SebChw / Actually-Robust-Training

Actually Robust Training - Tool Inspired by Andrej Karpathy "Recipe for training neural networks". It allows you to decompose your Deep Learning pipeline into modular and insightful "Steps". Additionally it has many features for testing and debugging neural nets.

MIT License

44 stars 0 forks source link

default force rerun to false #167

Closed kordc closed 1 year ago

kordc commented 1 year ago

As we're talking about it several times before I propose this solution. For now even when we add an extra info to the logger all steps will be re-run. As we provide the rerun flag I thik it'd be beneficial for everyone of we default the rerun to false and just infor the user about the need of the rerun when the model code was changed.

If you feel differently feel free to open the discussion below

155

SebChw commented 1 year ago

I completely agree. Just inform that they should rerun it