ZJW666/ProjectedGANCLC
0
1# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES. All rights reserved.2#3# NVIDIA CORPORATION and its licensors retain all intellectual property4# and proprietary rights in and to this software, related documentation5# and any modifications thereto. Any use, reproduction, disclosure or6# distribution of this software and related documentation without an express7# license agreement from NVIDIA CORPORATION is strictly prohibited.8 9import re10import contextlib11import numpy as np12import torch13import warnings14import dnnlib15 16#----------------------------------------------------------------------------17# Cached construction of constant tensors. Avoids CPU=>GPU copy when the18# same constant is used multiple times.19 20_constant_cache = dict()21 22def constant(value, shape=None, dtype=None, device=None, memory_format=None):23 value = np.asarray(value)24 if shape is not None:25 shape = tuple(shape)26 if dtype is None:27 dtype = torch.get_default_dtype()28 if device is None:29 device = torch.device('cpu')30 if memory_format is None:31 memory_format = torch.contiguous_format32 33 key = (value.shape, value.dtype, value.tobytes(), shape, dtype, device, memory_format)34 tensor = _constant_cache.get(key, None)35 if tensor is None:36 tensor = torch.as_tensor(value.copy(), dtype=dtype, device=device)37 if shape is not None:38 tensor, _ = torch.broadcast_tensors(tensor, torch.empty(shape))39 tensor = tensor.contiguous(memory_format=memory_format)40 _constant_cache[key] = tensor41 return tensor42 43#----------------------------------------------------------------------------44# Replace NaN/Inf with specified numerical values.45 46try:47 nan_to_num = torch.nan_to_num # 1.8.0a048except AttributeError:49 def nan_to_num(input, nan=0.0, posinf=None, neginf=None, *, out=None): # pylint: disable=redefined-builtin50 assert isinstance(input, torch.Tensor)51 if posinf is None:52 posinf = torch.finfo(input.dtype).max53 if neginf is None:54 neginf = torch.finfo(input.dtype).min55 assert nan == 056 return torch.clamp(input.unsqueeze(0).nansum(0), min=neginf, max=posinf, out=out)57 58#----------------------------------------------------------------------------59# Symbolic assert.60 61try:62 symbolic_assert = torch._assert # 1.8.0a0 # pylint: disable=protected-access63except AttributeError:64 symbolic_assert = torch.Assert # 1.7.065 66#----------------------------------------------------------------------------67# Context manager to temporarily suppress known warnings in torch.jit.trace().68# Note: Cannot use catch_warnings because of https://bugs.python.org/issue2967269 70@contextlib.contextmanager71def suppress_tracer_warnings():72 flt = ('ignore', None, torch.jit.TracerWarning, None, 0)73 warnings.filters.insert(0, flt)74 yield75 warnings.filters.remove(flt)76 77#----------------------------------------------------------------------------78# Assert that the shape of a tensor matches the given list of integers.79# None indicates that the size of a dimension is allowed to vary.80# Performs symbolic assertion when used in torch.jit.trace().81 82def assert_shape(tensor, ref_shape):83 if tensor.ndim != len(ref_shape):84 raise AssertionError(f'Wrong number of dimensions: got {tensor.ndim}, expected {len(ref_shape)}')85 for idx, (size, ref_size) in enumerate(zip(tensor.shape, ref_shape)):86 if ref_size is None:87 pass88 elif isinstance(ref_size, torch.Tensor):89 with suppress_tracer_warnings(): # as_tensor results are registered as constants90 symbolic_assert(torch.equal(torch.as_tensor(size), ref_size), f'Wrong size for dimension {idx}')91 elif isinstance(size, torch.Tensor):92 with suppress_tracer_warnings(): # as_tensor results are registered as constants93 symbolic_assert(torch.equal(size, torch.as_tensor(ref_size)), f'Wrong size for dimension {idx}: expected {ref_size}')94 elif size != ref_size:95 raise AssertionError(f'Wrong size for dimension {idx}: got {size}, expected {ref_size}')96 97#----------------------------------------------------------------------------98# Function decorator that calls torch.autograd.profiler.record_function().99 100def profiled_function(fn):101 def decorator(*args, **kwargs):102 with torch.autograd.profiler.record_function(fn.__name__):103 return fn(*args, **kwargs)104 decorator.__name__ = fn.__name__105 return decorator106 107#----------------------------------------------------------------------------108# Sampler for torch.utils.data.DataLoader that loops over the dataset109# indefinitely, shuffling items as it goes.110 111class InfiniteSampler(torch.utils.data.Sampler):112 def __init__(self, dataset, rank=0, num_replicas=1, shuffle=True, seed=0, window_size=0.5):113 assert len(dataset) > 0114 assert num_replicas > 0115 assert 0 <= rank < num_replicas116 assert 0 <= window_size <= 1117 super().__init__(dataset)118 self.dataset = dataset119 self.rank = rank120 self.num_replicas = num_replicas121 self.shuffle = shuffle122 self.seed = seed123 self.window_size = window_size124 125 def __iter__(self):126 order = np.arange(len(self.dataset))127 rnd = None128 window = 0129 if self.shuffle:130 rnd = np.random.RandomState(self.seed)131 rnd.shuffle(order)132 window = int(np.rint(order.size * self.window_size))133 134 idx = 0135 while True:136 i = idx % order.size137 if idx % self.num_replicas == self.rank:138 yield order[i]139 if window >= 2:140 j = (i - rnd.randint(window)) % order.size141 order[i], order[j] = order[j], order[i]142 idx += 1143 144#----------------------------------------------------------------------------145# Utilities for operating with torch.nn.Module parameters and buffers.146 147def params_and_buffers(module):148 assert isinstance(module, torch.nn.Module)149 return list(module.parameters()) + list(module.buffers())150 151def named_params_and_buffers(module):152 assert isinstance(module, torch.nn.Module)153 return list(module.named_parameters()) + list(module.named_buffers())154 155def copy_params_and_buffers(src_module, dst_module, require_all=False):156 assert isinstance(src_module, torch.nn.Module)157 assert isinstance(dst_module, torch.nn.Module)158 src_tensors = dict(named_params_and_buffers(src_module))159 for name, tensor in named_params_and_buffers(dst_module):160 assert (name in src_tensors) or (not require_all)161 if name in src_tensors:162 try:163 tensor.copy_(src_tensors[name].detach()).requires_grad_(tensor.requires_grad)164 except:165 continue166 167#----------------------------------------------------------------------------168# Context manager for easily enabling/disabling DistributedDataParallel169# synchronization.170 171@contextlib.contextmanager172def ddp_sync(module, sync):173 assert isinstance(module, torch.nn.Module)174 if sync or not isinstance(module, torch.nn.parallel.DistributedDataParallel):175 yield176 else:177 with module.no_sync():178 yield179 180#----------------------------------------------------------------------------181# Check DistributedDataParallel consistency across processes.182 183def check_ddp_consistency(module, ignore_regex=None):184 assert isinstance(module, torch.nn.Module)185 for name, tensor in named_params_and_buffers(module):186 fullname = type(module).__name__ + '.' + name187 if ignore_regex is not None and re.fullmatch(ignore_regex, fullname):188 continue189 tensor = tensor.detach()190 if tensor.is_floating_point():191 tensor = nan_to_num(tensor)192 other = tensor.clone()193 torch.distributed.broadcast(tensor=other, src=0)194 assert (tensor == other).all(), fullname195 196#----------------------------------------------------------------------------197# Print summary table of module hierarchy.198 199def print_module_summary(module, inputs, max_nesting=3, skip_redundant=True):200 assert isinstance(module, torch.nn.Module)201 assert not isinstance(module, torch.jit.ScriptModule)202 assert isinstance(inputs, (tuple, list))203 204 # Register hooks.205 entries = []206 nesting = [0]207 def pre_hook(_mod, _inputs):208 nesting[0] += 1209 def post_hook(mod, _inputs, outputs):210 nesting[0] -= 1211 if nesting[0] <= max_nesting:212 outputs = list(outputs) if isinstance(outputs, (tuple, list)) else [outputs]213 outputs = [t for t in outputs if isinstance(t, torch.Tensor)]214 entries.append(dnnlib.EasyDict(mod=mod, outputs=outputs))215 hooks = [mod.register_forward_pre_hook(pre_hook) for mod in module.modules()]216 hooks += [mod.register_forward_hook(post_hook) for mod in module.modules()]217 218 # Run module.219 outputs = module(*inputs)220 for hook in hooks:221 hook.remove()222 223 # Identify unique outputs, parameters, and buffers.224 tensors_seen = set()225 for e in entries:226 e.unique_params = [t for t in e.mod.parameters() if id(t) not in tensors_seen]227 e.unique_buffers = [t for t in e.mod.buffers() if id(t) not in tensors_seen]228 e.unique_outputs = [t for t in e.outputs if id(t) not in tensors_seen]229 tensors_seen |= {id(t) for t in e.unique_params + e.unique_buffers + e.unique_outputs}230 231 # Filter out redundant entries.232 if skip_redundant:233 entries = [e for e in entries if len(e.unique_params) or len(e.unique_buffers) or len(e.unique_outputs)]234 235 # Construct table.236 rows = [[type(module).__name__, 'Parameters', 'Buffers', 'Output shape', 'Datatype']]237 rows += [['---'] * len(rows[0])]238 param_total = 0239 buffer_total = 0240 submodule_names = {mod: name for name, mod in module.named_modules()}241 for e in entries:242 name = '<top-level>' if e.mod is module else submodule_names[e.mod]243 param_size = sum(t.numel() for t in e.unique_params)244 buffer_size = sum(t.numel() for t in e.unique_buffers)245 output_shapes = [str(list(t.shape)) for t in e.outputs]246 output_dtypes = [str(t.dtype).split('.')[-1] for t in e.outputs]247 rows += [[248 name + (':0' if len(e.outputs) >= 2 else ''),249 str(param_size) if param_size else '-',250 str(buffer_size) if buffer_size else '-',251 (output_shapes + ['-'])[0],252 (output_dtypes + ['-'])[0],253 ]]254 for idx in range(1, len(e.outputs)):255 rows += [[name + f':{idx}', '-', '-', output_shapes[idx], output_dtypes[idx]]]256 param_total += param_size257 buffer_total += buffer_size258 rows += [['---'] * len(rows[0])]259 rows += [['Total', str(param_total), str(buffer_total), '-', '-']]260 261 # Print table.262 widths = [max(len(cell) for cell in column) for column in zip(*rows)]263 print()264 for row in rows:265 print(' '.join(cell + ' ' * (width - len(cell)) for cell, width in zip(row, widths)))266 print()267 return outputs268 269#----------------------------------------------------------------------------270 271# Added by Katja272import os273 274def get_ckpt_path(run_dir):275 return os.path.join(run_dir, f'network-snapshot.pkl')276