""" Vision Transformer (ViT) in PyTorch
A PyTorch implement of Vision Transformers as described in:
'Exploring Plain Vision Transformer Backbones for Object Detection'
- https://arxiv.org/abs/2203.16527
'Segment Anything Model (SAM)'
- https://github.com/facebookresearch/segment-anything/
"""
import logging
from functools import partial
from typing import Callable, Optional, Tuple
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.checkpoint
from torch.jit import Final
from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
from timm.layers import PatchEmbed, Mlp, DropPath, PatchDropout, LayerNorm2d, ClassifierHead, NormMlpClassifierHead,\
Format, resample_abs_pos_embed_nhwc, RotaryEmbeddingCat, apply_rot_embed_cat, to_2tuple, use_fused_attn
from ._builder import build_model_with_cfg
from ._manipulate import checkpoint_seq
from ._registry import generate_default_cfgs, register_model
from ._features_fx import register_notrace_function
# model_registry will add each entrypoint fn to this
__all__ = ['VisionTransformerSAM']
_logger = logging.getLogger(__name__)
def get_rel_pos(q_size: int, k_size: int, rel_pos: torch.Tensor) -> torch.Tensor:
"""
Get relative positional embeddings according to the relative positions of
query and key sizes.
Args:
q_size (int): size of query q.
k_size (int): size of key k.
rel_pos (Tensor): relative position embeddings (L, C).
Returns:
Extracted positional embeddings according to relative positions.
"""
max_rel_dist = int(2 * max(q_size, k_size) - 1)
# Interpolate rel pos if needed.
if rel_pos.shape[0] != max_rel_dist:
# Interpolate rel pos.
rel_pos_resized = F.interpolate(
rel_pos.reshape(1, rel_pos.shape[0], -1).permute(0, 2, 1),
size=max_rel_dist,
mode="linear",
)
rel_pos_resized = rel_pos_resized.reshape(-1, max_rel_dist).permute(1, 0)
else:
rel_pos_resized = rel_pos
# Scale the coords with short length if shapes for q and k are different.
q_coords = torch.arange(q_size)[:, None] * max(k_size / q_size, 1.0)
k_coords = torch.arange(k_size)[None, :] * max(q_size / k_size, 1.0)
relative_coords = (q_coords - k_coords) + (k_size - 1) * max(q_size / k_size, 1.0)
return rel_pos_resized[relative_coords.long()]
register_notrace_function(get_rel_pos)
def get_decomposed_rel_pos_bias(
q: torch.Tensor,
rel_pos_h: torch.Tensor,
rel_pos_w: torch.Tensor,
q_size: Tuple[int, int],
k_size: Tuple[int, int],
) -> torch.Tensor:
"""
Calculate decomposed Relative Positional Embeddings from :paper:`mvitv2`.
https://github.com/facebookresearch/mvit/blob/19786631e330df9f3622e5402b4a419a263a2c80/mvit/models/attention.py
Args:
q (Tensor): query q in the attention layer with shape (B, q_h * q_w, C).
rel_pos_h (Tensor): relative position embeddings (Lh, C) for height axis.
rel_pos_w (Tensor): relative position embeddings (Lw, C) for width axis.
q_size (Tuple): spatial sequence size of query q with (q_h, q_w).
k_size (Tuple): spatial sequence size of key k with (k_h, k_w).
Returns:
bias (Tensor): attention bias to add to attention map
"""
q_h, q_w = q_size
k_h, k_w = k_size
Rh = get_rel_pos(q_h, k_h, rel_pos_h)
Rw = get_rel_pos(q_w, k_w, rel_pos_w)
B, _, dim = q.shape
r_q = q.reshape(B, q_h, q_w, dim)
rel_h = torch.einsum("bhwc,hkc->bhwk", r_q, Rh)
rel_w = torch.einsum("bhwc,wkc->bhwk", r_q, Rw)
attn_bias = rel_h[:, :, :, :, None] + rel_w[:, :, :, None, :]
return attn_bias.reshape(-1, q_h * q_w, k_h * k_w)
class Attention(nn.Module):
fused_attn: Final[bool]
def __init__(
self,
dim,
num_heads=8,
qkv_bias=True,
qk_norm=False,
attn_drop=0.,
proj_drop=0.,
norm_layer=nn.LayerNorm,
use_rel_pos: bool = False,
input_size: Optional[Tuple[int, int]] = None,
rope: Optional[nn.Module] = None,
):
super().__init__()
assert dim % num_heads == 0, 'dim should be divisible by num_heads'
self.num_heads = num_heads
self.head_dim = dim // num_heads
self.scale = self.head_dim ** -0.5
self.fused_attn = use_fused_attn()
self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
self.q_norm = norm_layer(self.head_dim) if qk_norm else nn.Identity()
self.k_norm = norm_layer(self.head_dim) if qk_norm else nn.Identity()
self.attn_drop = nn.Dropout(attn_drop)
self.proj = nn.Linear(dim, dim)
self.proj_drop = nn.Dropout(proj_drop)
self.use_rel_pos = use_rel_pos
if self.use_rel_pos:
assert rope is None
assert (
input_size is not None
), "Input size must be provided if using relative positional encoding."
# initialize relative positional embeddings
self.rel_pos_h = nn.Parameter(torch.zeros(
2 * input_size[0] - 1, self.head_dim))
self.rel_pos_w = nn.Parameter(torch.zeros(
2 * input_size[1] - 1, self.head_dim))
self.rope = rope
def forward(self, x):
B, H, W, _ = x.shape
N = H * W
x = x.reshape(B, N, -1)
qkv = self.qkv(x).view(B, N, 3, self.num_heads, -1).permute(2, 0, 3, 1, 4)
# qkv with shape (3, B, nHead, H * W, C)
q, k, v = qkv.reshape(3, B * self.num_heads, N, -1).unbind(0)
# q, k, v with shape (B * nHead, H * W, C)
q, k = self.q_norm(q), self.k_norm(k)
if self.use_rel_pos:
attn_bias = get_decomposed_rel_pos_bias(q, self.rel_pos_h, self.rel_pos_w, (H, W), (H, W))
else:
attn_bias = None
if self.rope is not None:
rope = self.rope.get_embed()
q = apply_rot_embed_cat(q, rope).type_as(v)
k = apply_rot_embed_cat(k, rope).type_as(v)
if self.fused_attn:
x = torch.nn.functional.scaled_dot_product_attention(
q, k, v,
attn_mask=attn_bias,
dropout_p=self.attn_drop.p if self.training else 0.,
)
else:
q = q * self.scale
attn = q @ k.transpose(-2, -1)
if attn_bias is not None:
attn = attn + attn_bias
attn = attn.softmax(dim=-1)
attn = self.attn_drop(attn)
x = attn @ v
x = x.view(B, self.num_heads, N, -1).transpose(1, 2).reshape(B, N, -1)
x = self.proj(x)
x = x.view(B, H, W, -1)
return x
class LayerScale(nn.Module):
def __init__(self, dim, init_values=1e-5, inplace=False):
super().__init__()
self.inplace = inplace
self.gamma = nn.Parameter(init_values * torch.ones(dim))
def forward(self, x):
return x.mul_(self.gamma) if self.inplace else x * self.gamma
class Block(nn.Module):
def __init__(
self,
dim,
num_heads,
mlp_ratio=4.,
qkv_bias=True,
qk_norm=False,
proj_drop=0.,
attn_drop=0.,
init_values=None,
drop_path=0.,
act_layer=nn.GELU,
norm_layer=nn.LayerNorm,
mlp_layer=Mlp,
use_rel_pos=False,
window_size=0,
input_size=None,
rope=None,
):
super().__init__()
self.window_size = window_size
self.norm1 = norm_layer(dim)
self.attn = Attention(
dim,
num_heads=num_heads,
qkv_bias=qkv_bias,
qk_norm=qk_norm,
attn_drop=attn_drop,
proj_drop=proj_drop,
norm_layer=norm_layer,
use_rel_pos=use_rel_pos,
input_size=input_size if window_size == 0 else (window_size, window_size),
rope=rope,
)
self.ls1 = LayerScale(dim, init_values=init_values) if init_values else nn.Identity()
self.drop_path1 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
self.norm2 = norm_layer(dim)
self.mlp = mlp_layer(
in_features=dim,
hidden_features=int(dim * mlp_ratio),
act_layer=act_layer,
drop=proj_drop,
)
self.ls2 = LayerScale(dim, init_values=init_values) if init_values else nn.Identity()
self.drop_path2 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
def forward(self, x):
B, H, W, _ = x.shape
shortcut = x
x = self.norm1(x)
# Window partition
pad_hw: Optional[Tuple[int, int]] = None
if self.window_size > 0:
x, pad_hw = window_partition(x, self.window_size)
x = self.drop_path1(self.ls1(self.attn(x)))
# Reverse window partition
if self.window_size > 0:
x = window_unpartition(x, self.window_size, (H, W), pad_hw)
x = shortcut + x
x = x.reshape(B, H * W, -1) # MLP is faster for N, L, C tensor
x = x + self.drop_path2(self.ls2(self.mlp(self.norm2(x))))
x = x.reshape(B, H, W, -1)
return x
def window_partition(x: torch.Tensor, window_size: int) -> Tuple[torch.Tensor, Tuple[int, int]]:
"""
Partition into non-overlapping windows with padding if needed.
Args:
x (tensor): input tokens with [B, H, W, C].
window_size (int): window size.
Returns:
windows: windows after partition with [B * num_windows, window_size, window_size, C].
(Hp, Wp): padded height and width before partition
"""
B, H, W, C = x.shape
pad_h = (window_size - H % window_size) % window_size
pad_w = (window_size - W % window_size) % window_size
x = F.pad(x, (0, 0, 0, pad_w, 0, pad_h))
Hp, Wp = H + pad_h, W + pad_w
x = x.view(B, Hp // window_size, window_size, Wp // window_size, window_size, C)
windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size, window_size, C)
return windows, (Hp, Wp)
def window_unpartition(
windows: torch.Tensor, window_size: int, hw: Tuple[int, int], pad_hw: Optional[Tuple[int, int]] = None,
) -> torch.Tensor:
"""
Window unpartition into original sequences and removing padding.
Args:
windows (tensor): input tokens with [B * num_windows, window_size, window_size, C].
window_size (int): window size.
pad_hw (Tuple): padded height and width (Hp, Wp).
hw (Tuple): original height and width (H, W) before padding.
Returns:
x: unpartitioned sequences with [B, H, W, C].
"""
Hp, Wp = pad_hw if pad_hw is not None else hw
H, W = hw
B = windows.shape[0] // (Hp * Wp // window_size // window_size)
x = windows.view(B, Hp // window_size, Wp // window_size, window_size, window_size, -1)
x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(B, Hp, Wp, -1)
x = x[:, :H, :W, :].contiguous()
return x
class VisionTransformerSAM(nn.Module):
""" Vision Transformer for Segment-Anything Model(SAM)
A PyTorch impl of : `Exploring Plain Vision Transformer Backbones for Object Detection` or `Segment Anything Model (SAM)`
- https://arxiv.org/abs/2010.11929
"""
def __init__(
self,
img_size: int = 1024,
patch_size: int = 16,
in_chans: int = 3,
num_classes: int = 768,
embed_dim: int = 768,
depth: int = 12,
num_heads: int = 12,
mlp_ratio: float = 4.,
qkv_bias: bool = True,
qk_norm: bool = False,
init_values: Optional[float] = None,
pre_norm: bool = False,
drop_rate: float = 0.,
pos_drop_rate: float = 0.,
patch_drop_rate: float = 0.,
proj_drop_rate: float = 0.,
attn_drop_rate: float = 0.,
drop_path_rate: float = 0.,
weight_init: str = '',
embed_layer: Callable = partial(
PatchEmbed, output_fmt=Format.NHWC, strict_img_size=False),
norm_layer: Optional[Callable] = nn.LayerNorm,
act_layer: Optional[Callable] = nn.GELU,
block_fn: Callable = Block,
mlp_layer: Callable = Mlp,
use_abs_pos: bool = True,
use_rel_pos: bool = False,
use_rope: bool = False,
window_size: int = 14,
global_attn_indexes: Tuple[int, ...] = (),
neck_chans: int = 256,
global_pool: str = 'avg',
head_hidden_size: Optional[int] = None,
ref_feat_shape: Optional[Tuple[Tuple[int, int], Tuple[int, int]]] = None
):
"""
Args:
img_size: Input image size.
patch_size: Patch size.
in_chans: Number of image input channels.
num_classes: Mumber of classes for classification head.
global_pool: Type of global pooling for final sequence (default: 'token').
embed_dim: Transformer embedding dimension.
depth: Depth of transformer.
num_heads: Number of attention heads.
mlp_ratio: Ratio of mlp hidden dim to embedding dim.
qkv_bias: Enable bias for qkv projections if True.
init_values: Layer-scale init values (layer-scale enabled if not None).
drop_rate: Head dropout rate.
pos_drop_rate: Position embedding dropout rate.
attn_drop_rate: Attention dropout rate.
drop_path_rate: Stochastic depth rate.
weight_init: Weight initialization scheme.
embed_layer: Patch embedding layer.
norm_layer: Normalization layer.
act_layer: MLP activation layer.
block_fn: Transformer block layer.
use_abs_pos: If True, use absolute positional embeddings.
use_rel_pos: If True, add relative positional embeddings to the attention map.
use_rope: If True, add rotary position embeddings to q/k in attention block.
window_size: Window size for window attention blocks. If 0, not use window attention.
global_attn_indexes: Indexes for blocks using global attention. Used when window_size > 0.
global_pool: Global pooling type.
head_hidden_size: If set, use NormMlpHead
ref_feat_shape: Tuple of reference feature shapes for ROPE, (global, local)
"""
super().__init__()
norm_layer = norm_layer or partial(nn.LayerNorm, eps=1e-6)
act_layer = act_layer or nn.GELU
self.num_classes = num_classes
self.global_pool = global_pool
# num_features for consistency with other models
self.num_features = self.embed_dim = embed_dim
self.grad_checkpointing = False
self.patch_embed = embed_layer(
img_size=img_size,
patch_size=patch_size,
in_chans=in_chans,
embed_dim=embed_dim,
bias=not pre_norm, # disable bias if pre-norm is used
)
grid_size = self.patch_embed.grid_size
if use_abs_pos:
# Initialize absolute positional embedding with pretrain image size.
self.pos_embed = nn.Parameter(torch.zeros(1, grid_size[0], grid_size[1], embed_dim))
else:
self.pos_embed = None
self.pos_drop = nn.Dropout(p=pos_drop_rate)
if patch_drop_rate > 0:
self.patch_drop = PatchDropout(
patch_drop_rate,
num_prefix_tokens=0,
)
else:
self.patch_drop = nn.Identity()
self.norm_pre = norm_layer(embed_dim) if pre_norm else nn.Identity()
if use_rope:
assert not use_rel_pos, "ROPE and relative pos embeddings should not be enabled at same time"
if ref_feat_shape is not None:
assert len(ref_feat_shape) == 2
ref_feat_shape_global = to_2tuple(ref_feat_shape[0])
ref_feat_shape_window = to_2tuple(ref_feat_shape[1])
else:
ref_feat_shape_global = ref_feat_shape_window = None
self.rope_global = RotaryEmbeddingCat(
embed_dim // num_heads,
in_pixels=False,
feat_shape=grid_size,
ref_feat_shape=ref_feat_shape_global,
)
self.rope_window = RotaryEmbeddingCat(
embed_dim // num_heads,
in_pixels=False,
feat_shape=to_2tuple(window_size),
ref_feat_shape=ref_feat_shape_window,
)
else:
self.rope_global = None
self.rope_window = None
# stochastic depth decay rule
dpr = [x.item() for x in torch.linspace(0, drop_path_rate, depth)]
self.blocks = nn.Sequential(*[
block_fn(
dim=embed_dim,
num_heads=num_heads,
mlp_ratio=mlp_ratio,
qkv_bias=qkv_bias,
qk_norm=qk_norm,
init_values=init_values,
proj_drop=proj_drop_rate,
attn_drop=attn_drop_rate,
drop_path=dpr[i],
norm_layer=norm_layer,
act_layer=act_layer,
mlp_layer=mlp_layer,
use_rel_pos=use_rel_pos,
window_size=window_size if i not in global_attn_indexes else 0,
input_size=grid_size,
rope=self.rope_window if i not in global_attn_indexes else self.rope_global,
)
for i in range(depth)])
if neck_chans:
self.neck = nn.Sequential(
nn.Conv2d(
embed_dim,
neck_chans,
kernel_size=1,
bias=False,
),
LayerNorm2d(neck_chans),
nn.Conv2d(
neck_chans,
neck_chans,
kernel_size=3,
padding=1,
bias=False,
),
LayerNorm2d(neck_chans),
)
self.num_features = neck_chans
else:
if head_hidden_size:
self.neck = nn.Identity()
else:
# should have a final norm with standard ClassifierHead
self.neck = LayerNorm2d(embed_dim)
neck_chans = embed_dim
# Classifier Head
if head_hidden_size:
self.head = NormMlpClassifierHead(
neck_chans,
num_classes,
hidden_size=head_hidden_size,
pool_type=global_pool,
drop_rate=drop_rate,
)
else:
self.head = ClassifierHead(
neck_chans,
num_classes,
pool_type=global_pool,
drop_rate=drop_rate,
)
@torch.jit.ignore
def no_weight_decay(self):
return {'pos_embed', 'dist_token'}
@torch.jit.ignore
def group_matcher(self, coarse=False):
return dict(
stem=r'^pos_embed|patch_embed', # stem and embed
blocks=[(r'^blocks\.(\d+)', None), (r'^norm', (99999,))]
)
@torch.jit.ignore
def set_grad_checkpointing(self, enable=True):
self.grad_checkpointing = enable
@torch.jit.ignore
def get_classifier(self):
return self.head
def reset_classifier(self, num_classes=0, global_pool=None):
self.head.reset(num_classes, global_pool)
def forward_features(self, x):
x = self.patch_embed(x)
if self.pos_embed is not None:
# dynamically resize abs pos embedding if needed
x = x + resample_abs_pos_embed_nhwc(self.pos_embed, x.shape[1:3])
x = self.pos_drop(x)
x = self.patch_drop(x)
x = self.norm_pre(x)
if self.grad_checkpointing and not torch.jit.is_scripting():
x = checkpoint_seq(self.blocks, x)
else:
x = self.blocks(x)
x = self.neck(x.permute(0, 3, 1, 2))
return x
def forward_head(self, x, pre_logits: bool = False):
return self.head(x, pre_logits=True) if pre_logits else self.head(x)
def forward(self, x):
x = self.forward_features(x)
x = self.forward_head(x)
return x
def checkpoint_filter_fn(
state_dict,
model,
):
""" Remap SAM checkpoints -> timm """
sam_checkpoint = 'image_encoder.patch_embed.proj.weight' in state_dict
out_dict = {}
for k, v in state_dict.items():
if k.startswith('image_encoder.'):
k = k[14:]
k = k.replace('mlp.lin', 'mlp.fc')
else:
if sam_checkpoint:
continue
out_dict[k] = v
return out_dict
def _cfg(url='', **kwargs):
return {
'url': url,
'num_classes': 1000, 'input_size': (3, 1024, 1024), 'pool_size': None,
'crop_pct': .9, 'interpolation': 'bicubic', 'fixed_input_size': True,
'mean': IMAGENET_INCEPTION_MEAN, 'std': IMAGENET_INCEPTION_STD,
'first_conv': 'patch_embed.proj', 'classifier': 'head.fc',
**kwargs
}
default_cfgs = generate_default_cfgs({
# Segment-Anyhing Model (SAM) pretrained - https://github.com/facebookresearch/segment-anything (no classifier head, for fine-tune/features only)
'samvit_base_patch16.sa1b': _cfg(
url='https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth',
hf_hub_id='timm/',
license='apache-2.0',
mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=0,
input_size=(3, 1024, 1024), crop_pct=1.0),
'samvit_large_patch16.sa1b': _cfg(
url='https://dl.fbaipublicfiles.com/segment_anything/sam_vit_l_0b3195.pth',
hf_hub_id='timm/',
license='apache-2.0',
mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=0,
input_size=(3, 1024, 1024), crop_pct=1.0),
'samvit_huge_patch16.sa1b': _cfg(
url='https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth',
hf_hub_id='timm/',
license='apache-2.0',
mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=0,
input_size=(3, 1024, 1024), crop_pct=1.0),
'samvit_base_patch16_224': _cfg(
mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=1000,
input_size=(3, 224, 224), crop_pct=0.9),
})
def _create_vision_transformer(variant, pretrained=False, **kwargs):
if kwargs.get('features_only', None):
raise RuntimeError(
'features_only not implemented for Vision Transformer models.')
return build_model_with_cfg(
VisionTransformerSAM,
variant,
pretrained,
pretrained_filter_fn=checkpoint_filter_fn,
**kwargs,
)
@register_model
def samvit_base_patch16(pretrained=False, **kwargs) -> VisionTransformerSAM:
""" ViT-B/16 for Segment-Anything
"""
model_args = dict(
patch_size=16, embed_dim=768, depth=12, num_heads=12, global_attn_indexes=[2, 5, 8, 11],
window_size=14, use_rel_pos=True, img_size=1024,
)
model = _create_vision_transformer(
'samvit_base_patch16', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def samvit_large_patch16(pretrained=False, **kwargs) -> VisionTransformerSAM:
""" ViT-L/16 for Segment-Anything
"""
model_args = dict(
patch_size=16, embed_dim=1024, depth=24, num_heads=16, global_attn_indexes=[5, 11, 17, 23],
window_size=14, use_rel_pos=True, img_size=1024,
)
model = _create_vision_transformer(
'samvit_large_patch16', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def samvit_huge_patch16(pretrained=False, **kwargs) -> VisionTransformerSAM:
""" ViT-H/16 for Segment-Anything
"""
model_args = dict(
patch_size=16, embed_dim=1280, depth=32, num_heads=16, global_attn_indexes=[7, 15, 23, 31],
window_size=14, use_rel_pos=True, img_size=1024,
)
model = _create_vision_transformer(
'samvit_huge_patch16', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def samvit_base_patch16_224(pretrained=False, **kwargs) -> VisionTransformerSAM:
""" ViT-B/16 based on samvit arch
"""
model_args = dict(
patch_size=16, embed_dim=768, depth=12, num_heads=12, global_attn_indexes=[2, 5, 8, 11],
window_size=14, use_rel_pos=True, use_abs_pos=False, img_size=224, neck_chans=None,
)
model = _create_vision_transformer(
'samvit_base_patch16_224', pretrained=pretrained, **dict(model_args, **kwargs))
return model