=False): | |
| """ | |
| grid_size: int of the grid height and width | |
| return: | |
| pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/ or w/o cls_token) | |
| """ | |
| grid_h = np.arange(grid_size, dtype=np.float32) | |
| grid_w = np.arange(grid_size, dtype=np.float32) | |
| grid = np.meshgrid(grid_w, grid_h) # here w goes first | |
| grid = np.stack(grid, axis=0) | |
| grid = grid.reshape([2, 1, grid_size, grid_size]) | |
| pos_embed = get_2d_sincos_pos_embed_from_grid(embed_dim, grid) | |
| if cls_token: | |
| pos_embed = np.concatenate([np.zeros([1, embed_dim]), pos_embed], axis=0) | |
| return pos_embed | |
| def get_2d_sincos_pos_embed_from_grid(embed_dim, grid): | |
| assert embed_dim % 2 == 0 | |
| # use half of dimensions to encode grid_h | |
| emb_h = get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[0]) # (H*W, D/2) | |
| emb_w = get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[1]) # (H*W, D/2) | |
| emb = np.concatenate([emb_h, emb_w], axis=1) # (H*W, D) | |
| return emb | |
| def get_1d_sincos_pos_embed_from_grid(embed_dim, pos): | |
| """ | |
| embed_dim: output dimension for each position | |
| pos: a list of positions to be encoded: size (M,) | |
| out: (M, D) | |
| """ | |
| assert embed_dim % 2 == 0 | |
| omega = np.arange(embed_dim // 2, dtype=np.float32) | |
| omega /= embed_dim / 2. | |
| omega = 1. / 10000**omega # (D/2,) | |
| pos = pos.reshape(-1) # (M,) | |
| out = np.einsum('m,d->md', pos, omega) # (M, D/2), outer product | |
| emb_sin = np.sin(out) # (M, D/2) | |
| emb_cos = np.cos(out) # (M, D/2) | |
| emb = np.concatenate([emb_sin, emb_cos], axis=1) # (M, D) | |
| return emb |