search

yang-song / score_sde_pytorch

PyTorch implementation for Score-Based Generative Modeling through Stochastic Differential Equations (ICLR 2021, Oral)
https://arxiv.org/abs/2011.13456
Apache License 2.0
1.58k stars 295 forks source link

why pytorch version uses tensorflow？ #60

Open scut-ken5 opened 4 weeks ago

scut-ken5 commented 4 weeks ago

The code below is from run_lib.py. I just wanna ask why there exists the tensorflow in the pytorch version of score_sde. Thanks for your answer import numpy as np import tensorflow as tf import tensorflow_gan as tfgan import logging from models import ddpm, ncsnv2, ncsnpp import losses import sampling from models import utils as mutils from models.ema import ExponentialMovingAverage import datasets import evaluation import likelihood import sde_lib from absl import flags import torch from torch.utils import tensorboard from torchvision.utils import make_grid, save_image from utils import save_checkpoint, restore_checkpoint

johndpope commented 3 weeks ago

at quick glance - this paper originated from google / tensorflow - so it seems the training / eval test sets that are baked into tensorflow got ported across.

"""Return training and evaluation/test datasets from config files.""" import jax import tensorflow as tf import tensorflow_datasets as tfds

plus tensorboard