File size: 13,052 Bytes
09b13b3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 |
#!/usr/bin/env python
# coding=utf-8
# Copyright (c) 2022 PyTorch contributors and The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions.
""" Training a Deep Convolutional Generative Adversarial Network (DCGAN) leveraging the 🤗 ecosystem.
Paper: https://arxiv.org/abs/1511.06434.
Based on PyTorch's official tutorial: https://pytorch.org/tutorials/beginner/dcgan_faces_tutorial.html.
"""
import argparse
import logging
import os
import sys
from pathlib import Path
import torch
import torch.nn as nn
from torch.utils.data import DataLoader
from torchvision.transforms import (CenterCrop, Compose, Normalize, Resize,
ToTensor, ToPILImage)
from torchvision.utils import save_image
from PIL import Image, ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
from accelerate import Accelerator
from modeling_dcgan import Discriminator, Generator
from datasets import load_dataset
from huggan.pytorch.metrics.inception import InceptionV3
from huggan.pytorch.metrics.fid_score import calculate_fretchet
import wandb
logger = logging.getLogger(__name__)
def parse_args(args=None):
parser = argparse.ArgumentParser()
parser.add_argument("--dataset", type=str, default="mnist", help="Dataset to load from the HuggingFace hub.")
parser.add_argument("--num_workers", type=int, default=0, help="Number of workers when loading data")
parser.add_argument("--batch_size", type=int, default=128, help="Batch size to use during training")
parser.add_argument(
"--image_size",
type=int,
default=64,
help="Spatial size to use when resizing images for training.",
)
parser.add_argument(
"--num_channels",
type=int,
default=3,
help="Number of channels in the training images. For color images this is 3.",
)
parser.add_argument("--latent_dim", type=int, default=100, help="Dimensionality of the latent space.")
parser.add_argument(
"--generator_hidden_size",
type=int,
default=64,
help="Hidden size of the generator's feature maps.",
)
parser.add_argument(
"--discriminator_hidden_size",
type=int,
default=64,
help="Hidden size of the discriminator's feature maps.",
)
parser.add_argument("--num_epochs", type=int, default=5, help="number of epochs of training")
parser.add_argument("--lr", type=float, default=0.0002, help="adam: learning rate")
parser.add_argument(
"--beta1",
type=float,
default=0.5,
help="adam: decay of first order momentum of gradient",
)
parser.add_argument("--fp16", action="store_true", help="If passed, will use FP16 training.")
parser.add_argument(
"--mixed_precision",
type=str,
default="no",
choices=["no", "fp16", "bf16"],
help="Whether to use mixed precision. Choose"
"between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10."
"and an Nvidia Ampere GPU.",
)
parser.add_argument("--cpu", action="store_true", help="If passed, will train on the CPU.")
parser.add_argument("--output_dir", type=Path, default=Path("./output"), help="Name of the directory to dump generated images during training.")
parser.add_argument("--wandb", action="store_true", help="If passed, will log to Weights and Biases.")
parser.add_argument(
"--logging_steps",
type=int,
default=50,
help="Number of steps between each logging",
)
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether to push the model to the HuggingFace hub after training.",
)
parser.add_argument(
"--model_name",
default=None,
type=str,
help="Name of the model on the hub.",
)
parser.add_argument(
"--organization_name",
default="huggan",
type=str,
help="Organization name to push to, in case args.push_to_hub is specified.",
)
args = parser.parse_args()
if args.push_to_hub:
assert args.output_dir is not None, "Need an `output_dir` to create a repo when `--push_to_hub` is passed."
assert args.model_name is not None, "Need a `model_name` to create a repo when `--push_to_hub` is passed."
if args.output_dir is not None:
os.makedirs(args.output_dir, exist_ok=True)
return args
# Custom weights initialization called on Generator and Discriminator
def weights_init(m):
classname = m.__class__.__name__
if classname.find("Conv") != -1:
nn.init.normal_(m.weight.data, 0.0, 0.02)
elif classname.find("BatchNorm") != -1:
nn.init.normal_(m.weight.data, 1.0, 0.02)
nn.init.constant_(m.bias.data, 0)
def training_function(config, args):
# Initialize the accelerator. We will let the accelerator handle device placement for us in this example.
accelerator = Accelerator(fp16=args.fp16, cpu=args.cpu, mixed_precision=args.mixed_precision)
# Setup logging, we only want one process per machine to log things on the screen.
# accelerator.is_local_main_process is only True for one process per machine.
logger.setLevel(logging.INFO if accelerator.is_local_main_process else logging.ERROR)
if accelerator.is_local_main_process:
# set up Weights and Biases if requested
if args.wandb:
import wandb
wandb.init(project=str(args.output_dir).split("/")[-1])
# Loss function
criterion = nn.BCELoss()
# Initialize generator and discriminator
generator = Generator(
num_channels=args.num_channels,
latent_dim=args.latent_dim,
hidden_size=args.generator_hidden_size,
)
discriminator = Discriminator(num_channels=args.num_channels, hidden_size=args.discriminator_hidden_size)
# Initialize weights
generator.apply(weights_init)
discriminator.apply(weights_init)
# Initialize Inceptionv3 (for FID metric)
model = InceptionV3()
# Initialize Inceptionv3 (for FID metric)
model = InceptionV3()
# Create batch of latent vectors that we will use to visualize
# the progression of the generator
fixed_noise = torch.randn(64, args.latent_dim, 1, 1, device=accelerator.device)
# Establish convention for real and fake labels during training
real_label = 1.0
fake_label = 0.0
# Setup Adam optimizers for both G and D
discriminator_optimizer = torch.optim.Adam(discriminator.parameters(), lr=args.lr, betas=(args.beta1, 0.999))
generator_optimizer = torch.optim.Adam(generator.parameters(), lr=args.lr, betas=(args.beta1, 0.999))
# Configure data loader
dataset = load_dataset(args.dataset)
transform = Compose(
[
Resize(args.image_size),
CenterCrop(args.image_size),
ToTensor(),
Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
]
)
def transforms(examples):
examples["pixel_values"] = [transform(image.convert("RGB")) for image in examples["image"]]
del examples["image"]
return examples
transformed_dataset = dataset.with_transform(transforms)
dataloader = DataLoader(
transformed_dataset["train"], batch_size=args.batch_size, shuffle=True, num_workers=args.num_workers
)
generator, discriminator, generator_optimizer, discriminator_optimizer, dataloader = accelerator.prepare(generator, discriminator, generator_optimizer, discriminator_optimizer, dataloader)
# ----------
# Training
# ----------
# Training Loop
# Lists to keep track of progress
img_list = []
logger.info("***** Running training *****")
logger.info(f" Num Epochs = {args.num_epochs}")
# For each epoch
for epoch in range(args.num_epochs):
# For each batch in the dataloader
for step, batch in enumerate(dataloader, 0):
############################
# (1) Update D network: maximize log(D(x)) + log(1 - D(G(z)))
###########################
## Train with all-real batch
discriminator.zero_grad()
# Format batch
real_cpu = batch["pixel_values"]
batch_size = real_cpu.size(0)
label = torch.full((batch_size,), real_label, dtype=torch.float, device=accelerator.device)
# Forward pass real batch through D
output = discriminator(real_cpu).view(-1)
# Calculate loss on all-real batch
errD_real = criterion(output, label)
# Calculate gradients for D in backward pass
accelerator.backward(errD_real)
D_x = output.mean().item()
## Train with all-fake batch
# Generate batch of latent vectors
noise = torch.randn(batch_size, args.latent_dim, 1, 1, device=accelerator.device)
# Generate fake image batch with G
fake = generator(noise)
label.fill_(fake_label)
# Classify all fake batch with D
output = discriminator(fake.detach()).view(-1)
# Calculate D's loss on the all-fake batch
errD_fake = criterion(output, label)
# Calculate the gradients for this batch, accumulated (summed) with previous gradients
accelerator.backward(errD_fake)
D_G_z1 = output.mean().item()
# Compute error of D as sum over the fake and the real batches
errD = errD_real + errD_fake
# Update D
discriminator_optimizer.step()
############################
# (2) Update G network: maximize log(D(G(z)))
###########################
generator.zero_grad()
label.fill_(real_label) # fake labels are real for generator cost
# Since we just updated D, perform another forward pass of all-fake batch through D
output = discriminator(fake).view(-1)
# Calculate G's loss based on this output
errG = criterion(output, label)
# Calculate gradients for G
accelerator.backward(errG)
D_G_z2 = output.mean().item()
# Update G
generator_optimizer.step()
# Log all results
if (step + 1) % args.logging_steps == 0:
errD.detach()
errG.detach()
if accelerator.state.num_processes > 1:
errD = accelerator.gather(errD).sum() / accelerator.state.num_processes
errG = accelerator.gather(errG).sum() / accelerator.state.num_processes
train_logs = {
"epoch": epoch,
"discriminator_loss": errD,
"generator_loss": errG,
"D_x": D_x,
"D_G_z1": D_G_z1,
"D_G_z2": D_G_z2,
}
log_str = ""
for k, v in train_logs.items():
log_str += "| {}: {:.3e}".format(k, v)
if accelerator.is_local_main_process:
logger.info(log_str)
if args.wandb:
wandb.log(train_logs)
# Check how the generator is doing by saving G's output on fixed_noise
if (step % 500 == 0) or ((epoch == args.num_epochs - 1) and (step == len(dataloader) - 1)):
with torch.no_grad():
fake_images = generator(fixed_noise).detach().cpu()
file_name = args.output_dir/f"iter_{step}.png"
save_image(fake_images.data[:25], file_name, nrow=5, normalize=True)
if accelerator.is_local_main_process and args.wandb:
wandb.log({'generated_examples': wandb.Image(str(file_name)) })
# Calculate FID metric
fid = calculate_fretchet(real_cpu, fake, model.to(accelerator.device))
logger.info(f"FID: {fid}")
if accelerator.is_local_main_process and args.wandb:
wandb.log({"FID": fid})
# Optionally push to hub
if accelerator.is_main_process and args.push_to_hub:
generator.module.push_to_hub(
repo_path_or_name=args.output_dir / args.model_name,
organization=args.organization_name,
)
def main():
args = parse_args()
print(args)
training_function({}, args)
if __name__ == "__main__":
main()
|