Kirill
Results
2
comments of
Kirill
> Try to reduce the learning rate, this should solve the problem. The models are usually trained with a large batchsize, eg.: 128 for efficient_det d3, when adjusting this parameter...
> > How do you know, that 128 is optimal for d3? How much images for d0? I have only 28 examples in my whole dataset, using batch_size=1 and learning_rate=0.001....