CLI identify¶
The help documentation is available in the terminal from:
peerannot identify --help
We created an interactive tool to compare the WAUM and AUM identifications on the CIFAR-10H dataset. Check it out by clicking this link!
peerannot identify¶
Identify ambiguous tasks using different methods available in peerannot identificationinfo
peerannot identify [OPTIONS] [FOLDERPATH]
Options
- --hard-labels <hard_labels>¶
Path to file of hard labels (only for AUM)
- -K, --n-classes <n_classes>¶
Number of classes
- -s, --method <method>¶
Method to find ambiguous tasks
- --labels <labels>¶
Path to file of crowdsourced answers
- --use-pleiss¶
Use Pleiss et. al (2020) margin instead of Yang’s
- Default:
False
- --model <model>¶
Name of neural network to use. The list is available at peerannot modelinfo
- --n-epochs <n_epochs>¶
Number of training epochs
- --alpha <alpha>¶
Cutoff hyperparameter
- --topk <topk>¶
Use TopK WAUM with k=XXX
- --n-params <n_params>¶
Number of parameters for the logistic regression only
- --lr <lr>¶
Learning rate
- --pretrained¶
Use torch available weights to initialize the network
- Default:
False
- --momentum <momentum>¶
Momentum for the optimizer
- --metadata_path <metadata_path>¶
Path to the metadata of the dataset if different than default
- --decay <decay>¶
Weight decay for the optimizer
- --img-size <img_size>¶
Size of image (square)
- --maxiter-DS <maxiter_ds>¶
Maximum number of iterations for the Dawid and Skene algorithm
- -optim, --optimizer <optimizer>¶
Optimizer for the neural network
- --data-augmentation¶
Perform data augmentation on training set with a random choice between RandomAffine(shear=15), RandomHorizontalFlip(0.5) and RandomResizedCrop
- Default:
False
- --freeze¶
Freeze all layers of the network except for the last one
- Default:
False
- --matrix-file <matrix_file>¶
Path to confusion matrices saved with an aggregation method like DS. If not provided, run DS model
- --hard-labels <hard_labels>¶
Path to file of hard labels
- --seed <seed>¶
random seed
Arguments
- FOLDERPATH¶
Optional argument