Spaces:
Running
on
Zero
Running
on
Zero
File size: 10,525 Bytes
6d08643 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 |
"""
adopted from
https://github.com/openai/improved-diffusion/blob/main/improved_diffusion/gaussian_diffusion.py
and
https://github.com/lucidrains/denoising-diffusion-pytorch/blob/7706bdfc6f527f58d33f84b7b522e61e6e3164b3/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py
and
https://github.com/openai/guided-diffusion/blob/0ba878e517b276c45d1195eb29f6f5f72659a05b/guided_diffusion/nn.py
thanks!
"""
import math
import torch
import torch.nn as nn
from einops import repeat
def make_beta_schedule(
schedule,
n_timestep,
linear_start=1e-4,
linear_end=2e-2,
):
if schedule == "linear":
betas = (
torch.linspace(
linear_start**0.5, linear_end**0.5, n_timestep, dtype=torch.float64
)
** 2
)
return betas.numpy()
def extract_into_tensor(a, t, x_shape):
b, *_ = t.shape
out = a.gather(-1, t)
return out.reshape(b, *((1,) * (len(x_shape) - 1)))
def mixed_checkpoint(func, inputs: dict, params, flag):
"""
Evaluate a function without caching intermediate activations, allowing for
reduced memory at the expense of extra compute in the backward pass. This differs from the original checkpoint function
borrowed from https://github.com/openai/guided-diffusion/blob/0ba878e517b276c45d1195eb29f6f5f72659a05b/guided_diffusion/nn.py in that
it also works with non-tensor inputs
:param func: the function to evaluate.
:param inputs: the argument dictionary to pass to `func`.
:param params: a sequence of parameters `func` depends on but does not
explicitly take as arguments.
:param flag: if False, disable gradient checkpointing.
"""
if flag:
tensor_keys = [key for key in inputs if isinstance(inputs[key], torch.Tensor)]
tensor_inputs = [
inputs[key] for key in inputs if isinstance(inputs[key], torch.Tensor)
]
non_tensor_keys = [
key for key in inputs if not isinstance(inputs[key], torch.Tensor)
]
non_tensor_inputs = [
inputs[key] for key in inputs if not isinstance(inputs[key], torch.Tensor)
]
args = tuple(tensor_inputs) + tuple(non_tensor_inputs) + tuple(params)
return MixedCheckpointFunction.apply(
func,
len(tensor_inputs),
len(non_tensor_inputs),
tensor_keys,
non_tensor_keys,
*args,
)
else:
return func(**inputs)
class MixedCheckpointFunction(torch.autograd.Function):
@staticmethod
def forward(
ctx,
run_function,
length_tensors,
length_non_tensors,
tensor_keys,
non_tensor_keys,
*args,
):
ctx.end_tensors = length_tensors
ctx.end_non_tensors = length_tensors + length_non_tensors
ctx.gpu_autocast_kwargs = {
"enabled": torch.is_autocast_enabled(),
"dtype": torch.get_autocast_gpu_dtype(),
"cache_enabled": torch.is_autocast_cache_enabled(),
}
assert (
len(tensor_keys) == length_tensors
and len(non_tensor_keys) == length_non_tensors
)
ctx.input_tensors = {
key: val for (key, val) in zip(tensor_keys, list(args[: ctx.end_tensors]))
}
ctx.input_non_tensors = {
key: val
for (key, val) in zip(
non_tensor_keys, list(args[ctx.end_tensors : ctx.end_non_tensors])
)
}
ctx.run_function = run_function
ctx.input_params = list(args[ctx.end_non_tensors :])
with torch.no_grad():
output_tensors = ctx.run_function(
**ctx.input_tensors, **ctx.input_non_tensors
)
return output_tensors
@staticmethod
def backward(ctx, *output_grads):
# additional_args = {key: ctx.input_tensors[key] for key in ctx.input_tensors if not isinstance(ctx.input_tensors[key],torch.Tensor)}
ctx.input_tensors = {
key: ctx.input_tensors[key].detach().requires_grad_(True)
for key in ctx.input_tensors
}
with torch.enable_grad(), torch.cuda.amp.autocast(**ctx.gpu_autocast_kwargs):
# Fixes a bug where the first op in run_function modifies the
# Tensor storage in place, which is not allowed for detach()'d
# Tensors.
shallow_copies = {
key: ctx.input_tensors[key].view_as(ctx.input_tensors[key])
for key in ctx.input_tensors
}
# shallow_copies.update(additional_args)
output_tensors = ctx.run_function(**shallow_copies, **ctx.input_non_tensors)
input_grads = torch.autograd.grad(
output_tensors,
list(ctx.input_tensors.values()) + ctx.input_params,
output_grads,
allow_unused=True,
)
del ctx.input_tensors
del ctx.input_params
del output_tensors
return (
(None, None, None, None, None)
+ input_grads[: ctx.end_tensors]
+ (None,) * (ctx.end_non_tensors - ctx.end_tensors)
+ input_grads[ctx.end_tensors :]
)
def checkpoint(func, inputs, params, flag):
"""
Evaluate a function without caching intermediate activations, allowing for
reduced memory at the expense of extra compute in the backward pass.
:param func: the function to evaluate.
:param inputs: the argument sequence to pass to `func`.
:param params: a sequence of parameters `func` depends on but does not
explicitly take as arguments.
:param flag: if False, disable gradient checkpointing.
"""
if flag:
args = tuple(inputs) + tuple(params)
return CheckpointFunction.apply(func, len(inputs), *args)
else:
return func(*inputs)
class CheckpointFunction(torch.autograd.Function):
@staticmethod
def forward(ctx, run_function, length, *args):
ctx.run_function = run_function
ctx.input_tensors = list(args[:length])
ctx.input_params = list(args[length:])
ctx.gpu_autocast_kwargs = {
"enabled": torch.is_autocast_enabled(),
"dtype": torch.get_autocast_gpu_dtype(),
"cache_enabled": torch.is_autocast_cache_enabled(),
}
with torch.no_grad():
output_tensors = ctx.run_function(*ctx.input_tensors)
return output_tensors
@staticmethod
def backward(ctx, *output_grads):
ctx.input_tensors = [x.detach().requires_grad_(True) for x in ctx.input_tensors]
with torch.enable_grad(), torch.cuda.amp.autocast(**ctx.gpu_autocast_kwargs):
# Fixes a bug where the first op in run_function modifies the
# Tensor storage in place, which is not allowed for detach()'d
# Tensors.
shallow_copies = [x.view_as(x) for x in ctx.input_tensors]
output_tensors = ctx.run_function(*shallow_copies)
input_grads = torch.autograd.grad(
output_tensors,
ctx.input_tensors + ctx.input_params,
output_grads,
allow_unused=True,
)
del ctx.input_tensors
del ctx.input_params
del output_tensors
return (None, None) + input_grads
def timestep_embedding(timesteps, dim, max_period=10000, repeat_only=False):
"""
Create sinusoidal timestep embeddings.
:param timesteps: a 1-D Tensor of N indices, one per batch element.
These may be fractional.
:param dim: the dimension of the output.
:param max_period: controls the minimum frequency of the embeddings.
:return: an [N x dim] Tensor of positional embeddings.
"""
if not repeat_only:
half = dim // 2
freqs = torch.exp(
-math.log(max_period)
* torch.arange(start=0, end=half, dtype=torch.float32)
/ half
).to(device=timesteps.device)
args = timesteps[:, None].float() * freqs[None]
embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
if dim % 2:
embedding = torch.cat(
[embedding, torch.zeros_like(embedding[:, :1])], dim=-1
)
else:
embedding = repeat(timesteps, "b -> b d", d=dim)
return embedding
def zero_module(module):
"""
Zero out the parameters of a module and return it.
"""
for p in module.parameters():
p.detach().zero_()
return module
def scale_module(module, scale):
"""
Scale the parameters of a module and return it.
"""
for p in module.parameters():
p.detach().mul_(scale)
return module
def mean_flat(tensor):
"""
Take the mean over all non-batch dimensions.
"""
return tensor.mean(dim=list(range(1, len(tensor.shape))))
def normalization(channels):
"""
Make a standard normalization layer.
:param channels: number of input channels.
:return: an nn.Module for normalization.
"""
return GroupNorm32(32, channels)
# PyTorch 1.7 has SiLU, but we support PyTorch 1.5.
class SiLU(nn.Module):
def forward(self, x):
return x * torch.sigmoid(x)
class GroupNorm32(nn.GroupNorm):
def forward(self, x):
# return super().forward(x.float()).type(x.dtype)
return super().forward(x)
def conv_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D convolution module.
"""
if dims == 1:
return nn.Conv1d(*args, **kwargs)
elif dims == 2:
return nn.Conv2d(*args, **kwargs)
elif dims == 3:
return nn.Conv3d(*args, **kwargs)
raise ValueError(f"unsupported dimensions: {dims}")
def linear(*args, **kwargs):
"""
Create a linear module.
"""
return nn.Linear(*args, **kwargs)
def avg_pool_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D average pooling module.
"""
if dims == 1:
return nn.AvgPool1d(*args, **kwargs)
elif dims == 2:
return nn.AvgPool2d(*args, **kwargs)
elif dims == 3:
return nn.AvgPool3d(*args, **kwargs)
raise ValueError(f"unsupported dimensions: {dims}")
|