Simple Neural Network Using PyTorch

huangapple go评论59阅读模式
英文:

Simple Neural Network Using Pytorch

问题

I want to build Simple Neural Network with pytorch.
And I want to teach this network.

the network has
y = w(weight) * x + b(bias)
with w = 3 and b = 0

so I have the data
x = [1,2,3,4,5,6]
y = [3,6,9,12,15,18]

But I have some problem while building this Simple Neural Network

import torch
import torch.nn as nn

class MyNeuralNetwork(nn.Module):
    def __init__(self):
        super(MyNeuralNetwork, self).__init__()
        self.layer=nn.Linear(in_features=1, out_features=1, bias=True)
        
        weight = torch.rand(1)
        bias = torch.rand(1)
        self.layer.weight = nn.Parameter(weight)
        self.layer.bias = nn.Parameter(bias)
    
    def forward(self, input):
        output = self.layer(input)
        return output
    
model = MyNeuralNetwork().to("cpu")
print(model)

print(f"weight : {model.layer.weight}")
print(f"bias : {model.layer.bias}")

input = torch.tensor([1.]).view(-1,1)
model(input)
# out = model(input)
# print(out)
# y = torch.tensor([3.,6.,9.,12.,15.,18.])

I have an error which says that
"RuntimeError: mat2 must be a matrix, got 1-D tensor"

What should I do to fix this problem?
Thanks OTL....

英文:

I want to build Simple Neural Network with pytorch.
And I want to teach this network.

the network has
y = w(weight) * x + b(bias)
with w = 3 and b = 0

so I have the data
x = [1,2,3,4,5,6]
y = [3,6,9,12,15,18]

But I have some problem while building this Simple Neural Network

import torch
import torch.nn as nn

class MyNeuralNetwork(nn.Module):
    def __init__(self):
        super(MyNeuralNetwork, self).__init__()
        self.layer=nn.Linear(in_features=1, out_features=1, bias=True)
        
        weight = torch.rand(1)
        bias = torch.rand(1)
        self.layer.weight = nn.Parameter(weight)
        self.layer.bias = nn.Parameter(bias)
    
    def forward(self, input):
        output = self.layer(input)
        return output
    
model = MyNeuralNetwork().to("cpu")
print(model)

print(f"weight : {model.layer.weight}")
print(f"bias : {model.layer.bias}")

input = torch.tensor([1.]).view(-1,1)
model(input)
# out = model(input)
# print(out)
# y = torch.tensor([3.,6.,9.,12.,15.,18.])

I have an error which says that
"RuntimeError: mat2 must be a matrix, got 1-D tensor"

What should I do to fix this problem?
Thanks OTL....

答案1

得分: 1

你的权重应该是大小为[1, 1],但你已经用一个一维权重张量进行了覆盖。我不知道为什么你要覆盖层的权重和偏差,但要么用正确的形状进行覆盖,要么如果你只是删除__init__中的4行代码应该可以工作,或者修改成这样:

self.layer.weight = nn.Parameter(torch.rand(1, 1))  # 2维矩阵
英文:

your weight should be of size [1, 1] which you have overrided with a 1d weight tensor. I do not know why you have overwrite the weight and bias of the layer, but either override with correct shape or if you just remove the 4 lines inside the init should work.or change but this:

self.layer.weight  = nn.Parameter(torch.rand(1, 1)) # 2d Matrix

huangapple
  • 本文由 发表于 2023年2月16日 16:29:54
  • 转载请务必保留本文链接:https://go.coder-hub.com/75469573.html
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定