Skip to content

Releases: mvinyard/torch-nets

v0.2.0

28 Feb 19:52
744b84b

Choose a tag to compare

What's Changed

Full Changelog: v0.1.0...v0.2.0

v0.2.0rc0

28 Feb 19:39

Choose a tag to compare

v0.2.0rc0 Pre-release
Pre-release

Full Changelog: v0.1.0...v0.2.0rc0

v0.1.0

06 Nov 17:53
1b03b81

Choose a tag to compare

What's Changed

Full Changelog: v0.0.5...v0.1.0

v0.1.0rc1

06 Nov 17:49
a106168

Choose a tag to compare

v0.1.0rc1 Pre-release
Pre-release

Further testing from v0.1.0rc0

Full Changelog: v0.1.0rc0...v0.1.0rc1

v0.1.0rc0

06 Nov 17:45
bb65772

Choose a tag to compare

v0.1.0rc0 Pre-release
Pre-release

What's Changed

Full Changelog: v0.0.5...v0.1.0rc0

v0.0.5

03 Aug 03:10

Choose a tag to compare

What's Changed

  • updated compatibility with ABCParse base class by @mvinyard in #6

Full Changelog: v0.0.4...v0.0.5

pre-release: v0.0.5rc0

03 Aug 02:57

Choose a tag to compare

Pre-release

Full Changelog: v0.0.4...v0.0.5rc0

v0.0.4

13 Apr 19:18
d7caaa9

Choose a tag to compare

What's Changed

Full Changelog: v0.0.3...v0.0.4

v0.0.4rc0

13 Apr 19:08

Choose a tag to compare

v0.0.4rc0 Pre-release
Pre-release

Full Changelog: v0.0.3...v0.0.4rc0

v0.0.3

12 Apr 17:41

Choose a tag to compare

Changes to core

  • Revised the TorchNet such that it inherits torch.nn.Sequential, which is more intuitive in creating network objects through this interface. (commit: 6fe33cb3)
  • Added the potential property to the TorchNet which serves as a boolean indicator flag (commit: 42d0940)
  • Encoder and Decoder were updated accordingly as they are derived from the TorchNet (commit: 1f45826)
  • Updated AugmentedTorchNet class, which is also derived from the TorchNet (commit: 2de1bde)

Added the plotting (.pl) module

Added the tools (.tl) module

  • torch_nets.tl.infer_network_architecture_from_state (commit: 357fc37)
    • This enables one to pass a state_dict - often saved as a ckpt_model.pt and return a network composed with these weights.
    • Along with the state_dict, dropout and activation are required to be passed with each layer, since these are not saved as part of the state_dict.

Fixes made during testing in pre-release:

v0.0.3rc0

  • Updated setup.py with additional imports (vinplots, which also brings in matplotlib)

v0.0.3rc1

  • Reverted names as an attribute of TorchNet (commit: c2abc11)

v0.0.3rc2

  • Add example notebook of classes available in the API (commit: 841bb58)

Still to do

  • Documentation! Quarto is the likely path forward.

Full Changelog: v0.0.2...v0.0.3