add vivit

Author: lucidrains

README.md

@@ -31,6 +31,7 @@
 - [Patch Merger](#patch-merger)
 - [Vision Transformer for Small Datasets](#vision-transformer-for-small-datasets)
 - [3D Vit](#3d-vit)
+- [ViVit](#vivit)
 - [Parallel ViT](#parallel-vit)
 - [Learnable Memory ViT](#learnable-memory-vit)
 - [Dino](#dino)
@@ -1022,6 +1023,34 @@ video = torch.randn(4, 3, 16, 128, 128) # (batch, channels, frames, height, widt
 preds = v(video) # (4, 1000)
 ```
 
+## ViViT
+
+<img src="./images/vivit.png" width="350px"></img>
+
+This <a href="https://arxiv.org/abs/2103.15691">paper</a> offers 3 different types of architectures for efficient attention of videos, with the main theme being factorizing the attention across space and time. This repository will offer the first variant, which is a spatial transformer followed by a temporal one.
+
+```python
+import torch
+from vit_pytorch.vivit import ViT
+
+v = ViT(
+    image_size = 128,          # image size
+    frames = 16,               # number of frames
+    image_patch_size = 16,     # image patch size
+    frame_patch_size = 2,      # frame patch size
+    num_classes = 1000,
+    dim = 1024,
+    spatial_depth = 6,         # depth of the spatial transformer
+    temporal_depth = 6,        # depth of the temporal transformer
+    heads = 8,
+    mlp_dim = 2048
+)
+
+video = torch.randn(4, 3, 16, 128, 128) # (batch, channels, frames, height, width)
+
+preds = v(video) # (4, 1000)
+```
+
 ## Parallel ViT
 
 <img src="./images/parallel-vit.png" width="350px"></img>
@@ -1805,6 +1834,16 @@ Coming from computer vision and new to transformers? Here are some resources tha
 
 ```
 
+```bibtex
+@article{Arnab2021ViViTAV,
+    title   = {ViViT: A Video Vision Transformer},
+    author  = {Anurag Arnab and Mostafa Dehghani and Georg Heigold and Chen Sun and Mario Lucic and Cordelia Schmid},
+    journal = {2021 IEEE/CVF International Conference on Computer Vision (ICCV)},
+    year    = {2021},
+    pages   = {6816-6826}
+}
+```
+
 ```bibtex
 @misc{vaswani2017attention,
     title   = {Attention Is All You Need},

images/vivit.png

@@ -3,7 +3,7 @@
 setup(
   name = 'vit-pytorch',
   packages = find_packages(exclude=['examples']),
-  version = '0.36.2',
+  version = '0.37.0',
   license='MIT',
   description = 'Vision Transformer (ViT) - Pytorch',
   long_description_content_type = 'text/markdown',

setup.py

@@ -0,0 +1,169 @@
+import torch
+from torch import nn
+
+from einops import rearrange, repeat
+from einops.layers.torch import Rearrange
+
+# helpers
+
+def pair(t):
+    return t if isinstance(t, tuple) else (t, t)
+
+# classes
+
+class PreNorm(nn.Module):
+    def __init__(self, dim, fn):
+        super().__init__()
+        self.norm = nn.LayerNorm(dim)
+        self.fn = fn
+    def forward(self, x, **kwargs):
+        return self.fn(self.norm(x), **kwargs)
+
+class FeedForward(nn.Module):
+    def __init__(self, dim, hidden_dim, dropout = 0.):
+        super().__init__()
+        self.net = nn.Sequential(
+            nn.Linear(dim, hidden_dim),
+            nn.GELU(),
+            nn.Dropout(dropout),
+            nn.Linear(hidden_dim, dim),
+            nn.Dropout(dropout)
+        )
+    def forward(self, x):
+        return self.net(x)
+
+class Attention(nn.Module):
+    def __init__(self, dim, heads = 8, dim_head = 64, dropout = 0.):
+        super().__init__()
+        inner_dim = dim_head *  heads
+        project_out = not (heads == 1 and dim_head == dim)
+
+        self.heads = heads
+        self.scale = dim_head ** -0.5
+
+        self.attend = nn.Softmax(dim = -1)
+        self.dropout = nn.Dropout(dropout)
+
+        self.to_qkv = nn.Linear(dim, inner_dim * 3, bias = False)
+
+        self.to_out = nn.Sequential(
+            nn.Linear(inner_dim, dim),
+            nn.Dropout(dropout)
+        ) if project_out else nn.Identity()
+
+    def forward(self, x):
+        qkv = self.to_qkv(x).chunk(3, dim = -1)
+        q, k, v = map(lambda t: rearrange(t, 'b n (h d) -> b h n d', h = self.heads), qkv)
+
+        dots = torch.matmul(q, k.transpose(-1, -2)) * self.scale
+
+        attn = self.attend(dots)
+        attn = self.dropout(attn)
+
+        out = torch.matmul(attn, v)
+        out = rearrange(out, 'b h n d -> b n (h d)')
+        return self.to_out(out)
+
+class Transformer(nn.Module):
+    def __init__(self, dim, depth, heads, dim_head, mlp_dim, dropout = 0.):
+        super().__init__()
+        self.layers = nn.ModuleList([])
+        for _ in range(depth):
+            self.layers.append(nn.ModuleList([
+                PreNorm(dim, Attention(dim, heads = heads, dim_head = dim_head, dropout = dropout)),
+                PreNorm(dim, FeedForward(dim, mlp_dim, dropout = dropout))
+            ]))
+    def forward(self, x):
+        for attn, ff in self.layers:
+            x = attn(x) + x
+            x = ff(x) + x
+        return x
+
+class ViT(nn.Module):
+    def __init__(
+        self,
+        *,
+        image_size,
+        image_patch_size,
+        frames,
+        frame_patch_size,
+        num_classes,
+        dim,
+        spatial_depth,
+        temporal_depth,
+        heads,
+        mlp_dim,
+        pool = 'cls',
+        channels = 3,
+        dim_head = 64,
+        dropout = 0.,
+        emb_dropout = 0.
+    ):
+        super().__init__()
+        image_height, image_width = pair(image_size)
+        patch_height, patch_width = pair(image_patch_size)
+
+        assert image_height % patch_height == 0 and image_width % patch_width == 0, 'Image dimensions must be divisible by the patch size.'
+        assert frames % frame_patch_size == 0, 'Frames must be divisible by frame patch size'
+
+        num_patches = (image_height // patch_height) * (image_width // patch_width) * (frames // frame_patch_size)
+        patch_dim = channels * patch_height * patch_width * frame_patch_size
+
+        assert pool in {'cls', 'mean'}, 'pool type must be either cls (cls token) or mean (mean pooling)'
+
+        self.to_patch_embedding = nn.Sequential(
+            Rearrange('b c (f pf) (h p1) (w p2) -> b f (h w) (p1 p2 pf c)', p1 = patch_height, p2 = patch_width, pf = frame_patch_size),
+            nn.Linear(patch_dim, dim),
+        )
+
+        self.pos_embedding = nn.Parameter(torch.randn(1, num_patches + 1, dim))
+        self.dropout = nn.Dropout(emb_dropout)
+        self.spatial_cls_token = nn.Parameter(torch.randn(1, 1, dim))
+        self.temporal_cls_token = nn.Parameter(torch.randn(1, 1, dim))
+
+        self.spatial_transformer = Transformer(dim, spatial_depth, heads, dim_head, mlp_dim, dropout)
+        self.temporal_transformer = Transformer(dim, temporal_depth, heads, dim_head, mlp_dim, dropout)
+
+        self.pool = pool
+        self.to_latent = nn.Identity()
+
+        self.mlp_head = nn.Sequential(
+            nn.LayerNorm(dim),
+            nn.Linear(dim, num_classes)
+        )
+
+    def forward(self, img):
+        x = self.to_patch_embedding(img)
+        b, f, n, _ = x.shape
+
+        spatial_cls_tokens = repeat(self.spatial_cls_token, '1 1 d -> b f 1 d', b = b, f = f)
+        x = torch.cat((spatial_cls_tokens, x), dim = 2)
+        x += self.pos_embedding[:, :(n + 1)]
+        x = self.dropout(x)
+
+        x = rearrange(x, 'b f n d -> (b f) n d')
+
+        # attend across space
+
+        x = self.spatial_transformer(x)
+
+        x = rearrange(x, '(b f) n d -> b f n d', b = b)
+
+        # excise out the spatial cls tokens for temporal attention
+
+        x = x[:, :, 0]
+
+        # append temporal CLS tokens
+
+        temporal_cls_tokens = repeat(self.temporal_cls_token, '1 1 d-> b 1 d', b = b)
+
+        x = torch.cat((temporal_cls_tokens, x), dim = 1)
+
+        # attend across time
+
+        x = self.temporal_transformer(x)
+
+        x = x.mean(dim = 1) if self.pool == 'mean' else x[:, 0]
+
+        x = self.to_latent(x)
+        return self.mlp_head(x)