To train the GoogLeNet model on datasets like Fashion-MNIST, the input images must be resized to a resolution of $96 \times 96$ pixels before initiating the training process. The following multi-framework snippets demonstrate how to initialize the model, configure the data loader with the appropriate image dimensions, and execute the training loop:

PyTorch Implementation:

model = GoogleNet(lr=0.01)
trainer = d2l.Trainer(max_epochs=10, num_gpus=1)
data = d2l.FashionMNIST(batch_size=128, resize=(96, 96))
model.apply_init([next(iter(data.get_dataloader(True)))[0]], d2l.init_cnn)
trainer.fit(model, data)

MXNet Implementation:

model = GoogleNet(lr=0.01)
trainer = d2l.Trainer(max_epochs=10, num_gpus=1)
data = d2l.FashionMNIST(batch_size=128, resize=(96, 96))
trainer.fit(model, data)

JAX Implementation:

model = GoogleNet(lr=0.01)
trainer = d2l.Trainer(max_epochs=10, num_gpus=1)
data = d2l.FashionMNIST(batch_size=128, resize=(96, 96))
trainer.fit(model, data)

TensorFlow Implementation:

trainer = d2l.Trainer(max_epochs=10)
data = d2l.FashionMNIST(batch_size=128, resize=(96, 96))
with d2l.try_gpu():
    model = GoogleNet(lr=0.01)
    trainer.fit(model, data)