x-transformers 1.30.8__py3-none-any.whl → 1.30.9__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -312,11 +312,9 @@ class CoPE(Module):
312
312
  self,
313
313
  dim,
314
314
  max_pos,
315
- soft_onehot = True,
316
- reverse = True
315
+ soft_onehot = False,
317
316
  ):
318
317
  super () . __init__ ()
319
- self.reverse = reverse
320
318
  self.max_pos = max_pos
321
319
  self.pos_emb = nn.Parameter(torch.zeros(max_pos, dim))
322
320
 
@@ -330,11 +328,7 @@ class CoPE(Module):
330
328
 
331
329
  gates = attn_logits.sigmoid()
332
330
 
333
- if self.reverse:
334
- pos = gates.flip(-1).cumsum(dim = -1).flip(-1)
335
- else:
336
- pos = gates.cumsum(dim = -1)
337
-
331
+ pos = gates.flip(-1).cumsum(dim = -1).flip(-1)
338
332
  pos = pos.clamp(max = self.max_pos - 1)
339
333
 
340
334
  logits_int = einsum('b h n d, p d -> b h n p', query, self.pos_emb)
@@ -775,6 +769,7 @@ class Attention(Module):
775
769
  rotary_embed_values = False,
776
770
  use_cope = False,
777
771
  cope_max_pos = 16,
772
+ cope_soft_onehot_pos = False,
778
773
  logit_softclamp_value = None,
779
774
  onnxable = False
780
775
  ):
@@ -858,7 +853,7 @@ class Attention(Module):
858
853
  assert causal, 'CoPE was designed for causal attention'
859
854
  assert not flash, 'CoPE is not flash attention compatible'
860
855
 
861
- cope = CoPE(dim_head, cope_max_pos)
856
+ cope = CoPE(dim_head, cope_max_pos, soft_onehot = cope_soft_onehot_pos)
862
857
 
863
858
  # attend class - includes core attention algorithm + talking heads
864
859
 
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: x-transformers
3
- Version: 1.30.8
3
+ Version: 1.30.9
4
4
  Summary: X-Transformers - Pytorch
5
5
  Home-page: https://github.com/lucidrains/x-transformers
6
6
  Author: Phil Wang
@@ -4,11 +4,11 @@ x_transformers/autoregressive_wrapper.py,sha256=uX8Mb0zLsQrZECt_9UGt35g7tC05Rk3n
4
4
  x_transformers/continuous.py,sha256=WO52n9lFAXv5-SGadi2cApGF8dkouN8QSTEOuC7erj8,6180
5
5
  x_transformers/dpo.py,sha256=LjvWgCkqTl-UuehrzQ8nkX5guLr4whYwsmm7SKSwdls,3450
6
6
  x_transformers/nonautoregressive_wrapper.py,sha256=ys_p8obc7lTeeodCqvkRKxOXQ1C9T3j5Jwr-JbVgnXk,10432
7
- x_transformers/x_transformers.py,sha256=5cof7yvOAfFviLh-luafmhtTJDemCPoy9rHHYjWxLu4,68338
7
+ x_transformers/x_transformers.py,sha256=V_kR77GhWNu3TuUetu3xryWoFliwTJdNKRX2lVWnsRc,68274
8
8
  x_transformers/xl_autoregressive_wrapper.py,sha256=DCx4n0_c1tFai4nOqaWVnqx2p9eutsZsDMiMP1ckxNU,4117
9
9
  x_transformers/xval.py,sha256=QE1ltYZTR_eGgIHPP2BrMWVWVLqMW-OpDZh87BSmQEg,8563
10
- x_transformers-1.30.8.dist-info/LICENSE,sha256=As9u198X-U-vph5noInuUfqsAG2zX_oXPHDmdjwlPPY,1066
11
- x_transformers-1.30.8.dist-info/METADATA,sha256=2L0SfGhrbLMjpRKLwTp1_YH1Amu3g_j1nEuWIuGNqrQ,661
12
- x_transformers-1.30.8.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
13
- x_transformers-1.30.8.dist-info/top_level.txt,sha256=hO6KGpFuGucRNEtRfme4A_rGcM53AKwGP7RVlRIxS5Q,15
14
- x_transformers-1.30.8.dist-info/RECORD,,
10
+ x_transformers-1.30.9.dist-info/LICENSE,sha256=As9u198X-U-vph5noInuUfqsAG2zX_oXPHDmdjwlPPY,1066
11
+ x_transformers-1.30.9.dist-info/METADATA,sha256=auoB6F1DCe054gY70tE1k048x6wQRtftk8-Pk6nJD-I,661
12
+ x_transformers-1.30.9.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
13
+ x_transformers-1.30.9.dist-info/top_level.txt,sha256=hO6KGpFuGucRNEtRfme4A_rGcM53AKwGP7RVlRIxS5Q,15
14
+ x_transformers-1.30.9.dist-info/RECORD,,