v1.0.0 Release

This is the stable v1.0.0 release. Below is a summary of changes and updates from the alpha pre-release of v1.0.0. For updates since v0.4, see the pre-release notes.

Thank you to the following users who contributed patches:

• Isabelle Garnreiter (@isagarnreiter)
• Michael Berger (@bergem1t)

Thank you also to Alex Williams (@ahwillia) for suggesting adding an example with fixed random seeds to the documentation.

New Features

• Add references to our published paper!
• Add logo to readthedocs
• Add setting seeds demonstration to Simple Example (#26)

Bug Fixes

• Make backwards compatible with previous weights formats (#24)
• Fix Simple Example so saving weights works in collab.
• Fix bug in get_effective_W_out when using output_connectivity.

v1.0.0-alpha Release

This is the alpha pre-release of the stable v1.0.0 release. Below is a summary of changes and updates from the previous release, v0.4.

New Features

• Exhaustive documentation now available!
• Now compatible with both TF1 and TF2 (still uses tensorflow graphs).
• Add fixed weights feature-- can choose not to train certain weights in W_rec, W_in, W_out.
• Revamps Simulation to make it relevant and functional.
• Adds new MatchToCategory task.
• Add support for custom loss and regularization functions.

Other New Features

• Added options for constant uniform, constant gaussian, glorot gaussian, glorot uniform initializations.
• Automatically build and initialize in get_weights, save, test, train, train_curric. No model.build() calls needed from now on.
• Added train_curric function.

Breaking Changes

• No long supports TF < 1.13.1
• Refactors RDM Task--> PerceptualDiscrimination & changes some functionality.
• Refactors Romo Task--> DelayedDiscrimination & changes some functionality.
• Refactors Curriculum's self.stopTraining --> self. stop_training. This change should not affect most users since it is a variable used internally only by curriculum.py and rnn.py
• Removes sussillo regularizer because it was specific to the transfer-function used. Can be defined and used via the custom regularizer.
• Remove get_initial_weights function. Use get_weights instead.

Other changes (may affect results)

• Default random initialization changed to glorot gaussian from constant uniform. This may affect results.
• Unexpected behavior in Dale's fixed and Dale's setup moved into WeightInitializer. This should not affect most users but may affect results.
• Default training iters changed from 10000 to 50000

Other Changes (should not affect users)

• BasicScan internal implementation slightly change. Should not affect users unless users inherited BasicScan to make a custom Model.
• Use get_effective to get the weight matrices to regularize. Should not affect users.

Bug Fixes

• Bug in binary cross entropy loss function fixed.

v0.4 Dev Release

Breaking Changes

• the batch_generator now has an additional params output
• Removed Basic Sigmoid (can be implemented with Basic + transfer_function = sigmoid

New Features

• Curriculum Learning
• get_weights function in RNN
• Add customizability to initializations

Small Bug Fixes / Quality Control

• Add tests
• Fix bugs in Romo
• Make Dale's Law train automatically
• Getting / Saving weights returns the effective weights
• Add better user warnings

Note

This version was never released on pip and is only available here.

Initial v0.3 Dev Release

Merge branch 'master' of https://github.com/dbehrlich/PsychRNN