Can't load in pre-trained model

[atcold]

Can't load in pre-trained model

[Atcold]

You have that (ref.)

def save_checkpoint(state, is_best, filename='checkpoint.pth.tar'):
    torch.save(state, filename)
    if is_best:
        shutil.copyfile(filename, 'model_best.pth.tar')

will save a file named model_best.pth.tar when is_best == True (ref.)

# remember best prec@1 and save checkpoint
is_best = prec1 > best_prec1
best_prec1 = max(prec1, best_prec1)
save_checkpoint({
    'epoch': epoch + 1,
    'state_dict': model.state_dict(),
    'best_prec1': best_prec1,
}, is_best)

So, I guess your training procedure has a decreasing accuracy over time trend?

Hold on, you get model.pth.tar??? I was expecting you to tell me you were getting checkpoint.pth.tar instead. What code are you actually running?