Contact: Maximilian Bachl, Alexander Hartl. Explores defenses against backdoors and poisoning attacks for Intrusion Detection Systems. Code for "EagerNet" is in the "eager" branch.
For the eager branch, we run:
./learn.py --dataroot CAIA_17.csv --function train_eager_stopping --method nn --lr 0.001 --batchSize 512 --eagerStoppingWeightingMethod "eager_linearly_increasing_weights" --nLayers 3 --layerSize 128
find errors:
Traceback (most recent call last):
File "./learn.py", line 654, in
globals()'%s_%s' % (opt.function, opt.method)
File "./learn.py", line 210, in train_eager_stopping_nn
_ = writer.log_dir
AttributeError: 'SummaryWriter' object has no attribute 'log_dir'
whenever I am trying to run $ ./learn.py --dataroot CAIA_backdoor_15.csv --method=rf --net <path to you random forest> --function prune_backdoor --backdoor --reduceValidationSet <fraction from 0 to 1 that indicates the fraction of the validation dataset that is going to be used> --nSteps 49 --pruneOnlyHarmless --depth
in the ~/Desktop/Research_papers/ids-backdoor-master directory, I am facing this error.