Features
1
Easy Access to ML Implementations
We pre-registered diverse preprocessing and learning algorithms from state-of-the-art toolboxes e.g. scikit-learn, keras and imbalanced learn, which you can choose to rapidly build custom pipelines.
2
Hyperparameter Optimization
With PHOTONAI you can seamlessly switch between diverse hyperparameter optimization strategies, such as (random) grid-search or bayesian optimization (scikit-optimize, smac3).
3
Model Sharing
PHOTONAI provides a standardized format for sharing and loading optimized models and pipelines across platforms with only one line of code.
4
Extended ML Pipeline
You can build custom sequences of processing and learning algorithms with a simple syntax. PHOTONAI offers extended pipeline functionality such as parallel sequences, custom callbacks in-between pipeline elements, AND- and OR-Operations, as well as the possibility to flexibly position data augmentation, class balancing or learning algorithms anywhere in the pipeline.
5
Automation
While you concentrate on selecting appropriate processing steps, learning algorithms, hyperparameters and training parameters, PHOTONAI automates the nested cross-validated optimization and evaluation loop for any custom pipeline.
6
Results Visualization
PHOTONAI comes with extensive logging of all information in the training, testing and hyperparameter optimization process. In addition, optimum performances and the hyperparameter optimization progress are visualized in the PHOTONAI Explorer.
PHOTONAI is especially well-suited in contexts such as applied machine learning research in medicine and the Life Sciences, which require iterative evaluation of novel approaches.
See our mission statement and give us a star on Github!
Code Example
imports ...
# DESIGN YOUR PIPELINE
my_pipe = Hyperpipe('basic_svm_pipe', # the name of your pipeline
# which optimizer PHOTON shall use
optimizer='sk_opt',
optimizer_params={'n_configurations': 10},
# the performance metrics of your interest
metrics=['accuracy', 'precision', 'recall', 'balanced_accuracy'],
# this metric declares the winner hyperparameter config
best_config_metric='accuracy',
# repeat hyperparameter optimization three times
outer_cv=KFold(n_splits=3),
# test each configuration five times respectively,
inner_cv=KFold(n_splits=5))
# all algorithms are accessed via keyword and imported from scikit-learn
# first normalize all features
my_pipe.add(PipelineElement('StandardScaler'))
# then do feature selection using a PCA
my_pipe += PipelineElement('PCA',
hyperparameters={'n_components': IntegerRange(5, 20)},
test_disabled=True)
# engage and optimize the good old SVM for Classification
my_pipe += PipelineElement('SVC',
hyperparameters={'kernel': Categorical(['rbf', 'linear']),
'C': FloatRange(0.5, 2)}, gamma='scale')
# train pipeline
X, y = load_breast_cancer(True)
my_pipe.fit(X, y)
Featured Examples
How To Handle Imbalanced Classes
Based on the popular imbalanced-learn library, learn how to handle class imbalance in your custom pipeline.
Show exampleCompeting Learning Algorithms
In case you are wondering which learning algorithm fits your data well, try the PHOTONAI OR-Element to compare two or more learning algorithms.
Show exampleSave and Share the optimized Model
Learn how to share your PHOTONAI model with the world, receive external validation and make other people use it in two lines of code.
Show exampleEASY ACCESS TO ESTABLISHED ML IMPLEMENTATIONS
Model Repository
Science needs exchange and cooperation.
We re-invented the wheel so many times in our work, that we feel the need to create a place where we can all share our best ideas - not as a paper, but in a form usable right away for everyone. Let others benefit from your work - benefit from others' work.
