├── .gitignore ├── docs ├── stylegan2-ada-teaser-1024x252.png ├── stylegan2-ada-training-curves.png ├── dataset-tool-help.txt ├── train-help.txt └── license.html ├── metrics ├── __init__.py ├── inception_score.py ├── frechet_inception_distance.py ├── kernel_inception_distance.py ├── precision_recall.py ├── perceptual_path_length.py ├── metric_main.py └── metric_utils.py ├── training ├── __init__.py ├── loss.py └── dataset.py ├── torch_utils ├── __init__.py ├── ops │ ├── __init__.py │ ├── bias_act.h │ ├── upfirdn2d.h │ ├── fma.py │ ├── grid_sample_gradfix.py │ ├── bias_act.cpp │ ├── upfirdn2d.cpp │ ├── bias_act.cu │ ├── conv2d_resample.py │ ├── conv2d_gradfix.py │ └── bias_act.py ├── custom_ops.py ├── persistence.py ├── training_stats.py ├── misc.py └── gen_utils.py ├── dnnlib └── __init__.py ├── Dockerfile ├── .github └── ISSUE_TEMPLATE │ └── bug_report.md ├── docker_run.sh ├── pytorch_ssim └── __init__.py ├── LICENSE.txt ├── sightseeding.py ├── experiments.py └── calc_metrics.py /.gitignore: -------------------------------------------------------------------------------- 1 | __pycache__/ 2 | .cache/ 3 | -------------------------------------------------------------------------------- /docs/stylegan2-ada-teaser-1024x252.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/PDillis/stylegan2-ada-pytorch/HEAD/docs/stylegan2-ada-teaser-1024x252.png -------------------------------------------------------------------------------- /docs/stylegan2-ada-training-curves.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/PDillis/stylegan2-ada-pytorch/HEAD/docs/stylegan2-ada-training-curves.png -------------------------------------------------------------------------------- /metrics/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | # empty 10 | -------------------------------------------------------------------------------- /training/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | # empty 10 | -------------------------------------------------------------------------------- /torch_utils/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | # empty 10 | -------------------------------------------------------------------------------- /torch_utils/ops/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | # empty 10 | -------------------------------------------------------------------------------- /dnnlib/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | from .util import EasyDict, make_cache_dir_path 10 | -------------------------------------------------------------------------------- /Dockerfile: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | FROM nvcr.io/nvidia/pytorch:20.12-py3 10 | 11 | ENV PYTHONDONTWRITEBYTECODE 1 12 | ENV PYTHONUNBUFFERED 1 13 | 14 | RUN pip install imageio-ffmpeg==0.4.3 pyspng==0.1.0 15 | 16 | WORKDIR /workspace 17 | 18 | # Unset TORCH_CUDA_ARCH_LIST and exec. This makes pytorch run-time 19 | # extension builds significantly faster as we only compile for the 20 | # currently active GPU configuration. 21 | RUN (printf '#!/bin/bash\nunset TORCH_CUDA_ARCH_LIST\nexec \"$@\"\n' >> /entry.sh) && chmod a+x /entry.sh 22 | ENTRYPOINT ["/entry.sh"] 23 | -------------------------------------------------------------------------------- /.github/ISSUE_TEMPLATE/bug_report.md: -------------------------------------------------------------------------------- 1 | --- 2 | name: Bug report 3 | about: Create a report to help us improve 4 | title: '' 5 | labels: '' 6 | assignees: '' 7 | 8 | --- 9 | 10 | **Describe the bug** 11 | A clear and concise description of what the bug is. 12 | 13 | **To Reproduce** 14 | Steps to reproduce the behavior: 15 | 1. In '...' directory, run command '...' 16 | 2. See error (please copy&paste full log and stacktraces). 17 | 18 | Please copy&paste text instead of screenshots for better searchability. 19 | 20 | **Expected behavior** 21 | A clear and concise description of what you expected to happen. 22 | 23 | **Screenshots** 24 | If applicable, add screenshots to help explain your problem. 25 | 26 | **Desktop (please complete the following information):** 27 | - OS: [e.g. Linux Ubuntu 20.04, Windows 10] 28 | - PyTorch version (e.g., pytorch 1.7.1) 29 | - CUDA toolkit version (e.g., CUDA 11.0) 30 | - NVIDIA driver version 31 | - GPU [e.g., Titan V, RTX 3090] 32 | - Docker: did you use Docker? If yes, specify docker image URL (e.g., nvcr.io/nvidia/pytorch:20.12-py3) 33 | 34 | **Additional context** 35 | Add any other context about the problem here. 36 | -------------------------------------------------------------------------------- /docker_run.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | 3 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 4 | # 5 | # NVIDIA CORPORATION and its licensors retain all intellectual property 6 | # and proprietary rights in and to this software, related documentation 7 | # and any modifications thereto. Any use, reproduction, disclosure or 8 | # distribution of this software and related documentation without an express 9 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 10 | 11 | set -e 12 | 13 | # Wrapper script for setting up `docker run` to properly 14 | # cache downloaded files, custom extension builds and 15 | # mount the source directory into the container and make it 16 | # run as non-root user. 17 | # 18 | # Use it like: 19 | # 20 | # ./docker_run.sh python generate.py --help 21 | # 22 | # To override the default `stylegan2ada:latest` image, run: 23 | # 24 | # IMAGE=my_image:v1.0 ./docker_run.sh python generate.py --help 25 | # 26 | 27 | rest=$@ 28 | 29 | IMAGE="${IMAGE:-sg2ada:latest}" 30 | 31 | CONTAINER_ID=$(docker inspect --format="{{.Id}}" ${IMAGE} 2> /dev/null) 32 | if [[ "${CONTAINER_ID}" ]]; then 33 | docker run --shm-size=2g --gpus all -it --rm -v `pwd`:/scratch --user $(id -u):$(id -g) \ 34 | --workdir=/scratch -e HOME=/scratch $IMAGE $@ 35 | else 36 | echo "Unknown container image: ${IMAGE}" 37 | exit 1 38 | fi 39 | -------------------------------------------------------------------------------- /torch_utils/ops/bias_act.h: -------------------------------------------------------------------------------- 1 | // Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | // 3 | // NVIDIA CORPORATION and its licensors retain all intellectual property 4 | // and proprietary rights in and to this software, related documentation 5 | // and any modifications thereto. Any use, reproduction, disclosure or 6 | // distribution of this software and related documentation without an express 7 | // license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | //------------------------------------------------------------------------ 10 | // CUDA kernel parameters. 11 | 12 | struct bias_act_kernel_params 13 | { 14 | const void* x; // [sizeX] 15 | const void* b; // [sizeB] or NULL 16 | const void* xref; // [sizeX] or NULL 17 | const void* yref; // [sizeX] or NULL 18 | const void* dy; // [sizeX] or NULL 19 | void* y; // [sizeX] 20 | 21 | int grad; 22 | int act; 23 | float alpha; 24 | float gain; 25 | float clamp; 26 | 27 | int sizeX; 28 | int sizeB; 29 | int stepB; 30 | int loopX; 31 | }; 32 | 33 | //------------------------------------------------------------------------ 34 | // CUDA kernel selection. 35 | 36 | template void* choose_bias_act_kernel(const bias_act_kernel_params& p); 37 | 38 | //------------------------------------------------------------------------ 39 | -------------------------------------------------------------------------------- /metrics/inception_score.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Inception Score (IS) from the paper "Improved techniques for training 10 | GANs". Matches the original implementation by Salimans et al. at 11 | https://github.com/openai/improved-gan/blob/master/inception_score/model.py""" 12 | 13 | import numpy as np 14 | from . import metric_utils 15 | 16 | #---------------------------------------------------------------------------- 17 | 18 | def compute_is(opts, num_gen, num_splits): 19 | # Direct TorchScript translation of http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz 20 | detector_url = 'https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/metrics/inception-2015-12-05.pt' 21 | detector_kwargs = dict(no_output_bias=True) # Match the original implementation by not applying bias in the softmax layer. 22 | 23 | gen_probs = metric_utils.compute_feature_stats_for_generator( 24 | opts=opts, detector_url=detector_url, detector_kwargs=detector_kwargs, 25 | capture_all=True, max_items=num_gen).get_all() 26 | 27 | if opts.rank != 0: 28 | return float('nan'), float('nan') 29 | 30 | scores = [] 31 | for i in range(num_splits): 32 | part = gen_probs[i * num_gen // num_splits : (i + 1) * num_gen // num_splits] 33 | kl = part * (np.log(part) - np.log(np.mean(part, axis=0, keepdims=True))) 34 | kl = np.mean(np.sum(kl, axis=1)) 35 | scores.append(np.exp(kl)) 36 | return float(np.mean(scores)), float(np.std(scores)) 37 | 38 | #---------------------------------------------------------------------------- 39 | -------------------------------------------------------------------------------- /torch_utils/ops/upfirdn2d.h: -------------------------------------------------------------------------------- 1 | // Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | // 3 | // NVIDIA CORPORATION and its licensors retain all intellectual property 4 | // and proprietary rights in and to this software, related documentation 5 | // and any modifications thereto. Any use, reproduction, disclosure or 6 | // distribution of this software and related documentation without an express 7 | // license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | #include 10 | 11 | //------------------------------------------------------------------------ 12 | // CUDA kernel parameters. 13 | 14 | struct upfirdn2d_kernel_params 15 | { 16 | const void* x; 17 | const float* f; 18 | void* y; 19 | 20 | int2 up; 21 | int2 down; 22 | int2 pad0; 23 | int flip; 24 | float gain; 25 | 26 | int4 inSize; // [width, height, channel, batch] 27 | int4 inStride; 28 | int2 filterSize; // [width, height] 29 | int2 filterStride; 30 | int4 outSize; // [width, height, channel, batch] 31 | int4 outStride; 32 | int sizeMinor; 33 | int sizeMajor; 34 | 35 | int loopMinor; 36 | int loopMajor; 37 | int loopX; 38 | int launchMinor; 39 | int launchMajor; 40 | }; 41 | 42 | //------------------------------------------------------------------------ 43 | // CUDA kernel specialization. 44 | 45 | struct upfirdn2d_kernel_spec 46 | { 47 | void* kernel; 48 | int tileOutW; 49 | int tileOutH; 50 | int loopMinor; 51 | int loopX; 52 | }; 53 | 54 | //------------------------------------------------------------------------ 55 | // CUDA kernel selection. 56 | 57 | template upfirdn2d_kernel_spec choose_upfirdn2d_kernel(const upfirdn2d_kernel_params& p); 58 | 59 | //------------------------------------------------------------------------ 60 | -------------------------------------------------------------------------------- /metrics/frechet_inception_distance.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Frechet Inception Distance (FID) from the paper 10 | "GANs trained by a two time-scale update rule converge to a local Nash 11 | equilibrium". Matches the original implementation by Heusel et al. at 12 | https://github.com/bioinf-jku/TTUR/blob/master/fid.py""" 13 | 14 | import numpy as np 15 | import scipy.linalg 16 | from . import metric_utils 17 | 18 | #---------------------------------------------------------------------------- 19 | 20 | def compute_fid(opts, max_real, num_gen): 21 | # Direct TorchScript translation of http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz 22 | detector_url = 'https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/metrics/inception-2015-12-05.pt' 23 | detector_kwargs = dict(return_features=True) # Return raw features before the softmax layer. 24 | 25 | mu_real, sigma_real = metric_utils.compute_feature_stats_for_dataset( 26 | opts=opts, detector_url=detector_url, detector_kwargs=detector_kwargs, 27 | rel_lo=0, rel_hi=0, capture_mean_cov=True, max_items=max_real).get_mean_cov() 28 | 29 | mu_gen, sigma_gen = metric_utils.compute_feature_stats_for_generator( 30 | opts=opts, detector_url=detector_url, detector_kwargs=detector_kwargs, 31 | rel_lo=0, rel_hi=1, capture_mean_cov=True, max_items=num_gen).get_mean_cov() 32 | 33 | if opts.rank != 0: 34 | return float('nan') 35 | 36 | m = np.square(mu_gen - mu_real).sum() 37 | s, _ = scipy.linalg.sqrtm(np.dot(sigma_gen, sigma_real), disp=False) # pylint: disable=no-member 38 | fid = np.real(m + np.trace(sigma_gen + sigma_real - s * 2)) 39 | return float(fid) 40 | 41 | #---------------------------------------------------------------------------- 42 | -------------------------------------------------------------------------------- /torch_utils/ops/fma.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Fused multiply-add, with slightly faster gradients than `torch.addcmul()`.""" 10 | 11 | import torch 12 | 13 | #---------------------------------------------------------------------------- 14 | 15 | def fma(a, b, c): # => a * b + c 16 | return _FusedMultiplyAdd.apply(a, b, c) 17 | 18 | #---------------------------------------------------------------------------- 19 | 20 | class _FusedMultiplyAdd(torch.autograd.Function): # a * b + c 21 | @staticmethod 22 | def forward(ctx, a, b, c): # pylint: disable=arguments-differ 23 | out = torch.addcmul(c, a, b) 24 | ctx.save_for_backward(a, b) 25 | ctx.c_shape = c.shape 26 | return out 27 | 28 | @staticmethod 29 | def backward(ctx, dout): # pylint: disable=arguments-differ 30 | a, b = ctx.saved_tensors 31 | c_shape = ctx.c_shape 32 | da = None 33 | db = None 34 | dc = None 35 | 36 | if ctx.needs_input_grad[0]: 37 | da = _unbroadcast(dout * b, a.shape) 38 | 39 | if ctx.needs_input_grad[1]: 40 | db = _unbroadcast(dout * a, b.shape) 41 | 42 | if ctx.needs_input_grad[2]: 43 | dc = _unbroadcast(dout, c_shape) 44 | 45 | return da, db, dc 46 | 47 | #---------------------------------------------------------------------------- 48 | 49 | def _unbroadcast(x, shape): 50 | extra_dims = x.ndim - len(shape) 51 | assert extra_dims >= 0 52 | dim = [i for i in range(x.ndim) if x.shape[i] > 1 and (i < extra_dims or shape[i - extra_dims] == 1)] 53 | if len(dim): 54 | x = x.sum(dim=dim, keepdim=True) 55 | if extra_dims: 56 | x = x.reshape(-1, *x.shape[extra_dims+1:]) 57 | assert x.shape == shape 58 | return x 59 | 60 | #---------------------------------------------------------------------------- 61 | -------------------------------------------------------------------------------- /metrics/kernel_inception_distance.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Kernel Inception Distance (KID) from the paper "Demystifying MMD 10 | GANs". Matches the original implementation by Binkowski et al. at 11 | https://github.com/mbinkowski/MMD-GAN/blob/master/gan/compute_scores.py""" 12 | 13 | import numpy as np 14 | from . import metric_utils 15 | 16 | #---------------------------------------------------------------------------- 17 | 18 | def compute_kid(opts, max_real, num_gen, num_subsets, max_subset_size): 19 | # Direct TorchScript translation of http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz 20 | detector_url = 'https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/metrics/inception-2015-12-05.pt' 21 | detector_kwargs = dict(return_features=True) # Return raw features before the softmax layer. 22 | 23 | real_features = metric_utils.compute_feature_stats_for_dataset( 24 | opts=opts, detector_url=detector_url, detector_kwargs=detector_kwargs, 25 | rel_lo=0, rel_hi=0, capture_all=True, max_items=max_real).get_all() 26 | 27 | gen_features = metric_utils.compute_feature_stats_for_generator( 28 | opts=opts, detector_url=detector_url, detector_kwargs=detector_kwargs, 29 | rel_lo=0, rel_hi=1, capture_all=True, max_items=num_gen).get_all() 30 | 31 | if opts.rank != 0: 32 | return float('nan') 33 | 34 | n = real_features.shape[1] 35 | m = min(min(real_features.shape[0], gen_features.shape[0]), max_subset_size) 36 | t = 0 37 | for _subset_idx in range(num_subsets): 38 | x = gen_features[np.random.choice(gen_features.shape[0], m, replace=False)] 39 | y = real_features[np.random.choice(real_features.shape[0], m, replace=False)] 40 | a = (x @ x.T / n + 1) ** 3 + (y @ y.T / n + 1) ** 3 41 | b = (x @ y.T / n + 1) ** 3 42 | t += (a.sum() - np.diag(a).sum()) / (m - 1) - b.sum() * 2 / m 43 | kid = t / num_subsets / m 44 | return float(kid) 45 | 46 | #---------------------------------------------------------------------------- 47 | -------------------------------------------------------------------------------- /docs/dataset-tool-help.txt: -------------------------------------------------------------------------------- 1 | Usage: dataset_tool.py [OPTIONS] 2 | 3 | Convert an image dataset into a dataset archive usable with StyleGAN2 ADA 4 | PyTorch. 5 | 6 | The input dataset format is guessed from the --source argument: 7 | 8 | --source *_lmdb/ - Load LSUN dataset 9 | --source cifar-10-python.tar.gz - Load CIFAR-10 dataset 10 | --source path/ - Recursively load all images from path/ 11 | --source dataset.zip - Recursively load all images from dataset.zip 12 | 13 | The output dataset format can be either an image folder or a zip archive. 14 | Specifying the output format and path: 15 | 16 | --dest /path/to/dir - Save output files under /path/to/dir 17 | --dest /path/to/dataset.zip - Save output files into /path/to/dataset.zip archive 18 | 19 | Images within the dataset archive will be stored as uncompressed PNG. 20 | 21 | Image scale/crop and resolution requirements: 22 | 23 | Output images must be square-shaped and they must all have the same power- 24 | of-two dimensions. 25 | 26 | To scale arbitrary input image size to a specific width and height, use 27 | the --width and --height options. Output resolution will be either the 28 | original input resolution (if --width/--height was not specified) or the 29 | one specified with --width/height. 30 | 31 | Use the --transform=center-crop or --transform=center-crop-wide options to 32 | apply a center crop transform on the input image. These options should be 33 | used with the --width and --height options. For example: 34 | 35 | python dataset_tool.py --source LSUN/raw/cat_lmdb --dest /tmp/lsun_cat \ 36 | --transform=center-crop-wide --width 512 --height=384 37 | 38 | Options: 39 | --source PATH Directory or archive name for input dataset 40 | [required] 41 | --dest PATH Output directory or archive name for output 42 | dataset [required] 43 | --max-images INTEGER Output only up to `max-images` images 44 | --resize-filter [box|lanczos] Filter to use when resizing images for 45 | output resolution [default: lanczos] 46 | --transform [center-crop|center-crop-wide] 47 | Input crop/resize mode 48 | --width INTEGER Output width 49 | --height INTEGER Output height 50 | --help Show this message and exit. 51 | -------------------------------------------------------------------------------- /pytorch_ssim/__init__.py: -------------------------------------------------------------------------------- 1 | # Code from Evan Su/Po-Hsun-Su: https://github.com/Po-Hsun-Su/pytorch-ssim 2 | 3 | import torch 4 | import torch.nn.functional as F 5 | from torch.autograd import Variable 6 | from math import exp 7 | 8 | 9 | def gaussian(window_size, sigma): 10 | gauss = torch.Tensor([exp(-(x - window_size//2)**2/float(2*sigma**2)) for x in range(window_size)]) 11 | return gauss/gauss.sum() 12 | 13 | 14 | def create_window(window_size, channel): 15 | _1D_window = gaussian(window_size, 1.5).unsqueeze(1) 16 | _2D_window = _1D_window.mm(_1D_window.t()).float().unsqueeze(0).unsqueeze(0) 17 | window = Variable(_2D_window.expand(channel, 1, window_size, window_size).contiguous()) 18 | return window 19 | 20 | 21 | def _ssim(img1, img2, window, window_size, channel, size_average=True): 22 | mu1 = F.conv2d(img1, window, padding=window_size//2, groups=channel) 23 | mu2 = F.conv2d(img2, window, padding=window_size//2, groups=channel) 24 | 25 | mu1_sq = mu1.pow(2) 26 | mu2_sq = mu2.pow(2) 27 | mu1_mu2 = mu1*mu2 28 | 29 | sigma1_sq = F.conv2d(img1 * img1, window, padding=window_size//2, groups=channel) - mu1_sq 30 | sigma2_sq = F.conv2d(img2 * img2, window, padding=window_size//2, groups=channel) - mu2_sq 31 | sigma12 = F.conv2d(img1 * img2, window, padding=window_size//2, groups=channel) - mu1_mu2 32 | 33 | C1 = 0.01**2 34 | C2 = 0.03**2 35 | 36 | ssim_map = ((2*mu1_mu2 + C1)*(2*sigma12 + C2))/((mu1_sq + mu2_sq + C1)*(sigma1_sq + sigma2_sq + C2)) 37 | 38 | if size_average: 39 | return ssim_map.mean() 40 | else: 41 | return ssim_map.mean(1).mean(1).mean(1) 42 | 43 | 44 | class SSIM(torch.nn.Module): 45 | def __init__(self, window_size = 11, size_average = True): 46 | super(SSIM, self).__init__() 47 | self.window_size = window_size 48 | self.size_average = size_average 49 | self.channel = 1 50 | self.window = create_window(window_size, self.channel) 51 | 52 | def forward(self, img1, img2): 53 | (_, channel, _, _) = img1.size() 54 | 55 | if channel == self.channel and self.window.data.type() == img1.data.type(): 56 | window = self.window 57 | else: 58 | window = create_window(self.window_size, channel) 59 | 60 | if img1.is_cuda: 61 | window = window.cuda(img1.get_device()) 62 | window = window.type_as(img1) 63 | 64 | self.window = window 65 | self.channel = channel 66 | 67 | return _ssim(img1, img2, window, self.window_size, channel, self.size_average) 68 | 69 | 70 | def ssim(img1, img2, window_size=11, size_average=True): 71 | (_, channel, _, _) = img1.size() 72 | window = create_window(window_size, channel) 73 | 74 | if img1.is_cuda: 75 | window = window.cuda(img1.get_device()) 76 | window = window.type_as(img1) 77 | 78 | return _ssim(img1, img2, window, window_size, channel, size_average) 79 | -------------------------------------------------------------------------------- /torch_utils/ops/grid_sample_gradfix.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Custom replacement for `torch.nn.functional.grid_sample` that 10 | supports arbitrarily high order gradients between the input and output. 11 | Only works on 2D images and assumes 12 | `mode='bilinear'`, `padding_mode='zeros'`, `align_corners=False`.""" 13 | 14 | import warnings 15 | import torch 16 | 17 | # pylint: disable=redefined-builtin 18 | # pylint: disable=arguments-differ 19 | # pylint: disable=protected-access 20 | 21 | #---------------------------------------------------------------------------- 22 | 23 | enabled = False # Enable the custom op by setting this to true. 24 | 25 | #---------------------------------------------------------------------------- 26 | 27 | def grid_sample(input, grid): 28 | if _should_use_custom_op(): 29 | return _GridSample2dForward.apply(input, grid) 30 | return torch.nn.functional.grid_sample(input=input, grid=grid, mode='bilinear', padding_mode='zeros', align_corners=False) 31 | 32 | #---------------------------------------------------------------------------- 33 | 34 | def _should_use_custom_op(): 35 | if not enabled: 36 | return False 37 | if any(torch.__version__.startswith(x) for x in ['1.7.', '1.8.', '1.9']): 38 | return True 39 | warnings.warn(f'grid_sample_gradfix not supported on PyTorch {torch.__version__}. Falling back to torch.nn.functional.grid_sample().') 40 | return False 41 | 42 | #---------------------------------------------------------------------------- 43 | 44 | class _GridSample2dForward(torch.autograd.Function): 45 | @staticmethod 46 | def forward(ctx, input, grid): 47 | assert input.ndim == 4 48 | assert grid.ndim == 4 49 | output = torch.nn.functional.grid_sample(input=input, grid=grid, mode='bilinear', padding_mode='zeros', align_corners=False) 50 | ctx.save_for_backward(input, grid) 51 | return output 52 | 53 | @staticmethod 54 | def backward(ctx, grad_output): 55 | input, grid = ctx.saved_tensors 56 | grad_input, grad_grid = _GridSample2dBackward.apply(grad_output, input, grid) 57 | return grad_input, grad_grid 58 | 59 | #---------------------------------------------------------------------------- 60 | 61 | class _GridSample2dBackward(torch.autograd.Function): 62 | @staticmethod 63 | def forward(ctx, grad_output, input, grid): 64 | op = torch._C._jit_get_operation('aten::grid_sampler_2d_backward') 65 | grad_input, grad_grid = op(grad_output, input, grid, 0, 0, False) 66 | ctx.save_for_backward(grid) 67 | return grad_input, grad_grid 68 | 69 | @staticmethod 70 | def backward(ctx, grad2_grad_input, grad2_grad_grid): 71 | _ = grad2_grad_grid # unused 72 | grid, = ctx.saved_tensors 73 | grad2_grad_output = None 74 | grad2_input = None 75 | grad2_grid = None 76 | 77 | if ctx.needs_input_grad[0]: 78 | grad2_grad_output = _GridSample2dForward.apply(grad2_grad_input, grid) 79 | 80 | assert not ctx.needs_input_grad[2] 81 | return grad2_grad_output, grad2_input, grad2_grid 82 | 83 | #---------------------------------------------------------------------------- 84 | -------------------------------------------------------------------------------- /metrics/precision_recall.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Precision/Recall (PR) from the paper "Improved Precision and Recall 10 | Metric for Assessing Generative Models". Matches the original implementation 11 | by Kynkaanniemi et al. at 12 | https://github.com/kynkaat/improved-precision-and-recall-metric/blob/master/precision_recall.py""" 13 | 14 | import torch 15 | from . import metric_utils 16 | 17 | #---------------------------------------------------------------------------- 18 | 19 | def compute_distances(row_features, col_features, num_gpus, rank, col_batch_size): 20 | assert 0 <= rank < num_gpus 21 | num_cols = col_features.shape[0] 22 | num_batches = ((num_cols - 1) // col_batch_size // num_gpus + 1) * num_gpus 23 | col_batches = torch.nn.functional.pad(col_features, [0, 0, 0, -num_cols % num_batches]).chunk(num_batches) 24 | dist_batches = [] 25 | for col_batch in col_batches[rank :: num_gpus]: 26 | dist_batch = torch.cdist(row_features.unsqueeze(0), col_batch.unsqueeze(0))[0] 27 | for src in range(num_gpus): 28 | dist_broadcast = dist_batch.clone() 29 | if num_gpus > 1: 30 | torch.distributed.broadcast(dist_broadcast, src=src) 31 | dist_batches.append(dist_broadcast.cpu() if rank == 0 else None) 32 | return torch.cat(dist_batches, dim=1)[:, :num_cols] if rank == 0 else None 33 | 34 | #---------------------------------------------------------------------------- 35 | 36 | def compute_pr(opts, max_real, num_gen, nhood_size, row_batch_size, col_batch_size): 37 | detector_url = 'https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/metrics/vgg16.pt' 38 | detector_kwargs = dict(return_features=True) 39 | 40 | real_features = metric_utils.compute_feature_stats_for_dataset( 41 | opts=opts, detector_url=detector_url, detector_kwargs=detector_kwargs, 42 | rel_lo=0, rel_hi=0, capture_all=True, max_items=max_real).get_all_torch().to(torch.float16).to(opts.device) 43 | 44 | gen_features = metric_utils.compute_feature_stats_for_generator( 45 | opts=opts, detector_url=detector_url, detector_kwargs=detector_kwargs, 46 | rel_lo=0, rel_hi=1, capture_all=True, max_items=num_gen).get_all_torch().to(torch.float16).to(opts.device) 47 | 48 | results = dict() 49 | for name, manifold, probes in [('precision', real_features, gen_features), ('recall', gen_features, real_features)]: 50 | kth = [] 51 | for manifold_batch in manifold.split(row_batch_size): 52 | dist = compute_distances(row_features=manifold_batch, col_features=manifold, num_gpus=opts.num_gpus, rank=opts.rank, col_batch_size=col_batch_size) 53 | kth.append(dist.to(torch.float32).kthvalue(nhood_size + 1).values.to(torch.float16) if opts.rank == 0 else None) 54 | kth = torch.cat(kth) if opts.rank == 0 else None 55 | pred = [] 56 | for probes_batch in probes.split(row_batch_size): 57 | dist = compute_distances(row_features=probes_batch, col_features=manifold, num_gpus=opts.num_gpus, rank=opts.rank, col_batch_size=col_batch_size) 58 | pred.append((dist <= kth).any(dim=1) if opts.rank == 0 else None) 59 | results[name] = float(torch.cat(pred).to(torch.float32).mean() if opts.rank == 0 else 'nan') 60 | return results['precision'], results['recall'] 61 | 62 | #---------------------------------------------------------------------------- 63 | -------------------------------------------------------------------------------- /docs/train-help.txt: -------------------------------------------------------------------------------- 1 | Usage: train.py [OPTIONS] 2 | 3 | Train a GAN using the techniques described in the paper "Training 4 | Generative Adversarial Networks with Limited Data". 5 | 6 | Examples: 7 | 8 | # Train with custom images using 1 GPU. 9 | python train.py --outdir=~/training-runs --data=~/my-image-folder 10 | 11 | # Train class-conditional CIFAR-10 using 2 GPUs. 12 | python train.py --outdir=~/training-runs --data=~/datasets/cifar10.zip \ 13 | --gpus=2 --cfg=cifar --cond=1 14 | 15 | # Transfer learn MetFaces from FFHQ using 4 GPUs. 16 | python train.py --outdir=~/training-runs --data=~/datasets/metfaces.zip \ 17 | --gpus=4 --cfg=paper1024 --mirror=1 --resume=ffhq1024 --snap=10 18 | 19 | # Reproduce original StyleGAN2 config F. 20 | python train.py --outdir=~/training-runs --data=~/datasets/ffhq.zip \ 21 | --gpus=8 --cfg=stylegan2 --mirror=1 --aug=noaug 22 | 23 | Base configs (--cfg): 24 | auto Automatically select reasonable defaults based on resolution 25 | and GPU count. Good starting point for new datasets. 26 | stylegan2 Reproduce results for StyleGAN2 config F at 1024x1024. 27 | paper256 Reproduce results for FFHQ and LSUN Cat at 256x256. 28 | paper512 Reproduce results for BreCaHAD and AFHQ at 512x512. 29 | paper1024 Reproduce results for MetFaces at 1024x1024. 30 | cifar Reproduce results for CIFAR-10 at 32x32. 31 | 32 | Transfer learning source networks (--resume): 33 | ffhq256 FFHQ trained at 256x256 resolution. 34 | ffhq512 FFHQ trained at 512x512 resolution. 35 | ffhq1024 FFHQ trained at 1024x1024 resolution. 36 | celebahq256 CelebA-HQ trained at 256x256 resolution. 37 | lsundog256 LSUN Dog trained at 256x256 resolution. 38 | Custom network pickle. 39 | 40 | Options: 41 | --outdir DIR Where to save the results [required] 42 | --gpus INT Number of GPUs to use [default: 1] 43 | --snap INT Snapshot interval [default: 50 ticks] 44 | --metrics LIST Comma-separated list or "none" [default: 45 | fid50k_full] 46 | --seed INT Random seed [default: 0] 47 | -n, --dry-run Print training options and exit 48 | --data PATH Training data (directory or zip) [required] 49 | --cond BOOL Train conditional model based on dataset 50 | labels [default: false] 51 | --subset INT Train with only N images [default: all] 52 | --mirror BOOL Enable dataset x-flips [default: false] 53 | --cfg [auto|stylegan2|paper256|paper512|paper1024|cifar] 54 | Base config [default: auto] 55 | --gamma FLOAT Override R1 gamma 56 | --kimg INT Override training duration 57 | --batch INT Override batch size 58 | --aug [noaug|ada|fixed] Augmentation mode [default: ada] 59 | --p FLOAT Augmentation probability for --aug=fixed 60 | --target FLOAT ADA target value for --aug=ada 61 | --augpipe [blit|geom|color|filter|noise|cutout|bg|bgc|bgcf|bgcfn|bgcfnc] 62 | Augmentation pipeline [default: bgc] 63 | --resume PKL Resume training [default: noresume] 64 | --freezed INT Freeze-D [default: 0 layers] 65 | --fp32 BOOL Disable mixed-precision training 66 | --nhwc BOOL Use NHWC memory format with FP16 67 | --nobench BOOL Disable cuDNN benchmarking 68 | --allow-tf32 BOOL Allow PyTorch to use TF32 internally 69 | --workers INT Override number of DataLoader workers 70 | --help Show this message and exit. 71 | -------------------------------------------------------------------------------- /LICENSE.txt: -------------------------------------------------------------------------------- 1 | Copyright (c) 2021, NVIDIA Corporation. All rights reserved. 2 | 3 | 4 | NVIDIA Source Code License for StyleGAN2 with Adaptive Discriminator Augmentation (ADA) 5 | 6 | 7 | ======================================================================= 8 | 9 | 1. Definitions 10 | 11 | "Licensor" means any person or entity that distributes its Work. 12 | 13 | "Software" means the original work of authorship made available under 14 | this License. 15 | 16 | "Work" means the Software and any additions to or derivative works of 17 | the Software that are made available under this License. 18 | 19 | The terms "reproduce," "reproduction," "derivative works," and 20 | "distribution" have the meaning as provided under U.S. copyright law; 21 | provided, however, that for the purposes of this License, derivative 22 | works shall not include works that remain separable from, or merely 23 | link (or bind by name) to the interfaces of, the Work. 24 | 25 | Works, including the Software, are "made available" under this License 26 | by including in or with the Work either (a) a copyright notice 27 | referencing the applicability of this License to the Work, or (b) a 28 | copy of this License. 29 | 30 | 2. License Grants 31 | 32 | 2.1 Copyright Grant. Subject to the terms and conditions of this 33 | License, each Licensor grants to you a perpetual, worldwide, 34 | non-exclusive, royalty-free, copyright license to reproduce, 35 | prepare derivative works of, publicly display, publicly perform, 36 | sublicense and distribute its Work and any resulting derivative 37 | works in any form. 38 | 39 | 3. Limitations 40 | 41 | 3.1 Redistribution. You may reproduce or distribute the Work only 42 | if (a) you do so under this License, (b) you include a complete 43 | copy of this License with your distribution, and (c) you retain 44 | without modification any copyright, patent, trademark, or 45 | attribution notices that are present in the Work. 46 | 47 | 3.2 Derivative Works. You may specify that additional or different 48 | terms apply to the use, reproduction, and distribution of your 49 | derivative works of the Work ("Your Terms") only if (a) Your Terms 50 | provide that the use limitation in Section 3.3 applies to your 51 | derivative works, and (b) you identify the specific derivative 52 | works that are subject to Your Terms. Notwithstanding Your Terms, 53 | this License (including the redistribution requirements in Section 54 | 3.1) will continue to apply to the Work itself. 55 | 56 | 3.3 Use Limitation. The Work and any derivative works thereof only 57 | may be used or intended for use non-commercially. Notwithstanding 58 | the foregoing, NVIDIA and its affiliates may use the Work and any 59 | derivative works commercially. As used herein, "non-commercially" 60 | means for research or evaluation purposes only. 61 | 62 | 3.4 Patent Claims. If you bring or threaten to bring a patent claim 63 | against any Licensor (including any claim, cross-claim or 64 | counterclaim in a lawsuit) to enforce any patents that you allege 65 | are infringed by any Work, then your rights under this License from 66 | such Licensor (including the grant in Section 2.1) will terminate 67 | immediately. 68 | 69 | 3.5 Trademarks. This License does not grant any rights to use any 70 | Licensor’s or its affiliates’ names, logos, or trademarks, except 71 | as necessary to reproduce the notices described in this License. 72 | 73 | 3.6 Termination. If you violate any term of this License, then your 74 | rights under this License (including the grant in Section 2.1) will 75 | terminate immediately. 76 | 77 | 4. Disclaimer of Warranty. 78 | 79 | THE WORK IS PROVIDED "AS IS" WITHOUT WARRANTIES OR CONDITIONS OF ANY 80 | KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WARRANTIES OR CONDITIONS OF 81 | MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE OR 82 | NON-INFRINGEMENT. YOU BEAR THE RISK OF UNDERTAKING ANY ACTIVITIES UNDER 83 | THIS LICENSE. 84 | 85 | 5. Limitation of Liability. 86 | 87 | EXCEPT AS PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL 88 | THEORY, WHETHER IN TORT (INCLUDING NEGLIGENCE), CONTRACT, OR OTHERWISE 89 | SHALL ANY LICENSOR BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY DIRECT, 90 | INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF 91 | OR RELATED TO THIS LICENSE, THE USE OR INABILITY TO USE THE WORK 92 | (INCLUDING BUT NOT LIMITED TO LOSS OF GOODWILL, BUSINESS INTERRUPTION, 93 | LOST PROFITS OR DATA, COMPUTER FAILURE OR MALFUNCTION, OR ANY OTHER 94 | COMMERCIAL DAMAGES OR LOSSES), EVEN IF THE LICENSOR HAS BEEN ADVISED OF 95 | THE POSSIBILITY OF SUCH DAMAGES. 96 | 97 | ======================================================================= 98 | -------------------------------------------------------------------------------- /torch_utils/ops/bias_act.cpp: -------------------------------------------------------------------------------- 1 | // Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | // 3 | // NVIDIA CORPORATION and its licensors retain all intellectual property 4 | // and proprietary rights in and to this software, related documentation 5 | // and any modifications thereto. Any use, reproduction, disclosure or 6 | // distribution of this software and related documentation without an express 7 | // license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | #include 10 | #include 11 | #include 12 | #include "bias_act.h" 13 | 14 | //------------------------------------------------------------------------ 15 | 16 | static bool has_same_layout(torch::Tensor x, torch::Tensor y) 17 | { 18 | if (x.dim() != y.dim()) 19 | return false; 20 | for (int64_t i = 0; i < x.dim(); i++) 21 | { 22 | if (x.size(i) != y.size(i)) 23 | return false; 24 | if (x.size(i) >= 2 && x.stride(i) != y.stride(i)) 25 | return false; 26 | } 27 | return true; 28 | } 29 | 30 | //------------------------------------------------------------------------ 31 | 32 | static torch::Tensor bias_act(torch::Tensor x, torch::Tensor b, torch::Tensor xref, torch::Tensor yref, torch::Tensor dy, int grad, int dim, int act, float alpha, float gain, float clamp) 33 | { 34 | // Validate arguments. 35 | TORCH_CHECK(x.is_cuda(), "x must reside on CUDA device"); 36 | TORCH_CHECK(b.numel() == 0 || (b.dtype() == x.dtype() && b.device() == x.device()), "b must have the same dtype and device as x"); 37 | TORCH_CHECK(xref.numel() == 0 || (xref.sizes() == x.sizes() && xref.dtype() == x.dtype() && xref.device() == x.device()), "xref must have the same shape, dtype, and device as x"); 38 | TORCH_CHECK(yref.numel() == 0 || (yref.sizes() == x.sizes() && yref.dtype() == x.dtype() && yref.device() == x.device()), "yref must have the same shape, dtype, and device as x"); 39 | TORCH_CHECK(dy.numel() == 0 || (dy.sizes() == x.sizes() && dy.dtype() == x.dtype() && dy.device() == x.device()), "dy must have the same dtype and device as x"); 40 | TORCH_CHECK(x.numel() <= INT_MAX, "x is too large"); 41 | TORCH_CHECK(b.dim() == 1, "b must have rank 1"); 42 | TORCH_CHECK(b.numel() == 0 || (dim >= 0 && dim < x.dim()), "dim is out of bounds"); 43 | TORCH_CHECK(b.numel() == 0 || b.numel() == x.size(dim), "b has wrong number of elements"); 44 | TORCH_CHECK(grad >= 0, "grad must be non-negative"); 45 | 46 | // Validate layout. 47 | TORCH_CHECK(x.is_non_overlapping_and_dense(), "x must be non-overlapping and dense"); 48 | TORCH_CHECK(b.is_contiguous(), "b must be contiguous"); 49 | TORCH_CHECK(xref.numel() == 0 || has_same_layout(xref, x), "xref must have the same layout as x"); 50 | TORCH_CHECK(yref.numel() == 0 || has_same_layout(yref, x), "yref must have the same layout as x"); 51 | TORCH_CHECK(dy.numel() == 0 || has_same_layout(dy, x), "dy must have the same layout as x"); 52 | 53 | // Create output tensor. 54 | const at::cuda::OptionalCUDAGuard device_guard(device_of(x)); 55 | torch::Tensor y = torch::empty_like(x); 56 | TORCH_CHECK(has_same_layout(y, x), "y must have the same layout as x"); 57 | 58 | // Initialize CUDA kernel parameters. 59 | bias_act_kernel_params p; 60 | p.x = x.data_ptr(); 61 | p.b = (b.numel()) ? b.data_ptr() : NULL; 62 | p.xref = (xref.numel()) ? xref.data_ptr() : NULL; 63 | p.yref = (yref.numel()) ? yref.data_ptr() : NULL; 64 | p.dy = (dy.numel()) ? dy.data_ptr() : NULL; 65 | p.y = y.data_ptr(); 66 | p.grad = grad; 67 | p.act = act; 68 | p.alpha = alpha; 69 | p.gain = gain; 70 | p.clamp = clamp; 71 | p.sizeX = (int)x.numel(); 72 | p.sizeB = (int)b.numel(); 73 | p.stepB = (b.numel()) ? (int)x.stride(dim) : 1; 74 | 75 | // Choose CUDA kernel. 76 | void* kernel; 77 | AT_DISPATCH_FLOATING_TYPES_AND_HALF(x.scalar_type(), "upfirdn2d_cuda", [&] 78 | { 79 | kernel = choose_bias_act_kernel(p); 80 | }); 81 | TORCH_CHECK(kernel, "no CUDA kernel found for the specified activation func"); 82 | 83 | // Launch CUDA kernel. 84 | p.loopX = 4; 85 | int blockSize = 4 * 32; 86 | int gridSize = (p.sizeX - 1) / (p.loopX * blockSize) + 1; 87 | void* args[] = {&p}; 88 | AT_CUDA_CHECK(cudaLaunchKernel(kernel, gridSize, blockSize, args, 0, at::cuda::getCurrentCUDAStream())); 89 | return y; 90 | } 91 | 92 | //------------------------------------------------------------------------ 93 | 94 | PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) 95 | { 96 | m.def("bias_act", &bias_act); 97 | } 98 | 99 | //------------------------------------------------------------------------ 100 | -------------------------------------------------------------------------------- /torch_utils/ops/upfirdn2d.cpp: -------------------------------------------------------------------------------- 1 | // Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | // 3 | // NVIDIA CORPORATION and its licensors retain all intellectual property 4 | // and proprietary rights in and to this software, related documentation 5 | // and any modifications thereto. Any use, reproduction, disclosure or 6 | // distribution of this software and related documentation without an express 7 | // license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | #include 10 | #include 11 | #include 12 | #include "upfirdn2d.h" 13 | 14 | //------------------------------------------------------------------------ 15 | 16 | static torch::Tensor upfirdn2d(torch::Tensor x, torch::Tensor f, int upx, int upy, int downx, int downy, int padx0, int padx1, int pady0, int pady1, bool flip, float gain) 17 | { 18 | // Validate arguments. 19 | TORCH_CHECK(x.is_cuda(), "x must reside on CUDA device"); 20 | TORCH_CHECK(f.device() == x.device(), "f must reside on the same device as x"); 21 | TORCH_CHECK(f.dtype() == torch::kFloat, "f must be float32"); 22 | TORCH_CHECK(x.numel() <= INT_MAX, "x is too large"); 23 | TORCH_CHECK(f.numel() <= INT_MAX, "f is too large"); 24 | TORCH_CHECK(x.dim() == 4, "x must be rank 4"); 25 | TORCH_CHECK(f.dim() == 2, "f must be rank 2"); 26 | TORCH_CHECK(f.size(0) >= 1 && f.size(1) >= 1, "f must be at least 1x1"); 27 | TORCH_CHECK(upx >= 1 && upy >= 1, "upsampling factor must be at least 1"); 28 | TORCH_CHECK(downx >= 1 && downy >= 1, "downsampling factor must be at least 1"); 29 | 30 | // Create output tensor. 31 | const at::cuda::OptionalCUDAGuard device_guard(device_of(x)); 32 | int outW = ((int)x.size(3) * upx + padx0 + padx1 - (int)f.size(1) + downx) / downx; 33 | int outH = ((int)x.size(2) * upy + pady0 + pady1 - (int)f.size(0) + downy) / downy; 34 | TORCH_CHECK(outW >= 1 && outH >= 1, "output must be at least 1x1"); 35 | torch::Tensor y = torch::empty({x.size(0), x.size(1), outH, outW}, x.options(), x.suggest_memory_format()); 36 | TORCH_CHECK(y.numel() <= INT_MAX, "output is too large"); 37 | 38 | // Initialize CUDA kernel parameters. 39 | upfirdn2d_kernel_params p; 40 | p.x = x.data_ptr(); 41 | p.f = f.data_ptr(); 42 | p.y = y.data_ptr(); 43 | p.up = make_int2(upx, upy); 44 | p.down = make_int2(downx, downy); 45 | p.pad0 = make_int2(padx0, pady0); 46 | p.flip = (flip) ? 1 : 0; 47 | p.gain = gain; 48 | p.inSize = make_int4((int)x.size(3), (int)x.size(2), (int)x.size(1), (int)x.size(0)); 49 | p.inStride = make_int4((int)x.stride(3), (int)x.stride(2), (int)x.stride(1), (int)x.stride(0)); 50 | p.filterSize = make_int2((int)f.size(1), (int)f.size(0)); 51 | p.filterStride = make_int2((int)f.stride(1), (int)f.stride(0)); 52 | p.outSize = make_int4((int)y.size(3), (int)y.size(2), (int)y.size(1), (int)y.size(0)); 53 | p.outStride = make_int4((int)y.stride(3), (int)y.stride(2), (int)y.stride(1), (int)y.stride(0)); 54 | p.sizeMajor = (p.inStride.z == 1) ? p.inSize.w : p.inSize.w * p.inSize.z; 55 | p.sizeMinor = (p.inStride.z == 1) ? p.inSize.z : 1; 56 | 57 | // Choose CUDA kernel. 58 | upfirdn2d_kernel_spec spec; 59 | AT_DISPATCH_FLOATING_TYPES_AND_HALF(x.scalar_type(), "upfirdn2d_cuda", [&] 60 | { 61 | spec = choose_upfirdn2d_kernel(p); 62 | }); 63 | 64 | // Set looping options. 65 | p.loopMajor = (p.sizeMajor - 1) / 16384 + 1; 66 | p.loopMinor = spec.loopMinor; 67 | p.loopX = spec.loopX; 68 | p.launchMinor = (p.sizeMinor - 1) / p.loopMinor + 1; 69 | p.launchMajor = (p.sizeMajor - 1) / p.loopMajor + 1; 70 | 71 | // Compute grid size. 72 | dim3 blockSize, gridSize; 73 | if (spec.tileOutW < 0) // large 74 | { 75 | blockSize = dim3(4, 32, 1); 76 | gridSize = dim3( 77 | ((p.outSize.y - 1) / blockSize.x + 1) * p.launchMinor, 78 | (p.outSize.x - 1) / (blockSize.y * p.loopX) + 1, 79 | p.launchMajor); 80 | } 81 | else // small 82 | { 83 | blockSize = dim3(256, 1, 1); 84 | gridSize = dim3( 85 | ((p.outSize.y - 1) / spec.tileOutH + 1) * p.launchMinor, 86 | (p.outSize.x - 1) / (spec.tileOutW * p.loopX) + 1, 87 | p.launchMajor); 88 | } 89 | 90 | // Launch CUDA kernel. 91 | void* args[] = {&p}; 92 | AT_CUDA_CHECK(cudaLaunchKernel(spec.kernel, gridSize, blockSize, args, 0, at::cuda::getCurrentCUDAStream())); 93 | return y; 94 | } 95 | 96 | //------------------------------------------------------------------------ 97 | 98 | PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) 99 | { 100 | m.def("upfirdn2d", &upfirdn2d); 101 | } 102 | 103 | //------------------------------------------------------------------------ 104 | -------------------------------------------------------------------------------- /metrics/perceptual_path_length.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Perceptual Path Length (PPL) from the paper "A Style-Based Generator 10 | Architecture for Generative Adversarial Networks". Matches the original 11 | implementation by Karras et al. at 12 | https://github.com/NVlabs/stylegan/blob/master/metrics/perceptual_path_length.py""" 13 | 14 | import copy 15 | import numpy as np 16 | import torch 17 | import dnnlib 18 | from . import metric_utils 19 | 20 | #---------------------------------------------------------------------------- 21 | 22 | # Spherical interpolation of a batch of vectors. 23 | def slerp(a, b, t): 24 | a = a / a.norm(dim=-1, keepdim=True) 25 | b = b / b.norm(dim=-1, keepdim=True) 26 | d = (a * b).sum(dim=-1, keepdim=True) 27 | p = t * torch.acos(d) 28 | c = b - d * a 29 | c = c / c.norm(dim=-1, keepdim=True) 30 | d = a * torch.cos(p) + c * torch.sin(p) 31 | d = d / d.norm(dim=-1, keepdim=True) 32 | return d 33 | 34 | #---------------------------------------------------------------------------- 35 | 36 | class PPLSampler(torch.nn.Module): 37 | def __init__(self, G, G_kwargs, epsilon, space, sampling, crop, vgg16): 38 | assert space in ['z', 'w'] 39 | assert sampling in ['full', 'end'] 40 | super().__init__() 41 | self.G = copy.deepcopy(G) 42 | self.G_kwargs = G_kwargs 43 | self.epsilon = epsilon 44 | self.space = space 45 | self.sampling = sampling 46 | self.crop = crop 47 | self.vgg16 = copy.deepcopy(vgg16) 48 | 49 | def forward(self, c): 50 | # Generate random latents and interpolation t-values. 51 | t = torch.rand([c.shape[0]], device=c.device) * (1 if self.sampling == 'full' else 0) 52 | z0, z1 = torch.randn([c.shape[0] * 2, self.G.z_dim], device=c.device).chunk(2) 53 | 54 | # Interpolate in W or Z. 55 | if self.space == 'w': 56 | w0, w1 = self.G.mapping(z=torch.cat([z0,z1]), c=torch.cat([c,c])).chunk(2) 57 | wt0 = w0.lerp(w1, t.unsqueeze(1).unsqueeze(2)) 58 | wt1 = w0.lerp(w1, t.unsqueeze(1).unsqueeze(2) + self.epsilon) 59 | else: # space == 'z' 60 | zt0 = slerp(z0, z1, t.unsqueeze(1)) 61 | zt1 = slerp(z0, z1, t.unsqueeze(1) + self.epsilon) 62 | wt0, wt1 = self.G.mapping(z=torch.cat([zt0,zt1]), c=torch.cat([c,c])).chunk(2) 63 | 64 | # Randomize noise buffers. 65 | for name, buf in self.G.named_buffers(): 66 | if name.endswith('.noise_const'): 67 | buf.copy_(torch.randn_like(buf)) 68 | 69 | # Generate images. 70 | img = self.G.synthesis(ws=torch.cat([wt0,wt1]), noise_mode='const', force_fp32=True, **self.G_kwargs) 71 | 72 | # Center crop. 73 | if self.crop: 74 | assert img.shape[2] == img.shape[3] 75 | c = img.shape[2] // 8 76 | img = img[:, :, c*3 : c*7, c*2 : c*6] 77 | 78 | # Downsample to 256x256. 79 | factor = self.G.img_resolution // 256 80 | if factor > 1: 81 | img = img.reshape([-1, img.shape[1], img.shape[2] // factor, factor, img.shape[3] // factor, factor]).mean([3, 5]) 82 | 83 | # Scale dynamic range from [-1,1] to [0,255]. 84 | img = (img + 1) * (255 / 2) 85 | if self.G.img_channels == 1: 86 | img = img.repeat([1, 3, 1, 1]) 87 | 88 | # Evaluate differential LPIPS. 89 | lpips_t0, lpips_t1 = self.vgg16(img, resize_images=False, return_lpips=True).chunk(2) 90 | dist = (lpips_t0 - lpips_t1).square().sum(1) / self.epsilon ** 2 91 | return dist 92 | 93 | #---------------------------------------------------------------------------- 94 | 95 | def compute_ppl(opts, num_samples, epsilon, space, sampling, crop, batch_size, jit=False): 96 | dataset = dnnlib.util.construct_class_by_name(**opts.dataset_kwargs) 97 | vgg16_url = 'https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/metrics/vgg16.pt' 98 | vgg16 = metric_utils.get_feature_detector(vgg16_url, num_gpus=opts.num_gpus, rank=opts.rank, verbose=opts.progress.verbose) 99 | 100 | # Setup sampler. 101 | sampler = PPLSampler(G=opts.G, G_kwargs=opts.G_kwargs, epsilon=epsilon, space=space, sampling=sampling, crop=crop, vgg16=vgg16) 102 | sampler.eval().requires_grad_(False).to(opts.device) 103 | if jit: 104 | c = torch.zeros([batch_size, opts.G.c_dim], device=opts.device) 105 | sampler = torch.jit.trace(sampler, [c], check_trace=False) 106 | 107 | # Sampling loop. 108 | dist = [] 109 | progress = opts.progress.sub(tag='ppl sampling', num_items=num_samples) 110 | for batch_start in range(0, num_samples, batch_size * opts.num_gpus): 111 | progress.update(batch_start) 112 | c = [dataset.get_label(np.random.randint(len(dataset))) for _i in range(batch_size)] 113 | c = torch.from_numpy(np.stack(c)).pin_memory().to(opts.device) 114 | x = sampler(c) 115 | for src in range(opts.num_gpus): 116 | y = x.clone() 117 | if opts.num_gpus > 1: 118 | torch.distributed.broadcast(y, src=src) 119 | dist.append(y) 120 | progress.update(num_samples) 121 | 122 | # Compute PPL. 123 | if opts.rank != 0: 124 | return float('nan') 125 | dist = torch.cat(dist)[:num_samples].cpu().numpy() 126 | lo = np.percentile(dist, 1, interpolation='lower') 127 | hi = np.percentile(dist, 99, interpolation='higher') 128 | ppl = np.extract(np.logical_and(dist >= lo, dist <= hi), dist).mean() 129 | return float(ppl) 130 | 131 | #---------------------------------------------------------------------------- 132 | -------------------------------------------------------------------------------- /torch_utils/custom_ops.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | import os 10 | import glob 11 | import torch 12 | import torch.utils.cpp_extension 13 | import importlib 14 | import hashlib 15 | import shutil 16 | from pathlib import Path 17 | 18 | from torch.utils.file_baton import FileBaton 19 | 20 | #---------------------------------------------------------------------------- 21 | # Global options. 22 | 23 | verbosity = 'brief' # Verbosity level: 'none', 'brief', 'full' 24 | 25 | #---------------------------------------------------------------------------- 26 | # Internal helper funcs. 27 | 28 | def _find_compiler_bindir(): 29 | patterns = [ 30 | 'C:/Program Files (x86)/Microsoft Visual Studio/*/Professional/VC/Tools/MSVC/*/bin/Hostx64/x64', 31 | 'C:/Program Files (x86)/Microsoft Visual Studio/*/BuildTools/VC/Tools/MSVC/*/bin/Hostx64/x64', 32 | 'C:/Program Files (x86)/Microsoft Visual Studio/*/Community/VC/Tools/MSVC/*/bin/Hostx64/x64', 33 | 'C:/Program Files (x86)/Microsoft Visual Studio */vc/bin', 34 | ] 35 | for pattern in patterns: 36 | matches = sorted(glob.glob(pattern)) 37 | if len(matches): 38 | return matches[-1] 39 | return None 40 | 41 | #---------------------------------------------------------------------------- 42 | # Main entry point for compiling and loading C++/CUDA plugins. 43 | 44 | _cached_plugins = dict() 45 | 46 | def get_plugin(module_name, sources, **build_kwargs): 47 | assert verbosity in ['none', 'brief', 'full'] 48 | 49 | # Already cached? 50 | if module_name in _cached_plugins: 51 | return _cached_plugins[module_name] 52 | 53 | # Print status. 54 | if verbosity == 'full': 55 | print(f'Setting up PyTorch plugin "{module_name}"...') 56 | elif verbosity == 'brief': 57 | print(f'Setting up PyTorch plugin "{module_name}"... ', end='', flush=True) 58 | 59 | try: # pylint: disable=too-many-nested-blocks 60 | # Make sure we can find the necessary compiler binaries. 61 | if os.name == 'nt' and os.system("where cl.exe >nul 2>nul") != 0: 62 | compiler_bindir = _find_compiler_bindir() 63 | if compiler_bindir is None: 64 | raise RuntimeError(f'Could not find MSVC/GCC/CLANG installation on this computer. Check _find_compiler_bindir() in "{__file__}".') 65 | os.environ['PATH'] += ';' + compiler_bindir 66 | 67 | # Compile and load. 68 | verbose_build = (verbosity == 'full') 69 | 70 | # Incremental build md5sum trickery. Copies all the input source files 71 | # into a cached build directory under a combined md5 digest of the input 72 | # source files. Copying is done only if the combined digest has changed. 73 | # This keeps input file timestamps and filenames the same as in previous 74 | # extension builds, allowing for fast incremental rebuilds. 75 | # 76 | # This optimization is done only in case all the source files reside in 77 | # a single directory (just for simplicity) and if the TORCH_EXTENSIONS_DIR 78 | # environment variable is set (we take this as a signal that the user 79 | # actually cares about this.) 80 | source_dirs_set = set(os.path.dirname(source) for source in sources) 81 | if len(source_dirs_set) == 1 and ('TORCH_EXTENSIONS_DIR' in os.environ): 82 | all_source_files = sorted(list(x for x in Path(list(source_dirs_set)[0]).iterdir() if x.is_file())) 83 | 84 | # Compute a combined hash digest for all source files in the same 85 | # custom op directory (usually .cu, .cpp, .py and .h files). 86 | hash_md5 = hashlib.md5() 87 | for src in all_source_files: 88 | with open(src, 'rb') as f: 89 | hash_md5.update(f.read()) 90 | build_dir = torch.utils.cpp_extension._get_build_directory(module_name, verbose=verbose_build) # pylint: disable=protected-access 91 | digest_build_dir = os.path.join(build_dir, hash_md5.hexdigest()) 92 | 93 | if not os.path.isdir(digest_build_dir): 94 | os.makedirs(digest_build_dir, exist_ok=True) 95 | baton = FileBaton(os.path.join(digest_build_dir, 'lock')) 96 | if baton.try_acquire(): 97 | try: 98 | for src in all_source_files: 99 | shutil.copyfile(src, os.path.join(digest_build_dir, os.path.basename(src))) 100 | finally: 101 | baton.release() 102 | else: 103 | # Someone else is copying source files under the digest dir, 104 | # wait until done and continue. 105 | baton.wait() 106 | digest_sources = [os.path.join(digest_build_dir, os.path.basename(x)) for x in sources] 107 | torch.utils.cpp_extension.load(name=module_name, build_directory=build_dir, 108 | verbose=verbose_build, sources=digest_sources, **build_kwargs) 109 | else: 110 | torch.utils.cpp_extension.load(name=module_name, verbose=verbose_build, sources=sources, **build_kwargs) 111 | module = importlib.import_module(module_name) 112 | 113 | except: 114 | if verbosity == 'brief': 115 | print('Failed!') 116 | raise 117 | 118 | # Print status and add to cache. 119 | if verbosity == 'full': 120 | print(f'Done setting up PyTorch plugin "{module_name}".') 121 | elif verbosity == 'brief': 122 | print('Done.') 123 | _cached_plugins[module_name] = module 124 | return module 125 | 126 | #---------------------------------------------------------------------------- 127 | -------------------------------------------------------------------------------- /docs/license.html: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | Nvidia Source Code License-NC 7 | 8 | 56 | 57 | 58 | 59 |

NVIDIA Source Code License for StyleGAN2 with Adaptive Discriminator Augmentation (ADA)

60 | 61 |
62 | 63 |

1. Definitions

64 | 65 |

“Licensor” means any person or entity that distributes its Work.

66 | 67 |

“Software” means the original work of authorship made available under 68 | this License.

69 | 70 |

“Work” means the Software and any additions to or derivative works of 71 | the Software that are made available under this License.

72 | 73 |

The terms “reproduce,” “reproduction,” “derivative works,” and 74 | “distribution” have the meaning as provided under U.S. copyright law; 75 | provided, however, that for the purposes of this License, derivative 76 | works shall not include works that remain separable from, or merely 77 | link (or bind by name) to the interfaces of, the Work.

78 | 79 |

Works, including the Software, are “made available” under this License 80 | by including in or with the Work either (a) a copyright notice 81 | referencing the applicability of this License to the Work, or (b) a 82 | copy of this License.

83 | 84 |

2. License Grants

85 | 86 |

2.1 Copyright Grant. Subject to the terms and conditions of this 87 | License, each Licensor grants to you a perpetual, worldwide, 88 | non-exclusive, royalty-free, copyright license to reproduce, 89 | prepare derivative works of, publicly display, publicly perform, 90 | sublicense and distribute its Work and any resulting derivative 91 | works in any form.

92 | 93 |

3. Limitations

94 | 95 |

3.1 Redistribution. You may reproduce or distribute the Work only 96 | if (a) you do so under this License, (b) you include a complete 97 | copy of this License with your distribution, and (c) you retain 98 | without modification any copyright, patent, trademark, or 99 | attribution notices that are present in the Work.

100 | 101 |

3.2 Derivative Works. You may specify that additional or different 102 | terms apply to the use, reproduction, and distribution of your 103 | derivative works of the Work (“Your Terms”) only if (a) Your Terms 104 | provide that the use limitation in Section 3.3 applies to your 105 | derivative works, and (b) you identify the specific derivative 106 | works that are subject to Your Terms. Notwithstanding Your Terms, 107 | this License (including the redistribution requirements in Section 108 | 3.1) will continue to apply to the Work itself.

109 | 110 |

3.3 Use Limitation. The Work and any derivative works thereof only may be used or intended for 111 | use non-commercially. Notwithstanding the foregoing, NVIDIA and its affiliates may use the Work 112 | and any derivative works commercially. As used herein, “non-commercially” means for research or 113 | evaluation purposes only. 114 | 115 |

3.4 Patent Claims. If you bring or threaten to bring a patent claim 116 | against any Licensor (including any claim, cross-claim or 117 | counterclaim in a lawsuit) to enforce any patents that you allege 118 | are infringed by any Work, then your rights under this License from 119 | such Licensor (including the grant in Section 2.1) will terminate immediately. 120 | 121 |

3.5 Trademarks. This License does not grant any rights to use any 122 | Licensor’s or its affiliates’ names, logos, or trademarks, except 123 | as necessary to reproduce the notices described in this License.

124 | 125 |

3.6 Termination. If you violate any term of this License, then your 126 | rights under this License (including the grant in Section 2.1) 127 | will terminate immediately.

128 | 129 |

4. Disclaimer of Warranty.

130 | 131 |

THE WORK IS PROVIDED “AS IS” WITHOUT WARRANTIES OR CONDITIONS OF ANY 132 | KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WARRANTIES OR CONDITIONS OF 133 | MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE OR 134 | NON-INFRINGEMENT. YOU BEAR THE RISK OF UNDERTAKING ANY ACTIVITIES UNDER 135 | THIS LICENSE.

136 | 137 |

5. Limitation of Liability.

138 | 139 |

EXCEPT AS PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL 140 | THEORY, WHETHER IN TORT (INCLUDING NEGLIGENCE), CONTRACT, OR OTHERWISE 141 | SHALL ANY LICENSOR BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY DIRECT, 142 | INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF 143 | OR RELATED TO THIS LICENSE, THE USE OR INABILITY TO USE THE WORK 144 | (INCLUDING BUT NOT LIMITED TO LOSS OF GOODWILL, BUSINESS INTERRUPTION, 145 | LOST PROFITS OR DATA, COMPUTER FAILURE OR MALFUNCTION, OR ANY OTHER 146 | COMMERCIAL DAMAGES OR LOSSES), EVEN IF THE LICENSOR HAS BEEN ADVISED OF 147 | THE POSSIBILITY OF SUCH DAMAGES.

148 | 149 |
150 |
151 | 152 | 153 | 154 | -------------------------------------------------------------------------------- /metrics/metric_main.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | import os 10 | import time 11 | import json 12 | import torch 13 | import dnnlib 14 | 15 | from . import metric_utils 16 | from . import frechet_inception_distance 17 | from . import kernel_inception_distance 18 | from . import precision_recall 19 | from . import perceptual_path_length 20 | from . import inception_score 21 | 22 | #---------------------------------------------------------------------------- 23 | 24 | _metric_dict = dict() # name => fn 25 | 26 | def register_metric(fn): 27 | assert callable(fn) 28 | _metric_dict[fn.__name__] = fn 29 | return fn 30 | 31 | def is_valid_metric(metric): 32 | return metric in _metric_dict 33 | 34 | def list_valid_metrics(): 35 | return list(_metric_dict.keys()) 36 | 37 | #---------------------------------------------------------------------------- 38 | 39 | def calc_metric(metric, **kwargs): # See metric_utils.MetricOptions for the full list of arguments. 40 | assert is_valid_metric(metric) 41 | opts = metric_utils.MetricOptions(**kwargs) 42 | 43 | # Calculate. 44 | start_time = time.time() 45 | results = _metric_dict[metric](opts) 46 | total_time = time.time() - start_time 47 | 48 | # Broadcast results. 49 | for key, value in list(results.items()): 50 | if opts.num_gpus > 1: 51 | value = torch.as_tensor(value, dtype=torch.float64, device=opts.device) 52 | torch.distributed.broadcast(tensor=value, src=0) 53 | value = float(value.cpu()) 54 | results[key] = value 55 | 56 | # Decorate with metadata. 57 | return dnnlib.EasyDict( 58 | results = dnnlib.EasyDict(results), 59 | metric = metric, 60 | total_time = total_time, 61 | total_time_str = dnnlib.util.format_time(total_time), 62 | num_gpus = opts.num_gpus, 63 | ) 64 | 65 | #---------------------------------------------------------------------------- 66 | 67 | def report_metric(result_dict, run_dir=None, snapshot_pkl=None): 68 | metric = result_dict['metric'] 69 | assert is_valid_metric(metric) 70 | if run_dir is not None and snapshot_pkl is not None: 71 | snapshot_pkl = os.path.relpath(snapshot_pkl, run_dir) 72 | 73 | jsonl_line = json.dumps(dict(result_dict, snapshot_pkl=snapshot_pkl, timestamp=time.time())) 74 | print(jsonl_line) 75 | if run_dir is not None and os.path.isdir(run_dir): 76 | with open(os.path.join(run_dir, f'metric-{metric}.jsonl'), 'at') as f: 77 | f.write(jsonl_line + '\n') 78 | 79 | #---------------------------------------------------------------------------- 80 | # Primary metrics. 81 | 82 | @register_metric 83 | def fid50k_full(opts): 84 | opts.dataset_kwargs.update(max_size=None, xflip=False) 85 | fid = frechet_inception_distance.compute_fid(opts, max_real=None, num_gen=50000) 86 | return dict(fid50k_full=fid) 87 | 88 | @register_metric 89 | def kid50k_full(opts): 90 | opts.dataset_kwargs.update(max_size=None, xflip=False) 91 | kid = kernel_inception_distance.compute_kid(opts, max_real=1000000, num_gen=50000, num_subsets=100, max_subset_size=1000) 92 | return dict(kid50k_full=kid) 93 | 94 | @register_metric 95 | def pr50k3_full(opts): 96 | opts.dataset_kwargs.update(max_size=None, xflip=False) 97 | precision, recall = precision_recall.compute_pr(opts, max_real=200000, num_gen=50000, nhood_size=3, row_batch_size=10000, col_batch_size=10000) 98 | return dict(pr50k3_full_precision=precision, pr50k3_full_recall=recall) 99 | 100 | @register_metric 101 | def ppl2_wend(opts): 102 | ppl = perceptual_path_length.compute_ppl(opts, num_samples=50000, epsilon=1e-4, space='w', sampling='end', crop=False, batch_size=2) 103 | return dict(ppl2_wend=ppl) 104 | 105 | @register_metric 106 | def is50k(opts): 107 | opts.dataset_kwargs.update(max_size=None, xflip=False) 108 | mean, std = inception_score.compute_is(opts, num_gen=50000, num_splits=10) 109 | return dict(is50k_mean=mean, is50k_std=std) 110 | 111 | #---------------------------------------------------------------------------- 112 | # Legacy metrics. 113 | 114 | @register_metric 115 | def fid50k(opts): 116 | opts.dataset_kwargs.update(max_size=None) 117 | fid = frechet_inception_distance.compute_fid(opts, max_real=50000, num_gen=50000) 118 | return dict(fid50k=fid) 119 | 120 | @register_metric 121 | def kid50k(opts): 122 | opts.dataset_kwargs.update(max_size=None) 123 | kid = kernel_inception_distance.compute_kid(opts, max_real=50000, num_gen=50000, num_subsets=100, max_subset_size=1000) 124 | return dict(kid50k=kid) 125 | 126 | @register_metric 127 | def pr50k3(opts): 128 | opts.dataset_kwargs.update(max_size=None) 129 | precision, recall = precision_recall.compute_pr(opts, max_real=50000, num_gen=50000, nhood_size=3, row_batch_size=10000, col_batch_size=10000) 130 | return dict(pr50k3_precision=precision, pr50k3_recall=recall) 131 | 132 | @register_metric 133 | def ppl_zfull(opts): 134 | ppl = perceptual_path_length.compute_ppl(opts, num_samples=50000, epsilon=1e-4, space='z', sampling='full', crop=True, batch_size=2) 135 | return dict(ppl_zfull=ppl) 136 | 137 | @register_metric 138 | def ppl_wfull(opts): 139 | ppl = perceptual_path_length.compute_ppl(opts, num_samples=50000, epsilon=1e-4, space='w', sampling='full', crop=True, batch_size=2) 140 | return dict(ppl_wfull=ppl) 141 | 142 | @register_metric 143 | def ppl_zend(opts): 144 | ppl = perceptual_path_length.compute_ppl(opts, num_samples=50000, epsilon=1e-4, space='z', sampling='end', crop=True, batch_size=2) 145 | return dict(ppl_zend=ppl) 146 | 147 | @register_metric 148 | def ppl_wend(opts): 149 | ppl = perceptual_path_length.compute_ppl(opts, num_samples=50000, epsilon=1e-4, space='w', sampling='end', crop=True, batch_size=2) 150 | return dict(ppl_wend=ppl) 151 | 152 | #---------------------------------------------------------------------------- 153 | -------------------------------------------------------------------------------- /torch_utils/ops/bias_act.cu: -------------------------------------------------------------------------------- 1 | // Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | // 3 | // NVIDIA CORPORATION and its licensors retain all intellectual property 4 | // and proprietary rights in and to this software, related documentation 5 | // and any modifications thereto. Any use, reproduction, disclosure or 6 | // distribution of this software and related documentation without an express 7 | // license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | #include 10 | #include "bias_act.h" 11 | 12 | //------------------------------------------------------------------------ 13 | // Helpers. 14 | 15 | template struct InternalType; 16 | template <> struct InternalType { typedef double scalar_t; }; 17 | template <> struct InternalType { typedef float scalar_t; }; 18 | template <> struct InternalType { typedef float scalar_t; }; 19 | 20 | //------------------------------------------------------------------------ 21 | // CUDA kernel. 22 | 23 | template 24 | __global__ void bias_act_kernel(bias_act_kernel_params p) 25 | { 26 | typedef typename InternalType::scalar_t scalar_t; 27 | int G = p.grad; 28 | scalar_t alpha = (scalar_t)p.alpha; 29 | scalar_t gain = (scalar_t)p.gain; 30 | scalar_t clamp = (scalar_t)p.clamp; 31 | scalar_t one = (scalar_t)1; 32 | scalar_t two = (scalar_t)2; 33 | scalar_t expRange = (scalar_t)80; 34 | scalar_t halfExpRange = (scalar_t)40; 35 | scalar_t seluScale = (scalar_t)1.0507009873554804934193349852946; 36 | scalar_t seluAlpha = (scalar_t)1.6732632423543772848170429916717; 37 | 38 | // Loop over elements. 39 | int xi = blockIdx.x * p.loopX * blockDim.x + threadIdx.x; 40 | for (int loopIdx = 0; loopIdx < p.loopX && xi < p.sizeX; loopIdx++, xi += blockDim.x) 41 | { 42 | // Load. 43 | scalar_t x = (scalar_t)((const T*)p.x)[xi]; 44 | scalar_t b = (p.b) ? (scalar_t)((const T*)p.b)[(xi / p.stepB) % p.sizeB] : 0; 45 | scalar_t xref = (p.xref) ? (scalar_t)((const T*)p.xref)[xi] : 0; 46 | scalar_t yref = (p.yref) ? (scalar_t)((const T*)p.yref)[xi] : 0; 47 | scalar_t dy = (p.dy) ? (scalar_t)((const T*)p.dy)[xi] : one; 48 | scalar_t yy = (gain != 0) ? yref / gain : 0; 49 | scalar_t y = 0; 50 | 51 | // Apply bias. 52 | ((G == 0) ? x : xref) += b; 53 | 54 | // linear 55 | if (A == 1) 56 | { 57 | if (G == 0) y = x; 58 | if (G == 1) y = x; 59 | } 60 | 61 | // relu 62 | if (A == 2) 63 | { 64 | if (G == 0) y = (x > 0) ? x : 0; 65 | if (G == 1) y = (yy > 0) ? x : 0; 66 | } 67 | 68 | // lrelu 69 | if (A == 3) 70 | { 71 | if (G == 0) y = (x > 0) ? x : x * alpha; 72 | if (G == 1) y = (yy > 0) ? x : x * alpha; 73 | } 74 | 75 | // tanh 76 | if (A == 4) 77 | { 78 | if (G == 0) { scalar_t c = exp(x); scalar_t d = one / c; y = (x < -expRange) ? -one : (x > expRange) ? one : (c - d) / (c + d); } 79 | if (G == 1) y = x * (one - yy * yy); 80 | if (G == 2) y = x * (one - yy * yy) * (-two * yy); 81 | } 82 | 83 | // sigmoid 84 | if (A == 5) 85 | { 86 | if (G == 0) y = (x < -expRange) ? 0 : one / (exp(-x) + one); 87 | if (G == 1) y = x * yy * (one - yy); 88 | if (G == 2) y = x * yy * (one - yy) * (one - two * yy); 89 | } 90 | 91 | // elu 92 | if (A == 6) 93 | { 94 | if (G == 0) y = (x >= 0) ? x : exp(x) - one; 95 | if (G == 1) y = (yy >= 0) ? x : x * (yy + one); 96 | if (G == 2) y = (yy >= 0) ? 0 : x * (yy + one); 97 | } 98 | 99 | // selu 100 | if (A == 7) 101 | { 102 | if (G == 0) y = (x >= 0) ? seluScale * x : (seluScale * seluAlpha) * (exp(x) - one); 103 | if (G == 1) y = (yy >= 0) ? x * seluScale : x * (yy + seluScale * seluAlpha); 104 | if (G == 2) y = (yy >= 0) ? 0 : x * (yy + seluScale * seluAlpha); 105 | } 106 | 107 | // softplus 108 | if (A == 8) 109 | { 110 | if (G == 0) y = (x > expRange) ? x : log(exp(x) + one); 111 | if (G == 1) y = x * (one - exp(-yy)); 112 | if (G == 2) { scalar_t c = exp(-yy); y = x * c * (one - c); } 113 | } 114 | 115 | // swish 116 | if (A == 9) 117 | { 118 | if (G == 0) 119 | y = (x < -expRange) ? 0 : x / (exp(-x) + one); 120 | else 121 | { 122 | scalar_t c = exp(xref); 123 | scalar_t d = c + one; 124 | if (G == 1) 125 | y = (xref > halfExpRange) ? x : x * c * (xref + d) / (d * d); 126 | else 127 | y = (xref > halfExpRange) ? 0 : x * c * (xref * (two - d) + two * d) / (d * d * d); 128 | yref = (xref < -expRange) ? 0 : xref / (exp(-xref) + one) * gain; 129 | } 130 | } 131 | 132 | // Apply gain. 133 | y *= gain * dy; 134 | 135 | // Clamp. 136 | if (clamp >= 0) 137 | { 138 | if (G == 0) 139 | y = (y > -clamp & y < clamp) ? y : (y >= 0) ? clamp : -clamp; 140 | else 141 | y = (yref > -clamp & yref < clamp) ? y : 0; 142 | } 143 | 144 | // Store. 145 | ((T*)p.y)[xi] = (T)y; 146 | } 147 | } 148 | 149 | //------------------------------------------------------------------------ 150 | // CUDA kernel selection. 151 | 152 | template void* choose_bias_act_kernel(const bias_act_kernel_params& p) 153 | { 154 | if (p.act == 1) return (void*)bias_act_kernel; 155 | if (p.act == 2) return (void*)bias_act_kernel; 156 | if (p.act == 3) return (void*)bias_act_kernel; 157 | if (p.act == 4) return (void*)bias_act_kernel; 158 | if (p.act == 5) return (void*)bias_act_kernel; 159 | if (p.act == 6) return (void*)bias_act_kernel; 160 | if (p.act == 7) return (void*)bias_act_kernel; 161 | if (p.act == 8) return (void*)bias_act_kernel; 162 | if (p.act == 9) return (void*)bias_act_kernel; 163 | return NULL; 164 | } 165 | 166 | //------------------------------------------------------------------------ 167 | // Template specializations. 168 | 169 | template void* choose_bias_act_kernel (const bias_act_kernel_params& p); 170 | template void* choose_bias_act_kernel (const bias_act_kernel_params& p); 171 | template void* choose_bias_act_kernel (const bias_act_kernel_params& p); 172 | 173 | //------------------------------------------------------------------------ 174 | -------------------------------------------------------------------------------- /training/loss.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | import numpy as np 10 | import torch 11 | from torch_utils import training_stats 12 | from torch_utils import misc 13 | from torch_utils.ops import conv2d_gradfix 14 | 15 | #---------------------------------------------------------------------------- 16 | 17 | class Loss: 18 | def accumulate_gradients(self, phase, real_img, real_c, gen_z, gen_c, sync, gain): # to be overridden by subclass 19 | raise NotImplementedError() 20 | 21 | #---------------------------------------------------------------------------- 22 | 23 | class StyleGAN2Loss(Loss): 24 | def __init__(self, device, G_mapping, G_synthesis, D, augment_pipe=None, style_mixing_prob=0.9, r1_gamma=10, pl_batch_shrink=2, pl_decay=0.01, pl_weight=2): 25 | super().__init__() 26 | self.device = device 27 | self.G_mapping = G_mapping 28 | self.G_synthesis = G_synthesis 29 | self.D = D 30 | self.augment_pipe = augment_pipe 31 | self.style_mixing_prob = style_mixing_prob 32 | self.r1_gamma = r1_gamma 33 | self.pl_batch_shrink = pl_batch_shrink 34 | self.pl_decay = pl_decay 35 | self.pl_weight = pl_weight 36 | self.pl_mean = torch.zeros([], device=device) 37 | 38 | def run_G(self, z, c, sync): 39 | with misc.ddp_sync(self.G_mapping, sync): 40 | ws = self.G_mapping(z, c) 41 | if self.style_mixing_prob > 0: 42 | with torch.autograd.profiler.record_function('style_mixing'): 43 | cutoff = torch.empty([], dtype=torch.int64, device=ws.device).random_(1, ws.shape[1]) 44 | cutoff = torch.where(torch.rand([], device=ws.device) < self.style_mixing_prob, cutoff, torch.full_like(cutoff, ws.shape[1])) 45 | ws[:, cutoff:] = self.G_mapping(torch.randn_like(z), c, skip_w_avg_update=True)[:, cutoff:] 46 | with misc.ddp_sync(self.G_synthesis, sync): 47 | img = self.G_synthesis(ws) 48 | return img, ws 49 | 50 | def run_D(self, img, c, sync): 51 | if self.augment_pipe is not None: 52 | img = self.augment_pipe(img) 53 | with misc.ddp_sync(self.D, sync): 54 | logits = self.D(img, c) 55 | return logits 56 | 57 | def accumulate_gradients(self, phase, real_img, real_c, gen_z, gen_c, sync, gain): 58 | assert phase in ['Gmain', 'Greg', 'Gboth', 'Dmain', 'Dreg', 'Dboth'] 59 | do_Gmain = (phase in ['Gmain', 'Gboth']) 60 | do_Dmain = (phase in ['Dmain', 'Dboth']) 61 | do_Gpl = (phase in ['Greg', 'Gboth']) and (self.pl_weight != 0) 62 | do_Dr1 = (phase in ['Dreg', 'Dboth']) and (self.r1_gamma != 0) 63 | 64 | # Gmain: Maximize logits for generated images. 65 | if do_Gmain: 66 | with torch.autograd.profiler.record_function('Gmain_forward'): 67 | gen_img, _gen_ws = self.run_G(gen_z, gen_c, sync=(sync and not do_Gpl)) # May get synced by Gpl. 68 | gen_logits = self.run_D(gen_img, gen_c, sync=False) 69 | training_stats.report('Loss/scores/fake', gen_logits) 70 | training_stats.report('Loss/signs/fake', gen_logits.sign()) 71 | loss_Gmain = torch.nn.functional.softplus(-gen_logits) # -log(sigmoid(gen_logits)) 72 | training_stats.report('Loss/G/loss', loss_Gmain) 73 | with torch.autograd.profiler.record_function('Gmain_backward'): 74 | loss_Gmain.mean().mul(gain).backward() 75 | 76 | # Gpl: Apply path length regularization. 77 | if do_Gpl: 78 | with torch.autograd.profiler.record_function('Gpl_forward'): 79 | batch_size = gen_z.shape[0] // self.pl_batch_shrink 80 | gen_img, gen_ws = self.run_G(gen_z[:batch_size], gen_c[:batch_size], sync=sync) 81 | pl_noise = torch.randn_like(gen_img) / np.sqrt(gen_img.shape[2] * gen_img.shape[3]) 82 | with torch.autograd.profiler.record_function('pl_grads'), conv2d_gradfix.no_weight_gradients(): 83 | pl_grads = torch.autograd.grad(outputs=[(gen_img * pl_noise).sum()], inputs=[gen_ws], create_graph=True, only_inputs=True)[0] 84 | pl_lengths = pl_grads.square().sum(2).mean(1).sqrt() 85 | pl_mean = self.pl_mean.lerp(pl_lengths.mean(), self.pl_decay) 86 | self.pl_mean.copy_(pl_mean.detach()) 87 | pl_penalty = (pl_lengths - pl_mean).square() 88 | training_stats.report('Loss/pl_penalty', pl_penalty) 89 | loss_Gpl = pl_penalty * self.pl_weight 90 | training_stats.report('Loss/G/reg', loss_Gpl) 91 | with torch.autograd.profiler.record_function('Gpl_backward'): 92 | (gen_img[:, 0, 0, 0] * 0 + loss_Gpl).mean().mul(gain).backward() 93 | 94 | # Dmain: Minimize logits for generated images. 95 | loss_Dgen = 0 96 | if do_Dmain: 97 | with torch.autograd.profiler.record_function('Dgen_forward'): 98 | gen_img, _gen_ws = self.run_G(gen_z, gen_c, sync=False) 99 | gen_logits = self.run_D(gen_img, gen_c, sync=False) # Gets synced by loss_Dreal. 100 | training_stats.report('Loss/scores/fake', gen_logits) 101 | training_stats.report('Loss/signs/fake', gen_logits.sign()) 102 | loss_Dgen = torch.nn.functional.softplus(gen_logits) # -log(1 - sigmoid(gen_logits)) 103 | with torch.autograd.profiler.record_function('Dgen_backward'): 104 | loss_Dgen.mean().mul(gain).backward() 105 | 106 | # Dmain: Maximize logits for real images. 107 | # Dr1: Apply R1 regularization. 108 | if do_Dmain or do_Dr1: 109 | name = 'Dreal_Dr1' if do_Dmain and do_Dr1 else 'Dreal' if do_Dmain else 'Dr1' 110 | with torch.autograd.profiler.record_function(name + '_forward'): 111 | real_img_tmp = real_img.detach().requires_grad_(do_Dr1) 112 | real_logits = self.run_D(real_img_tmp, real_c, sync=sync) 113 | training_stats.report('Loss/scores/real', real_logits) 114 | training_stats.report('Loss/signs/real', real_logits.sign()) 115 | 116 | loss_Dreal = 0 117 | if do_Dmain: 118 | loss_Dreal = torch.nn.functional.softplus(-real_logits) # -log(sigmoid(real_logits)) 119 | training_stats.report('Loss/D/loss', loss_Dgen + loss_Dreal) 120 | 121 | loss_Dr1 = 0 122 | if do_Dr1: 123 | with torch.autograd.profiler.record_function('r1_grads'), conv2d_gradfix.no_weight_gradients(): 124 | r1_grads = torch.autograd.grad(outputs=[real_logits.sum()], inputs=[real_img_tmp], create_graph=True, only_inputs=True)[0] 125 | r1_penalty = r1_grads.square().sum([1,2,3]) 126 | loss_Dr1 = r1_penalty * (self.r1_gamma / 2) 127 | training_stats.report('Loss/r1_penalty', r1_penalty) 128 | training_stats.report('Loss/D/reg', loss_Dr1) 129 | 130 | with torch.autograd.profiler.record_function(name + '_backward'): 131 | (real_logits * 0 + loss_Dreal + loss_Dr1).mean().mul(gain).backward() 132 | 133 | #---------------------------------------------------------------------------- 134 | -------------------------------------------------------------------------------- /torch_utils/ops/conv2d_resample.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """2D convolution with optional up/downsampling.""" 10 | 11 | import torch 12 | 13 | from .. import misc 14 | from . import conv2d_gradfix 15 | from . import upfirdn2d 16 | from .upfirdn2d import _parse_padding 17 | from .upfirdn2d import _get_filter_size 18 | 19 | #---------------------------------------------------------------------------- 20 | 21 | def _get_weight_shape(w): 22 | with misc.suppress_tracer_warnings(): # this value will be treated as a constant 23 | shape = [int(sz) for sz in w.shape] 24 | misc.assert_shape(w, shape) 25 | return shape 26 | 27 | #---------------------------------------------------------------------------- 28 | 29 | def _conv2d_wrapper(x, w, stride=1, padding=0, groups=1, transpose=False, flip_weight=True): 30 | """Wrapper for the underlying `conv2d()` and `conv_transpose2d()` implementations. 31 | """ 32 | out_channels, in_channels_per_group, kh, kw = _get_weight_shape(w) 33 | 34 | # Flip weight if requested. 35 | if not flip_weight: # conv2d() actually performs correlation (flip_weight=True) not convolution (flip_weight=False). 36 | w = w.flip([2, 3]) 37 | 38 | # Workaround performance pitfall in cuDNN 8.0.5, triggered when using 39 | # 1x1 kernel + memory_format=channels_last + less than 64 channels. 40 | if kw == 1 and kh == 1 and stride == 1 and padding in [0, [0, 0], (0, 0)] and not transpose: 41 | if x.stride()[1] == 1 and min(out_channels, in_channels_per_group) < 64: 42 | if out_channels <= 4 and groups == 1: 43 | in_shape = x.shape 44 | x = w.squeeze(3).squeeze(2) @ x.reshape([in_shape[0], in_channels_per_group, -1]) 45 | x = x.reshape([in_shape[0], out_channels, in_shape[2], in_shape[3]]) 46 | else: 47 | x = x.to(memory_format=torch.contiguous_format) 48 | w = w.to(memory_format=torch.contiguous_format) 49 | x = conv2d_gradfix.conv2d(x, w, groups=groups) 50 | return x.to(memory_format=torch.channels_last) 51 | 52 | # Otherwise => execute using conv2d_gradfix. 53 | op = conv2d_gradfix.conv_transpose2d if transpose else conv2d_gradfix.conv2d 54 | return op(x, w, stride=stride, padding=padding, groups=groups) 55 | 56 | #---------------------------------------------------------------------------- 57 | 58 | @misc.profiled_function 59 | def conv2d_resample(x, w, f=None, up=1, down=1, padding=0, groups=1, flip_weight=True, flip_filter=False): 60 | r"""2D convolution with optional up/downsampling. 61 | 62 | Padding is performed only once at the beginning, not between the operations. 63 | 64 | Args: 65 | x: Input tensor of shape 66 | `[batch_size, in_channels, in_height, in_width]`. 67 | w: Weight tensor of shape 68 | `[out_channels, in_channels//groups, kernel_height, kernel_width]`. 69 | f: Low-pass filter for up/downsampling. Must be prepared beforehand by 70 | calling upfirdn2d.setup_filter(). None = identity (default). 71 | up: Integer upsampling factor (default: 1). 72 | down: Integer downsampling factor (default: 1). 73 | padding: Padding with respect to the upsampled image. Can be a single number 74 | or a list/tuple `[x, y]` or `[x_before, x_after, y_before, y_after]` 75 | (default: 0). 76 | groups: Split input channels into N groups (default: 1). 77 | flip_weight: False = convolution, True = correlation (default: True). 78 | flip_filter: False = convolution, True = correlation (default: False). 79 | 80 | Returns: 81 | Tensor of the shape `[batch_size, num_channels, out_height, out_width]`. 82 | """ 83 | # Validate arguments. 84 | assert isinstance(x, torch.Tensor) and (x.ndim == 4) 85 | assert isinstance(w, torch.Tensor) and (w.ndim == 4) and (w.dtype == x.dtype) 86 | assert f is None or (isinstance(f, torch.Tensor) and f.ndim in [1, 2] and f.dtype == torch.float32) 87 | assert isinstance(up, int) and (up >= 1) 88 | assert isinstance(down, int) and (down >= 1) 89 | assert isinstance(groups, int) and (groups >= 1) 90 | out_channels, in_channels_per_group, kh, kw = _get_weight_shape(w) 91 | fw, fh = _get_filter_size(f) 92 | px0, px1, py0, py1 = _parse_padding(padding) 93 | 94 | # Adjust padding to account for up/downsampling. 95 | if up > 1: 96 | px0 += (fw + up - 1) // 2 97 | px1 += (fw - up) // 2 98 | py0 += (fh + up - 1) // 2 99 | py1 += (fh - up) // 2 100 | if down > 1: 101 | px0 += (fw - down + 1) // 2 102 | px1 += (fw - down) // 2 103 | py0 += (fh - down + 1) // 2 104 | py1 += (fh - down) // 2 105 | 106 | # Fast path: 1x1 convolution with downsampling only => downsample first, then convolve. 107 | if kw == 1 and kh == 1 and (down > 1 and up == 1): 108 | x = upfirdn2d.upfirdn2d(x=x, f=f, down=down, padding=[px0,px1,py0,py1], flip_filter=flip_filter) 109 | x = _conv2d_wrapper(x=x, w=w, groups=groups, flip_weight=flip_weight) 110 | return x 111 | 112 | # Fast path: 1x1 convolution with upsampling only => convolve first, then upsample. 113 | if kw == 1 and kh == 1 and (up > 1 and down == 1): 114 | x = _conv2d_wrapper(x=x, w=w, groups=groups, flip_weight=flip_weight) 115 | x = upfirdn2d.upfirdn2d(x=x, f=f, up=up, padding=[px0,px1,py0,py1], gain=up**2, flip_filter=flip_filter) 116 | return x 117 | 118 | # Fast path: downsampling only => use strided convolution. 119 | if down > 1 and up == 1: 120 | x = upfirdn2d.upfirdn2d(x=x, f=f, padding=[px0,px1,py0,py1], flip_filter=flip_filter) 121 | x = _conv2d_wrapper(x=x, w=w, stride=down, groups=groups, flip_weight=flip_weight) 122 | return x 123 | 124 | # Fast path: upsampling with optional downsampling => use transpose strided convolution. 125 | if up > 1: 126 | if groups == 1: 127 | w = w.transpose(0, 1) 128 | else: 129 | w = w.reshape(groups, out_channels // groups, in_channels_per_group, kh, kw) 130 | w = w.transpose(1, 2) 131 | w = w.reshape(groups * in_channels_per_group, out_channels // groups, kh, kw) 132 | px0 -= kw - 1 133 | px1 -= kw - up 134 | py0 -= kh - 1 135 | py1 -= kh - up 136 | pxt = max(min(-px0, -px1), 0) 137 | pyt = max(min(-py0, -py1), 0) 138 | x = _conv2d_wrapper(x=x, w=w, stride=up, padding=[pyt,pxt], groups=groups, transpose=True, flip_weight=(not flip_weight)) 139 | x = upfirdn2d.upfirdn2d(x=x, f=f, padding=[px0+pxt,px1+pxt,py0+pyt,py1+pyt], gain=up**2, flip_filter=flip_filter) 140 | if down > 1: 141 | x = upfirdn2d.upfirdn2d(x=x, f=f, down=down, flip_filter=flip_filter) 142 | return x 143 | 144 | # Fast path: no up/downsampling, padding supported by the underlying implementation => use plain conv2d. 145 | if up == 1 and down == 1: 146 | if px0 == px1 and py0 == py1 and px0 >= 0 and py0 >= 0: 147 | return _conv2d_wrapper(x=x, w=w, padding=[py0,px0], groups=groups, flip_weight=flip_weight) 148 | 149 | # Fallback: Generic reference implementation. 150 | x = upfirdn2d.upfirdn2d(x=x, f=(f if up > 1 else None), up=up, padding=[px0,px1,py0,py1], gain=up**2, flip_filter=flip_filter) 151 | x = _conv2d_wrapper(x=x, w=w, groups=groups, flip_weight=flip_weight) 152 | if down > 1: 153 | x = upfirdn2d.upfirdn2d(x=x, f=f, down=down, flip_filter=flip_filter) 154 | return x 155 | 156 | #---------------------------------------------------------------------------- 157 | -------------------------------------------------------------------------------- /torch_utils/ops/conv2d_gradfix.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Custom replacement for `torch.nn.functional.conv2d` that supports 10 | arbitrarily high order gradients with zero performance penalty.""" 11 | 12 | import warnings 13 | import contextlib 14 | import torch 15 | 16 | # pylint: disable=redefined-builtin 17 | # pylint: disable=arguments-differ 18 | # pylint: disable=protected-access 19 | 20 | #---------------------------------------------------------------------------- 21 | 22 | enabled = False # Enable the custom op by setting this to true. 23 | weight_gradients_disabled = False # Forcefully disable computation of gradients with respect to the weights. 24 | 25 | @contextlib.contextmanager 26 | def no_weight_gradients(): 27 | global weight_gradients_disabled 28 | old = weight_gradients_disabled 29 | weight_gradients_disabled = True 30 | yield 31 | weight_gradients_disabled = old 32 | 33 | #---------------------------------------------------------------------------- 34 | 35 | def conv2d(input, weight, bias=None, stride=1, padding=0, dilation=1, groups=1): 36 | if _should_use_custom_op(input): 37 | return _conv2d_gradfix(transpose=False, weight_shape=weight.shape, stride=stride, padding=padding, output_padding=0, dilation=dilation, groups=groups).apply(input, weight, bias) 38 | return torch.nn.functional.conv2d(input=input, weight=weight, bias=bias, stride=stride, padding=padding, dilation=dilation, groups=groups) 39 | 40 | def conv_transpose2d(input, weight, bias=None, stride=1, padding=0, output_padding=0, groups=1, dilation=1): 41 | if _should_use_custom_op(input): 42 | return _conv2d_gradfix(transpose=True, weight_shape=weight.shape, stride=stride, padding=padding, output_padding=output_padding, groups=groups, dilation=dilation).apply(input, weight, bias) 43 | return torch.nn.functional.conv_transpose2d(input=input, weight=weight, bias=bias, stride=stride, padding=padding, output_padding=output_padding, groups=groups, dilation=dilation) 44 | 45 | #---------------------------------------------------------------------------- 46 | 47 | def _should_use_custom_op(input): 48 | assert isinstance(input, torch.Tensor) 49 | if (not enabled) or (not torch.backends.cudnn.enabled): 50 | return False 51 | if input.device.type != 'cuda': 52 | return False 53 | if any(torch.__version__.startswith(x) for x in ['1.7.', '1.8.', '1.9']): 54 | return True 55 | warnings.warn(f'conv2d_gradfix not supported on PyTorch {torch.__version__}. Falling back to torch.nn.functional.conv2d().') 56 | return False 57 | 58 | def _tuple_of_ints(xs, ndim): 59 | xs = tuple(xs) if isinstance(xs, (tuple, list)) else (xs,) * ndim 60 | assert len(xs) == ndim 61 | assert all(isinstance(x, int) for x in xs) 62 | return xs 63 | 64 | #---------------------------------------------------------------------------- 65 | 66 | _conv2d_gradfix_cache = dict() 67 | 68 | def _conv2d_gradfix(transpose, weight_shape, stride, padding, output_padding, dilation, groups): 69 | # Parse arguments. 70 | ndim = 2 71 | weight_shape = tuple(weight_shape) 72 | stride = _tuple_of_ints(stride, ndim) 73 | padding = _tuple_of_ints(padding, ndim) 74 | output_padding = _tuple_of_ints(output_padding, ndim) 75 | dilation = _tuple_of_ints(dilation, ndim) 76 | 77 | # Lookup from cache. 78 | key = (transpose, weight_shape, stride, padding, output_padding, dilation, groups) 79 | if key in _conv2d_gradfix_cache: 80 | return _conv2d_gradfix_cache[key] 81 | 82 | # Validate arguments. 83 | assert groups >= 1 84 | assert len(weight_shape) == ndim + 2 85 | assert all(stride[i] >= 1 for i in range(ndim)) 86 | assert all(padding[i] >= 0 for i in range(ndim)) 87 | assert all(dilation[i] >= 0 for i in range(ndim)) 88 | if not transpose: 89 | assert all(output_padding[i] == 0 for i in range(ndim)) 90 | else: # transpose 91 | assert all(0 <= output_padding[i] < max(stride[i], dilation[i]) for i in range(ndim)) 92 | 93 | # Helpers. 94 | common_kwargs = dict(stride=stride, padding=padding, dilation=dilation, groups=groups) 95 | def calc_output_padding(input_shape, output_shape): 96 | if transpose: 97 | return [0, 0] 98 | return [ 99 | input_shape[i + 2] 100 | - (output_shape[i + 2] - 1) * stride[i] 101 | - (1 - 2 * padding[i]) 102 | - dilation[i] * (weight_shape[i + 2] - 1) 103 | for i in range(ndim) 104 | ] 105 | 106 | # Forward & backward. 107 | class Conv2d(torch.autograd.Function): 108 | @staticmethod 109 | def forward(ctx, input, weight, bias): 110 | assert weight.shape == weight_shape 111 | if not transpose: 112 | output = torch.nn.functional.conv2d(input=input, weight=weight, bias=bias, **common_kwargs) 113 | else: # transpose 114 | output = torch.nn.functional.conv_transpose2d(input=input, weight=weight, bias=bias, output_padding=output_padding, **common_kwargs) 115 | ctx.save_for_backward(input, weight) 116 | return output 117 | 118 | @staticmethod 119 | def backward(ctx, grad_output): 120 | input, weight = ctx.saved_tensors 121 | grad_input = None 122 | grad_weight = None 123 | grad_bias = None 124 | 125 | if ctx.needs_input_grad[0]: 126 | p = calc_output_padding(input_shape=input.shape, output_shape=grad_output.shape) 127 | grad_input = _conv2d_gradfix(transpose=(not transpose), weight_shape=weight_shape, output_padding=p, **common_kwargs).apply(grad_output, weight, None) 128 | assert grad_input.shape == input.shape 129 | 130 | if ctx.needs_input_grad[1] and not weight_gradients_disabled: 131 | grad_weight = Conv2dGradWeight.apply(grad_output, input) 132 | assert grad_weight.shape == weight_shape 133 | 134 | if ctx.needs_input_grad[2]: 135 | grad_bias = grad_output.sum([0, 2, 3]) 136 | 137 | return grad_input, grad_weight, grad_bias 138 | 139 | # Gradient with respect to the weights. 140 | class Conv2dGradWeight(torch.autograd.Function): 141 | @staticmethod 142 | def forward(ctx, grad_output, input): 143 | op = torch._C._jit_get_operation('aten::cudnn_convolution_backward_weight' if not transpose else 'aten::cudnn_convolution_transpose_backward_weight') 144 | flags = [torch.backends.cudnn.benchmark, torch.backends.cudnn.deterministic, torch.backends.cudnn.allow_tf32] 145 | grad_weight = op(weight_shape, grad_output, input, padding, stride, dilation, groups, *flags) 146 | assert grad_weight.shape == weight_shape 147 | ctx.save_for_backward(grad_output, input) 148 | return grad_weight 149 | 150 | @staticmethod 151 | def backward(ctx, grad2_grad_weight): 152 | grad_output, input = ctx.saved_tensors 153 | grad2_grad_output = None 154 | grad2_input = None 155 | 156 | if ctx.needs_input_grad[0]: 157 | grad2_grad_output = Conv2d.apply(input, grad2_grad_weight, None) 158 | assert grad2_grad_output.shape == grad_output.shape 159 | 160 | if ctx.needs_input_grad[1]: 161 | p = calc_output_padding(input_shape=input.shape, output_shape=grad_output.shape) 162 | grad2_input = _conv2d_gradfix(transpose=(not transpose), weight_shape=weight_shape, output_padding=p, **common_kwargs).apply(grad_output, grad2_grad_weight, None) 163 | assert grad2_input.shape == input.shape 164 | 165 | return grad2_grad_output, grad2_input 166 | 167 | _conv2d_gradfix_cache[key] = Conv2d 168 | return Conv2d 169 | 170 | #---------------------------------------------------------------------------- 171 | -------------------------------------------------------------------------------- /sightseeding.py: -------------------------------------------------------------------------------- 1 | import os 2 | from typing import List, Union 3 | import click 4 | 5 | import dnnlib 6 | import legacy 7 | 8 | import torch 9 | 10 | import numpy as np 11 | from torch_utils.gen_utils import parse_fps, compress_video, make_run_dir, w_to_img, create_image_grid, interpolate, \ 12 | num_range 13 | 14 | os.environ['PYGAME_HIDE_SUPPORT_PROMPT'] = "hide" 15 | import moviepy.editor 16 | 17 | 18 | # ---------------------------------------------------------------------------- 19 | 20 | 21 | def _parse_seeds(s: str) -> List[int]: 22 | """ 23 | Helper function for parsing seeds. With a, b, c,... as ints, then s can be either: 24 | * a comma-separated list of numbers: 'a,b,c,d,...' 25 | * a range-like of numbers: 'a-d' 26 | * a combination of both: 'a,b,c-d,a-e,f...' 27 | 28 | The returned list will be the numbers in this range, in order as the user entered 29 | them, without deleting repeated values. 30 | """ 31 | nums = num_range(s, False) 32 | return nums 33 | 34 | 35 | # ---------------------------------------------------------------------------- 36 | 37 | 38 | @click.command() 39 | @click.pass_context 40 | @click.option('--network', '-net', 'network_pkl', help='Network pickle filename', required=True) 41 | @click.option('--seeds', '-s', type=_parse_seeds, help='List of seeds to visit in order ("a,b,c", "a-b", "a,b-c,d,e-f,a", ...', required=True) 42 | @click.option('--class', 'class_idx', type=int, help='Class label (unconditional if not specified)') 43 | @click.option('--trunc', 'truncation_psi', type=float, help='Truncation psi', default=1, show_default=True) 44 | @click.option('--noise-mode', help='Noise mode', type=click.Choice(['const', 'random', 'none']), default='const', show_default=True) 45 | @click.option('--seed-sec', '-sec', type=float, help='Number of seconds between each seed transition', default=5.0, show_default=True) 46 | @click.option('--interp-type', '-interp', type=click.Choice(['linear', 'spherical']), help='Type of interpolation in Z or W', default='spherical', show_default=True) 47 | @click.option('--interp-in-z', is_flag=True, help='Add flag to interpolate in Z instead of in W') 48 | @click.option('--smooth', is_flag=True, help='Add flag to smooth the transition between the latent vectors') 49 | @click.option('--fps', type=parse_fps, help='Video FPS.', default=30, show_default=True) 50 | @click.option('--outdir', type=click.Path(file_okay=False), help='Directory path to save the results', default=os.path.join(os.getcwd(), 'out'), show_default=True, metavar='DIR') 51 | @click.option('--desc', type=str, help='Additional description for the directory name where', default='', show_default=True) 52 | @click.option('--compress', is_flag=True, help='Add flag to compress the final mp4 file via ffmpeg-python (same resolution, lower file size)') 53 | def sightseeding( 54 | ctx: click.Context, 55 | network_pkl: Union[str, os.PathLike], 56 | seeds: List[int], 57 | class_idx: int, 58 | truncation_psi: float, 59 | noise_mode: str, 60 | seed_sec: float, 61 | interp_type: str, 62 | interp_in_z: bool, 63 | smooth: bool, 64 | fps: int, 65 | outdir: Union[str, os.PathLike], 66 | desc: str, 67 | compress: bool, 68 | ): 69 | """ 70 | Examples: 71 | 72 | # Will go from seeds 0 through 5, coming to the starting one in the end; the transition between each pair of seeds 73 | taking 7.5 seconds, spherically (and smoothly) interpolating in W, compressing the final video with ffmpeg-python 74 | 75 | python sightseeding.py --seeds=0-5,0 --seed-sec=7.5 --smooth --compress \ 76 | --network=https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/afhqwild.pkl 77 | """ 78 | # Sanity check: 79 | if len(seeds) < 2: 80 | ctx.fail('Please enter more than one seed to interpolate between!') 81 | 82 | print(f'Loading networks from "{network_pkl}"...') 83 | device = torch.device('cuda') 84 | with dnnlib.util.open_url(network_pkl) as f: 85 | G = legacy.load_network_pkl(f)['G_ema'].to(device) 86 | 87 | # Get the average dlatent 88 | w_avg = G.mapping.w_avg 89 | 90 | # Create the run dir with the given name description 91 | desc = f'{desc}-sightseeding' if len(desc) != 0 else 'sightseeding' 92 | desc = f'{desc}-{interp_type}-smooth' if smooth else f'{desc}-{interp_type}' 93 | desc = f'{desc}-in-Z' if interp_in_z else f'{desc}-in-W' 94 | run_dir = make_run_dir(outdir, desc) 95 | 96 | # Number of steps to take between each latent vector 97 | n_steps = int(np.rint(seed_sec * fps)) 98 | # Total number of frames 99 | num_frames = int(n_steps * (len(seeds) - 1)) 100 | # Total video length in seconds 101 | duration_sec = num_frames / fps 102 | 103 | # TODO: use labels (the following will be ignored for now) 104 | label = torch.zeros([1, G.c_dim], device=device) 105 | if G.c_dim != 0: 106 | if class_idx is None: 107 | ctx.fail('Must specify class label with --class when using a conditional network') 108 | label[:, class_idx] = 1 109 | else: 110 | if class_idx is not None: 111 | print('warn: --class=lbl ignored when running on an unconditional network') 112 | 113 | # Generate the random vectors from each seed 114 | print('Generating Z vectors...') 115 | all_z = np.stack([np.random.RandomState(seed).randn(G.z_dim).astype(np.float32) for seed in seeds]) 116 | # If user wants to interpolate in Z 117 | if interp_in_z: 118 | print(f'Interpolating in Z...(interpolation type: {interp_type})') 119 | src_z = np.empty([0] + list(all_z.shape[1:]), dtype=np.float32) 120 | for i in range(len(all_z) - 1): 121 | # We interpolate between each pair of latents 122 | interp = interpolate(all_z[i], all_z[i + 1], n_steps, interp_type, smooth) 123 | # Append it to our source 124 | src_z = np.append(src_z, interp, axis=0) 125 | # Convert to dlatent vectors 126 | print('Generating W vectors...') 127 | src_w = G.mapping(torch.from_numpy(src_z).to(device), None) 128 | 129 | # Otherwise, interpolation is done in W 130 | else: 131 | print(f'Interpolating in W... (interpolation type: {interp_type})') 132 | print('Generating W vectors...') 133 | all_w = G.mapping(torch.from_numpy(all_z).to(device), None).cpu() 134 | src_w = np.empty([0] + list(all_w.shape[1:]), dtype=np.float32) 135 | for i in range(len(all_w) - 1): 136 | # We interpolate between each pair of dlatents 137 | interp = interpolate(all_w[i], all_w[i + 1], n_steps, interp_type, smooth) 138 | # Append it to our source 139 | src_w = np.append(src_w, interp, axis=0) 140 | src_w = torch.from_numpy(src_w).to(device) 141 | 142 | # Do the truncation trick 143 | src_w = w_avg + (src_w - w_avg) * truncation_psi 144 | 145 | # Auxiliary function for moviepy 146 | def make_frame(t): 147 | frame_idx = int(np.clip(np.round(t * fps), 0, num_frames - 1)) 148 | w = src_w[frame_idx].unsqueeze(0) # [18, 512] -> [1, 18, 512] 149 | image = w_to_img(G, w, noise_mode) 150 | # Generate the grid for this timestamp 151 | grid = create_image_grid(image, (1, 1)) 152 | # grayscale => RGB 153 | if grid.shape[2] == 1: 154 | grid = grid.repeat(3, 2) 155 | return grid 156 | 157 | # Generate video using make_frame 158 | print('Generating sightseeding video...') 159 | videoclip = moviepy.editor.VideoClip(make_frame, duration=duration_sec) 160 | videoclip.set_duration(duration_sec) 161 | mp4_name = '-'.join(map(str, seeds)) # Make it clear by the file name what is the path taken 162 | mp4_name = f'{mp4_name}-sightseeding' if len(mp4_name) < 50 else 'sightseeding' # arbitrary rule of mine 163 | 164 | # Set the video parameters (change if you like) 165 | final_video = os.path.join(run_dir, f'{mp4_name}.mp4') 166 | videoclip.write_videofile(final_video, fps=fps, codec='libx264', bitrate='16M') 167 | 168 | # Compress the video (lower file size, same resolution) 169 | if compress: 170 | compress_video(original_video=final_video, original_video_name=mp4_name, outdir=run_dir, ctx=ctx) 171 | 172 | 173 | # ---------------------------------------------------------------------------- 174 | 175 | if __name__ == '__main__': 176 | sightseeding() 177 | -------------------------------------------------------------------------------- /experiments.py: -------------------------------------------------------------------------------- 1 | import click 2 | from typing import Union, Optional 3 | import os 4 | 5 | from torch_utils.gen_utils import parse_slowdown, parse_fps, make_run_dir, w_to_img, create_image_grid, save_config, \ 6 | double_slowdown, compress_video 7 | 8 | import numpy as np 9 | import scipy 10 | import torch 11 | 12 | import dnnlib 13 | import legacy 14 | 15 | os.environ['PYGAME_HIDE_SUPPORT_PROMPT'] = 'hide' 16 | import moviepy.editor 17 | 18 | # ---------------------------------------------------------------------------- 19 | 20 | 21 | @click.group() 22 | def main(): 23 | pass 24 | 25 | 26 | # ---------------------------------------------------------------------------- 27 | 28 | 29 | @main.command(name='mirror-video') 30 | @click.pass_context 31 | @click.option('--network', 'network_pkl', help='Network pickle filename', required=True) 32 | @click.option('--seed', type=int, help='Random seed to use', required=True) 33 | @click.option('--trunc', 'truncation_psi', type=float, help='Truncation psi', default=1, show_default=True) 34 | @click.option('--class', 'class_idx', type=int, help='Class label (unconditional if not specified)') 35 | @click.option('--noise-mode', help='Noise mode', type=click.Choice(['const', 'random', 'none']), default='const', show_default=True) 36 | @click.option('--slowdown', type=parse_slowdown, help='Slow down the video by this amount; will be approximated to the nearest power of 2', default='1', show_default=True) 37 | @click.option('--duration-sec', '-sec', type=float, help='Duration length of the video', default=30.0, show_default=True) 38 | @click.option('--fps', type=parse_fps, help='Video FPS.', default=30, show_default=True) 39 | @click.option('--compress', is_flag=True, help='Add flag to compress the final mp4 file with ffmpeg-python (same resolution, lower file size)') 40 | @click.option('--outdir', type=click.Path(file_okay=False), help='Directory path to save the results', default=os.path.join(os.getcwd(), 'out'), show_default=True, metavar='DIR') 41 | @click.option('--description', '-desc', type=str, help='Description name for the directory path to save results', default='', show_default=True) 42 | def mirror_random_video( 43 | ctx: click.Context, 44 | network_pkl: Union[str, os.PathLike], 45 | seed: Optional[int], 46 | truncation_psi: float, 47 | class_idx: Optional[int], 48 | noise_mode: str, 49 | slowdown: int, 50 | duration_sec: float, 51 | fps: int, 52 | outdir: Union[str, os.PathLike], 53 | description: str, 54 | compress: bool, 55 | smoothing_sec: Optional[float] = 3.0 # for Gaussian blur; won't be a parameter, change at own risk 56 | ): 57 | """ 58 | Generate a random interpolation video using a pretrained network. 59 | 60 | Examples: 61 | 62 | \b 63 | # Generate a 30-second long, truncated MetFaces video at 30 FPS: 64 | python generate.py random-video --seed=0 --trunc=0.7 \\ 65 | --network=https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/metfaces.pkl 66 | """ 67 | print(f'Loading networks from "{network_pkl}"...') 68 | device = torch.device('cuda') 69 | with dnnlib.util.open_url(network_pkl) as f: 70 | G = legacy.load_network_pkl(f)['G_ema'].to(device) # type: ignore 71 | 72 | # Create the run dir with the given name description; add slowdown if different than the default (1) 73 | description = 'mirror-video' if len(description) == 0 else description 74 | description = f'{description}-{slowdown}xslowdown' if slowdown != 1 else description 75 | run_dir = make_run_dir(outdir, description) 76 | 77 | # Number of frames in the video and its total duration in seconds 78 | num_frames = int(np.rint(duration_sec * fps)) 79 | total_duration = duration_sec * slowdown 80 | 81 | print('Generating latent vectors...') 82 | # TODO: let another helper function handle each case, we will use it for the grid 83 | # If there's more than one seed provided and the shape isn't specified by the user 84 | grid_size = (2, 1) 85 | # Shape of the latents to generate 86 | shape = [num_frames, G.z_dim] 87 | # Get the z latents 88 | all_latents = np.random.RandomState(seed).randn(*shape).astype(np.float32) 89 | 90 | # Let's smooth out the random latents so that now they form a loop (and are correctly generated in a 512-dim space) 91 | all_latents = scipy.ndimage.gaussian_filter(all_latents, sigma=[smoothing_sec * fps, 0], mode='wrap') 92 | all_latents /= np.sqrt(np.mean(np.square(all_latents))) 93 | 94 | # Save the configuration used 95 | ctx.obj = { 96 | 'network_pkl': network_pkl, 97 | 'seed': seed, 98 | 'truncation_psi': truncation_psi, 99 | 'class_idx': class_idx, 100 | 'noise_mode': noise_mode, 101 | 'slowdown': slowdown, 102 | 'duration_sec': duration_sec, 103 | 'video_fps': fps, 104 | 'run_dir': run_dir, 105 | 'description': description, 106 | 'compress': compress, 107 | 'smoothing_sec': smoothing_sec 108 | } 109 | # Save the run configuration 110 | save_config(ctx=ctx, run_dir=run_dir) 111 | 112 | # Labels. 113 | label = torch.zeros([1, G.c_dim], device=device) 114 | if G.c_dim != 0: 115 | if class_idx is None: 116 | ctx.fail('Must specify class label with --class when using a conditional network') 117 | label[:, class_idx] = 1 118 | else: 119 | if class_idx is not None: 120 | print('warn: --class=lbl ignored when running on an unconditional network') 121 | 122 | # Name of the video will change if we use slowdown 123 | mp4_name = f'mirror-seed-{seed}-{slowdown}xslowdown' if slowdown != 1 else f'mirror-seed-{seed}' 124 | 125 | # Let's slowdown the video, if so desired 126 | while slowdown > 1: 127 | all_latents, duration_sec, num_frames = double_slowdown(latents=all_latents, 128 | duration=duration_sec, 129 | frames=num_frames) 130 | slowdown //= 2 131 | 132 | # Map to W and do truncation trick 133 | w_avg = G.mapping.w_avg 134 | all_w = G.mapping(torch.from_numpy(all_latents).to(device), None) 135 | all_w = w_avg + (all_w - w_avg) * truncation_psi 136 | 137 | # Mirror 138 | w_mirror = w_avg + (all_w - w_avg) * (-truncation_psi) 139 | 140 | def make_frame(t): 141 | frame_idx = int(np.clip(np.round(t * fps), 0, num_frames - 1)) 142 | w = all_w[frame_idx].unsqueeze(0) 143 | w_m = w_mirror[frame_idx].unsqueeze(0) 144 | dlatent = torch.cat((w, w_m), axis=0) 145 | # Get the images with the labels 146 | images = w_to_img(G, dlatent, noise_mode) 147 | # Generate the grid for this timestamp 148 | grid = create_image_grid(images, grid_size) 149 | # Grayscale => RGB 150 | if grid.shape[2] == 1: 151 | grid = grid.repeat(3, 2) 152 | return grid 153 | 154 | # Generate video using the respective make_frame function 155 | videoclip = moviepy.editor.VideoClip(make_frame, duration=duration_sec) 156 | videoclip.set_duration(total_duration) 157 | 158 | # Change the video parameters (codec, bitrate) if you so desire 159 | final_video = os.path.join(run_dir, f'{mp4_name}.mp4') 160 | videoclip.write_videofile(final_video, fps=fps, codec='libx264', bitrate='16M') 161 | 162 | # Compress the video (lower file size, same resolution) 163 | if compress: 164 | compress_video(original_video=final_video, original_video_name=mp4_name, outdir=run_dir, ctx=ctx) 165 | 166 | 167 | # ---------------------------------------------------------------------------- 168 | 169 | 170 | def project(): 171 | a = np.random.RandomState(0).randn(1, 512) 172 | a = G.mapping(torch.from_numpy(a).to(device), None) 173 | b = torch.from_numpy(np.load('path')).to(device) 174 | 175 | proj_ab = b * torch.sum(a * b) / (b.square().sum()) 176 | proj_perp = a - proj_ab 177 | 178 | img_a = w_to_img(G, a)[0] 179 | img_b = w_to_img(G, b)[0] 180 | img_projab = w_to_img(G, proj_ab)[0] 181 | img_projperp = w_to_img(G, proj_perp)[0] 182 | 183 | 184 | # ---------------------------------------------------------------------------- 185 | 186 | 187 | def n_pendulum(): 188 | pass 189 | 190 | 191 | # ---------------------------------------------------------------------------- 192 | 193 | 194 | def circular(): 195 | pass 196 | 197 | 198 | # ---------------------------------------------------------------------------- 199 | 200 | 201 | if __name__ == '__main__': 202 | main() 203 | -------------------------------------------------------------------------------- /calc_metrics.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Calculate quality metrics for previous training run or pretrained network pickle.""" 10 | 11 | import os 12 | import click 13 | import json 14 | import tempfile 15 | import copy 16 | import torch 17 | import dnnlib 18 | 19 | import legacy 20 | from metrics import metric_main 21 | from metrics import metric_utils 22 | from torch_utils import training_stats 23 | from torch_utils import custom_ops 24 | from torch_utils import misc 25 | 26 | #---------------------------------------------------------------------------- 27 | 28 | def subprocess_fn(rank, args, temp_dir): 29 | dnnlib.util.Logger(should_flush=True) 30 | 31 | # Init torch.distributed. 32 | if args.num_gpus > 1: 33 | init_file = os.path.abspath(os.path.join(temp_dir, '.torch_distributed_init')) 34 | if os.name == 'nt': 35 | init_method = 'file:///' + init_file.replace('\\', '/') 36 | torch.distributed.init_process_group(backend='gloo', init_method=init_method, rank=rank, world_size=args.num_gpus) 37 | else: 38 | init_method = f'file://{init_file}' 39 | torch.distributed.init_process_group(backend='nccl', init_method=init_method, rank=rank, world_size=args.num_gpus) 40 | 41 | # Init torch_utils. 42 | sync_device = torch.device('cuda', rank) if args.num_gpus > 1 else None 43 | training_stats.init_multiprocessing(rank=rank, sync_device=sync_device) 44 | if rank != 0 or not args.verbose: 45 | custom_ops.verbosity = 'none' 46 | 47 | # Print network summary. 48 | device = torch.device('cuda', rank) 49 | torch.backends.cudnn.benchmark = True 50 | torch.backends.cuda.matmul.allow_tf32 = False 51 | torch.backends.cudnn.allow_tf32 = False 52 | G = copy.deepcopy(args.G).eval().requires_grad_(False).to(device) 53 | if rank == 0 and args.verbose: 54 | z = torch.empty([1, G.z_dim], device=device) 55 | c = torch.empty([1, G.c_dim], device=device) 56 | misc.print_module_summary(G, [z, c]) 57 | 58 | # Calculate each metric. 59 | for metric in args.metrics: 60 | if rank == 0 and args.verbose: 61 | print(f'Calculating {metric}...') 62 | progress = metric_utils.ProgressMonitor(verbose=args.verbose) 63 | result_dict = metric_main.calc_metric(metric=metric, G=G, dataset_kwargs=args.dataset_kwargs, 64 | num_gpus=args.num_gpus, rank=rank, device=device, progress=progress) 65 | if rank == 0: 66 | metric_main.report_metric(result_dict, run_dir=args.run_dir, snapshot_pkl=args.network_pkl) 67 | if rank == 0 and args.verbose: 68 | print() 69 | 70 | # Done. 71 | if rank == 0 and args.verbose: 72 | print('Exiting...') 73 | 74 | #---------------------------------------------------------------------------- 75 | 76 | class CommaSeparatedList(click.ParamType): 77 | name = 'list' 78 | 79 | def convert(self, value, param, ctx): 80 | _ = param, ctx 81 | if value is None or value.lower() == 'none' or value == '': 82 | return [] 83 | return value.split(',') 84 | 85 | #---------------------------------------------------------------------------- 86 | 87 | @click.command() 88 | @click.pass_context 89 | @click.option('network_pkl', '--network', help='Network pickle filename or URL', metavar='PATH', required=True) 90 | @click.option('--metrics', help='Comma-separated list or "none"', type=CommaSeparatedList(), default='fid50k_full', show_default=True) 91 | @click.option('--data', help='Dataset to evaluate metrics against (directory or zip) [default: same as training data]', metavar='PATH') 92 | @click.option('--mirror', help='Whether the dataset was augmented with x-flips during training [default: look up]', type=bool, metavar='BOOL') 93 | @click.option('--gpus', help='Number of GPUs to use', type=int, default=1, metavar='INT', show_default=True) 94 | @click.option('--verbose', help='Print optional information', type=bool, default=True, metavar='BOOL', show_default=True) 95 | 96 | def calc_metrics(ctx, network_pkl, metrics, data, mirror, gpus, verbose): 97 | """Calculate quality metrics for previous training run or pretrained network pickle. 98 | 99 | Examples: 100 | 101 | \b 102 | # Previous training run: look up options automatically, save result to JSONL file. 103 | python calc_metrics.py --metrics=pr50k3_full \\ 104 | --network=~/training-runs/00000-ffhq10k-res64-auto1/network-snapshot-000000.pkl 105 | 106 | \b 107 | # Pre-trained network pickle: specify dataset explicitly, print result to stdout. 108 | python calc_metrics.py --metrics=fid50k_full --data=~/datasets/ffhq.zip --mirror=1 \\ 109 | --network=https://nvlabs-fi-cdn.nvidia.com/stylegan2-ada-pytorch/pretrained/ffhq.pkl 110 | 111 | Available metrics: 112 | 113 | \b 114 | ADA paper: 115 | fid50k_full Frechet inception distance against the full dataset. 116 | kid50k_full Kernel inception distance against the full dataset. 117 | pr50k3_full Precision and recall againt the full dataset. 118 | is50k Inception score for CIFAR-10. 119 | 120 | \b 121 | StyleGAN and StyleGAN2 papers: 122 | fid50k Frechet inception distance against 50k real images. 123 | kid50k Kernel inception distance against 50k real images. 124 | pr50k3 Precision and recall against 50k real images. 125 | ppl2_wend Perceptual path length in W at path endpoints against full image. 126 | ppl_zfull Perceptual path length in Z for full paths against cropped image. 127 | ppl_wfull Perceptual path length in W for full paths against cropped image. 128 | ppl_zend Perceptual path length in Z at path endpoints against cropped image. 129 | ppl_wend Perceptual path length in W at path endpoints against cropped image. 130 | """ 131 | dnnlib.util.Logger(should_flush=True) 132 | 133 | # Validate arguments. 134 | args = dnnlib.EasyDict(metrics=metrics, num_gpus=gpus, network_pkl=network_pkl, verbose=verbose) 135 | if not all(metric_main.is_valid_metric(metric) for metric in args.metrics): 136 | ctx.fail('\n'.join(['--metrics can only contain the following values:'] + metric_main.list_valid_metrics())) 137 | if not args.num_gpus >= 1: 138 | ctx.fail('--gpus must be at least 1') 139 | 140 | # Load network. 141 | if not dnnlib.util.is_url(network_pkl, allow_file_urls=True) and not os.path.isfile(network_pkl): 142 | ctx.fail('--network must point to a file or URL') 143 | if args.verbose: 144 | print(f'Loading network from "{network_pkl}"...') 145 | with dnnlib.util.open_url(network_pkl, verbose=args.verbose) as f: 146 | network_dict = legacy.load_network_pkl(f) 147 | args.G = network_dict['G_ema'] # subclass of torch.nn.Module 148 | 149 | # Initialize dataset options. 150 | if data is not None: 151 | args.dataset_kwargs = dnnlib.EasyDict(class_name='training.dataset.ImageFolderDataset', path=data) 152 | elif network_dict['training_set_kwargs'] is not None: 153 | args.dataset_kwargs = dnnlib.EasyDict(network_dict['training_set_kwargs']) 154 | else: 155 | ctx.fail('Could not look up dataset options; please specify --data') 156 | 157 | # Finalize dataset options. 158 | args.dataset_kwargs.resolution = args.G.img_resolution 159 | args.dataset_kwargs.use_labels = (args.G.c_dim != 0) 160 | if mirror is not None: 161 | args.dataset_kwargs.xflip = mirror 162 | 163 | # Print dataset options. 164 | if args.verbose: 165 | print('Dataset options:') 166 | print(json.dumps(args.dataset_kwargs, indent=2)) 167 | 168 | # Locate run dir. 169 | args.run_dir = None 170 | if os.path.isfile(network_pkl): 171 | pkl_dir = os.path.dirname(network_pkl) 172 | if os.path.isfile(os.path.join(pkl_dir, 'training_options.json')): 173 | args.run_dir = pkl_dir 174 | 175 | # Launch processes. 176 | if args.verbose: 177 | print('Launching processes...') 178 | torch.multiprocessing.set_start_method('spawn') 179 | with tempfile.TemporaryDirectory() as temp_dir: 180 | if args.num_gpus == 1: 181 | subprocess_fn(rank=0, args=args, temp_dir=temp_dir) 182 | else: 183 | torch.multiprocessing.spawn(fn=subprocess_fn, args=(args, temp_dir), nprocs=args.num_gpus) 184 | 185 | #---------------------------------------------------------------------------- 186 | 187 | if __name__ == "__main__": 188 | calc_metrics() # pylint: disable=no-value-for-parameter 189 | 190 | #---------------------------------------------------------------------------- 191 | -------------------------------------------------------------------------------- /training/dataset.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | import os 10 | import numpy as np 11 | import zipfile 12 | import PIL.Image 13 | import json 14 | import torch 15 | import dnnlib 16 | 17 | try: 18 | import pyspng 19 | except ImportError: 20 | pyspng = None 21 | 22 | 23 | # ---------------------------------------------------------------------------- 24 | 25 | 26 | class Dataset(torch.utils.data.Dataset): 27 | def __init__(self, 28 | name, # Name of the dataset. 29 | raw_shape, # Shape of the raw image data (NCHW). 30 | max_size = None, # Artificially limit the size of the dataset. None = no limit. Applied before xflip. 31 | use_labels = False, # Enable conditioning labels? False = label dimension is zero. 32 | xflip = False, # Artificially double the size of the dataset via x-flips. Applied after max_size. 33 | yflip = False, # Artificially double the size of the dataset via y-flips. Applied after xflip. 34 | random_seed = 0, # Random seed to use when applying max_size. 35 | ): 36 | self._name = name 37 | self._raw_shape = list(raw_shape) 38 | self._use_labels = use_labels 39 | self._raw_labels = None 40 | self._label_shape = None 41 | 42 | # Apply max_size. 43 | self._raw_idx = np.arange(self._raw_shape[0], dtype=np.int64) 44 | if (max_size is not None) and (self._raw_idx.size > max_size): 45 | np.random.RandomState(random_seed).shuffle(self._raw_idx) 46 | self._raw_idx = np.sort(self._raw_idx[:max_size]) 47 | 48 | # Apply xflip. 49 | self._xflip = np.zeros(self._raw_idx.size, dtype=np.uint8) 50 | if xflip: 51 | self._raw_idx = np.tile(self._raw_idx, 2) # double the size of images 52 | self._xflip = np.concatenate([self._xflip, np.ones_like(self._xflip)]) 53 | 54 | # Apply yflip. 55 | self._yflip = np.zeros(self._raw_idx.size, dtype=np.uint8) 56 | if yflip: 57 | self._raw_idx = np.tile(self._raw_idx, 2) # double the size of images 58 | self._yflip = np.concatenate([self._yflip, np.ones_like(self._yflip)]) 59 | self._xflip = np.tile(self._xflip, 2) # double the indices for xflip, otherwise we get out of bounds 60 | 61 | def _get_raw_labels(self): 62 | if self._raw_labels is None: 63 | self._raw_labels = self._load_raw_labels() if self._use_labels else None 64 | if self._raw_labels is None: 65 | self._raw_labels = np.zeros([self._raw_shape[0], 0], dtype=np.float32) 66 | assert isinstance(self._raw_labels, np.ndarray) 67 | assert self._raw_labels.shape[0] == self._raw_shape[0] 68 | assert self._raw_labels.dtype in [np.float32, np.int64] 69 | if self._raw_labels.dtype == np.int64: 70 | assert self._raw_labels.ndim == 1 71 | assert np.all(self._raw_labels >= 0) 72 | return self._raw_labels 73 | 74 | def close(self): # to be overridden by subclass 75 | pass 76 | 77 | def _load_raw_image(self, raw_idx): # to be overridden by subclass 78 | raise NotImplementedError 79 | 80 | def _load_raw_labels(self): # to be overridden by subclass 81 | raise NotImplementedError 82 | 83 | def __getstate__(self): 84 | return dict(self.__dict__, _raw_labels=None) 85 | 86 | def __del__(self): 87 | try: 88 | self.close() 89 | except: 90 | pass 91 | 92 | def __len__(self): 93 | return self._raw_idx.size 94 | 95 | def __getitem__(self, idx): 96 | image = self._load_raw_image(self._raw_idx[idx]) 97 | assert isinstance(image, np.ndarray) 98 | assert list(image.shape) == self.image_shape 99 | assert image.dtype == np.uint8 100 | if self._xflip[idx]: 101 | assert image.ndim == 3 # CHW 102 | image = image[:, :, ::-1] 103 | if self._yflip[idx]: 104 | assert image.ndim == 3 # CHW 105 | image = image[:, ::-1, :] 106 | return image.copy(), self.get_label(idx) 107 | 108 | def get_label(self, idx): 109 | label = self._get_raw_labels()[self._raw_idx[idx]] 110 | if label.dtype == np.int64: 111 | onehot = np.zeros(self.label_shape, dtype=np.float32) 112 | onehot[label] = 1 113 | label = onehot 114 | return label.copy() 115 | 116 | def get_details(self, idx): 117 | d = dnnlib.EasyDict() 118 | d.raw_idx = int(self._raw_idx[idx]) 119 | d.xflip = (int(self._xflip[idx]) != 0) 120 | d.yflip = (int(self._yflip[idx]) != 0) 121 | d.raw_label = self._get_raw_labels()[d.raw_idx].copy() 122 | return d 123 | 124 | @property 125 | def name(self): 126 | return self._name 127 | 128 | @property 129 | def image_shape(self): 130 | return list(self._raw_shape[1:]) 131 | 132 | @property 133 | def num_channels(self): 134 | assert len(self.image_shape) == 3 # CHW 135 | return self.image_shape[0] 136 | 137 | @property 138 | def resolution(self): 139 | assert len(self.image_shape) == 3 # CHW 140 | assert self.image_shape[1] == self.image_shape[2] 141 | return self.image_shape[1] 142 | 143 | @property 144 | def label_shape(self): 145 | if self._label_shape is None: 146 | raw_labels = self._get_raw_labels() 147 | if raw_labels.dtype == np.int64: 148 | self._label_shape = [int(np.max(raw_labels)) + 1] 149 | else: 150 | self._label_shape = raw_labels.shape[1:] 151 | return list(self._label_shape) 152 | 153 | @property 154 | def label_dim(self): 155 | assert len(self.label_shape) == 1 156 | return self.label_shape[0] 157 | 158 | @property 159 | def has_labels(self): 160 | return any(x != 0 for x in self.label_shape) 161 | 162 | @property 163 | def has_onehot_labels(self): 164 | return self._get_raw_labels().dtype == np.int64 165 | 166 | 167 | # ---------------------------------------------------------------------------- 168 | 169 | 170 | class ImageFolderDataset(Dataset): 171 | def __init__(self, 172 | path, # Path to directory or zip. 173 | resolution = None, # Ensure specific resolution, None = highest available. 174 | **super_kwargs, # Additional arguments for the Dataset base class. 175 | ): 176 | self._path = path 177 | self._zipfile = None 178 | 179 | if os.path.isdir(self._path): 180 | self._type = 'dir' 181 | self._all_fnames = {os.path.relpath(os.path.join(root, fname), start=self._path) for root, _dirs, files in os.walk(self._path) for fname in files} 182 | elif self._file_ext(self._path) == '.zip': 183 | self._type = 'zip' 184 | self._all_fnames = set(self._get_zipfile().namelist()) 185 | else: 186 | raise IOError('Path must point to a directory or zip') 187 | 188 | PIL.Image.init() 189 | self._image_fnames = sorted(fname for fname in self._all_fnames if self._file_ext(fname) in PIL.Image.EXTENSION) 190 | if len(self._image_fnames) == 0: 191 | raise IOError('No image files found in the specified path') 192 | 193 | name = os.path.splitext(os.path.basename(self._path))[0] 194 | raw_shape = [len(self._image_fnames)] + list(self._load_raw_image(0).shape) 195 | if resolution is not None and (raw_shape[2] != resolution or raw_shape[3] != resolution): 196 | raise IOError('Image files do not match the specified resolution') 197 | super().__init__(name=name, raw_shape=raw_shape, **super_kwargs) 198 | 199 | @staticmethod 200 | def _file_ext(fname): 201 | return os.path.splitext(fname)[1].lower() 202 | 203 | def _get_zipfile(self): 204 | assert self._type == 'zip' 205 | if self._zipfile is None: 206 | self._zipfile = zipfile.ZipFile(self._path) 207 | return self._zipfile 208 | 209 | def _open_file(self, fname): 210 | if self._type == 'dir': 211 | return open(os.path.join(self._path, fname), 'rb') 212 | if self._type == 'zip': 213 | return self._get_zipfile().open(fname, 'r') 214 | return None 215 | 216 | def close(self): 217 | try: 218 | if self._zipfile is not None: 219 | self._zipfile.close() 220 | finally: 221 | self._zipfile = None 222 | 223 | def __getstate__(self): 224 | return dict(super().__getstate__(), _zipfile=None) 225 | 226 | def _load_raw_image(self, raw_idx): 227 | fname = self._image_fnames[raw_idx] 228 | with self._open_file(fname) as f: 229 | if pyspng is not None and self._file_ext(fname) == '.png': 230 | image = pyspng.load(f.read()) 231 | else: 232 | image = np.array(PIL.Image.open(f)) 233 | if image.ndim == 2: 234 | image = image[:, :, np.newaxis] # HW => HWC 235 | image = image.transpose(2, 0, 1) # HWC => CHW 236 | return image 237 | 238 | def _load_raw_labels(self): 239 | fname = 'dataset.json' 240 | if fname not in self._all_fnames: 241 | return None 242 | with self._open_file(fname) as f: 243 | labels = json.load(f)['labels'] 244 | if labels is None: 245 | return None 246 | labels = dict(labels) 247 | labels = [labels[fname.replace('\\', '/')] for fname in self._image_fnames] 248 | labels = np.array(labels) 249 | labels = labels.astype({1: np.int64, 2: np.float32}[labels.ndim]) 250 | return labels 251 | 252 | -------------------------------------------------------------------------------- /torch_utils/persistence.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Facilities for pickling Python code alongside other data. 10 | 11 | The pickled code is automatically imported into a separate Python module 12 | during unpickling. This way, any previously exported pickles will remain 13 | usable even if the original code is no longer available, or if the current 14 | version of the code is not consistent with what was originally pickled.""" 15 | 16 | import sys 17 | import pickle 18 | import io 19 | import inspect 20 | import copy 21 | import uuid 22 | import types 23 | import dnnlib 24 | 25 | #---------------------------------------------------------------------------- 26 | 27 | _version = 6 # internal version number 28 | _decorators = set() # {decorator_class, ...} 29 | _import_hooks = [] # [hook_function, ...] 30 | _module_to_src_dict = dict() # {module: src, ...} 31 | _src_to_module_dict = dict() # {src: module, ...} 32 | 33 | #---------------------------------------------------------------------------- 34 | 35 | def persistent_class(orig_class): 36 | r"""Class decorator that extends a given class to save its source code 37 | when pickled. 38 | 39 | Example: 40 | 41 | from torch_utils import persistence 42 | 43 | @persistence.persistent_class 44 | class MyNetwork(torch.nn.Module): 45 | def __init__(self, num_inputs, num_outputs): 46 | super().__init__() 47 | self.fc = MyLayer(num_inputs, num_outputs) 48 | ... 49 | 50 | @persistence.persistent_class 51 | class MyLayer(torch.nn.Module): 52 | ... 53 | 54 | When pickled, any instance of `MyNetwork` and `MyLayer` will save its 55 | source code alongside other internal state (e.g., parameters, buffers, 56 | and submodules). This way, any previously exported pickle will remain 57 | usable even if the class definitions have been modified or are no 58 | longer available. 59 | 60 | The decorator saves the source code of the entire Python module 61 | containing the decorated class. It does *not* save the source code of 62 | any imported modules. Thus, the imported modules must be available 63 | during unpickling, also including `torch_utils.persistence` itself. 64 | 65 | It is ok to call functions defined in the same module from the 66 | decorated class. However, if the decorated class depends on other 67 | classes defined in the same module, they must be decorated as well. 68 | This is illustrated in the above example in the case of `MyLayer`. 69 | 70 | It is also possible to employ the decorator just-in-time before 71 | calling the constructor. For example: 72 | 73 | cls = MyLayer 74 | if want_to_make_it_persistent: 75 | cls = persistence.persistent_class(cls) 76 | layer = cls(num_inputs, num_outputs) 77 | 78 | As an additional feature, the decorator also keeps track of the 79 | arguments that were used to construct each instance of the decorated 80 | class. The arguments can be queried via `obj.init_args` and 81 | `obj.init_kwargs`, and they are automatically pickled alongside other 82 | object state. A typical use case is to first unpickle a previous 83 | instance of a persistent class, and then upgrade it to use the latest 84 | version of the source code: 85 | 86 | with open('old_pickle.pkl', 'rb') as f: 87 | old_net = pickle.load(f) 88 | new_net = MyNetwork(*old_obj.init_args, **old_obj.init_kwargs) 89 | misc.copy_params_and_buffers(old_net, new_net, require_all=True) 90 | """ 91 | assert isinstance(orig_class, type) 92 | if is_persistent(orig_class): 93 | return orig_class 94 | 95 | assert orig_class.__module__ in sys.modules 96 | orig_module = sys.modules[orig_class.__module__] 97 | orig_module_src = _module_to_src(orig_module) 98 | 99 | class Decorator(orig_class): 100 | _orig_module_src = orig_module_src 101 | _orig_class_name = orig_class.__name__ 102 | 103 | def __init__(self, *args, **kwargs): 104 | super().__init__(*args, **kwargs) 105 | self._init_args = copy.deepcopy(args) 106 | self._init_kwargs = copy.deepcopy(kwargs) 107 | assert orig_class.__name__ in orig_module.__dict__ 108 | _check_pickleable(self.__reduce__()) 109 | 110 | @property 111 | def init_args(self): 112 | return copy.deepcopy(self._init_args) 113 | 114 | @property 115 | def init_kwargs(self): 116 | return dnnlib.EasyDict(copy.deepcopy(self._init_kwargs)) 117 | 118 | def __reduce__(self): 119 | fields = list(super().__reduce__()) 120 | fields += [None] * max(3 - len(fields), 0) 121 | if fields[0] is not _reconstruct_persistent_obj: 122 | meta = dict(type='class', version=_version, module_src=self._orig_module_src, class_name=self._orig_class_name, state=fields[2]) 123 | fields[0] = _reconstruct_persistent_obj # reconstruct func 124 | fields[1] = (meta,) # reconstruct args 125 | fields[2] = None # state dict 126 | return tuple(fields) 127 | 128 | Decorator.__name__ = orig_class.__name__ 129 | _decorators.add(Decorator) 130 | return Decorator 131 | 132 | #---------------------------------------------------------------------------- 133 | 134 | def is_persistent(obj): 135 | r"""Test whether the given object or class is persistent, i.e., 136 | whether it will save its source code when pickled. 137 | """ 138 | try: 139 | if obj in _decorators: 140 | return True 141 | except TypeError: 142 | pass 143 | return type(obj) in _decorators # pylint: disable=unidiomatic-typecheck 144 | 145 | #---------------------------------------------------------------------------- 146 | 147 | def import_hook(hook): 148 | r"""Register an import hook that is called whenever a persistent object 149 | is being unpickled. A typical use case is to patch the pickled source 150 | code to avoid errors and inconsistencies when the API of some imported 151 | module has changed. 152 | 153 | The hook should have the following signature: 154 | 155 | hook(meta) -> modified meta 156 | 157 | `meta` is an instance of `dnnlib.EasyDict` with the following fields: 158 | 159 | type: Type of the persistent object, e.g. `'class'`. 160 | version: Internal version number of `torch_utils.persistence`. 161 | module_src Original source code of the Python module. 162 | class_name: Class name in the original Python module. 163 | state: Internal state of the object. 164 | 165 | Example: 166 | 167 | @persistence.import_hook 168 | def wreck_my_network(meta): 169 | if meta.class_name == 'MyNetwork': 170 | print('MyNetwork is being imported. I will wreck it!') 171 | meta.module_src = meta.module_src.replace("True", "False") 172 | return meta 173 | """ 174 | assert callable(hook) 175 | _import_hooks.append(hook) 176 | 177 | #---------------------------------------------------------------------------- 178 | 179 | def _reconstruct_persistent_obj(meta): 180 | r"""Hook that is called internally by the `pickle` module to unpickle 181 | a persistent object. 182 | """ 183 | meta = dnnlib.EasyDict(meta) 184 | meta.state = dnnlib.EasyDict(meta.state) 185 | for hook in _import_hooks: 186 | meta = hook(meta) 187 | assert meta is not None 188 | 189 | assert meta.version == _version 190 | module = _src_to_module(meta.module_src) 191 | 192 | assert meta.type == 'class' 193 | orig_class = module.__dict__[meta.class_name] 194 | decorator_class = persistent_class(orig_class) 195 | obj = decorator_class.__new__(decorator_class) 196 | 197 | setstate = getattr(obj, '__setstate__', None) 198 | if callable(setstate): 199 | setstate(meta.state) # pylint: disable=not-callable 200 | else: 201 | obj.__dict__.update(meta.state) 202 | return obj 203 | 204 | #---------------------------------------------------------------------------- 205 | 206 | def _module_to_src(module): 207 | r"""Query the source code of a given Python module. 208 | """ 209 | src = _module_to_src_dict.get(module, None) 210 | if src is None: 211 | src = inspect.getsource(module) 212 | _module_to_src_dict[module] = src 213 | _src_to_module_dict[src] = module 214 | return src 215 | 216 | def _src_to_module(src): 217 | r"""Get or create a Python module for the given source code. 218 | """ 219 | module = _src_to_module_dict.get(src, None) 220 | if module is None: 221 | module_name = "_imported_module_" + uuid.uuid4().hex 222 | module = types.ModuleType(module_name) 223 | sys.modules[module_name] = module 224 | _module_to_src_dict[module] = src 225 | _src_to_module_dict[src] = module 226 | exec(src, module.__dict__) # pylint: disable=exec-used 227 | return module 228 | 229 | #---------------------------------------------------------------------------- 230 | 231 | def _check_pickleable(obj): 232 | r"""Check that the given object is pickleable, raising an exception if 233 | it is not. This function is expected to be considerably more efficient 234 | than actually pickling the object. 235 | """ 236 | def recurse(obj): 237 | if isinstance(obj, (list, tuple, set)): 238 | return [recurse(x) for x in obj] 239 | if isinstance(obj, dict): 240 | return [[recurse(x), recurse(y)] for x, y in obj.items()] 241 | if isinstance(obj, (str, int, float, bool, bytes, bytearray)): 242 | return None # Python primitive types are pickleable. 243 | if f'{type(obj).__module__}.{type(obj).__name__}' in ['numpy.ndarray', 'torch.Tensor']: 244 | return None # NumPy arrays and PyTorch tensors are pickleable. 245 | if is_persistent(obj): 246 | return None # Persistent objects are pickleable, by virtue of the constructor check. 247 | return obj 248 | with io.BytesIO() as f: 249 | pickle.dump(recurse(obj), f) 250 | 251 | #---------------------------------------------------------------------------- 252 | -------------------------------------------------------------------------------- /torch_utils/ops/bias_act.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Custom PyTorch ops for efficient bias and activation.""" 10 | 11 | import os 12 | import warnings 13 | import numpy as np 14 | import torch 15 | import dnnlib 16 | import traceback 17 | 18 | from .. import custom_ops 19 | from .. import misc 20 | 21 | #---------------------------------------------------------------------------- 22 | 23 | activation_funcs = { 24 | 'linear': dnnlib.EasyDict(func=lambda x, **_: x, def_alpha=0, def_gain=1, cuda_idx=1, ref='', has_2nd_grad=False), 25 | 'relu': dnnlib.EasyDict(func=lambda x, **_: torch.nn.functional.relu(x), def_alpha=0, def_gain=np.sqrt(2), cuda_idx=2, ref='y', has_2nd_grad=False), 26 | 'lrelu': dnnlib.EasyDict(func=lambda x, alpha, **_: torch.nn.functional.leaky_relu(x, alpha), def_alpha=0.2, def_gain=np.sqrt(2), cuda_idx=3, ref='y', has_2nd_grad=False), 27 | 'tanh': dnnlib.EasyDict(func=lambda x, **_: torch.tanh(x), def_alpha=0, def_gain=1, cuda_idx=4, ref='y', has_2nd_grad=True), 28 | 'sigmoid': dnnlib.EasyDict(func=lambda x, **_: torch.sigmoid(x), def_alpha=0, def_gain=1, cuda_idx=5, ref='y', has_2nd_grad=True), 29 | 'elu': dnnlib.EasyDict(func=lambda x, **_: torch.nn.functional.elu(x), def_alpha=0, def_gain=1, cuda_idx=6, ref='y', has_2nd_grad=True), 30 | 'selu': dnnlib.EasyDict(func=lambda x, **_: torch.nn.functional.selu(x), def_alpha=0, def_gain=1, cuda_idx=7, ref='y', has_2nd_grad=True), 31 | 'softplus': dnnlib.EasyDict(func=lambda x, **_: torch.nn.functional.softplus(x), def_alpha=0, def_gain=1, cuda_idx=8, ref='y', has_2nd_grad=True), 32 | 'swish': dnnlib.EasyDict(func=lambda x, **_: torch.sigmoid(x) * x, def_alpha=0, def_gain=np.sqrt(2), cuda_idx=9, ref='x', has_2nd_grad=True), 33 | } 34 | 35 | #---------------------------------------------------------------------------- 36 | 37 | _inited = False 38 | _plugin = None 39 | _null_tensor = torch.empty([0]) 40 | 41 | def _init(): 42 | global _inited, _plugin 43 | if not _inited: 44 | _inited = True 45 | sources = ['bias_act.cpp', 'bias_act.cu'] 46 | sources = [os.path.join(os.path.dirname(__file__), s) for s in sources] 47 | try: 48 | _plugin = custom_ops.get_plugin('bias_act_plugin', sources=sources, extra_cuda_cflags=['--use_fast_math']) 49 | except: 50 | warnings.warn('Failed to build CUDA kernels for bias_act. Falling back to slow reference implementation. Details:\n\n' + traceback.format_exc()) 51 | return _plugin is not None 52 | 53 | #---------------------------------------------------------------------------- 54 | 55 | def bias_act(x, b=None, dim=1, act='linear', alpha=None, gain=None, clamp=None, impl='cuda'): 56 | r"""Fused bias and activation function. 57 | 58 | Adds bias `b` to activation tensor `x`, evaluates activation function `act`, 59 | and scales the result by `gain`. Each of the steps is optional. In most cases, 60 | the fused op is considerably more efficient than performing the same calculation 61 | using standard PyTorch ops. It supports first and second order gradients, 62 | but not third order gradients. 63 | 64 | Args: 65 | x: Input activation tensor. Can be of any shape. 66 | b: Bias vector, or `None` to disable. Must be a 1D tensor of the same type 67 | as `x`. The shape must be known, and it must match the dimension of `x` 68 | corresponding to `dim`. 69 | dim: The dimension in `x` corresponding to the elements of `b`. 70 | The value of `dim` is ignored if `b` is not specified. 71 | act: Name of the activation function to evaluate, or `"linear"` to disable. 72 | Can be e.g. `"relu"`, `"lrelu"`, `"tanh"`, `"sigmoid"`, `"swish"`, etc. 73 | See `activation_funcs` for a full list. `None` is not allowed. 74 | alpha: Shape parameter for the activation function, or `None` to use the default. 75 | gain: Scaling factor for the output tensor, or `None` to use default. 76 | See `activation_funcs` for the default scaling of each activation function. 77 | If unsure, consider specifying 1. 78 | clamp: Clamp the output values to `[-clamp, +clamp]`, or `None` to disable 79 | the clamping (default). 80 | impl: Name of the implementation to use. Can be `"ref"` or `"cuda"` (default). 81 | 82 | Returns: 83 | Tensor of the same shape and datatype as `x`. 84 | """ 85 | assert isinstance(x, torch.Tensor) 86 | assert impl in ['ref', 'cuda'] 87 | if impl == 'cuda' and x.device.type == 'cuda' and _init(): 88 | return _bias_act_cuda(dim=dim, act=act, alpha=alpha, gain=gain, clamp=clamp).apply(x, b) 89 | return _bias_act_ref(x=x, b=b, dim=dim, act=act, alpha=alpha, gain=gain, clamp=clamp) 90 | 91 | #---------------------------------------------------------------------------- 92 | 93 | @misc.profiled_function 94 | def _bias_act_ref(x, b=None, dim=1, act='linear', alpha=None, gain=None, clamp=None): 95 | """Slow reference implementation of `bias_act()` using standard TensorFlow ops. 96 | """ 97 | assert isinstance(x, torch.Tensor) 98 | assert clamp is None or clamp >= 0 99 | spec = activation_funcs[act] 100 | alpha = float(alpha if alpha is not None else spec.def_alpha) 101 | gain = float(gain if gain is not None else spec.def_gain) 102 | clamp = float(clamp if clamp is not None else -1) 103 | 104 | # Add bias. 105 | if b is not None: 106 | assert isinstance(b, torch.Tensor) and b.ndim == 1 107 | assert 0 <= dim < x.ndim 108 | assert b.shape[0] == x.shape[dim] 109 | x = x + b.reshape([-1 if i == dim else 1 for i in range(x.ndim)]) 110 | 111 | # Evaluate activation function. 112 | alpha = float(alpha) 113 | x = spec.func(x, alpha=alpha) 114 | 115 | # Scale by gain. 116 | gain = float(gain) 117 | if gain != 1: 118 | x = x * gain 119 | 120 | # Clamp. 121 | if clamp >= 0: 122 | x = x.clamp(-clamp, clamp) # pylint: disable=invalid-unary-operand-type 123 | return x 124 | 125 | #---------------------------------------------------------------------------- 126 | 127 | _bias_act_cuda_cache = dict() 128 | 129 | def _bias_act_cuda(dim=1, act='linear', alpha=None, gain=None, clamp=None): 130 | """Fast CUDA implementation of `bias_act()` using custom ops. 131 | """ 132 | # Parse arguments. 133 | assert clamp is None or clamp >= 0 134 | spec = activation_funcs[act] 135 | alpha = float(alpha if alpha is not None else spec.def_alpha) 136 | gain = float(gain if gain is not None else spec.def_gain) 137 | clamp = float(clamp if clamp is not None else -1) 138 | 139 | # Lookup from cache. 140 | key = (dim, act, alpha, gain, clamp) 141 | if key in _bias_act_cuda_cache: 142 | return _bias_act_cuda_cache[key] 143 | 144 | # Forward op. 145 | class BiasActCuda(torch.autograd.Function): 146 | @staticmethod 147 | def forward(ctx, x, b): # pylint: disable=arguments-differ 148 | ctx.memory_format = torch.channels_last if x.ndim > 2 and x.stride()[1] == 1 else torch.contiguous_format 149 | x = x.contiguous(memory_format=ctx.memory_format) 150 | b = b.contiguous() if b is not None else _null_tensor 151 | y = x 152 | if act != 'linear' or gain != 1 or clamp >= 0 or b is not _null_tensor: 153 | y = _plugin.bias_act(x, b, _null_tensor, _null_tensor, _null_tensor, 0, dim, spec.cuda_idx, alpha, gain, clamp) 154 | ctx.save_for_backward( 155 | x if 'x' in spec.ref or spec.has_2nd_grad else _null_tensor, 156 | b if 'x' in spec.ref or spec.has_2nd_grad else _null_tensor, 157 | y if 'y' in spec.ref else _null_tensor) 158 | return y 159 | 160 | @staticmethod 161 | def backward(ctx, dy): # pylint: disable=arguments-differ 162 | dy = dy.contiguous(memory_format=ctx.memory_format) 163 | x, b, y = ctx.saved_tensors 164 | dx = None 165 | db = None 166 | 167 | if ctx.needs_input_grad[0] or ctx.needs_input_grad[1]: 168 | dx = dy 169 | if act != 'linear' or gain != 1 or clamp >= 0: 170 | dx = BiasActCudaGrad.apply(dy, x, b, y) 171 | 172 | if ctx.needs_input_grad[1]: 173 | db = dx.sum([i for i in range(dx.ndim) if i != dim]) 174 | 175 | return dx, db 176 | 177 | # Backward op. 178 | class BiasActCudaGrad(torch.autograd.Function): 179 | @staticmethod 180 | def forward(ctx, dy, x, b, y): # pylint: disable=arguments-differ 181 | ctx.memory_format = torch.channels_last if dy.ndim > 2 and dy.stride()[1] == 1 else torch.contiguous_format 182 | dx = _plugin.bias_act(dy, b, x, y, _null_tensor, 1, dim, spec.cuda_idx, alpha, gain, clamp) 183 | ctx.save_for_backward( 184 | dy if spec.has_2nd_grad else _null_tensor, 185 | x, b, y) 186 | return dx 187 | 188 | @staticmethod 189 | def backward(ctx, d_dx): # pylint: disable=arguments-differ 190 | d_dx = d_dx.contiguous(memory_format=ctx.memory_format) 191 | dy, x, b, y = ctx.saved_tensors 192 | d_dy = None 193 | d_x = None 194 | d_b = None 195 | d_y = None 196 | 197 | if ctx.needs_input_grad[0]: 198 | d_dy = BiasActCudaGrad.apply(d_dx, x, b, y) 199 | 200 | if spec.has_2nd_grad and (ctx.needs_input_grad[1] or ctx.needs_input_grad[2]): 201 | d_x = _plugin.bias_act(d_dx, b, x, y, dy, 2, dim, spec.cuda_idx, alpha, gain, clamp) 202 | 203 | if spec.has_2nd_grad and ctx.needs_input_grad[2]: 204 | d_b = d_x.sum([i for i in range(d_x.ndim) if i != dim]) 205 | 206 | return d_dy, d_x, d_b, d_y 207 | 208 | # Add to cache. 209 | _bias_act_cuda_cache[key] = BiasActCuda 210 | return BiasActCuda 211 | 212 | #---------------------------------------------------------------------------- 213 | -------------------------------------------------------------------------------- /torch_utils/training_stats.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | """Facilities for reporting and collecting training statistics across 10 | multiple processes and devices. The interface is designed to minimize 11 | synchronization overhead as well as the amount of boilerplate in user 12 | code.""" 13 | 14 | import re 15 | import numpy as np 16 | import torch 17 | import dnnlib 18 | 19 | from . import misc 20 | 21 | #---------------------------------------------------------------------------- 22 | 23 | _num_moments = 3 # [num_scalars, sum_of_scalars, sum_of_squares] 24 | _reduce_dtype = torch.float32 # Data type to use for initial per-tensor reduction. 25 | _counter_dtype = torch.float64 # Data type to use for the internal counters. 26 | _rank = 0 # Rank of the current process. 27 | _sync_device = None # Device to use for multiprocess communication. None = single-process. 28 | _sync_called = False # Has _sync() been called yet? 29 | _counters = dict() # Running counters on each device, updated by report(): name => device => torch.Tensor 30 | _cumulative = dict() # Cumulative counters on the CPU, updated by _sync(): name => torch.Tensor 31 | 32 | #---------------------------------------------------------------------------- 33 | 34 | def init_multiprocessing(rank, sync_device): 35 | r"""Initializes `torch_utils.training_stats` for collecting statistics 36 | across multiple processes. 37 | 38 | This function must be called after 39 | `torch.distributed.init_process_group()` and before `Collector.update()`. 40 | The call is not necessary if multi-process collection is not needed. 41 | 42 | Args: 43 | rank: Rank of the current process. 44 | sync_device: PyTorch device to use for inter-process 45 | communication, or None to disable multi-process 46 | collection. Typically `torch.device('cuda', rank)`. 47 | """ 48 | global _rank, _sync_device 49 | assert not _sync_called 50 | _rank = rank 51 | _sync_device = sync_device 52 | 53 | #---------------------------------------------------------------------------- 54 | 55 | @misc.profiled_function 56 | def report(name, value): 57 | r"""Broadcasts the given set of scalars to all interested instances of 58 | `Collector`, across device and process boundaries. 59 | 60 | This function is expected to be extremely cheap and can be safely 61 | called from anywhere in the training loop, loss function, or inside a 62 | `torch.nn.Module`. 63 | 64 | Warning: The current implementation expects the set of unique names to 65 | be consistent across processes. Please make sure that `report()` is 66 | called at least once for each unique name by each process, and in the 67 | same order. If a given process has no scalars to broadcast, it can do 68 | `report(name, [])` (empty list). 69 | 70 | Args: 71 | name: Arbitrary string specifying the name of the statistic. 72 | Averages are accumulated separately for each unique name. 73 | value: Arbitrary set of scalars. Can be a list, tuple, 74 | NumPy array, PyTorch tensor, or Python scalar. 75 | 76 | Returns: 77 | The same `value` that was passed in. 78 | """ 79 | if name not in _counters: 80 | _counters[name] = dict() 81 | 82 | elems = torch.as_tensor(value) 83 | if elems.numel() == 0: 84 | return value 85 | 86 | elems = elems.detach().flatten().to(_reduce_dtype) 87 | moments = torch.stack([ 88 | torch.ones_like(elems).sum(), 89 | elems.sum(), 90 | elems.square().sum(), 91 | ]) 92 | assert moments.ndim == 1 and moments.shape[0] == _num_moments 93 | moments = moments.to(_counter_dtype) 94 | 95 | device = moments.device 96 | if device not in _counters[name]: 97 | _counters[name][device] = torch.zeros_like(moments) 98 | _counters[name][device].add_(moments) 99 | return value 100 | 101 | #---------------------------------------------------------------------------- 102 | 103 | def report0(name, value): 104 | r"""Broadcasts the given set of scalars by the first process (`rank = 0`), 105 | but ignores any scalars provided by the other processes. 106 | See `report()` for further details. 107 | """ 108 | report(name, value if _rank == 0 else []) 109 | return value 110 | 111 | #---------------------------------------------------------------------------- 112 | 113 | class Collector: 114 | r"""Collects the scalars broadcasted by `report()` and `report0()` and 115 | computes their long-term averages (mean and standard deviation) over 116 | user-defined periods of time. 117 | 118 | The averages are first collected into internal counters that are not 119 | directly visible to the user. They are then copied to the user-visible 120 | state as a result of calling `update()` and can then be queried using 121 | `mean()`, `std()`, `as_dict()`, etc. Calling `update()` also resets the 122 | internal counters for the next round, so that the user-visible state 123 | effectively reflects averages collected between the last two calls to 124 | `update()`. 125 | 126 | Args: 127 | regex: Regular expression defining which statistics to 128 | collect. The default is to collect everything. 129 | keep_previous: Whether to retain the previous averages if no 130 | scalars were collected on a given round 131 | (default: True). 132 | """ 133 | def __init__(self, regex='.*', keep_previous=True): 134 | self._regex = re.compile(regex) 135 | self._keep_previous = keep_previous 136 | self._cumulative = dict() 137 | self._moments = dict() 138 | self.update() 139 | self._moments.clear() 140 | 141 | def names(self): 142 | r"""Returns the names of all statistics broadcasted so far that 143 | match the regular expression specified at construction time. 144 | """ 145 | return [name for name in _counters if self._regex.fullmatch(name)] 146 | 147 | def update(self): 148 | r"""Copies current values of the internal counters to the 149 | user-visible state and resets them for the next round. 150 | 151 | If `keep_previous=True` was specified at construction time, the 152 | operation is skipped for statistics that have received no scalars 153 | since the last update, retaining their previous averages. 154 | 155 | This method performs a number of GPU-to-CPU transfers and one 156 | `torch.distributed.all_reduce()`. It is intended to be called 157 | periodically in the main training loop, typically once every 158 | N training steps. 159 | """ 160 | if not self._keep_previous: 161 | self._moments.clear() 162 | for name, cumulative in _sync(self.names()): 163 | if name not in self._cumulative: 164 | self._cumulative[name] = torch.zeros([_num_moments], dtype=_counter_dtype) 165 | delta = cumulative - self._cumulative[name] 166 | self._cumulative[name].copy_(cumulative) 167 | if float(delta[0]) != 0: 168 | self._moments[name] = delta 169 | 170 | def _get_delta(self, name): 171 | r"""Returns the raw moments that were accumulated for the given 172 | statistic between the last two calls to `update()`, or zero if 173 | no scalars were collected. 174 | """ 175 | assert self._regex.fullmatch(name) 176 | if name not in self._moments: 177 | self._moments[name] = torch.zeros([_num_moments], dtype=_counter_dtype) 178 | return self._moments[name] 179 | 180 | def num(self, name): 181 | r"""Returns the number of scalars that were accumulated for the given 182 | statistic between the last two calls to `update()`, or zero if 183 | no scalars were collected. 184 | """ 185 | delta = self._get_delta(name) 186 | return int(delta[0]) 187 | 188 | def mean(self, name): 189 | r"""Returns the mean of the scalars that were accumulated for the 190 | given statistic between the last two calls to `update()`, or NaN if 191 | no scalars were collected. 192 | """ 193 | delta = self._get_delta(name) 194 | if int(delta[0]) == 0: 195 | return float('nan') 196 | return float(delta[1] / delta[0]) 197 | 198 | def std(self, name): 199 | r"""Returns the standard deviation of the scalars that were 200 | accumulated for the given statistic between the last two calls to 201 | `update()`, or NaN if no scalars were collected. 202 | """ 203 | delta = self._get_delta(name) 204 | if int(delta[0]) == 0 or not np.isfinite(float(delta[1])): 205 | return float('nan') 206 | if int(delta[0]) == 1: 207 | return float(0) 208 | mean = float(delta[1] / delta[0]) 209 | raw_var = float(delta[2] / delta[0]) 210 | return np.sqrt(max(raw_var - np.square(mean), 0)) 211 | 212 | def as_dict(self): 213 | r"""Returns the averages accumulated between the last two calls to 214 | `update()` as an `dnnlib.EasyDict`. The contents are as follows: 215 | 216 | dnnlib.EasyDict( 217 | NAME = dnnlib.EasyDict(num=FLOAT, mean=FLOAT, std=FLOAT), 218 | ... 219 | ) 220 | """ 221 | stats = dnnlib.EasyDict() 222 | for name in self.names(): 223 | stats[name] = dnnlib.EasyDict(num=self.num(name), mean=self.mean(name), std=self.std(name)) 224 | return stats 225 | 226 | def __getitem__(self, name): 227 | r"""Convenience getter. 228 | `collector[name]` is a synonym for `collector.mean(name)`. 229 | """ 230 | return self.mean(name) 231 | 232 | #---------------------------------------------------------------------------- 233 | 234 | def _sync(names): 235 | r"""Synchronize the global cumulative counters across devices and 236 | processes. Called internally by `Collector.update()`. 237 | """ 238 | if len(names) == 0: 239 | return [] 240 | global _sync_called 241 | _sync_called = True 242 | 243 | # Collect deltas within current rank. 244 | deltas = [] 245 | device = _sync_device if _sync_device is not None else torch.device('cpu') 246 | for name in names: 247 | delta = torch.zeros([_num_moments], dtype=_counter_dtype, device=device) 248 | for counter in _counters[name].values(): 249 | delta.add_(counter.to(device)) 250 | counter.copy_(torch.zeros_like(counter)) 251 | deltas.append(delta) 252 | deltas = torch.stack(deltas) 253 | 254 | # Sum deltas across ranks. 255 | if _sync_device is not None: 256 | torch.distributed.all_reduce(deltas) 257 | 258 | # Update cumulative values. 259 | deltas = deltas.cpu() 260 | for idx, name in enumerate(names): 261 | if name not in _cumulative: 262 | _cumulative[name] = torch.zeros([_num_moments], dtype=_counter_dtype) 263 | _cumulative[name].add_(deltas[idx]) 264 | 265 | # Return name-value pairs. 266 | return [(name, _cumulative[name]) for name in names] 267 | 268 | #---------------------------------------------------------------------------- 269 | -------------------------------------------------------------------------------- /torch_utils/misc.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | import re 10 | import contextlib 11 | import numpy as np 12 | import torch 13 | import warnings 14 | import dnnlib 15 | 16 | #---------------------------------------------------------------------------- 17 | # Cached construction of constant tensors. Avoids CPU=>GPU copy when the 18 | # same constant is used multiple times. 19 | 20 | _constant_cache = dict() 21 | 22 | def 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_format 32 | 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] = tensor 41 | return tensor 42 | 43 | #---------------------------------------------------------------------------- 44 | # Replace NaN/Inf with specified numerical values. 45 | 46 | try: 47 | nan_to_num = torch.nan_to_num # 1.8.0a0 48 | except AttributeError: 49 | def nan_to_num(input, nan=0.0, posinf=None, neginf=None, *, out=None): # pylint: disable=redefined-builtin 50 | assert isinstance(input, torch.Tensor) 51 | if posinf is None: 52 | posinf = torch.finfo(input.dtype).max 53 | if neginf is None: 54 | neginf = torch.finfo(input.dtype).min 55 | assert nan == 0 56 | return torch.clamp(input.unsqueeze(0).nansum(0), min=neginf, max=posinf, out=out) 57 | 58 | #---------------------------------------------------------------------------- 59 | # Symbolic assert. 60 | 61 | try: 62 | symbolic_assert = torch._assert # 1.8.0a0 # pylint: disable=protected-access 63 | except AttributeError: 64 | symbolic_assert = torch.Assert # 1.7.0 65 | 66 | #---------------------------------------------------------------------------- 67 | # Context manager to suppress known warnings in torch.jit.trace(). 68 | 69 | class suppress_tracer_warnings(warnings.catch_warnings): 70 | def __enter__(self): 71 | super().__enter__() 72 | warnings.simplefilter('ignore', category=torch.jit.TracerWarning) 73 | return self 74 | 75 | #---------------------------------------------------------------------------- 76 | # Assert that the shape of a tensor matches the given list of integers. 77 | # None indicates that the size of a dimension is allowed to vary. 78 | # Performs symbolic assertion when used in torch.jit.trace(). 79 | 80 | def assert_shape(tensor, ref_shape): 81 | if tensor.ndim != len(ref_shape): 82 | raise AssertionError(f'Wrong number of dimensions: got {tensor.ndim}, expected {len(ref_shape)}') 83 | for idx, (size, ref_size) in enumerate(zip(tensor.shape, ref_shape)): 84 | if ref_size is None: 85 | pass 86 | elif isinstance(ref_size, torch.Tensor): 87 | with suppress_tracer_warnings(): # as_tensor results are registered as constants 88 | symbolic_assert(torch.equal(torch.as_tensor(size), ref_size), f'Wrong size for dimension {idx}') 89 | elif isinstance(size, torch.Tensor): 90 | with suppress_tracer_warnings(): # as_tensor results are registered as constants 91 | symbolic_assert(torch.equal(size, torch.as_tensor(ref_size)), f'Wrong size for dimension {idx}: expected {ref_size}') 92 | elif size != ref_size: 93 | raise AssertionError(f'Wrong size for dimension {idx}: got {size}, expected {ref_size}') 94 | 95 | #---------------------------------------------------------------------------- 96 | # Function decorator that calls torch.autograd.profiler.record_function(). 97 | 98 | def profiled_function(fn): 99 | def decorator(*args, **kwargs): 100 | with torch.autograd.profiler.record_function(fn.__name__): 101 | return fn(*args, **kwargs) 102 | decorator.__name__ = fn.__name__ 103 | return decorator 104 | 105 | #---------------------------------------------------------------------------- 106 | # Sampler for torch.utils.data.DataLoader that loops over the dataset 107 | # indefinitely, shuffling items as it goes. 108 | 109 | class InfiniteSampler(torch.utils.data.Sampler): 110 | def __init__(self, dataset, rank=0, num_replicas=1, shuffle=True, seed=0, window_size=0.5): 111 | assert len(dataset) > 0 112 | assert num_replicas > 0 113 | assert 0 <= rank < num_replicas 114 | assert 0 <= window_size <= 1 115 | super().__init__(dataset) 116 | self.dataset = dataset 117 | self.rank = rank 118 | self.num_replicas = num_replicas 119 | self.shuffle = shuffle 120 | self.seed = seed 121 | self.window_size = window_size 122 | 123 | def __iter__(self): 124 | order = np.arange(len(self.dataset)) 125 | rnd = None 126 | window = 0 127 | if self.shuffle: 128 | rnd = np.random.RandomState(self.seed) 129 | rnd.shuffle(order) 130 | window = int(np.rint(order.size * self.window_size)) 131 | 132 | idx = 0 133 | while True: 134 | i = idx % order.size 135 | if idx % self.num_replicas == self.rank: 136 | yield order[i] 137 | if window >= 2: 138 | j = (i - rnd.randint(window)) % order.size 139 | order[i], order[j] = order[j], order[i] 140 | idx += 1 141 | 142 | #---------------------------------------------------------------------------- 143 | # Utilities for operating with torch.nn.Module parameters and buffers. 144 | 145 | def params_and_buffers(module): 146 | assert isinstance(module, torch.nn.Module) 147 | return list(module.parameters()) + list(module.buffers()) 148 | 149 | def named_params_and_buffers(module): 150 | assert isinstance(module, torch.nn.Module) 151 | return list(module.named_parameters()) + list(module.named_buffers()) 152 | 153 | def copy_params_and_buffers(src_module, dst_module, require_all=False): 154 | assert isinstance(src_module, torch.nn.Module) 155 | assert isinstance(dst_module, torch.nn.Module) 156 | src_tensors = {name: tensor for name, tensor in named_params_and_buffers(src_module)} 157 | for name, tensor in named_params_and_buffers(dst_module): 158 | assert (name in src_tensors) or (not require_all) 159 | if name in src_tensors: 160 | tensor.copy_(src_tensors[name].detach()).requires_grad_(tensor.requires_grad) 161 | 162 | #---------------------------------------------------------------------------- 163 | # Context manager for easily enabling/disabling DistributedDataParallel 164 | # synchronization. 165 | 166 | @contextlib.contextmanager 167 | def ddp_sync(module, sync): 168 | assert isinstance(module, torch.nn.Module) 169 | if sync or not isinstance(module, torch.nn.parallel.DistributedDataParallel): 170 | yield 171 | else: 172 | with module.no_sync(): 173 | yield 174 | 175 | #---------------------------------------------------------------------------- 176 | # Check DistributedDataParallel consistency across processes. 177 | 178 | def check_ddp_consistency(module, ignore_regex=None): 179 | assert isinstance(module, torch.nn.Module) 180 | for name, tensor in named_params_and_buffers(module): 181 | fullname = type(module).__name__ + '.' + name 182 | if ignore_regex is not None and re.fullmatch(ignore_regex, fullname): 183 | continue 184 | tensor = tensor.detach() 185 | other = tensor.clone() 186 | torch.distributed.broadcast(tensor=other, src=0) 187 | assert (nan_to_num(tensor) == nan_to_num(other)).all(), fullname 188 | 189 | #---------------------------------------------------------------------------- 190 | # Print summary table of module hierarchy. 191 | 192 | def print_module_summary(module, inputs, max_nesting=3, skip_redundant=True): 193 | assert isinstance(module, torch.nn.Module) 194 | assert not isinstance(module, torch.jit.ScriptModule) 195 | assert isinstance(inputs, (tuple, list)) 196 | 197 | # Register hooks. 198 | entries = [] 199 | nesting = [0] 200 | def pre_hook(_mod, _inputs): 201 | nesting[0] += 1 202 | def post_hook(mod, _inputs, outputs): 203 | nesting[0] -= 1 204 | if nesting[0] <= max_nesting: 205 | outputs = list(outputs) if isinstance(outputs, (tuple, list)) else [outputs] 206 | outputs = [t for t in outputs if isinstance(t, torch.Tensor)] 207 | entries.append(dnnlib.EasyDict(mod=mod, outputs=outputs)) 208 | hooks = [mod.register_forward_pre_hook(pre_hook) for mod in module.modules()] 209 | hooks += [mod.register_forward_hook(post_hook) for mod in module.modules()] 210 | 211 | # Run module. 212 | outputs = module(*inputs) 213 | for hook in hooks: 214 | hook.remove() 215 | 216 | # Identify unique outputs, parameters, and buffers. 217 | tensors_seen = set() 218 | for e in entries: 219 | e.unique_params = [t for t in e.mod.parameters() if id(t) not in tensors_seen] 220 | e.unique_buffers = [t for t in e.mod.buffers() if id(t) not in tensors_seen] 221 | e.unique_outputs = [t for t in e.outputs if id(t) not in tensors_seen] 222 | tensors_seen |= {id(t) for t in e.unique_params + e.unique_buffers + e.unique_outputs} 223 | 224 | # Filter out redundant entries. 225 | if skip_redundant: 226 | entries = [e for e in entries if len(e.unique_params) or len(e.unique_buffers) or len(e.unique_outputs)] 227 | 228 | # Construct table. 229 | rows = [[type(module).__name__, 'Parameters', 'Buffers', 'Output shape', 'Datatype']] 230 | rows += [['---'] * len(rows[0])] 231 | param_total = 0 232 | buffer_total = 0 233 | submodule_names = {mod: name for name, mod in module.named_modules()} 234 | for e in entries: 235 | name = '' if e.mod is module else submodule_names[e.mod] 236 | param_size = sum(t.numel() for t in e.unique_params) 237 | buffer_size = sum(t.numel() for t in e.unique_buffers) 238 | output_shapes = [str(list(e.outputs[0].shape)) for t in e.outputs] 239 | output_dtypes = [str(t.dtype).split('.')[-1] for t in e.outputs] 240 | rows += [[ 241 | name + (':0' if len(e.outputs) >= 2 else ''), 242 | str(param_size) if param_size else '-', 243 | str(buffer_size) if buffer_size else '-', 244 | (output_shapes + ['-'])[0], 245 | (output_dtypes + ['-'])[0], 246 | ]] 247 | for idx in range(1, len(e.outputs)): 248 | rows += [[name + f':{idx}', '-', '-', output_shapes[idx], output_dtypes[idx]]] 249 | param_total += param_size 250 | buffer_total += buffer_size 251 | rows += [['---'] * len(rows[0])] 252 | rows += [['Total', str(param_total), str(buffer_total), '-', '-']] 253 | 254 | # Print table. 255 | widths = [max(len(cell) for cell in column) for column in zip(*rows)] 256 | print() 257 | for row in rows: 258 | print(' '.join(cell + ' ' * (width - len(cell)) for cell, width in zip(row, widths))) 259 | print() 260 | return outputs 261 | 262 | #---------------------------------------------------------------------------- 263 | -------------------------------------------------------------------------------- /metrics/metric_utils.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # 3 | # NVIDIA CORPORATION and its licensors retain all intellectual property 4 | # and proprietary rights in and to this software, related documentation 5 | # and any modifications thereto. Any use, reproduction, disclosure or 6 | # distribution of this software and related documentation without an express 7 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | 9 | import os 10 | import time 11 | import hashlib 12 | import pickle 13 | import copy 14 | import uuid 15 | import numpy as np 16 | import torch 17 | import dnnlib 18 | 19 | #---------------------------------------------------------------------------- 20 | 21 | class MetricOptions: 22 | def __init__(self, G=None, G_kwargs={}, dataset_kwargs={}, num_gpus=1, rank=0, device=None, progress=None, cache=True): 23 | assert 0 <= rank < num_gpus 24 | self.G = G 25 | self.G_kwargs = dnnlib.EasyDict(G_kwargs) 26 | self.dataset_kwargs = dnnlib.EasyDict(dataset_kwargs) 27 | self.num_gpus = num_gpus 28 | self.rank = rank 29 | self.device = device if device is not None else torch.device('cuda', rank) 30 | self.progress = progress.sub() if progress is not None and rank == 0 else ProgressMonitor() 31 | self.cache = cache 32 | 33 | #---------------------------------------------------------------------------- 34 | 35 | _feature_detector_cache = dict() 36 | 37 | def get_feature_detector_name(url): 38 | return os.path.splitext(url.split('/')[-1])[0] 39 | 40 | def get_feature_detector(url, device=torch.device('cpu'), num_gpus=1, rank=0, verbose=False): 41 | assert 0 <= rank < num_gpus 42 | key = (url, device) 43 | if key not in _feature_detector_cache: 44 | is_leader = (rank == 0) 45 | if not is_leader and num_gpus > 1: 46 | torch.distributed.barrier() # leader goes first 47 | with dnnlib.util.open_url(url, verbose=(verbose and is_leader)) as f: 48 | _feature_detector_cache[key] = torch.jit.load(f).eval().to(device) 49 | if is_leader and num_gpus > 1: 50 | torch.distributed.barrier() # others follow 51 | return _feature_detector_cache[key] 52 | 53 | #---------------------------------------------------------------------------- 54 | 55 | class FeatureStats: 56 | def __init__(self, capture_all=False, capture_mean_cov=False, max_items=None): 57 | self.capture_all = capture_all 58 | self.capture_mean_cov = capture_mean_cov 59 | self.max_items = max_items 60 | self.num_items = 0 61 | self.num_features = None 62 | self.all_features = None 63 | self.raw_mean = None 64 | self.raw_cov = None 65 | 66 | def set_num_features(self, num_features): 67 | if self.num_features is not None: 68 | assert num_features == self.num_features 69 | else: 70 | self.num_features = num_features 71 | self.all_features = [] 72 | self.raw_mean = np.zeros([num_features], dtype=np.float64) 73 | self.raw_cov = np.zeros([num_features, num_features], dtype=np.float64) 74 | 75 | def is_full(self): 76 | return (self.max_items is not None) and (self.num_items >= self.max_items) 77 | 78 | def append(self, x): 79 | x = np.asarray(x, dtype=np.float32) 80 | assert x.ndim == 2 81 | if (self.max_items is not None) and (self.num_items + x.shape[0] > self.max_items): 82 | if self.num_items >= self.max_items: 83 | return 84 | x = x[:self.max_items - self.num_items] 85 | 86 | self.set_num_features(x.shape[1]) 87 | self.num_items += x.shape[0] 88 | if self.capture_all: 89 | self.all_features.append(x) 90 | if self.capture_mean_cov: 91 | x64 = x.astype(np.float64) 92 | self.raw_mean += x64.sum(axis=0) 93 | self.raw_cov += x64.T @ x64 94 | 95 | def append_torch(self, x, num_gpus=1, rank=0): 96 | assert isinstance(x, torch.Tensor) and x.ndim == 2 97 | assert 0 <= rank < num_gpus 98 | if num_gpus > 1: 99 | ys = [] 100 | for src in range(num_gpus): 101 | y = x.clone() 102 | torch.distributed.broadcast(y, src=src) 103 | ys.append(y) 104 | x = torch.stack(ys, dim=1).flatten(0, 1) # interleave samples 105 | self.append(x.cpu().numpy()) 106 | 107 | def get_all(self): 108 | assert self.capture_all 109 | return np.concatenate(self.all_features, axis=0) 110 | 111 | def get_all_torch(self): 112 | return torch.from_numpy(self.get_all()) 113 | 114 | def get_mean_cov(self): 115 | assert self.capture_mean_cov 116 | mean = self.raw_mean / self.num_items 117 | cov = self.raw_cov / self.num_items 118 | cov = cov - np.outer(mean, mean) 119 | return mean, cov 120 | 121 | def save(self, pkl_file): 122 | with open(pkl_file, 'wb') as f: 123 | pickle.dump(self.__dict__, f) 124 | 125 | @staticmethod 126 | def load(pkl_file): 127 | with open(pkl_file, 'rb') as f: 128 | s = dnnlib.EasyDict(pickle.load(f)) 129 | obj = FeatureStats(capture_all=s.capture_all, max_items=s.max_items) 130 | obj.__dict__.update(s) 131 | return obj 132 | 133 | #---------------------------------------------------------------------------- 134 | 135 | class ProgressMonitor: 136 | def __init__(self, tag=None, num_items=None, flush_interval=1000, verbose=False, progress_fn=None, pfn_lo=0, pfn_hi=1000, pfn_total=1000): 137 | self.tag = tag 138 | self.num_items = num_items 139 | self.verbose = verbose 140 | self.flush_interval = flush_interval 141 | self.progress_fn = progress_fn 142 | self.pfn_lo = pfn_lo 143 | self.pfn_hi = pfn_hi 144 | self.pfn_total = pfn_total 145 | self.start_time = time.time() 146 | self.batch_time = self.start_time 147 | self.batch_items = 0 148 | if self.progress_fn is not None: 149 | self.progress_fn(self.pfn_lo, self.pfn_total) 150 | 151 | def update(self, cur_items): 152 | assert (self.num_items is None) or (cur_items <= self.num_items) 153 | if (cur_items < self.batch_items + self.flush_interval) and (self.num_items is None or cur_items < self.num_items): 154 | return 155 | cur_time = time.time() 156 | total_time = cur_time - self.start_time 157 | time_per_item = (cur_time - self.batch_time) / max(cur_items - self.batch_items, 1) 158 | if (self.verbose) and (self.tag is not None): 159 | print(f'{self.tag:<19s} items {cur_items:<7d} time {dnnlib.util.format_time(total_time):<12s} ms/item {time_per_item*1e3:.2f}') 160 | self.batch_time = cur_time 161 | self.batch_items = cur_items 162 | 163 | if (self.progress_fn is not None) and (self.num_items is not None): 164 | self.progress_fn(self.pfn_lo + (self.pfn_hi - self.pfn_lo) * (cur_items / self.num_items), self.pfn_total) 165 | 166 | def sub(self, tag=None, num_items=None, flush_interval=1000, rel_lo=0, rel_hi=1): 167 | return ProgressMonitor( 168 | tag = tag, 169 | num_items = num_items, 170 | flush_interval = flush_interval, 171 | verbose = self.verbose, 172 | progress_fn = self.progress_fn, 173 | pfn_lo = self.pfn_lo + (self.pfn_hi - self.pfn_lo) * rel_lo, 174 | pfn_hi = self.pfn_lo + (self.pfn_hi - self.pfn_lo) * rel_hi, 175 | pfn_total = self.pfn_total, 176 | ) 177 | 178 | #---------------------------------------------------------------------------- 179 | 180 | def compute_feature_stats_for_dataset(opts, detector_url, detector_kwargs, rel_lo=0, rel_hi=1, batch_size=64, data_loader_kwargs=None, max_items=None, **stats_kwargs): 181 | dataset = dnnlib.util.construct_class_by_name(**opts.dataset_kwargs) 182 | if data_loader_kwargs is None: 183 | data_loader_kwargs = dict(pin_memory=True, num_workers=3, prefetch_factor=2) 184 | 185 | # Try to lookup from cache. 186 | cache_file = None 187 | if opts.cache: 188 | # Choose cache file name. 189 | args = dict(dataset_kwargs=opts.dataset_kwargs, detector_url=detector_url, detector_kwargs=detector_kwargs, stats_kwargs=stats_kwargs) 190 | md5 = hashlib.md5(repr(sorted(args.items())).encode('utf-8')) 191 | cache_tag = f'{dataset.name}-{get_feature_detector_name(detector_url)}-{md5.hexdigest()}' 192 | cache_file = dnnlib.make_cache_dir_path('gan-metrics', cache_tag + '.pkl') 193 | 194 | # Check if the file exists (all processes must agree). 195 | flag = os.path.isfile(cache_file) if opts.rank == 0 else False 196 | if opts.num_gpus > 1: 197 | flag = torch.as_tensor(flag, dtype=torch.float32, device=opts.device) 198 | torch.distributed.broadcast(tensor=flag, src=0) 199 | flag = (float(flag.cpu()) != 0) 200 | 201 | # Load. 202 | if flag: 203 | return FeatureStats.load(cache_file) 204 | 205 | # Initialize. 206 | num_items = len(dataset) 207 | if max_items is not None: 208 | num_items = min(num_items, max_items) 209 | stats = FeatureStats(max_items=num_items, **stats_kwargs) 210 | progress = opts.progress.sub(tag='dataset features', num_items=num_items, rel_lo=rel_lo, rel_hi=rel_hi) 211 | detector = get_feature_detector(url=detector_url, device=opts.device, num_gpus=opts.num_gpus, rank=opts.rank, verbose=progress.verbose) 212 | 213 | # Main loop. 214 | item_subset = [(i * opts.num_gpus + opts.rank) % num_items for i in range((num_items - 1) // opts.num_gpus + 1)] 215 | for images, _labels in torch.utils.data.DataLoader(dataset=dataset, sampler=item_subset, batch_size=batch_size, **data_loader_kwargs): 216 | if images.shape[1] == 1: 217 | images = images.repeat([1, 3, 1, 1]) 218 | features = detector(images.to(opts.device), **detector_kwargs) 219 | stats.append_torch(features, num_gpus=opts.num_gpus, rank=opts.rank) 220 | progress.update(stats.num_items) 221 | 222 | # Save to cache. 223 | if cache_file is not None and opts.rank == 0: 224 | os.makedirs(os.path.dirname(cache_file), exist_ok=True) 225 | temp_file = cache_file + '.' + uuid.uuid4().hex 226 | stats.save(temp_file) 227 | os.replace(temp_file, cache_file) # atomic 228 | return stats 229 | 230 | #---------------------------------------------------------------------------- 231 | 232 | def compute_feature_stats_for_generator(opts, detector_url, detector_kwargs, rel_lo=0, rel_hi=1, batch_size=64, batch_gen=None, jit=False, **stats_kwargs): 233 | if batch_gen is None: 234 | batch_gen = min(batch_size, 4) 235 | assert batch_size % batch_gen == 0 236 | 237 | # Setup generator and load labels. 238 | G = copy.deepcopy(opts.G).eval().requires_grad_(False).to(opts.device) 239 | dataset = dnnlib.util.construct_class_by_name(**opts.dataset_kwargs) 240 | 241 | # Image generation func. 242 | def run_generator(z, c): 243 | img = G(z=z, c=c, **opts.G_kwargs) 244 | img = (img * 127.5 + 128).clamp(0, 255).to(torch.uint8) 245 | return img 246 | 247 | # JIT. 248 | if jit: 249 | z = torch.zeros([batch_gen, G.z_dim], device=opts.device) 250 | c = torch.zeros([batch_gen, G.c_dim], device=opts.device) 251 | run_generator = torch.jit.trace(run_generator, [z, c], check_trace=False) 252 | 253 | # Initialize. 254 | stats = FeatureStats(**stats_kwargs) 255 | assert stats.max_items is not None 256 | progress = opts.progress.sub(tag='generator features', num_items=stats.max_items, rel_lo=rel_lo, rel_hi=rel_hi) 257 | detector = get_feature_detector(url=detector_url, device=opts.device, num_gpus=opts.num_gpus, rank=opts.rank, verbose=progress.verbose) 258 | 259 | # Main loop. 260 | while not stats.is_full(): 261 | images = [] 262 | for _i in range(batch_size // batch_gen): 263 | z = torch.randn([batch_gen, G.z_dim], device=opts.device) 264 | c = [dataset.get_label(np.random.randint(len(dataset))) for _i in range(batch_gen)] 265 | c = torch.from_numpy(np.stack(c)).pin_memory().to(opts.device) 266 | images.append(run_generator(z, c)) 267 | images = torch.cat(images) 268 | if images.shape[1] == 1: 269 | images = images.repeat([1, 3, 1, 1]) 270 | features = detector(images, **detector_kwargs) 271 | stats.append_torch(features, num_gpus=opts.num_gpus, rank=opts.rank) 272 | progress.update(stats.num_items) 273 | return stats 274 | 275 | #---------------------------------------------------------------------------- 276 | -------------------------------------------------------------------------------- /torch_utils/gen_utils.py: -------------------------------------------------------------------------------- 1 | import os 2 | import re 3 | import json 4 | 5 | from typing import List, Tuple, Union, Optional 6 | from collections import OrderedDict 7 | from locale import atof 8 | 9 | import click 10 | import numpy as np 11 | import torch 12 | 13 | 14 | # ---------------------------------------------------------------------------- 15 | 16 | 17 | def create_image_grid(images: np.ndarray, grid_size: Optional[Tuple[int, int]] = None): 18 | """ 19 | Create a grid with the fed images 20 | Args: 21 | images (np.array): array of images 22 | grid_size (tuple(int)): size of grid (grid_width, grid_height) 23 | Returns: 24 | grid (np.array): image grid of size grid_size 25 | """ 26 | # Sanity check 27 | assert images.ndim == 3 or images.ndim == 4, f'Images has {images.ndim} dimensions (shape: {images.shape})!' 28 | num, img_h, img_w, c = images.shape 29 | # If user specifies the grid shape, use it 30 | if grid_size is not None: 31 | grid_w, grid_h = tuple(grid_size) 32 | # If one of the sides is None, then we must infer it (this was divine inspiration) 33 | if grid_w is None: 34 | grid_w = num // grid_h + min(num % grid_h, 1) 35 | elif grid_h is None: 36 | grid_h = num // grid_w + min(num % grid_w, 1) 37 | 38 | # Otherwise, we can infer it by the number of images (priority is given to grid_w) 39 | else: 40 | grid_w = max(int(np.ceil(np.sqrt(num))), 1) 41 | grid_h = max((num - 1) // grid_w + 1, 1) 42 | 43 | # Sanity check 44 | assert grid_w * grid_h >= num, 'Number of rows and columns must be greater than the number of images!' 45 | # Get the grid 46 | grid = np.zeros([grid_h * img_h, grid_w * img_h] + list(images.shape[-1:]), dtype=images.dtype) 47 | # Paste each image in the grid 48 | for idx in range(num): 49 | x = (idx % grid_w) * img_w 50 | y = (idx // grid_w) * img_h 51 | grid[y:y + img_h, x:x + img_w, ...] = images[idx] 52 | return grid 53 | 54 | 55 | # ---------------------------------------------------------------------------- 56 | 57 | 58 | def parse_fps(fps: Union[str, int]) -> int: 59 | """Return FPS for the video; at worst, video will be 1 FPS, but no lower.""" 60 | if isinstance(fps, int): 61 | return max(fps, 1) 62 | try: 63 | fps = int(atof(fps)) 64 | return max(fps, 1) 65 | except ValueError: 66 | print(f'Typo in "--fps={fps}", will use default value of 30') 67 | return 30 68 | 69 | 70 | def num_range(s: str, remove_repeated: bool = True) -> List[int]: 71 | """ 72 | Extended helper function from the original (original is contained here). 73 | Accept a comma separated list of numbers 'a,b,c', a range 'a-c', or a combination 74 | of both 'a,b-c', 'a-b,c', 'a,b-c,d,e-f,...', and return as a list of ints. 75 | """ 76 | str_list = s.split(',') 77 | nums = [] 78 | range_re = re.compile(r'^(\d+)-(\d+)$') 79 | for el in str_list: 80 | match = range_re.match(el) 81 | if match: 82 | # Sanity check 1: accept ranges 'a-b' or 'b-a', with a<=b 83 | lower, upper = int(match.group(1)), int(match.group(2)) 84 | if lower <= upper: 85 | r = list(range(lower, upper + 1)) 86 | else: 87 | r = list(range(upper, lower + 1)) 88 | # We will extend nums as r is also a list 89 | nums.extend(r) 90 | else: 91 | # It's a single number, so just append it (if it's an int) 92 | try: 93 | nums.append(int(atof(el))) 94 | except ValueError: 95 | continue # we ignore bad values 96 | # Sanity check 2: delete repeating numbers by default, but keep order given by user 97 | if remove_repeated: 98 | nums = list(OrderedDict.fromkeys(nums)) 99 | return nums 100 | 101 | 102 | def parse_slowdown(slowdown: Union[str, int]) -> int: 103 | """Function to parse the 'slowdown' parameter by the user. Will approximate to the nearest power of 2.""" 104 | # TODO: slowdown should be any int 105 | if not isinstance(slowdown, int): 106 | try: 107 | slowdown = atof(slowdown) 108 | except ValueError: 109 | print(f'Typo in "{slowdown}"; will use default value of 1') 110 | slowdown = 1 111 | assert slowdown > 0, '"slowdown" cannot be negative!' 112 | # Let's approximate slowdown to the closest power of 2 (nothing happens if it's already a power of 2) 113 | slowdown = 2**int(np.rint(np.log2(slowdown))) 114 | return max(slowdown, 1) # Guard against 0.5, 0.25, ... cases 115 | 116 | 117 | # ---------------------------------------------------------------------------- 118 | 119 | 120 | def compress_video( 121 | original_video: Union[str, os.PathLike], 122 | original_video_name: Union[str, os.PathLike], 123 | outdir: Union[str, os.PathLike], 124 | ctx: click.Context) -> None: 125 | """ Helper function to compress the original_video using ffmpeg-python. moviepy creates huge videos, so use 126 | ffmpeg to 'compress' it (won't be perfect, 'compression' will depend on the video dimensions). ffmpeg 127 | can also be used to e.g. resize the video, make a GIF, save all frames in the video to the outdir, etc. 128 | """ 129 | try: 130 | import ffmpeg 131 | except (ModuleNotFoundError, ImportError): 132 | ctx.fail('Missing ffmpeg! Install it via "pip install ffmpeg-python"') 133 | 134 | print('Compressing the video...') 135 | resized_video_name = os.path.join(outdir, f'{original_video_name}-compressed.mp4') 136 | ffmpeg.input(original_video).output(resized_video_name).run(capture_stdout=True, capture_stderr=True) 137 | print('Success!') 138 | 139 | 140 | # ---------------------------------------------------------------------------- 141 | 142 | 143 | def interpolation_checks( 144 | t: Union[float, np.ndarray], 145 | v0: np.ndarray, 146 | v1: np.ndarray) -> Tuple[Union[float, np.ndarray], np.ndarray, np.ndarray]: 147 | """Tests for the interpolation functions""" 148 | # Make sure 0.0<=t<=1.0 149 | assert np.min(t) >= 0.0 and np.max(t) <= 1.0 150 | # Guard against v0 and v1 not being NumPy arrays 151 | if not isinstance(v0, np.ndarray) or not isinstance(v1, np.ndarray): 152 | v0 = np.array(v0) 153 | v1 = np.array(v1) 154 | assert v0.shape == v1.shape, f'Incompatible shapes! v0: {v0.shape}, v1: {v1.shape}' 155 | return t, v0, v1 156 | 157 | 158 | def lerp( 159 | t: Union[float, np.ndarray], 160 | v0: Union[float, list, tuple, np.ndarray], 161 | v1: Union[float, list, tuple, np.ndarray]) -> np.ndarray: 162 | """ 163 | Linear interpolation between v0 (starting) and v1 (final) vectors; for optimal results, 164 | use t as an np.ndarray to return all results at once via broadcasting 165 | """ 166 | t, v0, v1 = interpolation_checks(t, v0, v1) 167 | v2 = (1.0 - t) * v0 + t * v1 168 | return v2 169 | 170 | 171 | def slerp( 172 | t: Union[float, np.ndarray], 173 | v0: Union[float, list, tuple, np.ndarray], 174 | v1: Union[float, list, tuple, np.ndarray], 175 | dot_threshold: float = 0.9995) -> np.ndarray: 176 | """ 177 | Spherical linear interpolation between v0 (starting) and v1 (final) vectors; for optimal 178 | results, use t as an np.ndarray to return all results at once via broadcasting. 179 | 180 | dot_threshold is the threshold for considering if the two vectors are collinear (not recommended to alter). 181 | 182 | Adapted from the Python code at: https://en.wikipedia.org/wiki/Slerp (at the time, now no longer available). 183 | Most likely taken from Jonathan Blow's code in C++: 184 | http://number-none.com/product/Understanding%20Slerp,%20Then%20Not%20Using%20It 185 | """ 186 | t, v0, v1 = interpolation_checks(t, v0, v1) 187 | # Copy vectors to reuse them later 188 | v0_copy = np.copy(v0) 189 | v1_copy = np.copy(v1) 190 | # Normalize the vectors to get the directions and angles 191 | v0 = v0 / np.linalg.norm(v0) 192 | v1 = v1 / np.linalg.norm(v1) 193 | # Dot product with the normalized vectors (can't always use np.dot, so we use the definition) 194 | dot = np.sum(v0 * v1) 195 | # If it's ~1, vectors are ~colineal, so use lerp 196 | if np.abs(dot) > dot_threshold: 197 | return lerp(t, v0, v1) 198 | # Stay within domain of arccos 199 | dot = np.clip(dot, -1.0, 1.0) 200 | # Calculate initial angle between v0 and v1 201 | theta_0 = np.arccos(dot) 202 | sin_theta_0 = np.sin(theta_0) 203 | # Divide the angle into t steps 204 | theta_t = theta_0 * t 205 | sin_theta_t = np.sin(theta_t) 206 | # Finish the slerp algorithm 207 | s0 = np.sin(theta_0 - theta_t) / sin_theta_0 208 | s1 = sin_theta_t / sin_theta_0 209 | v2 = s0 * v0_copy + s1 * v1_copy 210 | return v2 211 | 212 | 213 | def interpolate( 214 | v0: Union[float, list, tuple, np.ndarray], 215 | v1: Union[float, list, tuple, np.ndarray], 216 | n_steps: int, 217 | interp_type: str = 'spherical', 218 | smooth: bool = False) -> np.ndarray: 219 | """ 220 | Interpolation function between two vectors, v0 and v1. We will either do a 'linear' or 'spherical' interpolation, 221 | taking n_steps. The steps can be 'smooth'-ed out, so that the transition between vectors isn't too drastic. 222 | """ 223 | t_array = np.linspace(0, 1, num=n_steps, endpoint=False) 224 | # TODO: have a dictionary with easing functions that contains my 'smooth' one (might be useful for someone else) 225 | if smooth: 226 | # Smooth out the interpolation with a polynomial of order 3 (cubic function f) 227 | # Constructed f by setting f'(0) = f'(1) = 0, and f(0) = 0, f(1) = 1 => f(t) = -2t^3+3t^2 = t^2 (3-2t) 228 | t_array = t_array ** 2 * (3 - 2 * t_array) # One line thanks to NumPy arrays 229 | # TODO: this might be possible to optimize by using the fact they're numpy arrays, but haven't found a nice way yet 230 | funcs_dict = {'linear': lerp, 'spherical': slerp} 231 | vectors = np.array([funcs_dict[interp_type](t, v0, v1) for t in t_array], dtype=np.float32) 232 | return vectors 233 | 234 | 235 | # ---------------------------------------------------------------------------- 236 | 237 | 238 | def double_slowdown(latents: np.ndarray, duration: float, frames: int) -> Tuple[np.ndarray, float, int]: 239 | """ 240 | Auxiliary function to slow down the video by 2x. We return the new latents, duration, and frames of the video 241 | """ 242 | # Make an empty latent vector with double the amount of frames, but keep the others the same 243 | z = np.empty(np.multiply(latents.shape, [2, 1, 1]), dtype=np.float32) 244 | # In the even frames, populate it with the latents 245 | for i in range(len(latents)): 246 | z[2 * i] = latents[i] 247 | # Interpolate in the odd frames 248 | for i in range(1, len(z), 2): 249 | # slerp between (t=0.5) even frames; for the last frame, we loop to the first one (z[0]) 250 | z[i] = slerp(0.5, z[i - 1], z[i + 1]) if i != len(z) - 1 else slerp(0.5, z[0], z[i - 1]) 251 | # TODO: we could change this to any slowdown: slerp(1/slowdown, ...), and we return z, slowdown * duration, ... 252 | # Return the new latents, and the respective new duration and number of frames 253 | return z, 2 * duration, 2 * frames 254 | 255 | 256 | # ---------------------------------------------------------------------------- 257 | 258 | 259 | def z_to_img(G, latents: torch.Tensor, label: torch.Tensor, truncation_psi: float, noise_mode: str = 'const'): 260 | """ 261 | Get an image/np.ndarray from a latent Z using G, the label, truncation_psi, and noise_mode. The shape 262 | of the output image/np.ndarray will be [len(dlatents), G.img_resolution, G.img_resolution, G.img_channels] 263 | """ 264 | assert isinstance(latents, torch.Tensor), f'latents should be a torch.Tensor!: "{type(latents)}"' 265 | if len(latents.shape) == 1: 266 | latents = latents.unsqueeze(0) # An individual latent => [1, G.z_dim] 267 | img = G(z=latents, c=label, truncation_psi=truncation_psi, noise_mode=noise_mode) 268 | img = (img + 1) * 255 / 2 # [-1.0, 1.0] -> [0.0, 255.0] 269 | img = img.permute(0, 2, 3, 1).clamp(0, 255).to(torch.uint8).cpu().numpy() # NCWH => NWHC 270 | return img 271 | 272 | 273 | def w_to_img(G, dlatents: torch.Tensor, noise_mode: str = 'const') -> np.ndarray: 274 | """ 275 | Get an image/np.ndarray from a dlatent W using G and the selected noise_mode. The final shape of the 276 | returned image will be [len(dlatents), G.img_resolution, G.img_resolution, G.img_channels]. 277 | """ 278 | assert isinstance(dlatents, torch.Tensor), f'dlatents should be a torch.Tensor!: "{type(dlatents)}"' 279 | if len(dlatents.shape) == 2: 280 | dlatents = dlatents.unsqueeze(0) # An individual dlatent => [1, G.mapping.num_ws, G.mapping.w_dim] 281 | synth_image = G.synthesis(dlatents, noise_mode=noise_mode) 282 | synth_image = (synth_image + 1) * 255/2 # [-1.0, 1.0] -> [0.0, 255.0] 283 | synth_image = synth_image.permute(0, 2, 3, 1).clamp(0, 255).to(torch.uint8).cpu().numpy() # NCWH => NWHC 284 | return synth_image 285 | 286 | 287 | def get_w_from_seed(G, device: torch.device, seed: int, truncation_psi: float) -> torch.Tensor: 288 | """Get the dlatent from a random seed, using the truncation trick (this could be optional)""" 289 | z = np.random.RandomState(seed).randn(1, G.z_dim) 290 | w = G.mapping(torch.from_numpy(z).to(device), None) 291 | w_avg = G.mapping.w_avg 292 | w = w_avg + (w - w_avg) * truncation_psi 293 | 294 | return w 295 | 296 | 297 | def get_w_from_file(file: Union[str, os.PathLike], return_ext: bool = False) -> Tuple[np.ndarray, Optional[str]]: 298 | """Get dlatent (w) from a .npy or .npz file""" 299 | filename, file_extension = os.path.splitext(file) 300 | assert file_extension in ['.npy', '.npz'], f'"{file}" has wrong file format! Use either ".npy" or ".npz"' 301 | if file_extension == '.npy': 302 | r = (np.load(file), '.npy') if return_ext else np.load(file) 303 | return r 304 | r = (np.load(file)['w'], '.npz') if return_ext else np.load(file)['w'] 305 | return r 306 | 307 | 308 | # ---------------------------------------------------------------------------- 309 | 310 | 311 | def save_config(ctx: click.Context, run_dir: Union[str, os.PathLike], save_name: str = 'config.json') -> None: 312 | """Save the configuration stored in ctx.obj into a JSON file at the output directory.""" 313 | with open(os.path.join(run_dir, save_name), 'w') as f: 314 | json.dump(ctx.obj, f, indent=4, sort_keys=True) 315 | 316 | 317 | # ---------------------------------------------------------------------------- 318 | 319 | 320 | def make_run_dir(outdir: Union[str, os.PathLike], desc: str, dry_run: bool = False) -> str: 321 | """Reject modernity, return to automatically create the run dir.""" 322 | # Pick output directory. 323 | prev_run_dirs = [] 324 | if os.path.isdir(outdir): # sanity check, but click.Path() should clear this one 325 | prev_run_dirs = [x for x in os.listdir(outdir) if os.path.isdir(os.path.join(outdir, x))] 326 | prev_run_ids = [re.match(r'^\d+', x) for x in prev_run_dirs] 327 | prev_run_ids = [int(x.group()) for x in prev_run_ids if x is not None] 328 | cur_run_id = max(prev_run_ids, default=-1) + 1 # start with 00000 329 | run_dir = os.path.join(outdir, f'{cur_run_id:05d}-{desc}') 330 | assert not os.path.exists(run_dir) # make sure it doesn't already exist 331 | 332 | # Don't create the dir if it's a dry-run 333 | if not dry_run: 334 | print('Creating output directory...') 335 | os.makedirs(run_dir) 336 | return run_dir 337 | --------------------------------------------------------------------------------