This project is mirrored from Pull mirroring failed .
Last successful update .
  1. 18 5月, 2021 3 次提交
  2. 17 5月, 2021 1 次提交
  3. 15 5月, 2021 3 次提交
  4. 14 5月, 2021 3 次提交
  5. 13 5月, 2021 1 次提交
  6. 11 5月, 2021 2 次提交
    • Taylor Cramer's avatar
      Statically intern TFF types for the life of the Python runtime · 83e23100
      Taylor Cramer 创作于
      This change removes the WeakValueDictionary which was causing weird behavior
      where lookups would occur after parts of the Python standard library had
      already destroyed themselves. `atexit` hooks to clear the dictionary appeared
      to fix this at first, but it regressed again (for unknown reasons, this time
      on a smaller subset of programs).
      TFF types themselves are not particularly large in memory, and are frequently
      long-lived anyways (many are in global-scope objects), so the longer lifetimes
      are likely not a significant downside.
      PiperOrigin-RevId: 373052202
    • Shanshan Wu's avatar
      Removes the `._asdict` in the p13n example as it is not needed. · d78c1745
      Shanshan Wu 创作于
      PiperOrigin-RevId: 372951650
  7. 07 5月, 2021 2 次提交
  8. 06 5月, 2021 1 次提交
  9. 05 5月, 2021 3 次提交
  10. 04 5月, 2021 3 次提交
    • Zachary Charles's avatar
      Remove `tff.simulation.build_uniform_client_sampling_fn`. · bba5f42f
      Zachary Charles 创作于
      This function is only a small wrapper around the more flexible `tff.simulation.build_uniform_sampling_fn`, which should be used instead.
      PiperOrigin-RevId: 371815302
    • Zachary Garrett's avatar
      Change the result of `tff.simulation.build_uniform_sampling_fn` to... · d5244c75
      Zachary Garrett 创作于
      Change the result of `tff.simulation.build_uniform_sampling_fn` to additionally be paratermized by `size`.
      This allows `functools.partial(fn, size=X)` to replicate previous behavior,
      while also allowing new usages with variable sizes.
      PiperOrigin-RevId: 371797271
    • Galen Andrew's avatar
      TransformingClientData now works only with transformation functions that can... · efc0d351
      Galen Andrew 创作于
      TransformingClientData now works only with transformation functions that can be serialized as tf.computation.
      Also more general and less cumbersome method to split clients into pseudoclients: the user can specify a function to generate a list of pseudoclient ids for each client id. This allows, for example, differing numbers of pseudoclients for different clients.
      Also updates infinite_emnist to work with the new transforming_client_data.
      PiperOrigin-RevId: 371748160
  11. 01 5月, 2021 1 次提交
  12. 30 4月, 2021 3 次提交
  13. 29 4月, 2021 2 次提交
  14. 28 4月, 2021 5 次提交
  15. 27 4月, 2021 7 次提交