Skip to content
Snippets Groups Projects
Select Git revision
  • cl/332000035
  • cl/343353741
  • cl/346635895
  • cl/347081719
  • cl/347401844
  • cl/347907220
  • cl/348027488
  • cl/348050964
  • cl/348070543
  • cl/348098438
  • cl/348138051
  • cl/348556585
  • cl/348851214
  • cl/348866975
  • cl/349624352
  • cl/350701164
  • cl/352033008
  • cl/353341137
  • cl/354238630
  • cl/354591074
  • v0.68.0
  • v0.67.0
  • v0.66.0
  • v0.65.0
  • v0.64.0
  • v0.63.0
  • v0.62.0
  • v0.61.0
  • v0.60.0
  • v0.59.0
  • v0.58.0
  • v0.57.0
  • v0.56.0
  • v0.55.0
  • v0.54.0
  • v0.53.0
  • v0.52.0
  • v0.51.0
  • v0.50.0
  • v0.49.0
40 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.030Jan29282724232221201817161514131098763230Dec27262320191817161312111096543227Nov2625242221201918161514131287654131Oct302928252423221716151411109874325Sep24201917161312111096543230Aug292827262322212019151413121312987652129Jul262524232220191817Update type checking in `canonical_form_utils` to handle the all the `CanonicalForm` attributes consistently.Reduce strictness of comparing execution time metric to 1 millisecond instead of 0.01 milliseconds.Convert record to list if it cannot be converted to dict in default type_fns for build_dp_aggregate.Extract the logic to insert a fake client output into a helper function.Remove unnecessary type check on the return value of `consolidate_and_extract_local_processing`.Update `check_tensorflow_compatible_type` to have a more generic error message.Add `contains_called_intrinsic` helper function.Be compatible with cutomized keras layer in TFF.Update benchmarks to run.Update transformations to support the `secure_sum` intrinsic.Remove unused test utility.Implements `deduplicate_called_graphs`.Internal.Close ConcurrentExecutor thread at end of program rather than in close()Implements `generate_tensorflow_representing_block`.Delete cache across executor closeRemoved the special casing of extracting a reference to the top level lambda.Removes ReferenceExecutor-only benchmark, as this is no longer the default.Centralized training for stackoverflow model.Remove string comparison from unit test.Remove extra indent in the doc.This CL creates a `build_personalization_eval` API which builds the TFF computation for evaluating personalization strategies.Adds experimental capability to pin more than one client on a TFF bottom stack.Integrate sizing_executor to be available when running trainingUpdated process for creating Docker images of the Remote Executor ServiceAdd ClientData convenience helpersUpdated the ConcurrentExecutor to re-initialize the event loop after close() in case it is used in multiple computations.Adjusts semantics of `group_block_locals_by_namespace` to return parallel structure of variable names bound to these comps.Add `secure_sum` intrinsic to TFF.Fix order-dependent test.Update references to simple_fedavg in `Using TFF for Federated Learning Research`.Implements helper to generate simple TensorFlow representing calling a function on a given argument.Implements _construct_unbound_reference_classes function.Adds "Model and update compression" to tff_for_research.md.Added executor to calculate size of values passed throughUpdate tutorials to save the outputs for the tutorials that run.Use uncompiled Keras model in simple_fedavg.Use inline tff decorator in simple_fedavg.Update the docstring of TupleCalledGraphs to reflect the performance guaranteesExplicitly export files needed by other packages