remove all chkpt files from logs to save space

add simple conv model with relu and batchnorm

Revert "change conv model to use activation function and rerun a training run" as it degrades performance

change conv model to use activation function and rerun a training run

add non-pretrained-focus-liteNN model

rerun and add logs for focusResNet101_pretrained_150

rerun and add logs for focusResNet_150

rerun and add logs for focusMSE_150

rerun and add logs for focusMAE_150

rerun and add logs for focusConvMSE_besthyp_150

rerun and add logs for focusConvMSE_150

remove old runs to rerun all experiments

add results of simple cnn benchmark

add runs for first experiments

remove all runs due to results being not useful (error in loss function)

remove all checkpoint files from earlier runs

run experiments comparing MSE and MAE