flexibly handle hybrid module outputs

setup.py

@@ -3,7 +3,7 @@
 setup(
   name = 'x-transformers',
   packages = find_packages(exclude=['examples']),
-  version = '1.44.0',
+  version = '1.44.2',
   license='MIT',
   description = 'X-Transformers - Pytorch',
   author = 'Phil Wang',

tests/test_x_transformers.py

@@ -617,7 +617,7 @@ def test_hyper_connections(tanh):
 def test_hybrid():
     from torch.nn import GRU
 
-    model = TransformerWrapper(
+    dec = TransformerWrapper(
         num_tokens = 20000,
         max_seq_len = 1024,
         attn_layers = Decoder(
@@ -631,4 +631,19 @@ def test_hybrid():
 
     x = torch.randint(0, 20000, (2, 1024))
 
-    embed = model(x)
+    embed = dec(x)
+
+    enc = TransformerWrapper(
+        num_tokens = 20000,
+        max_seq_len = 1024,
+        attn_layers = Encoder(
+            dim = 128,
+            depth = 6,
+            heads = 8,
+            attn_dim_head = 64,
+            attn_hybrid_module = GRU(128, 64 * 4, batch_first = True, bidirectional = True)
+        )
+    )
+
+    mask = torch.randint(0, 2, (2, 1024)).bool()
+    embed = enc(x, mask = mask)

x_transformers/x_transformers.py

@@ -7,10 +7,11 @@
 from packaging import version
 
 import torch
+from torch.amp import autocast
 import torch.nn.functional as F
 from torch import nn, einsum, Tensor
+from torch.utils._pytree import tree_flatten
 from torch.nn import Module, ModuleList, ModuleDict
-from torch.amp import autocast
 
 from functools import partial, wraps
 from collections import namedtuple
@@ -1138,6 +1139,7 @@ def __init__(
         selective = False,
         custom_attn_fn: Callable | None = None,
         hybrid_module: Module | None = None,
+        hybrid_mask_kwarg: str | None = None,
         one_kv_head = False,
         kv_heads = None,
         shared_kv = False,
@@ -1341,6 +1343,8 @@ def __init__(
 
         self.hybrid_module = deepcopy(hybrid_module) if exists(hybrid_module) else None
 
+        self.hybrid_mask_kwarg = hybrid_mask_kwarg # for bidirectional, can forward `mask` into the hybrid module and let it handle variable lengths
+
         # output dimension by default same as input, but can be overridden
 
         dim_out = default(dim_out, dim)
@@ -1592,7 +1596,21 @@ def forward(
         # hybrid module
 
         if exists(self.hybrid_module):
-            hybrid_out, _ = self.hybrid_module(x)
+
+            # hybrid input
+
+            hybrid_forward_kwargs = dict()
+
+            if not self.causal and exists(self.hybrid_mask_kwarg):
+                hybrid_forward_kwargs = {self.hybrid_mask_kwarg: mask}
+
+            # hybrid forward
+
+            hybrid_outputs = self.hybrid_module(x, **hybrid_forward_kwargs)
+
+            # handle hybrid out
+
+            (hybrid_out, *rest_hybrid_outs), _ = tree_flatten(hybrid_outputs)
             out = 0.5 * (out + hybrid_out)
 
         # alphafold2 styled gating of the values