import cwru
data = cwru.CWRU("12DriveEndFault", "1797", 384)
Then, you can use data.X_train
, data.y_train
, data.X_test
, data.y_test
, data.labels
, data.nclasses
to train and evaluate your data driven fault diagnostic algrithms.
Not have a local copy of CWRU Bearing data on your machine? Don't be worry, These .mat
files required by your code will be automatically downloaded from Case Western Reserve University Bearing Data Center, renamed properly with their corresponding class, stored in ~/Datasets/CWRU
directory to avoid downloading in future.
The constructor of class cwru.CWRU
has three arguments:
- exp: experiment, supporting
"12DriveEndFault", "12FanEndFault", "48DriveEndFault"
- rpm: rpm during testing
- length: length of the signal slice, namely
X_train.shape[1]
Install by pip:
$ pip install --user cwru
Or install manually:
$ python setup.py install
You can run unit test by:
$ python -m unittest discover