Learners¶
Classes for machine learning
SceneClassifier¶
SceneClassifierGMM¶
Scene classifier with GMM. This learner is using sklearn.mixture.GaussianMixture implementation. See
documentation.
SceneClassifierGMM(\*args, \*\*kwargs) |
Scene classifier with GMM |
SceneClassifierGMM.learn(data, annotations) |
Learn based on data and annotations |
SceneClassifierGMM.predict(feature_data) |
Predict frame probabilities for given feature matrix |
SceneClassifierMLP¶
Scene classifier with MLP. This learner is a simple MLP based learner using Keras neural network implementation and sequential API. See documentation.
SceneClassifierMLP(\*args, \*\*kwargs) |
Scene classifier with MLP |
SceneClassifierMLP.learn(data, annotations) |
Learn based on data and annotations |
SceneClassifierMLP.predict(feature_data) |
Predict frame probabilities for given feature matrix |
SceneClassifierKerasSequential¶
Scene classifier with Keras sequential API (see documentation). This learner can be used for more advanced network structures than SceneClassifierMLP.
SceneClassifierKerasSequential(\*args, \*\*kwargs) |
Sequential Keras model for Acoustic scene classification |
SceneClassifierKerasSequential.learn(data, ...) |
Learn based on data and annotations |
SceneClassifierKerasSequential.predict(...) |
Predict frame probabilities for given feature matrix |
EventDetector¶
EventDetector(\*args, \*\*kwargs) |
Event detector (Frame classifier / Multi-class - Multi-label) |
EventDetectorGMM¶
EventDetectorGMM(\*args, \*\*kwargs) |
|
EventDetectorGMM.learn(data, annotations[, ...]) |
Learn based on data and annotations |
EventDetectorGMM.predict(feature_data) |
EventDetectorMLP¶
EventDetectorMLP(\*args, \*\*kwargs) |
Simple MLP based sequential Keras model for Sound Event Detection |
EventDetectorMLP.learn(data, annotations[, ...]) |
Learn based on data and annotations |
EventDetectorMLP.predict(feature_data) |
EventDetectorKerasSequential¶
EventDetectorKerasSequential(\*args, \*\*kwargs) |
Sequential Keras model for Sound Event Detection |
EventDetectorKerasSequential.learn(annotations) |
Learn based on data and annotations |
EventDetectorKerasSequential.predict(...) |
LearnerContainer - Base class¶
LearnerContainer(\*args, \*\*kwargs) |
Constructor |
LearnerContainer.class_labels |
Class labels |
LearnerContainer.method |
Learner method label |
LearnerContainer.params |
Parameters |
LearnerContainer.feature_masker |
Feature masker instance |
LearnerContainer.feature_normalizer |
Feature normalizer instance |
LearnerContainer.feature_stacker |
Feature stacker instance |
LearnerContainer.feature_aggregator |
Feature aggregator instance |
LearnerContainer.model |
Acoustic model |
LearnerContainer.set_seed([seed]) |
Set randomization seeds |
LearnerContainer.learner_params |
Get learner parameters from parameter container |