solve task 2

Task 2. Write Classification Neural Network with 10 output neurons using pytorch Your network must have methods: . forward() - performs forward pass Your network must return 10 values. Think which lo Your network must have 10 output neuron such that you can classify 10 different classes I import torch import torch.optin as optin import torchvision . trans sport Datatoanstorms from sklearn. model_selection import train_test_split device = ' cuda' if torch. cuda. is_available() else "cpu" transform = transforms. Compose( [ transforms. Normalize( (0.5. 0.5. 0.5). (0.5, 8.5, 8.5)) full_dataset = torchvision. datasets. CIFARIO( roots'./data". train=True. crain_size = intle.s * len(full_dataset)) train_dataset, test_dataset = torch. utils . data. random_split( full_dataset, [train_size, test_size]) test Loader bataloader train dataset, batch size-32, shuffle-Truel Files already downloaded and verified [ ] import ratplot lib.pyplot as pit import torchvision * inshow( 159); nping = ing. nunpy ( ) pit. inshowinp. transpose(nping. (1. 2, o]); et show_train_image(train_loader) images , labels = next (datalter ) inshow( torchvision. utils.make_grid( tra Print( 'Labels , " .join(t {classes [labels].item()1}' for j in range( ten( labels)))) lasses = 1: ' automobile ' 21 ' bird' . 53 'dog $1 'frog's 9: ' truck - show_train_image(train_loader) 20 -KY 20 Labels: horse dog bird truck dog deer deer frog airplane ship airplane cat d mobile bird dog cat [ ] import torch import torch.nn as an class MultiClassClassificationNet(nn.Modute]: * your cuba lassclassificationNet, self)._init_( def forward(self. x): return x think which Loss function you need to use if you output 1 neuron and it is between a and 1 * https: / / pytorch.org/ docs/ stable/generated/ torch. no. CrossEntropyloss. html or this? criterion = = your code goes here * Training Loop is given to Post ( ). tol er = optim. 560 (net. parameters(), lr=8. eal, momentum=.9) or epoch in range(21 or to data in enumerate( train_loader, 8): tabets = Labels for device labels = labels. long() optimizer.zero_grad() loss - criterion ( output optimizer.step() print(f 'Epoch {epoch + 1), Loss: {loss.item()}") correct = 8 total = e with torching in test_ loader labels = data is = labels . toldevice le,"_ labels. long() -. predicted = torch.max(outputs. data, 1) correct += ( predicted == labels). sum().iten( ) work on the test inagesi {lee * correct / total)9') Epoch 10