init_params.py 1.9 KB
Newer Older
Z
zhaoting 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
# Copyright 2020 Huawei Technologies Co., Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ============================================================================
"""Parameters utils"""

from mindspore import Tensor
from mindspore.common.initializer import initializer, TruncatedNormal

def init_net_param(network, initialize_mode='TruncatedNormal'):
    """Init the parameters in net."""
    params = network.trainable_params()
    for p in params:
        if isinstance(p.data, Tensor) and 'beta' not in p.name and 'gamma' not in p.name and 'bias' not in p.name:
            if initialize_mode == 'TruncatedNormal':
                p.set_parameter_data(initializer(TruncatedNormal(0.03), p.data.shape(), p.data.dtype()))
            else:
                p.set_parameter_data(initialize_mode, p.data.shape(), p.data.dtype())


def load_backbone_params(network, param_dict):
    """Init the parameters from pre-train model, default is mobilenetv2."""
    for _, param in net.parameters_and_names():
        param_name = param.name.replace('network.backbone.', '')
        name_split = param_name.split('.')
        if 'features_1' in param_name:
            param_name = param_name.replace('features_1', 'features')
        if 'features_2' in param_name:
            param_name = '.'.join(['features', str(int(name_split[1]) + 14)] + name_split[2:])
        if param_name in param_dict:
            param.set_parameter_data(param_dict[param_name].data)