Home › Claude › aifoundations › alexnet_cnn_imagenet › AlexNet was trained for roughly how many epochs …
AlexNet was trained for roughly how many epochs on ImageNet?
AAbout 10 epochs
BAbout 30 epochs
CAbout 90 epochs
DAbout 300 epochs
Answer & Solution
Correct answer: C. About 90 epochs
Roughly 90 cycles (epochs) through 1.2 million training images, taking 5 to 6 days on two NVIDIA GTX 580 GPUs.
Related questions
Image similarity in AlexNet's last hidden layer is measured byThe SGD update rule for AlexNet usesRemoving any single middle convolutional layer of AlexNet costs roughlyAt test time dropout is approximated byAlexNet uses overlapping pooling withAlexNet's second form of data augmentation isAlexNet's first form of data augmentation isWhy are ReLU networks faster to train than tanh networks?