DL-Art-School/codes/models/eval/flow_gaussian_nll.py

43 lines
1.5 KiB
Python
Raw Normal View History

2020-12-02 21:09:54 +00:00
import os
import torch
import os.path as osp
import torchvision
from torch.utils.data import DataLoader
from tqdm import tqdm
2020-12-02 21:09:54 +00:00
import models.eval.evaluator as evaluator
from pytorch_fid import fid_score
# Evaluate how close to true Gaussian a flow network predicts in a "normal" pass given a LQ/HQ image pair.
from data.image_folder_dataset import ImageFolderDataset
from models.archs.srflow_orig.flow import GaussianDiag
class FlowGaussianNll(evaluator.Evaluator):
def __init__(self, model, opt_eval, env):
super().__init__(model, opt_eval, env)
self.batch_sz = opt_eval['batch_size']
self.dataset = ImageFolderDataset(opt_eval['dataset'])
self.dataloader = DataLoader(self.dataset, self.batch_sz)
def perform_eval(self):
total_zs = 0
z_loss = 0
self.model.eval()
2020-12-02 21:09:54 +00:00
with torch.no_grad():
print("Evaluating FlowGaussianNll..")
for batch in tqdm(self.dataloader):
dev = self.env['device']
z, _, _ = self.model(gt=batch['hq'].to(dev),
lr=batch['lq'].to(dev),
2020-12-02 21:09:54 +00:00
epses=[],
reverse=False,
add_gt_noise=False)
for z_ in z:
z_loss += GaussianDiag.logp(None, None, z_).mean()
total_zs += 1
self.model.train()
2020-12-02 21:09:54 +00:00
return {"gaussian_diff": z_loss / total_zs}