网络编程
位置：首页>> 网络编程>> Python编程>> pytorch获取模型某一层参数名及参数值方式

pytorch获取模型某一层参数名及参数值方式

作者：HuiYu-Li　　发布时间：2022-07-03 06:10:28　

标签：pytorch,模型,参数名,参数值

1、Motivation：

I wanna modify the value of some param;

I wanna check the value of some param.

The needed function:

2、state_dict() #generator type

model.modules()#generator type

named_parameters()#OrderDict type


from torch import nn
import torch
#creat a simple model
model = nn.Sequential(
 nn.Conv3d(1,16,kernel_size=1),
 nn.Conv3d(16,2,kernel_size=1))#tend to print the W of this layer
input = torch.randn([1,1,16,256,256])
if torch.cuda.is_available():
 print('cuda is avaliable')
 model.cuda()
 input = input.cuda()
#打印某一层的参数名
for name in model.state_dict():
 print(name)
#Then I konw that the name of target layer is '1.weight'

#schemem1(recommended)
print(model.state_dict()['1.weight'])

#scheme2
params = list(model.named_parameters())#get the index by debuging
print(params[2][0])#name
print(params[2][1].data)#data

#scheme3
params = {}#change the tpye of 'generator' into dict
for name,param in model.named_parameters():
params[name] = param.detach().cpu().numpy()
print(params['0.weight'])

#scheme4
for layer in model.modules():
if(isinstance(layer,nn.Conv3d)):
 print(layer.weight)

#打印每一层的参数名和参数值
#schemem1(recommended)
for name,param in model.named_parameters():
 print(name，param)

#scheme2
for name in model.state_dict():
 print(name)
 print(model.state_dict()[name])

来源：https://blog.csdn.net/weixin_44058333/article/details/92691656

0
投稿

猜你喜欢

手机版 网络编程 asp之家 www.aspxhome.com