validation.py

import torch
from torch.autograd import Variable
import time
import sys

from utils import *


def val_epoch(epoch, data_loader, model, criterion, opt, logger):
    print('validation at epoch {}'.format(epoch))

    model.eval()

    batch_time = AverageMeter()
    data_time = AverageMeter()
    losses = AverageMeter()
    top1 = AverageMeter()
    top5 = AverageMeter()

    end_time = time.time()
    for i, (inputs, targets) in enumerate(data_loader):
        data_time.update(time.time() - end_time)

        if not opt.no_cuda:
            targets = targets.cuda()
        with torch.no_grad():
            inputs = Variable(inputs)
            targets = Variable(targets)
        outputs = model(inputs)
        loss = criterion(outputs, targets)
        prec1, prec5 = calculate_accuracy(outputs.data, targets.data, topk=(1, 5))
        top1.update(prec1, inputs.size(0))
        top5.update(prec5, inputs.size(0))

        losses.update(loss.data, inputs.size(0))

        batch_time.update(time.time() - end_time)
        end_time = time.time()

        if i % 10 == 0:
            print('Epoch: [{0}][{1}/{2}]\t'
                  'Time {batch_time.val:.5f} ({batch_time.avg:.5f})\t'
                  'Data {data_time.val:.5f} ({data_time.avg:.5f})\t'
                  'Loss {loss.val:.4f} ({loss.avg:.4f})\t'
                  'Prec@1 {top1.val:.5f} ({top1.avg:.5f})\t'
                  'Prec@5 {top5.val:.5f} ({top5.avg:.5f})'.format(
                epoch,
                i + 1,
                len(data_loader),
                batch_time=batch_time,
                data_time=data_time,
                loss=losses,
                top1=top1,
                top5=top5))

    logger.log({'epoch': epoch,
                'loss': losses.avg.item(),
                'prec1': top1.avg.item(),
                'prec5': top5.avg.item()})

    return losses.avg.item(), top1.avg.item()