Python 使用NumPy保存/加载MXNet模型参数

如何将MXNet模型的参数保存到NumPy文件（.npy）中？完成此操作后，如何将这些参数从.npy文件加载回模型中

下面是一个使用MXNet API保存MXNet模型参数的简单示例

import mxnet as mx
from mxnet import gluon
from mxnet.gluon.model_zoo import vision
import numpy as np

num_gpus = 0
ctx = [mx.gpu(i) for i in range(num_gpus)] if num_gpus > 0 else [mx.cpu()]
resnet = vision.resnet50_v2(pretrained=True, ctx=ctx)

parameters = resnet.collect_params()
resnet.save_parameters('model.params')

import mxnet as mx
from mxnet import gluon
from mxnet.gluon.model_zoo import vision
import numpy as np

num_gpus = 0
ctx = [mx.gpu(i) for i in range(num_gpus)] if num_gpus > 0 else [mx.cpu()]
resnet = vision.resnet50_v2(pretrained=True, ctx=ctx)

resnet.load_parameters('model.params', ctx=ctx)

使用MXNet API将参数从文件加载回模型的最小示例

import mxnet as mx
from mxnet import gluon
from mxnet.gluon.model_zoo import vision
import numpy as np

num_gpus = 0
ctx = [mx.gpu(i) for i in range(num_gpus)] if num_gpus > 0 else [mx.cpu()]
resnet = vision.resnet50_v2(pretrained=True, ctx=ctx)

parameters = resnet.collect_params()
resnet.save_parameters('model.params')

import mxnet as mx
from mxnet import gluon
from mxnet.gluon.model_zoo import vision
import numpy as np

num_gpus = 0
ctx = [mx.gpu(i) for i in range(num_gpus)] if num_gpus > 0 else [mx.cpu()]
resnet = vision.resnet50_v2(pretrained=True, ctx=ctx)

resnet.load_parameters('model.params', ctx=ctx)

在上述两个示例中，我都使用MXNETAPI来保存/加载模型参数。与此相反，我希望使用numpy保存/加载模型，然后将这些参数用于我的MXNet模型。我该怎么做