unexpected keyword argument trying to instantiate a class inheriting from torch.nn.Module

Question

I have seen similar questions, but most seem a little more involved. My problem seems to me to be very straightforward, yet I cannot figure it out. I am simply trying to define a class and then instantiate it, but the arguments passed to the constructor are not recognized.

import torch
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
from torch.utils.data import DataLoader
import torchvision.transforms as transforms

# fully connected network
class NN(nn.Module):
    def __int__(self, in_size, num_class):
        super(NN, self).__init__()
        self.fc1 = nn.Linear(in_size, 50)
        self.fc2 = nn.Linear(50, num_class)

    def forward(self, x):
        x = F.relu(self.fc1(x))
        x = self.fc2(x)
        return x

# initialize network
model = NN(in_size=input_size, num_class=num_classes) 

I get the error: __init__() got an unexpected keyword argument 'in_size' I am using Python 3.1, PyTorch 1.7.1, using PyCharm on macOS Monterey. Thank you

Answer 1

You have a typo: it should be __init__ instead of __int__ .