mirror of
https://github.com/vladmandic/automatic
synced 2026-09-20 01:31:13 +02:00
fix slow hypernetworks enum
This commit is contained in:
@@ -1,5 +1,4 @@
|
||||
import datetime
|
||||
import glob
|
||||
import html
|
||||
import os
|
||||
from collections import deque
|
||||
@@ -35,19 +34,14 @@ class HypernetworkModule(torch.nn.Module):
|
||||
def __init__(self, dim, state_dict=None, layer_structure=None, activation_func=None, weight_init='Normal',
|
||||
add_layer_norm=False, activate_output=False, dropout_structure=None):
|
||||
super().__init__()
|
||||
|
||||
self.multiplier = 1.0
|
||||
|
||||
assert layer_structure is not None, "layer_structure must not be None"
|
||||
assert layer_structure[0] == 1, "Multiplier Sequence should start with size 1!"
|
||||
assert layer_structure[-1] == 1, "Multiplier Sequence should end with size 1!"
|
||||
|
||||
linears = []
|
||||
for i in range(len(layer_structure) - 1):
|
||||
|
||||
# Add a fully-connected layer
|
||||
linears.append(torch.nn.Linear(int(dim * layer_structure[i]), int(dim * layer_structure[i+1])))
|
||||
|
||||
# Add an activation func except last layer
|
||||
if activation_func == "linear" or activation_func is None or (i >= len(layer_structure) - 2 and not activate_output):
|
||||
pass
|
||||
@@ -55,20 +49,16 @@ class HypernetworkModule(torch.nn.Module):
|
||||
linears.append(self.activation_dict[activation_func]())
|
||||
else:
|
||||
raise RuntimeError(f'hypernetwork uses an unsupported activation function: {activation_func}')
|
||||
|
||||
# Add layer normalization
|
||||
if add_layer_norm:
|
||||
linears.append(torch.nn.LayerNorm(int(dim * layer_structure[i+1])))
|
||||
|
||||
# Everything should be now parsed into dropout structure, and applied here.
|
||||
# Since we only have dropouts after layers, dropout structure should start with 0 and end with 0.
|
||||
if dropout_structure is not None and dropout_structure[i+1] > 0:
|
||||
assert 0 < dropout_structure[i+1] < 1, "Dropout probability should be 0 or float between 0 and 1!"
|
||||
linears.append(torch.nn.Dropout(p=dropout_structure[i+1]))
|
||||
# Code explanation : [1, 2, 1] -> dropout is missing when last_layer_dropout is false. [1, 2, 2, 1] -> [0, 0.3, 0, 0], when its True, [0, 0.3, 0.3, 0].
|
||||
|
||||
self.linear = torch.nn.Sequential(*linears)
|
||||
|
||||
if state_dict is not None:
|
||||
self.fix_old_state_dict(state_dict)
|
||||
self.load_state_dict(state_dict)
|
||||
@@ -102,12 +92,10 @@ class HypernetworkModule(torch.nn.Module):
|
||||
'linear2.bias': 'linear.1.bias',
|
||||
'linear2.weight': 'linear.1.weight',
|
||||
}
|
||||
|
||||
for fr, to in changes.items():
|
||||
x = state_dict.get(fr, None)
|
||||
if x is None:
|
||||
continue
|
||||
|
||||
del state_dict[fr]
|
||||
state_dict[to] = x
|
||||
|
||||
@@ -162,7 +150,6 @@ class Hypernetwork:
|
||||
self.optimizer_name = None
|
||||
self.optimizer_state_dict = None
|
||||
self.optional_info = None
|
||||
|
||||
for size in enable_sizes or []:
|
||||
self.layers[size] = (
|
||||
HypernetworkModule(size, None, self.layer_structure, self.activation_func, self.weight_init,
|
||||
@@ -210,10 +197,8 @@ class Hypernetwork:
|
||||
def save(self, filename):
|
||||
state_dict = {}
|
||||
optimizer_saved_dict = {}
|
||||
|
||||
for k, v in self.layers.items():
|
||||
state_dict[k] = (v[0].state_dict(), v[1].state_dict())
|
||||
|
||||
state_dict['step'] = self.step
|
||||
state_dict['name'] = self.name
|
||||
state_dict['layer_structure'] = self.layer_structure
|
||||
@@ -227,10 +212,8 @@ class Hypernetwork:
|
||||
state_dict['dropout_structure'] = self.dropout_structure
|
||||
state_dict['last_layer_dropout'] = (self.dropout_structure[-2] != 0) if self.dropout_structure is not None else self.last_layer_dropout
|
||||
state_dict['optional_info'] = self.optional_info if self.optional_info else None
|
||||
|
||||
if self.optimizer_name is not None:
|
||||
optimizer_saved_dict['optimizer_name'] = self.optimizer_name
|
||||
|
||||
torch.save(state_dict, filename)
|
||||
if shared.opts.save_optimizer_state and self.optimizer_state_dict:
|
||||
optimizer_saved_dict['hash'] = self.shorthash()
|
||||
@@ -241,10 +224,8 @@ class Hypernetwork:
|
||||
self.filename = filename
|
||||
if self.name is None:
|
||||
self.name = os.path.splitext(os.path.basename(filename))[0]
|
||||
|
||||
with progress.open(filename, 'rb', description=f'Loading hypernetwork: [cyan]{filename}', auto_refresh=True) as f:
|
||||
state_dict = torch.load(f, map_location='cpu')
|
||||
|
||||
self.layer_structure = state_dict.get('layer_structure', [1, 2, 1])
|
||||
self.optional_info = state_dict.get('optional_info', None)
|
||||
self.activation_func = state_dict.get('activation_func', None)
|
||||
@@ -257,11 +238,9 @@ class Hypernetwork:
|
||||
# Dropout structure should have same length as layer structure, Every digits should be in [0,1), and last digit must be 0.
|
||||
if self.dropout_structure is None:
|
||||
self.dropout_structure = parse_dropout_structure(self.layer_structure, self.use_dropout, self.last_layer_dropout)
|
||||
|
||||
if shared.opts.print_hypernet_extra:
|
||||
if self.optional_info is not None:
|
||||
print(f" INFO:\n {self.optional_info}\n")
|
||||
|
||||
print(f" Layer structure: {self.layer_structure}")
|
||||
print(f" Activation function: {self.activation_func}")
|
||||
print(f" Weight initialization: {self.weight_init}")
|
||||
@@ -269,9 +248,7 @@ class Hypernetwork:
|
||||
print(f" Dropout usage: {self.use_dropout}" )
|
||||
print(f" Activate last layer: {self.activate_output}")
|
||||
print(f" Dropout structure: {self.dropout_structure}")
|
||||
|
||||
optimizer_saved_dict = torch.load(self.filename + '.optim', map_location='cpu') if os.path.exists(self.filename + '.optim') else {}
|
||||
|
||||
if self.shorthash() == optimizer_saved_dict.get('hash', None):
|
||||
self.optimizer_state_dict = optimizer_saved_dict.get('optimizer_state_dict', None)
|
||||
else:
|
||||
@@ -285,7 +262,6 @@ class Hypernetwork:
|
||||
self.optimizer_name = "AdamW"
|
||||
if shared.opts.print_hypernet_extra:
|
||||
print("No saved optimizer exists in checkpoint")
|
||||
|
||||
for size, sd in state_dict.items():
|
||||
if type(size) == int:
|
||||
self.layers[size] = (
|
||||
@@ -294,7 +270,6 @@ class Hypernetwork:
|
||||
HypernetworkModule(size, sd[1], self.layer_structure, self.activation_func, self.weight_init,
|
||||
self.add_layer_norm, self.activate_output, self.dropout_structure),
|
||||
)
|
||||
|
||||
self.name = state_dict.get('name', self.name)
|
||||
self.step = state_dict.get('step', 0)
|
||||
self.sd_checkpoint = state_dict.get('sd_checkpoint', None)
|
||||
@@ -303,54 +278,49 @@ class Hypernetwork:
|
||||
|
||||
def shorthash(self):
|
||||
sha256 = hashes.sha256(self.filename, f'hypernet/{self.name}')
|
||||
|
||||
return sha256[0:10] if sha256 else None
|
||||
|
||||
|
||||
def list_hypernetworks(path):
|
||||
res = {}
|
||||
for filename in sorted(glob.iglob(os.path.join(path, '**/*.pt'), recursive=True), key=str.lower):
|
||||
name = os.path.splitext(os.path.basename(filename))[0]
|
||||
# Prevent a hypothetical "None.pt" from being listed.
|
||||
if name != "None":
|
||||
res[name] = filename
|
||||
def list_folder(folder):
|
||||
for filename in os.listdir(folder):
|
||||
fn = os.path.join(folder, filename)
|
||||
if os.path.isfile(fn) and fn.lower().endswith(".pt"):
|
||||
name = os.path.splitext(os.path.basename(fn))[0]
|
||||
res[name] = filename
|
||||
elif os.path.isdir(fn) and not fn.startswith('.'):
|
||||
list_folder(fn)
|
||||
|
||||
list_folder(path)
|
||||
return res
|
||||
|
||||
|
||||
def load_hypernetwork(name):
|
||||
path = shared.hypernetworks.get(name, None)
|
||||
|
||||
if path is None:
|
||||
return None
|
||||
|
||||
hypernetwork = Hypernetwork()
|
||||
|
||||
try:
|
||||
hypernetwork.load(path)
|
||||
except Exception as e:
|
||||
errors.display(e, f'hypernetwork load: {path}')
|
||||
return None
|
||||
|
||||
return hypernetwork
|
||||
|
||||
|
||||
def load_hypernetworks(names, multipliers=None):
|
||||
already_loaded = {}
|
||||
|
||||
for hypernetwork in shared.loaded_hypernetworks:
|
||||
if hypernetwork.name in names:
|
||||
already_loaded[hypernetwork.name] = hypernetwork
|
||||
|
||||
shared.loaded_hypernetworks.clear()
|
||||
|
||||
for i, name in enumerate(names):
|
||||
hypernetwork = already_loaded.get(name, None)
|
||||
if hypernetwork is None:
|
||||
hypernetwork = load_hypernetwork(name)
|
||||
|
||||
if hypernetwork is None:
|
||||
continue
|
||||
|
||||
hypernetwork.set_multiplier(multipliers[i] if multipliers else 1.0)
|
||||
shared.loaded_hypernetworks.append(hypernetwork)
|
||||
|
||||
@@ -368,14 +338,11 @@ def find_closest_hypernetwork_name(search: str):
|
||||
|
||||
def apply_single_hypernetwork(hypernetwork, context_k, context_v, layer=None):
|
||||
hypernetwork_layers = (hypernetwork.layers if hypernetwork is not None else {}).get(context_k.shape[2], None)
|
||||
|
||||
if hypernetwork_layers is None:
|
||||
return context_k, context_v
|
||||
|
||||
if layer is not None:
|
||||
layer.hyper_k = hypernetwork_layers[0]
|
||||
layer.hyper_v = hypernetwork_layers[1]
|
||||
|
||||
context_k = devices.cond_cast_unet(hypernetwork_layers[0](devices.cond_cast_float(context_k)))
|
||||
context_v = devices.cond_cast_unet(hypernetwork_layers[1](devices.cond_cast_float(context_v)))
|
||||
return context_k, context_v
|
||||
@@ -386,33 +353,25 @@ def apply_hypernetworks(hypernetworks, context, layer=None):
|
||||
context_v = context
|
||||
for hypernetwork in hypernetworks:
|
||||
context_k, context_v = apply_single_hypernetwork(hypernetwork, context_k, context_v, layer)
|
||||
|
||||
return context_k, context_v
|
||||
|
||||
|
||||
def attention_CrossAttention_forward(self, x, context=None, mask=None):
|
||||
h = self.heads
|
||||
|
||||
q = self.to_q(x)
|
||||
context = default(context, x)
|
||||
|
||||
context_k, context_v = apply_hypernetworks(shared.loaded_hypernetworks, context, self)
|
||||
k = self.to_k(context_k)
|
||||
v = self.to_v(context_v)
|
||||
|
||||
q, k, v = (rearrange(t, 'b n (h d) -> (b h) n d', h=h) for t in (q, k, v))
|
||||
|
||||
sim = einsum('b i d, b j d -> b i j', q, k) * self.scale
|
||||
|
||||
if mask is not None:
|
||||
mask = rearrange(mask, 'b ... -> b (...)')
|
||||
max_neg_value = -torch.finfo(sim.dtype).max
|
||||
mask = repeat(mask, 'b j -> (b h) () j', h=h)
|
||||
sim.masked_fill_(~mask, max_neg_value)
|
||||
|
||||
# attention, what we cannot get enough of
|
||||
attn = sim.softmax(dim=-1)
|
||||
|
||||
out = einsum('b i j, b j d -> b i d', attn, v)
|
||||
out = rearrange(out, '(b h) n d -> b n (h d)', h=h)
|
||||
return self.to_out(out)
|
||||
@@ -421,7 +380,6 @@ def attention_CrossAttention_forward(self, x, context=None, mask=None):
|
||||
def stack_conds(conds):
|
||||
if len(conds) == 1:
|
||||
return torch.stack(conds)
|
||||
|
||||
# same as in reconstruct_multicond_batch
|
||||
token_count = max([x.shape[0] for x in conds])
|
||||
for i in range(len(conds)):
|
||||
@@ -429,7 +387,6 @@ def stack_conds(conds):
|
||||
last_vector = conds[i][-1:]
|
||||
last_vector_repeated = last_vector.repeat([token_count - conds[i].shape[0], 1])
|
||||
conds[i] = torch.vstack([conds[i], last_vector_repeated])
|
||||
|
||||
return torch.stack(conds)
|
||||
|
||||
|
||||
@@ -464,19 +421,15 @@ def create_hypernetwork(name, enable_sizes, overwrite_old, layer_structure=None,
|
||||
# Remove illegal characters from name.
|
||||
name = "".join( x for x in name if (x.isalnum() or x in "._- "))
|
||||
assert name, "Name cannot be empty!"
|
||||
|
||||
fn = os.path.join(shared.opts.hypernetwork_dir, f"{name}.pt")
|
||||
if not overwrite_old:
|
||||
assert not os.path.exists(fn), f"file {fn} already exists"
|
||||
|
||||
if type(layer_structure) == str:
|
||||
layer_structure = [float(x.strip()) for x in layer_structure.split(",")]
|
||||
|
||||
if use_dropout and dropout_structure and type(dropout_structure) == str:
|
||||
dropout_structure = [float(x.strip()) for x in dropout_structure.split(",")]
|
||||
else:
|
||||
dropout_structure = [0] * len(layer_structure)
|
||||
|
||||
hypernet = modules.hypernetworks.hypernetwork.Hypernetwork(
|
||||
name=name,
|
||||
enable_sizes=[int(x) for x in enable_sizes],
|
||||
|
||||
Reference in New Issue
Block a user