def output_attention(val_loader, model, epoch, use_cuda, save_dir):
batch_time = AverageMeter()
data_time = AverageMeter()
top1 = AverageMeter()
top5 = AverageMeter()
model.eval()
end = time.time()
bar = Bar('Processing', max=len(val_loader))
fw = open(os.path.join(save_dir, 'attention.txt'), 'w')
for batch_idx, (inputs, targets) in enumerate(val_loader):
data_time.update(time.time() - end)
if use_cuda:
inputs, targets = inputs.cuda(), targets.cuda()
inputs, targets = torch.autograd.Variable(inputs, volatile=True), torch.autograd.Variable(targets)
probs, attention = model(inputs)
bs, c, w, h = attention.size()
attention = attention.sum(1)
attention = attention.cpu().data.numpy()
attention = attention.reshape((bs, -1))
for index in range(bs):
hot = ''
for j in range(w * h):
hot += '{:.3f} '.format(attention[index][j])
hot += '\n'
fw.write(hot)
prec1, prec5 = accuracy(probs.data, targets.data, topk=(1,5))
top1.update(prec1[0], inputs.size(0))
top5.update(prec5[0], inputs.size(0))
bar.shuffix = '({batch}/{size}) Data: {data:.3f}s | Batch: {bt:.3f}s | Total: {total:} | ETA: {eta:} | top1: {top1: .4f} | top5: {top5: .4f}'.format(
batch=batch_idx + 1,
size=len(val_loader),
data=data_time.avg,
bt=batch_time.avg,
total=bar.elapsed_td,
eta=bar.eta_td,
top1=top1.avg,
top5=top5.avg
)
bar.next()
bar.finish()
fw.close()
def output_prob(val_loader, model, epoch, use_cuda, save_dir, image_names,
criterion):
global avg_acc
batch_time = AverageMeter()
data_time = AverageMeter()
top1 = AverageMeter()
top5 = AverageMeter()
model.eval()
end = time.time()
bar = Bar('Processing', max=len(val_loader))
if not os.path.exists(save_dir):
os.mkdir(save_dir)
fw = open(os.path.join(save_dir, 'test_top10_probability_336.txt'), 'w')
submit_result = []
count = 0
class_preds = []
class_targets = []
for batch_idx, (inputs, targets, weights) in enumerate(val_loader):
data_time.update(time.time() - end)
if use_cuda:
inputs, targets = inputs.cuda(), targets.cuda()
inputs, targets = torch.autograd.Variable(
inputs, volatile=True), torch.autograd.Variable(targets)
bs, ncrops, c, h, w = inputs.size(0), inputs.size(1), inputs.size(
2), inputs.size(3), inputs.size(4)
inputs = inputs.view(-1, c, h, w)
outputs, sub_outputs = model(inputs)
outputs = outputs.view(bs, ncrops, -1).mean(1)
targets = torch.squeeze(targets)
#loss = criterion(outputs, targets)
prec1, prec5 = accuracy(outputs.data, targets.data, topk=(1, 5))
probs = torch.nn.functional.softmax(outputs)
topk_probs, topk_preds = probs.topk(20, 1, True, True)
pred_float, pred = outputs.data.topk(5, 1, True, True)
pred = pred.cpu().numpy()
pred = pred.reshape((-1))
output_probs = topk_probs.data.cpu().numpy()
output_preds = topk_preds.data.cpu().numpy()
real_targets = targets.cpu().data.numpy()
#real_targets = real_targets.reshape(-1)
#loss_num = loss.data
for index in range(outputs.data.size(0)):
probability = str(real_targets[index])
#probability += ' {} {:.3f}'.format(image_names[count], loss_num[index])
for i in range(20):
probability += ' {}:{:.3f}'.format(output_preds[index][i],
output_probs[index][i])
fw.write(probability)
fw.write('\n')
submit_each = []
submit_each.append(image_names[count])
submit_each.extend(list(pred[index * 5:(index + 1) * 5]))
submit_result.append(submit_each)
count += 1
class_preds.append(list(pred[index * 5:(index + 1) * 5]))
class_targets.append(real_targets[index])
top1.update(prec1[0], inputs.size(0))
top5.update(prec5[0], inputs.size(0))
batch_time.update(time.time() - end)
end = time.time()
bar.shuffix = '({batch}/{size}) | Batch: {bt:.3f}s | Total: {total:} | ETA: {eta:} | top1: {top1: .4f} | top5: {top5: .4f}'.format(
batch=batch_idx + 1,
size=len(val_loader),
bt=batch_time.avg,
total=bar.elapsed_td,
eta=bar.eta_td,
top1=top1.avg,
top5=top5.avg)
bar.next()
bar.finish()
df = pd.DataFrame(submit_result,
columns=[
'image_id', 'predicted_0', 'predicted_1',
'predicted_2', 'predicted_3', 'predicted_4'
])
df.to_csv(os.path.join(save_dir, 'test_280.csv'),
index=False,
header=False,
sep=' ')
class_acc1, class_acc5 = class_accuracy(np.array(class_targets),
np.array(class_preds))
print('average per class accuracy: top1: {:.4f} top5: {:.4f}'.format(
class_acc1, class_acc5))
fw.close()
