Embeddings with Variational Inference Substantial API changes have been made to facilitate variational inference and group-specific embeddings.
0.3.0
Not secure
Update to TensorFlow 2.0 idioms The internal code has been updated to use TensorFlow 2.0 style. All old idioms, like explicit sessions, have been removed. As part of the transition, the core of the model is now implemented using a tf.keras model composed of layers.
0.2.2
Not secure
Patch for loading old models. Changed deprecated variable use in active selection generator. Reworked preprocess module. Reworked visualize module to more closely align with Matplotlib conventions. Added timing information to verbose output for psiz.models.fit, psiz.models.posterior_samples and psiz.generators.generate methods.
0.2.1
Not secure
Patch for evaluate method.
0.2.0
Not secure
Updated getter/setter API. Setting variables now uses a flatter dot-notation syntax. The trainability of parameters is modified using the `trainable` method which replaces `freeze` and `thaw`.
0.1.0
Not secure
Initial alpha release This alpha release contains all of the core functionality necessary for inferring psychological embeddings and reproducing the results of the paper "Obtaining psychological embeddings through joint kernel and metric learning" (Roads & Mozer, in press).
This release relies on TensorFlow 1.14. The code base is gradually being moved towards Tensorflow 2.0. Once Tensorflow 2.0 is officially released, a new release of this package will be made.