x-transformers 1.30.11__py3-none-any.whl → 1.30.14__py3-none-any.whl

x_transformers/x_transformers.py

@@ -80,6 +80,13 @@ class equals():
     def __call__(self, x, *args, **kwargs):
         return x == self.val
 
+class Identity(Module):
+    def __init__(self, *args, **kwargs):
+        super().__init__()
+
+    def __call__(self, x, *args, **kwargs):
+        return x
+
 def Sequential(*modules):
     return nn.Sequential(*filter(exists, modules))
 
@@ -1129,6 +1136,7 @@ class AttentionLayers(Module):
         use_scalenorm = False,
         use_rmsnorm = False,
         use_simple_rmsnorm = False,
+        no_pre_or_postnorm = False,
         alibi_pos_bias = False,
         alibi_num_heads = None,
         rel_pos_bias = False,
@@ -1358,9 +1366,18 @@ class AttentionLayers(Module):
             residual_fn = GRUGating if gate_residual else Residual
             residual = residual_fn(dim, scale_residual = scale_residual, scale_residual_constant = scale_residual_constant)
 
-            pre_branch_norm = norm_fn() if pre_norm else None
-            post_branch_norm = norm_fn() if sandwich_norm else None
-            post_main_norm = norm_fn() if not pre_norm else None
+            # for the peri-layernorm config from https://arxiv.org/abs/2405.16039
+            # must be paired with qk norm
+
+            layer_norm_fn = norm_fn
+            if no_pre_or_postnorm:
+                layer_norm_fn = Identity
+
+            # all normalizations of the layer
+
+            pre_branch_norm = layer_norm_fn() if pre_norm else None
+            post_branch_norm = layer_norm_fn() if sandwich_norm else None
+            post_main_norm = layer_norm_fn() if not pre_norm else None
 
             norms = ModuleList([
                 pre_branch_norm,

{x_transformers-1.30.11.dist-info → x_transformers-1.30.14.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: x-transformers
-Version: 1.30.11
+Version: 1.30.14
 Summary: X-Transformers - Pytorch
 Home-page: https://github.com/lucidrains/x-transformers
 Author: Phil Wang

{x_transformers-1.30.11.dist-info → x_transformers-1.30.14.dist-info}/RECORD

@@ -4,11 +4,11 @@ x_transformers/autoregressive_wrapper.py,sha256=uX8Mb0zLsQrZECt_9UGt35g7tC05Rk3n
 x_transformers/continuous.py,sha256=WO52n9lFAXv5-SGadi2cApGF8dkouN8QSTEOuC7erj8,6180
 x_transformers/dpo.py,sha256=LjvWgCkqTl-UuehrzQ8nkX5guLr4whYwsmm7SKSwdls,3450
 x_transformers/nonautoregressive_wrapper.py,sha256=ys_p8obc7lTeeodCqvkRKxOXQ1C9T3j5Jwr-JbVgnXk,10432
-x_transformers/x_transformers.py,sha256=97WREYHfGljhrveGNUUSEWk2xbFvKdM52QXW7cnoBpk,69019
+x_transformers/x_transformers.py,sha256=yCLWe5g0iFDBE0lSnoB9Up0bYHzfIFL0f7zhNTbuoS0,69513
 x_transformers/xl_autoregressive_wrapper.py,sha256=DCx4n0_c1tFai4nOqaWVnqx2p9eutsZsDMiMP1ckxNU,4117
 x_transformers/xval.py,sha256=QE1ltYZTR_eGgIHPP2BrMWVWVLqMW-OpDZh87BSmQEg,8563
-x_transformers-1.30.11.dist-info/LICENSE,sha256=As9u198X-U-vph5noInuUfqsAG2zX_oXPHDmdjwlPPY,1066
-x_transformers-1.30.11.dist-info/METADATA,sha256=kpv8_mEb4DU4zL_oyLwzbWJ7z8WV7mf7c466Hehn-6c,662
-x_transformers-1.30.11.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-x_transformers-1.30.11.dist-info/top_level.txt,sha256=hO6KGpFuGucRNEtRfme4A_rGcM53AKwGP7RVlRIxS5Q,15
-x_transformers-1.30.11.dist-info/RECORD,,
+x_transformers-1.30.14.dist-info/LICENSE,sha256=As9u198X-U-vph5noInuUfqsAG2zX_oXPHDmdjwlPPY,1066
+x_transformers-1.30.14.dist-info/METADATA,sha256=0Y4a5bhVub1knobIGu7EnygkAGeiXV7EviJDkDaktBg,662
+x_transformers-1.30.14.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
+x_transformers-1.30.14.dist-info/top_level.txt,sha256=hO6KGpFuGucRNEtRfme4A_rGcM53AKwGP7RVlRIxS5Q,15
+x_transformers-1.30.14.dist-info/RECORD,,