1. MNIST Inference (Live WebGPU Handwritten Digit Recognition)
Draw a digit (0-9) on the canvas below. The neural network infers the digit in real-time utilizing the AMEVA-Forge WebGPU backend and pre-trained weights.
Drawing Canvas (280x280)
10-Class Softmax Probabilities
Model Architecture:
Linear(784, 256) → ReLU → Linear(256, 10) → Softmax
AMEVA-Forge Python Neural Network Definition (Running on WebGPU)
[ PyTorch Drop-in API ]import forge as torch
import forge.nn as nn
# 1. Define PyTorch-compatible Neural Network
class MNISTModel(nn.Module):
def __init__(self):
super().__init__()
self.net = nn.Sequential(
nn.Linear(784, 256),
nn.ReLU(),
nn.Linear(256, 10)
)
def forward(self, x):
return self.net(x)
# 2. Deploy to WebGPU device and infer in real-time
model = MNISTModel().to("gpu")
x_gpu = torch.tensor(canvas_28x28_pixels, device="gpu")
logits = model(x_gpu)
predicted_digit = torch.argmax(logits)