app.py

import torch
import torch.nn as nn
import matplotlib.pyplot as plt
import gradio as gr

# Synthetic data
torch.manual_seed(0)
X = torch.linspace(0, 1, 100).unsqueeze(1)
y = 2 * X + 3 + 0.1 * torch.randn_like(X)

# Model
class LinearModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(1, 1)

    def forward(self, x):
        return self.linear(x)

# Training loop
def train(optimizer_name):
    model = LinearModel()
    criterion = nn.MSELoss()
    optimizer = {
        'SGD': torch.optim.SGD(model.parameters(), lr=0.1),
        'Adam': torch.optim.Adam(model.parameters(), lr=0.1),
        'RMSProp': torch.optim.RMSprop(model.parameters(), lr=0.1),
        'Adagrad': torch.optim.Adagrad(model.parameters(), lr=0.1)
    }[optimizer_name]

    losses = []
    for epoch in range(100):
        optimizer.zero_grad()
        output = model(X)
        loss = criterion(output, y)
        loss.backward()
        optimizer.step()
        losses.append(loss.item())
    return losses

# Gradio plot
def plot_optimizer(opt_name):
    losses = train(opt_name)
    plt.figure(figsize=(6, 4))
    plt.plot(losses, label=opt_name)
    plt.xlabel('Epoch')
    plt.ylabel('Loss')
    plt.title(f'{opt_name} Loss Curve')
    plt.grid(True)
    plt.legend()
    return plt.gcf()

demo = gr.Interface(
    fn=plot_optimizer,
    inputs=gr.Dropdown(['SGD', 'Adam', 'RMSProp', 'Adagrad'], label="Choose Optimizer"),
    outputs=gr.Plot(label="Loss Curve"),
    title="Optimizer Comparison on Linear Regression",
    description="Select an optimizer to visualize its convergence on a simple linear regression task.'SGD':, 'Adam', 'RMSProp':,'Adagrad':"
)

if __name__ == "__main__":
    demo.launch()