https://github.com/facebookresearch/moco/blob/colab-notebook/colab/moco_cifar10_demo.ipynb
def train(net, data_loader, train_optimizer, epoch, args):
net.train()
adjust_learning_rate(optimizer, epoch, args)
adjust_learning_rate is called inside train method and therefore it should make use of the same train_optimizer.
Could this training in the notebook be with wrong results?
https://github.com/facebookresearch/moco/blob/colab-notebook/colab/moco_cifar10_demo.ipynb
def train(net, data_loader, train_optimizer, epoch, args):
net.train()
adjust_learning_rate(optimizer, epoch, args)
adjust_learning_rate is called inside train method and therefore it should make use of the same train_optimizer.
Could this training in the notebook be with wrong results?