0181fe9fc837b8cdcc612375f2f174c5652ef87a,example/mpii.py,,validate,#Any#Any#Any#Any#Any#,194

Before Change

        // generate predictions
        preds = final_preds(output[-1].data.cpu(), meta["center"], meta["scale"], [64, 64])
        for n in range(output[-1].size(0)):
            predictions[meta["index"][n], :, :] = preds[n, :, :]

After Change

        score_map = output[-1].data.cpu()
        if flip:
            flip_input_var = torch.autograd.Variable(
                    torch.from_numpy(fliplr(inputs.clone().numpy())).float().cuda(), 
                    volatile=True
                )
            flip_output_var = model(flip_input_var)
            flip_output = flip_back(flip_output_var[-1].data.cpu())
            score_map += flip_output



        loss = 0

Instances