Skip to content

disable neonpixel optimizations on M1 hardware #414

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

Merged
lstein merged 3 commits into from
Sep 7, 2022
Merged

disable neonpixel optimizations on M1 hardware #414

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
79e0e75
disable neonpixel optimizations on M1 hardware
lstein Sep 7, 2022
fb24b67
fix typo that was causing random noise images on m1
lstein Sep 7, 2022
d3b4c89
Merge branch 'development' into neonsecret-optimizations
lstein Sep 7, 2022
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
34 changes: 20 additions & 14 deletions ldm/modules/attention.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,9 +1,10 @@
from inspect import isfunction
import math
from inspect import isfunction

import torch
import torch.nn.functional as F
from torch import nn, einsum
from einops import rearrange, repeat
from torch import nn, einsum

from ldm.modules.diffusionmodules.util import checkpoint

Expand All @@ -13,7 +14,7 @@ def exists(val):


def uniq(arr):
return{el: True for el in arr}.keys()
return {el: True for el in arr}.keys()


def default(val, d):
Expand Down Expand Up @@ -82,14 +83,14 @@ def __init__(self, dim, heads=4, dim_head=32):
super().__init__()
self.heads = heads
hidden_dim = dim_head * heads
self.to_qkv = nn.Conv2d(dim, hidden_dim * 3, 1, bias = False)
self.to_qkv = nn.Conv2d(dim, hidden_dim * 3, 1, bias=False)
self.to_out = nn.Conv2d(hidden_dim, dim, 1)

def forward(self, x):
b, c, h, w = x.shape
qkv = self.to_qkv(x)
q, k, v = rearrange(qkv, 'b (qkv heads c) h w -> qkv b heads c (h w)', heads = self.heads, qkv=3)
k = k.softmax(dim=-1)
q, k, v = rearrange(qkv, 'b (qkv heads c) h w -> qkv b heads c (h w)', heads=self.heads, qkv=3)
k = k.softmax(dim=-1)
context = torch.einsum('bhdn,bhen->bhde', k, v)
out = torch.einsum('bhde,bhdn->bhen', context, q)
out = rearrange(out, 'b heads c (h w) -> b (heads c) h w', heads=self.heads, h=h, w=w)
Expand Down Expand Up @@ -131,12 +132,12 @@ def forward(self, x):
v = self.v(h_)

# compute attention
b,c,h,w = q.shape
b, c, h, w = q.shape
q = rearrange(q, 'b c h w -> b (h w) c')
k = rearrange(k, 'b c h w -> b c (h w)')
w_ = torch.einsum('bij,bjk->bik', q, k)

w_ = w_ * (int(c)**(-0.5))
w_ = w_ * (int(c) ** (-0.5))
w_ = torch.nn.functional.softmax(w_, dim=2)

# attend to values
Expand All @@ -146,7 +147,7 @@ def forward(self, x):
h_ = rearrange(h_, 'b c (h w) -> b c h w', h=h)
h_ = self.proj_out(h_)

return x+h_
return x + h_


class CrossAttention(nn.Module):
Expand Down Expand Up @@ -174,6 +175,7 @@ def forward(self, x, context=None, mask=None):
context = default(context, x)
k = self.to_k(context)
v = self.to_v(context)
device_type = x.device.type
del context, x

q, k, v = map(lambda t: rearrange(t, 'b n (h d) -> (b h) n d', h=h), (q, k, v))
Expand All @@ -188,9 +190,11 @@ def forward(self, x, context=None, mask=None):
sim.masked_fill_(~mask, max_neg_value)
del mask

# attention, what we cannot get enough of, by halves
sim[4:] = sim[4:].softmax(dim=-1)
sim[:4] = sim[:4].softmax(dim=-1)
if device_type == 'mps': #special case for M1 - disable neonsecret optimization
sim = sim.softmax(dim=-1)
else:
sim[4:] = sim[4:].softmax(dim=-1)
sim[:4] = sim[:4].softmax(dim=-1)

sim = einsum('b i j, b j d -> b i d', sim, v)
sim = rearrange(sim, '(b h) n d -> b n (h d)', h=h)
Expand All @@ -200,7 +204,8 @@ def forward(self, x, context=None, mask=None):
class BasicTransformerBlock(nn.Module):
def __init__(self, dim, n_heads, d_head, dropout=0., context_dim=None, gated_ff=True, checkpoint=True):
super().__init__()
self.attn1 = CrossAttention(query_dim=dim, heads=n_heads, dim_head=d_head, dropout=dropout) # is a self-attention
self.attn1 = CrossAttention(query_dim=dim, heads=n_heads, dim_head=d_head,
dropout=dropout) # is a self-attention
self.ff = FeedForward(dim, dropout=dropout, glu=gated_ff)
self.attn2 = CrossAttention(query_dim=dim, context_dim=context_dim,
heads=n_heads, dim_head=d_head, dropout=dropout) # is self-attn if context is none
Expand Down Expand Up @@ -228,6 +233,7 @@ class SpatialTransformer(nn.Module):
Then apply standard transformer action.
Finally, reshape to image
"""

def __init__(self, in_channels, n_heads, d_head,
depth=1, dropout=0., context_dim=None):
super().__init__()
Expand All @@ -243,7 +249,7 @@ def __init__(self, in_channels, n_heads, d_head,

self.transformer_blocks = nn.ModuleList(
[BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim)
for d in range(depth)]
for d in range(depth)]
)

self.proj_out = zero_module(nn.Conv2d(inner_dim,
Expand Down