Error load models

[microvnn]

sys.path.append('./vendors/Sphereface')
config_path = os.path.join('./vendors', 'Sphereface',
'20220428_232053', 'config.yml')
with open(config_path, 'r') as f:
test_config = yaml.load(f, yaml.SafeLoader)

build model

bkb_net = build_from_cfg(
test_config['model']['backbone']['net'],
'model.backbone',
)
bkb_net.eval()
checkpoint = os.path.join('./vendors', 'Sphereface',
'20220428_232053', 'models', 'backbone_80000.pth')
state_dict = torch.load(checkpoint, map_location=torch.device('cpu'))
bkb_net.load_state_dict(state_dict)
print(bkb_net)

Please help me.

Result:
RuntimeError: Error(s) in loading state_dict for SFNet_deprecated:
Missing key(s) in state_dict: "layer1.0.conv1.weight", "layer1.0.conv1.bias", "layer1.1.conv1.weight", "layer1.1.conv1.bias", "layer1.1.conv2.weight", "layer1.1.conv2.bias", "layer2.0.conv1.weight", "layer2.0.conv1.bias", "layer2.1.conv1.weight", "layer2.1.conv1.bias", "layer2.1.conv2.weight", "layer2.1.conv2.bias", "layer2.2.conv1.weight", "layer2.2.conv1.bias", "layer2.2.conv2.weight", "layer2.2.conv2.bias", "layer3.0.conv1.weight", "layer3.0.conv1.bias", "layer3.1.conv1.weight", "layer3.1.conv1.bias", "layer3.1.conv2.weight", "layer3.1.conv2.bias", "layer3.2.conv1.weight", "layer3.2.conv1.bias", "layer3.2.conv2.weight", "layer3.2.conv2.bias", "layer3.3.conv1.weight", "layer3.3.conv1.bias", "layer3.3.conv2.weight", "layer3.3.conv2.bias", "layer3.4.conv1.weight", "layer3.4.conv1.bias", "layer3.4.conv2.weight", "layer3.4.conv2.bias", "layer4.0.conv1.weight", "layer4.0.conv1.bias", "layer4.1.conv1.weight", "layer4.1.conv1.bias", "layer4.1.conv2.weight", "layer4.1.conv2.bias", "fc.weight", "fc.bias".
Unexpected key(s) in state_dict: "module.layer1.0.conv1.weight", "module.layer1.0.conv1.bias", "module.layer1.1.conv1.weight", "module.layer1.1.conv1.bias", "module.layer1.1.conv2.weight", "module.layer1.1.conv2.bias", "module.layer2.0.conv1.weight", "module.layer2.0.conv1.bias", "module.layer2.1.conv1.weight", "module.layer2.1.conv1.bias", "module.layer2.1.conv2.weight", "module.layer2.1.conv2.bias", "module.layer2.2.conv1.weight", "module.layer2.2.conv1.bias", "module.layer2.2.conv2.weight", "module.layer2.2.conv2.bias", "module.layer3.0.conv1.weight", "module.layer3.0.conv1.bias", "module.layer3.1.conv1.weight", "module.layer3.1.conv1.bias", "module.layer3.1.conv2.weight", "module.layer3.1.conv2.bias", "module.layer3.2.conv1.weight", "module.layer3.2.conv1.bias", "module.layer3.2.conv2.weight", "module.layer3.2.conv2.bias", "module.layer3.3.conv1.weight", "module.layer3.3.conv1.bias", "module.layer3.3.conv2.weight", "module.layer3.3.conv2.bias", "module.layer3.4.conv1.weight", "module.layer3.4.conv1.bias", "module.layer3.4.conv2.weight", "module.layer3.4.conv2.bias", "module.layer4.0.conv1.weight", "module.layer4.0.conv1.bias", "module.layer4.1.conv1.weight", "module.layer4.1.conv1.bias", "module.layer4.1.conv2.weight", "module.layer4.1.conv2.bias",
"module.fc.weight", "module.fc.bias".

[ydwen]

Hey microvnn,

You can try wrapping the model to DDP before loading the parameters.

example:
ddp_model = DDP(model, device_ids=[rank])