Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.
Added a MultiProblem class for Multitask Learning. Thanks @urvashik !
Added decoding option to pass through the features dictionary to predictions. Thanks @rsepassi !
Enabled MLEngine path to use Cloud TPUs. Thanks @rsepassi !
Added a simple One-Hot Symbol modality. Thanks @mbz !
Added Cleverhans integration. Thanks @aidangomez !
Problem definitions added for:
Model additions:
Notable bug fixes:
Many more fixes, tests and work on Model based RL, Transfomer, Video and other models and problems.
Great thanks to everyone!
registry.hparams
now returns an HParams
object instead of a function that returns an HParams
objectMultistepAdamOptimizer
thanks to @fstahlbergVideoProblem
pylintrc
and lint tests to Travis CI--random_seed
is unset by default now. Set it to an integer value to get reproducible results.WikisumWeb
data generation fixesRTransformer
model, a recurrent TransformerROC_AUC
metric thanks to @jjtan--problems
command-line flag renamed to --problem
hparams.problems
renamed to hparams.problem_hparams
and hparams.problem_instances
renamed to hparams.problem
(and neither are lists now)TransformerScorer
model which return log prob of targets on infer
bfloat16
weights and activations on TPUcommon_layers
--checkpoint_path
supported in interactive decodingVideoProblem
base classSquad
and SquadConcat
problem for question answering (and relevant base class)bfloat16
support for Transformer
on TPUsSigmoidClassLabelModality
for binary classificationImageTransformer
on TPU