torch.nn.Linear(in_features, out_features, bias=True) discription 1

torch.nn.Linear(in_features, out_features, bias=True) discription

import torch
import torch.nn as nn

x = torch.tensor([[1.0, -1.0],
                  [0.0,  1.0],
                  [0.0,  0.0]])

in_features = x.shape[1]  # = 2
out_features = 2

m = nn.Linear(in_features, out_features)
%%%%
results would be 
>>> m.weight
tensor([[-0.4500,  0.5856],
        [-0.1807, -0.4963]])

>>> m.bias
tensor([ 0.2223, -0.6114])
%%%%

Here is what the above code is Doing:
1. We create a tensor x with shape (3, 2)
2. We create a Linear layer with in_features=2 and out_features=2
3. We apply the layer to x

The layer has randomly initialized weights and biases.

Let’s look at the weights and biases:

Similar Posts