Bug Report for handwritten-digit-classifier

[vsancnaj]

Bug Report for https://neetcode.io/problems/handwritten-digit-classifier

Please describe the bug below and include any steps to reproduce the bug or screenshots if possible.

I applied this code and it marked it as wrong answer:

class Solution(nn.Module):

def init(self):

super().init()

torch.manual_seed(0)

# flatten the image to 1d vector

# self.flatten = nn.Flatten()

self.first = nn.Linear((28*28),512)

self.second = nn.ReLU()

self.third = nn.Dropout(p=0.2)

self.fourth = nn.Linear(512,10)

self.final = nn.Sigmoid()

# Define the architecture here

def forward(self, images: TensorType[float]) -> TensorType[float]:

torch.manual_seed(0)

# x = self.flatten(images)

x = self.first(images) # linear projection

x = self.second(x) # ReLU

x = self.third(x) # Dropout

x = self.fourth(x) # logits (shape: batch × 10)

x = self.final(x) # probabilities

return torch.round(x, decimals=4)

# Return the model's prediction to 4 decimal places

But when I commented out and applied the solution it still marked it as wrong