Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Some little changes #74

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Some little changes #74

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
076160a
Remove the unused parameter 'pad_value'
vasiliyeskin Aug 5, 2021
2f968ea
Remove the extra commas in the Performer and Attention definitions
vasiliyeskin Aug 5, 2021
7dadc2e
Merge branch 'lucidrains:main' into main
vasiliyeskin Aug 23, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -94,7 +94,7 @@ enc_dec = PerformerEncDec(
dec_num_tokens = 20000,
dec_depth = 6,
dec_heads = 8,
dec_max_seq_len = TGT_SEQ_LEN,
dec_max_seq_len = TGT_SEQ_LEN
)

src = torch.randint(0, 20000, (1, SRC_SEQ_LEN))
Expand Down Expand Up @@ -124,7 +124,7 @@ from performer_pytorch import SelfAttention
attn = SelfAttention(
dim = 512,
heads = 8,
causal = False,
causal = False
).cuda()

x = torch.randn(1, 1024, 512).cuda()
Expand Down
3 changes: 1 addition & 2 deletions performer_pytorch/autoregressive_wrapper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,9 +26,8 @@ def top_k(logits, thres = 0.9):
return probs

class AutoregressiveWrapper(nn.Module):
def __init__(self, net, ignore_index = 0, pad_value = 0):
def __init__(self, net, ignore_index = 0):
super().__init__()
self.pad_value = pad_value
self.ignore_index = ignore_index

self.net = net
Expand Down
3 changes: 1 addition & 2 deletions performer_pytorch/performer_enc_dec.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,6 @@ def __init__(
self,
dim,
ignore_index = 0,
pad_value = 0,
tie_token_embeds = False,
no_projection = False,
**kwargs
Expand All @@ -65,7 +64,7 @@ def __init__(
enc.token_emb = dec.token_emb

self.enc = enc
self.dec = AutoregressiveWrapper(dec, ignore_index = ignore_index, pad_value = pad_value)
self.dec = AutoregressiveWrapper(dec, ignore_index = ignore_index)

@torch.no_grad()
def generate(self, seq_in, seq_out_start, seq_len, **kwargs):
Expand Down