shubhomoydas/ad_examples

A collection of anomaly detection methods (iid/point-based, graph and time series) including active learning for anomaly detection/discovery, bayesian rule-mining, description for diversity/explanation/interpretability. Analysis of incorporating label feedback with ensemble and tree-based detectors. Includes adversarial attacks with Graph Convolutional Network.

ad_examples's Issues

Full Api Documentation Require

nowadays, docs with rarely interpretation is difficult to understand the algorithm,
Such as bayes rule sets and other Underdogs have little references.
If can provide a common introduce in interface level may be good for programmers
who have little information of algorithm to start

Fixed imports for test_add notebook

Hello,

Below is the code for fixing import issues within the test_aad ipython notebook.

`import logging
 import numpy as np

from ad_examples.common.gen_samples import get_synthetic_samples
from ad_examples.aad.forest_description import CompactDescriber, MinimumVolumeCoverDescriber, \
BayesianRulesetsDescriber, get_region_memberships
from ad_examples.aad.aad_globals import AadOpts, get_aad_command_args
from ad_examples.aad.demo_aad import get_debug_args, detect_anomalies_and_describe
from ad_examples.common.utils import configure_logger

logger = logging.getLogger(__name__)

# Prepare the aad arguments. It is easier to first create the parsed args and
# then create the actual AadOpts from the args
args = get_aad_command_args(debug=True, debug_args=get_debug_args())

# configure_logger(args)  # create a ./temp folder in case you need debug logs
opts = AadOpts(args)
logger.debug(opts.str_opts())

np.random.seed(opts.randseed)

# load synthetic (toy 2) dataset
x, y = get_synthetic_samples(stype=2)

# run interactive anomaly detection loop
model, x_transformed, queried, ridxs_counts, region_extents = detect_anomalies_and_describe(x, y, opts)`

module sklearn.externals.six was removed in the version 0.23

Hello,

Running the code with recent sklearn versions leads to the issues like the following:

  File "/home/matwey/lab/venv/lib/python3.6/site-packages/ad_examples/aad/aad_support.py", line 9, in <module>
    from .forest_aad_detector import *
  File "/home/matwey/lab/venv/lib/python3.6/site-packages/ad_examples/aad/forest_aad_detector.py", line 14, in <module>
    from .random_split_trees import *
  File "/home/matwey/lab/venv/lib/python3.6/site-packages/ad_examples/aad/random_split_trees.py", line 21, in <module>
    from sklearn.externals import six
ImportError: cannot import name 'six'

Get number positivity

I have a rule that throws numbers between -50 and 50 randomly, is there any way to predict the sign (positive or negative) of the next release with at least 90% accuracy based on a historical record?

Real dataset example require in the project

Now example are simulated data and may not have real application scene which
may have more effect on the understand of streaming scene, could you provide some
real examples ?

how to use various methods in the jupyter notebook

I'm trying to use your code in Jupyter notebook's Python kernel.
I only know how to install using pip, can you tell me how to use it?