python中mxnet_mxnet Python 笔记

1、查看mxnet的版本

import mxnet as mx

mx.__version__

2、扩展nd的维度

image_data = mx.random.normal(shape = (3, 112, 112))

image_data.expand_dims(axis = 0)

3、装载网络结构

import mxnet as mx

train_net = mx.gluon.model_zoo.get_model("ResNet34_V1",pretrianed = False)

4、修改预装载的网络结构

from mxnet.gluon import nn

train_net = ....(参考3)

conv_0 = nn.Conv2d(channels = 64, kernel_size = (3, 3), strides =(1,1), padding = (1, 1), use_bais = False)

conv_0.collect_params().initialize(mx.init.Xavier())

train_net.features.register_child(conv_0, "0")

train_net.features.hybridize()

5、画出网络结构

x =mx.sym.var('data')

mx.viz.plot_network(train_net(x), shape = {'data':(1, 3, 112, 112)} ).view()

6、上下文

num_gpus = 0

ctx =