Spaces:
Running
Running
File size: 8,671 Bytes
e17e8cc |
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 |
"""This code is taken from <https://github.com/alexandre01/deepsvg>
by Alexandre Carlier, Martin Danelljan, Alexandre Alahi and Radu Timofte
from the paper >https://arxiv.org/pdf/2007.11301.pdf>
"""
import shutil
import torch
import torch.nn as nn
import os
import random
import numpy as np
import glob
def save_ckpt(checkpoint_dir, model, cfg=None, optimizer=None, scheduler_lr=None, scheduler_warmup=None,
stats=None, train_vars=None):
if is_multi_gpu(model):
model = model.module
state = {
"model": model.state_dict()
}
if optimizer is not None:
state["optimizer"] = optimizer.state_dict()
if scheduler_lr is not None:
state["scheduler_lr"] = scheduler_lr.state_dict()
if scheduler_warmup is not None:
state["scheduler_warmup"] = scheduler_warmup.state_dict()
if cfg is not None:
state["cfg"] = cfg.to_dict()
if stats is not None:
state["stats"] = stats.to_dict()
if train_vars is not None:
state["train_vars"] = train_vars.to_dict()
checkpoint_path = os.path.join(checkpoint_dir, "{:06d}.pth.tar".format(stats.step))
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
torch.save(state, checkpoint_path)
if stats.is_best():
best_model_path = os.path.join(checkpoint_dir, "best.pth.tar")
shutil.copyfile(checkpoint_path, best_model_path)
def save_ckpt_list(checkpoint_dir, model, cfg=None, optimizers=None, scheduler_lrs=None, scheduler_warmups=None,
stats=None, train_vars=None):
if is_multi_gpu(model):
model = model.module
state = {
"model": model.state_dict()
}
if optimizers is not None:
state["optimizers"] = [optimizer.state_dict() if optimizer is not None else optimizer for optimizer in optimizers]
if scheduler_lrs is not None:
state["scheduler_lrs"] = [scheduler_lr.state_dict() if scheduler_lr is not None else scheduler_lr for scheduler_lr in scheduler_lrs]
if scheduler_warmups is not None:
state["scheduler_warmups"] = [scheduler_warmup.state_dict() if scheduler_warmup is not None else None for scheduler_warmup in scheduler_warmups]
if cfg is not None:
state["cfg"] = cfg.to_dict()
if stats is not None:
state["stats"] = stats.to_dict()
if train_vars is not None:
state["train_vars"] = train_vars.to_dict()
checkpoint_path = os.path.join(checkpoint_dir, "{:06d}.pth.tar".format(stats.step))
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
torch.save(state, checkpoint_path)
if stats.is_best():
best_model_path = os.path.join(checkpoint_dir, "best.pth.tar")
shutil.copyfile(checkpoint_path, best_model_path)
def load_ckpt(checkpoint_dir, model, cfg=None, optimizer=None, scheduler_lr=None, scheduler_warmup=None,
stats=None, train_vars=None):
if not os.path.exists(checkpoint_dir):
return False
if os.path.isfile(checkpoint_dir):
checkpoint_path = checkpoint_dir
else:
ckpts_paths = sorted(glob.glob(os.path.join(checkpoint_dir, "./[0-9]*.pth.tar")))
if not ckpts_paths:
return False
checkpoint_path = ckpts_paths[-1]
state = torch.load(checkpoint_path, map_location=torch.device('cpu'))
if is_multi_gpu(model):
model = model.module
model.load_state_dict(state["model"], strict=False)
if optimizer is not None:
optimizer.load_state_dict(state["optimizer"])
if scheduler_lr is not None:
scheduler_lr.load_state_dict(state["scheduler_lr"])
if scheduler_warmup is not None:
scheduler_warmup.load_state_dict(state["scheduler_warmup"])
if cfg is not None:
cfg.load_dict(state["cfg"])
if stats is not None:
stats.load_dict(state["stats"])
if train_vars is not None:
train_vars.load_dict(state["train_vars"])
return True
def load_ckpt_list(checkpoint_dir, model, cfg=None, optimizers=None, scheduler_lrs=None, scheduler_warmups=None,
stats=None, train_vars=None):
if not os.path.exists(checkpoint_dir):
return False
if os.path.isfile(checkpoint_dir):
checkpoint_path = checkpoint_dir
else:
ckpts_paths = sorted(glob.glob(os.path.join(checkpoint_dir, "./[0-9]*.pth.tar")))
if not ckpts_paths:
return False
checkpoint_path = ckpts_paths[-1]
state = torch.load(checkpoint_path, map_location=torch.device('cpu'))
if is_multi_gpu(model):
model = model.module
model.load_state_dict(state["model"], strict=False)
for optimizer, scheduler_lr, scheduler_warmup, optimizer_sd, scheduler_lr_sd, scheduler_warmups_sd in zip(optimizers, scheduler_lrs, scheduler_warmups, state["optimizers"], state["scheduler_lrs"], state["scheduler_warmups"]):
if optimizer is not None and optimizer_sd is not None:
optimizer.load_state_dict(optimizer_sd)
if scheduler_lr is not None and scheduler_lr_sd is not None:
scheduler_lr.load_state_dict(scheduler_lr_sd)
if scheduler_warmup is not None and scheduler_warmups_sd is not None:
scheduler_warmup.load_state_dict(scheduler_warmups_sd)
if cfg is not None and state["cfg"] is not None:
cfg.load_dict(state["cfg"])
if stats is not None and state["stats"] is not None:
stats.load_dict(state["stats"])
if train_vars is not None and state["train_vars"] is not None:
train_vars.load_dict(state["train_vars"])
return True
def load_model(checkpoint_path, model):
state = torch.load(checkpoint_path, map_location=torch.device('cpu'))
if is_multi_gpu(model):
model = model.module
model.load_state_dict(state["model"], strict=False)
def is_multi_gpu(model):
return isinstance(model, nn.DataParallel)
def count_parameters(model):
return sum(p.numel() for p in model.parameters() if p.requires_grad)
def pad_sequence(sequences, batch_first=False, padding_value=0, max_len=None):
r"""Pad a list of variable length Tensors with ``padding_value``
``pad_sequence`` stacks a list of Tensors along a new dimension,
and pads them to equal length. For example, if the input is list of
sequences with size ``L x *`` and if batch_first is False, and ``T x B x *``
otherwise.
`B` is batch size. It is equal to the number of elements in ``sequences``.
`T` is length of the longest sequence.
`L` is length of the sequence.
`*` is any number of trailing dimensions, including none.
Example:
>>> from torch.nn.utils.rnn import pad_sequence
>>> a = torch.ones(25, 300)
>>> b = torch.ones(22, 300)
>>> c = torch.ones(15, 300)
>>> pad_sequence([a, b, c]).size()
torch.Size([25, 3, 300])
Note:
This function returns a Tensor of size ``T x B x *`` or ``B x T x *``
where `T` is the length of the longest sequence. This function assumes
trailing dimensions and type of all the Tensors in sequences are same.
Arguments:
sequences (list[Tensor]): list of variable length sequences.
batch_first (bool, optional): output will be in ``B x T x *`` if True, or in
``T x B x *`` otherwise
padding_value (float, optional): value for padded elements. Default: 0.
Returns:
Tensor of size ``T x B x *`` if :attr:`batch_first` is ``False``.
Tensor of size ``B x T x *`` otherwise
"""
# assuming trailing dimensions and type of all the Tensors
# in sequences are same and fetching those from sequences[0]
max_size = sequences[0].size()
trailing_dims = max_size[1:]
if max_len is None:
max_len = max([s.size(0) for s in sequences])
if batch_first:
out_dims = (len(sequences), max_len) + trailing_dims
else:
out_dims = (max_len, len(sequences)) + trailing_dims
out_tensor = sequences[0].data.new(*out_dims).fill_(padding_value)
for i, tensor in enumerate(sequences):
length = tensor.size(0)
# use index notation to prevent duplicate references to the tensor
if batch_first:
out_tensor[i, :length, ...] = tensor
else:
out_tensor[:length, i, ...] = tensor
return out_tensor
def set_seed(_seed=42):
random.seed(_seed)
np.random.seed(_seed)
torch.manual_seed(_seed)
torch.cuda.manual_seed(_seed)
torch.cuda.manual_seed_all(_seed)
os.environ['PYTHONHASHSEED'] = str(_seed)
def infinite_range(start_idx=0):
while True:
yield start_idx
start_idx += 1
|