Open gfrogat opened 6 years ago
In this context it might be interesting to compare the learning / fine tuning of an image classifier on summer and winter images.
We would need a dataset with summer and winter images for this use case.
we can illustrate similar effects by experimenting on different random folds, i.e. cross validation.
Most people don't train models from scratch but fine-tune a pretrained model instead.
The resulting dynamics of the model might therefore be different. On a different note: is not completely clear whether the observed oscillations are due to changes in the convnet filters or in the fully connected parts of the network