Function load_and_assign_npz_dict

tensorlayer/files/utils.py:2077–2112

Restore the parameters saved by ``tl.files.save_npz_dict()``. Parameters ------------- name : str The name of the `.npz` file. network : :class:`Model` The network to be assigned. skip : boolean If 'skip' == True, loaded weights whose name is not found in

(name='model.npz', network=None, skip=False)

def load_and_assign_npz_dict(name='model.npz', network=None, skip=False):
    """Restore the parameters saved by ``tl.files.save_npz_dict()``.

    Parameters
    -------------
    name : str
        The name of the `.npz` file.
    network : :class:`Model`
        The network to be assigned.
    skip : boolean
        If 'skip' == True, loaded weights whose name is not found in network's weights will be skipped.
        If 'skip' is False, error will be raised when mismatch is found. Default False.

    """
    if not os.path.exists(name):
        logging.error("file {} doesn't exist.".format(name))
        return False

    weights = np.load(name, allow_pickle=True)
    if len(weights.keys()) != len(set(weights.keys())):
        raise Exception("Duplication in model npz_dict %s" % name)

    net_weights_name = [w.name for w in network.all_weights]

    for key in weights.keys():
        if key not in net_weights_name:
            if skip:
                logging.warning("Weights named '%s' not found in network. Skip it." % key)
            else:
                raise RuntimeError(
                    "Weights named '%s' not found in network. Hint: set argument skip=Ture "
                    "if you want to skip redundant or mismatch weights." % key
                )
        else:
            assign_tf_variable(network.all_weights[net_weights_name.index(key)], weights[key])
    logging.info("[*] Model restored from npz_dict %s" % name)


def save_ckpt(mode_name='model.ckpt', save_dir='checkpoint', var_list=None, global_step=None, printable=False):