a-PyTorch-Tutorial-to-Image-Captioning
a-PyTorch-Tutorial-to-Image-Captioning copied to clipboard
What is the traing v.s. validation loss on your trained model
Thanks for the tutorial and code implementation!
I am trying to train the model on my custom dataset and I encountered high loss on both training dataset and validation dataset (~5.8-7.0 loss). But oddly the accuracy is fair (40-60%) given the high loss.
May I ask you what is the loss value you got by the end of your trained model. It could be a reference to me whether my loss value makes sense/ could be a metrics to the performance.
hi, I had nearly the same loss(5.). Do you think the model learns or not? My accuracy is about %50