State-of-the-Art Deep Learning scripts organized by models - easy to train and deploy with reproducible accuracy and performance on enterprise-grade infrastructure.
When running a benchmark on flowers dataset which only has 5 classes, I got a size mismatch error regardless of --num-classes I provided.
Apparently, classes argument was not forwarded to the ResNet initialization. Once fixed, I managed to successfully run training on the dataset with 5 classes.
When running a benchmark on flowers dataset which only has 5 classes, I got a size mismatch error regardless of
--num-classes
I provided.Apparently,
classes
argument was not forwarded to theResNet
initialization. Once fixed, I managed to successfully run training on the dataset with 5 classes.