File size: 2,322 Bytes
be51ac0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
import torch
import torch.nn as nn
import torch.optim as optim
import torchvision
import torchvision.transforms as transforms
import gradio as gr
import numpy as np
from PIL import Image

# ANN model
class ANN(nn.Module):
    def __init__(self):
        super(ANN, self).__init__()
        self.fc1 = nn.Linear(28 * 28, 128)
        self.fc2 = nn.Linear(128, 128)
        self.fc3 = nn.Linear(128, 128)
        self.fc4 = nn.Linear(128, 10)
        self.relu = nn.ReLU()

    def forward(self, x):
        x = x.view(-1, 28 * 28)
        x = self.relu(self.fc1(x))
        x = self.relu(self.fc2(x))
        x = self.relu(self.fc3(x))
        x = self.fc4(x)
        return x

# Load dataset
transform = transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize((0.5,), (0.5,))
])
trainset = torchvision.datasets.MNIST(root='./data', train=True, download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=128, shuffle=True)

# Train model
model = ANN()
loss_fn = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)

def train_model(epochs=1):
    model.train()
    for epoch in range(epochs):
        running_loss = 0.0
        for images, labels in trainloader:
            optimizer.zero_grad()
            outputs = model(images)
            loss = loss_fn(outputs, labels)
            loss.backward()
            optimizer.step()
            running_loss += loss.item()
        print(f"Epoch {epoch+1}, Loss: {running_loss / len(trainloader):.4f}")
    model.eval()

# Call it once at the start
train_model(epochs=1)

# Inference
def predict_digit(img):
    img = img.convert('L').resize((28, 28))  # grayscale and resize
    img = np.array(img).astype(np.float32)
    img = (img - 127.5) / 127.5  # normalize to [-1, 1]
    img_tensor = torch.tensor(img).unsqueeze(0).unsqueeze(0)

    with torch.no_grad():
        output = model(img_tensor)
        _, predicted = torch.max(output, 1)
    return f"Prediction: {predicted.item()}"

gr.Interface(
    fn=predict_digit,
    inputs=gr.Image(image_mode="L", shape=(280, 280), invert_colors=True, source="canvas"),
    outputs="text",
    title="MNIST Digit Recognizer (MLP)",
    description="Draw a digit and the model will try to predict it after training for 1 epoch on MNIST."
).launch()