You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In DinoV2-vs-ResnetClassification notebook. In Resnet model1 training step - optimizer used is initialized with model.parameters(), which is of DinoV2 model. Shouldn't we use a dedicated optimizer for model1 training like optimizer1 = optim.Adam(model1.parameters(), lr=0.000001). The Resnet accuracy improves to approx 90% using this...
BR
The text was updated successfully, but these errors were encountered:
Hi @purnasai
In DinoV2-vs-ResnetClassification notebook. In Resnet model1 training step - optimizer used is initialized with model.parameters(), which is of DinoV2 model. Shouldn't we use a dedicated optimizer for model1 training like optimizer1 = optim.Adam(model1.parameters(), lr=0.000001). The Resnet accuracy improves to approx 90% using this...
BR
The text was updated successfully, but these errors were encountered: