self_supervision_benchmark/modeling/jigsaw/resnet_jigsaw_finetune_linear.py [179:208]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    )
    model.StopGradient(blob_in, blob_in)
    resize_c3 = model.AveragePool(
        blob_in, blob_in + '_s5k13_resize', stride=5, kernel=13, pad=0
    )
    bn_c3 = model.SpatialBN(
        resize_c3, resize_c3 + '_bn', 512, epsilon=cfg.MODEL.BN_EPSILON,
        momentum=cfg.MODEL.BN_MOMENTUM, is_test=test_mode
    )
    if cfg.MODEL.BN_NO_SCALE_SHIFT:
        model.param_init_net.ConstantFill(
            [bn_c3 + '_s'], bn_c3 + '_s', value=1.0
        )
        model.param_init_net.ConstantFill(
            [bn_c3 + '_b'], bn_c3 + '_b', value=0.0
        )
    fc_conv3 = model.FC(
        bn_c3, 'conv3_cls', 512 * 4 * 4, num_classes,
        weight_init=('GaussianFill', {'std': 0.01}),
        bias_init=('ConstantFill', {'value': 0.0}),
    )
    model.net.Alias(fc_conv3, 'pred_conv3')
    if not cfg.MODEL.EXTRACT_FEATURES_ONLY:
        model.Accuracy([fc_conv3, labels], 'accuracy_conv3')
        if split == 'train':
            softmax, loss_c3 = model.SoftmaxWithLoss(
                ['pred_conv3', labels],
                ['softmax_conv3', 'loss_conv3'], scale=scale
            )
        elif split in ['test', 'val']:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



self_supervision_benchmark/modeling/supervised/resnet_supervised_finetune_linear.py [173:202]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    )
    model.StopGradient(blob_in, blob_in)
    resize_c3 = model.AveragePool(
        blob_in, blob_in + '_s5k13_resize', stride=5, kernel=13, pad=0
    )
    bn_c3 = model.SpatialBN(
        resize_c3, resize_c3 + '_bn', 512, epsilon=cfg.MODEL.BN_EPSILON,
        momentum=cfg.MODEL.BN_MOMENTUM, is_test=test_mode
    )
    if cfg.MODEL.BN_NO_SCALE_SHIFT:
        model.param_init_net.ConstantFill(
            [bn_c3 + '_s'], bn_c3 + '_s', value=1.0
        )
        model.param_init_net.ConstantFill(
            [bn_c3 + '_b'], bn_c3 + '_b', value=0.0
        )
    fc_conv3 = model.FC(
        bn_c3, 'conv3_cls', 512 * 4 * 4, num_classes,
        weight_init=('GaussianFill', {'std': 0.01}),
        bias_init=('ConstantFill', {'value': 0.0}),
    )
    model.net.Alias(fc_conv3, 'pred_conv3')
    if not cfg.MODEL.EXTRACT_FEATURES_ONLY:
        model.Accuracy([fc_conv3, labels], 'accuracy_conv3')
        if split == 'train':
            softmax, loss_c3 = model.SoftmaxWithLoss(
                ['pred_conv3', labels],
                ['softmax_conv3', 'loss_conv3'], scale=scale
            )
        elif split in ['test', 'val']:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -