Add relative pos embed option to LambdaLayer, fix last transpose/reshape.

timm/models/byoanet.py

@@ -62,7 +62,7 @@ default_cfgs = {
     'lambda_resnet50ts': _cfg(
         url='',
         min_input_size=(3, 128, 128), input_size=(3, 256, 256), pool_size=(8, 8)),
-    'lambda_resnet26rt_256': _cfg(
+    'lambda_resnet26rpt_256': _cfg(
         url='',
         fixed_input_size=True, input_size=(3, 256, 256), pool_size=(8, 8)),
 }
@@ -218,7 +218,7 @@ model_cfgs = dict(
         self_attn_layer='lambda',
         self_attn_kwargs=dict(r=9)
     ),
-    lambda_resnet26rt_256=ByoModelCfg(
+    lambda_resnet26rpt_256=ByoModelCfg(
         blocks=(
             ByoBlockCfg(type='bottle', d=2, c=256, s=1, gs=0, br=0.25),
             ByoBlockCfg(type='bottle', d=2, c=512, s=2, gs=0, br=0.25),
@@ -321,8 +321,8 @@ def lambda_resnet50ts(pretrained=False, **kwargs):
 
 
 @register_model
-def lambda_resnet26rt_256(pretrained=False, **kwargs):
+def lambda_resnet26rpt_256(pretrained=False, **kwargs):
     """ Lambda-ResNet-26-R-T. Lambda layers w/ rel pos embed in last two stages.
     """
     kwargs.setdefault('img_size', 256)
-    return _create_byoanet('lambda_resnet26rt_256', pretrained=pretrained, **kwargs)
+    return _create_byoanet('lambda_resnet26rpt_256', pretrained=pretrained, **kwargs)

timm/models/layers/lambda_layer.py

@@ -24,18 +24,30 @@ import torch
 from torch import nn
 import torch.nn.functional as F
 
+from .helpers import to_2tuple
 from .weight_init import trunc_normal_
 
 
+def rel_pos_indices(size):
+    size = to_2tuple(size)
+    pos = torch.stack(torch.meshgrid(torch.arange(size[0]), torch.arange(size[1]))).flatten(1)
+    rel_pos = pos[:, None, :] - pos[:, :, None]
+    rel_pos[0] += size[0] - 1
+    rel_pos[1] += size[1] - 1
+    return rel_pos  # 2, H * W, H * W
+
+
 class LambdaLayer(nn.Module):
-    """Lambda Layer w/ lambda conv position embedding
+    """Lambda Layer
 
     Paper: `LambdaNetworks: Modeling Long-Range Interactions Without Attention`
         - https://arxiv.org/abs/2102.08602
+
+    NOTE: intra-depth parameter 'u' is fixed at 1. It did not appear worth the complexity to add.
     """
     def __init__(
             self,
-            dim, dim_out=None, stride=1, num_heads=4, dim_head=16, r=7, qkv_bias=False):
+            dim, dim_out=None, feat_size=None, stride=1, num_heads=4, dim_head=16, r=7, qkv_bias=False):
         super().__init__()
         self.dim = dim
         self.dim_out = dim_out or dim
@@ -43,7 +55,6 @@ class LambdaLayer(nn.Module):
         self.num_heads = num_heads
         assert self.dim_out % num_heads == 0, ' should be divided by num_heads'
         self.dim_v = self.dim_out // num_heads  # value depth 'v'
-        self.r = r  # relative position neighbourhood (lambda conv kernel size)
 
         self.qkv = nn.Conv2d(
             dim,
@@ -52,8 +63,19 @@ class LambdaLayer(nn.Module):
         self.norm_q = nn.BatchNorm2d(num_heads * dim_head)
         self.norm_v = nn.BatchNorm2d(self.dim_v)
 
-        # NOTE currently only supporting the local lambda convolutions for positional
-        self.conv_lambda = nn.Conv3d(1, dim_head, (r, r, 1), padding=(r // 2, r // 2, 0))
+        if r is not None:
+            # local lambda convolution for pos
+            self.conv_lambda = nn.Conv3d(1, dim_head, (r, r, 1), padding=(r // 2, r // 2, 0))
+            self.pos_emb = None
+            self.rel_pos = None
+        else:
+            # relative pos embedding
+            assert feat_size is not None
+            feat_size = to_2tuple(feat_size)
+            rel_size = [2 * s - 1 for s in feat_size]
+            self.conv_lambda = None
+            self.pos_emb = nn.Parameter(torch.zeros(rel_size[0], rel_size[1], self.dim_k))
+            self.register_buffer('rel_pos_indices', rel_pos_indices(feat_size), persistent=False)
 
         self.pool = nn.AvgPool2d(2, 2) if stride == 2 else nn.Identity()
 
@@ -61,12 +83,14 @@ class LambdaLayer(nn.Module):
 
     def reset_parameters(self):
         trunc_normal_(self.qkv.weight, std=self.dim ** -0.5)
-        trunc_normal_(self.conv_lambda.weight, std=self.dim_k ** -0.5)
+        if self.conv_lambda is not None:
+            trunc_normal_(self.conv_lambda.weight, std=self.dim_k ** -0.5)
+        if self.pos_emb is not None:
+            trunc_normal_(self.pos_emb, std=.02)
 
     def forward(self, x):
         B, C, H, W = x.shape
         M = H * W
-
         qkv = self.qkv(x)
         q, k, v = torch.split(qkv, [
             self.num_heads * self.dim_k, self.dim_k, self.dim_v], dim=1)
@@ -77,10 +101,15 @@ class LambdaLayer(nn.Module):
         content_lam = k @ v  # B, K, V
         content_out = q @ content_lam.unsqueeze(1)  # B, num_heads, M, V
 
-        position_lam = self.conv_lambda(v.reshape(B, 1, H, W, self.dim_v))  # B, H, W, V, K
-        position_lam = position_lam.reshape(B, 1, self.dim_k, H * W, self.dim_v).transpose(2, 3)  # B, 1, M, K, V
+        if self.pos_emb is None:
+            position_lam = self.conv_lambda(v.reshape(B, 1, H, W, self.dim_v))  # B, H, W, V, K
+            position_lam = position_lam.reshape(B, 1, self.dim_k, H * W, self.dim_v).transpose(2, 3)  # B, 1, M, K, V
+        else:
+            # FIXME relative pos embedding path not fully verified
+            pos_emb = self.pos_emb[self.rel_pos_indices[0], self.rel_pos_indices[1]].expand(B, -1, -1, -1)
+            position_lam = (pos_emb.transpose(-1, -2) @ v.unsqueeze(1)).unsqueeze(1)  # B, 1, M, K, V
         position_out = (q.unsqueeze(-2) @ position_lam).squeeze(-2)  # B, num_heads, M, V
 
-        out = (content_out + position_out).transpose(3, 1).reshape(B, C, H, W)  # B, C (num_heads * V), H, W
+        out = (content_out + position_out).transpose(-1, -2).reshape(B, C, H, W)  # B, C (num_heads * V), H, W
         out = self.pool(out)
         return out