train_vqvae.py and vqvae.py now supports distributed training. You can use --n_gpu [NUM_GPUS] arguments for train_vqvae.py to use [NUM_GPUS] during training. Python >= 3.6 PyTorch >= 1.1 lmdb (for storing extra
(args.path, transform=transform) loader = DataLoader(dataset, batch_size=128, shuffle=False, num_workers=4) model = VQVAE() model.load_state_dict(torch.load(args.ckpt)) model = model.to(device) model.eval() map_size = 100 * 1024 * 1024 * 1024 env = lmdb.open(args.name, map_...
Implementation of Generating Diverse High-Fidelity Images with VQ-VAE-2 in PyTorch - Added VQ-VAE checkpoint · rosinality/vq-vae-2-pytorch@ede65ce
Feature request for AMP support in VQ-VAE training. So far, I tried naively modifying the train function in train_vqvae.py like so: # ... for i, (img, label) in enumerate(loader): model.zero_grad() img = img.to(device) with torch.cuda.amp.autocast(): out, latent_loss = ...
Breadcrumbs vq-vae-2-pytorch / train_vqvae.pyTop File metadata and controls Code Blame executable file· 152 lines (113 loc) · 4.2 KB Raw import argparse import sys import os import torch from torch import nn, optim from torch.utils.data import DataLoader from torchvision import datasets,...
Note: This is a training sample  Binary file added BIN +1.44 MB stage1_sample.png Unable to render rich display Invalid image source. 0 comments on commit 6acd33a Please sign in to comment. Footer...