Skip to content

Releases: kundajelab/deeplift

Added support for tanh activations and intermediate-layer sigmoid activations

11 Nov 09:28
fa4957b
Compare
Choose a tag to compare

Corresponds to PR #117 (sigmoid as the output layer was already supported)

Faster dinucleotide shuffling

18 Aug 04:50
Compare
Choose a tag to compare
Pre-release

Thanks to @atseng95 for PR #109, which greatly speeds up dinucleotide shuffling. This tag also incorporates the change from PR #105 (which has support for supplying pre-generated shuffled references, and was tagged as v0.6.11.0), as well as the small fix in PR #101 (which allows the user to recover if they accidentally set an invalid task index, and was tagged as v0.6.10.1)

Fix to avoid redundant resetting of mxts

26 Jan 19:54
d0d5f36
Compare
Choose a tag to compare

Corresponds to the fix in #96 by @berleon, who found that it greatly improves the compilation time for models that have layers with multiple inputs.

Fix for models that don't have biases, target layer index error now a warning

14 Jan 20:50
667f00b
Compare
Choose a tag to compare

Corresponds to features implemented in PR #93. Two features: (1) has a fix for loading models that don't have biases, and (2) a message about the target layer that would previously be thrown as a runtime error now just results in a warning message being printed, as there are legitimate situations where that edge case can occur. See #92 for the issue that prompted the changes.

Ability to reuse same shuffled references for multiple tasks

07 Jan 23:30
0888551
Compare
Choose a tag to compare

This feature was requested in #83 and was implemented in PR https://github.com/kundajelab/deeplift/pull/84/files. I forgot to merge it into the master branch at the time and am doing so now. The genomics notebook was updated to use this feature (and also updated to python 3) in #94

Added ability to provide a random state for dinucleotide shuffling

18 Oct 15:22
Compare
Choose a tag to compare

(so that the random state doesn't have to be controlled by setting the numpy random seed externally, which doesn't always play will with jupyter notebooks)

Support for dinucleotide shuffling directly on one-hot encoded sequences

08 Mar 04:13
f143790
Compare
Choose a tag to compare

Corresponding to feature added by @annashcherbina in #78

Coping with Keras 2.2.3 breaking change + GlobalAveragePooling layer

12 Dec 20:03
9cbe05b
Compare
Choose a tag to compare
  • Coping with Keras 2.2.3 breaking change as described in this pull request #69
  • Release also includes GlobalAveragePooling layer added by @AnjaSei in #68

Python 3 fix in dinuc shuffle

14 Sep 08:00
beb8923
Compare
Choose a tag to compare
Pre-release

This pull request: #62

Upgraded to work with latest tensorflow

27 Aug 22:53
9fdb6b3
Compare
Choose a tag to compare
Pre-release

(Tensorflow 1.10.1). Also updated some of the tests to work with Keras 2.2.