This project is mirrored from https://github.com/tensorflow/federated.
Pull mirroring failed .
Last successful update .
Last successful update .
- 20 11月, 2021 1 次提交
-
-
由 Zheng Xu 创作于
PiperOrigin-RevId: 411090036
-
- 15 9月, 2021 1 次提交
-
-
由 Zachary Garrett 创作于
PiperOrigin-RevId: 396615371
-
- 11 9月, 2021 1 次提交
-
-
由 Zachary Garrett 创作于
The `.weights` attribute and `TrainableModel` no longer exist; handling both model types now makes the example much more complex and prohibits usefulness. Additionally: - Fix backticks in pydoc. - Rename "fields" to "attributes" in pydoc. - Replace federated_value with more preferred federated_eval. - Replace collections.namedtuple usage with attr.s classes. - Improve determinism for unittests. PiperOrigin-RevId: 395949705
-
- 21 7月, 2021 1 次提交
-
-
由 Zachary Garrett 创作于
The predominant usage was to obtain the `tff.learning.ModelWeights` from a model. Going forward code should use the `tff.learning.ModelWeights.from_model()` to do this. PiperOrigin-RevId: 385882563
-
- 08 4月, 2021 1 次提交
-
-
由 Zachary Garrett 创作于
Remove the `tff.utils` package now that it contains nothing. PiperOrigin-RevId: 367275852
-
- 05 4月, 2021 1 次提交
-
-
由 Zachary Garrett 创作于
Remove `tff.utils.assign`. This is trivially replaced with `tf.nest.map_structure` calls. PiperOrigin-RevId: 366789208
-
- 02 3月, 2021 1 次提交
-
-
由 Zheng Xu 创作于
PiperOrigin-RevId: 360341050
-
- 05 11月, 2020 1 次提交
-
-
由 Michael Reneer 创作于
Do not use or import the ABCs from `collections`, this is deprecated since Python 3.3, and in 3.9 it will stop working. These ABCs have moved to 'collections.abc' and should be used from this package. PiperOrigin-RevId: 340678321
-
- 02 10月, 2020 1 次提交
-
-
由 Zheng Xu 创作于
(1) use skip_multi_gpu decorator or 'nokokoro' flag until we fix `dataset.reduce` issue in general (2) update simple_fedavg client training to be compatible with our multi-gpu simulation recommendation. PiperOrigin-RevId: 334881654
-
- 16 7月, 2020 1 次提交
-
-
由 Michael Reneer 创作于
PiperOrigin-RevId: 321431928
-
- 08 7月, 2020 1 次提交
-
-
由 Zheng Xu 创作于
PiperOrigin-RevId: 320026302
-
- 30 6月, 2020 1 次提交
-
-
由 Zheng Xu 创作于
Refactor `forward_pass` in `simple_fedavg` to be compatible with tff.learning. Clean `simple_fedavg` test to clearly distinguish `tff.learning.Model` and the standalone wrapper. PiperOrigin-RevId: 318841174
-
- 05 6月, 2020 1 次提交
-
-
由 Zheng Xu 创作于
PiperOrigin-RevId: 314810821
-
- 28 5月, 2020 1 次提交
-
-
由 Michael Reneer 创作于
PiperOrigin-RevId: 313418919
-
- 26 5月, 2020 1 次提交
-
-
由 Zheng Xu 创作于
PiperOrigin-RevId: 313189545
-
- 03 4月, 2020 1 次提交
-
-
由 Zheng Xu 创作于
Remove the dependency on sample_batch in simple_fedavg. The standalone model wrapper is simplified and compatible with the recent change in keras and `tff.learning.Model`. PiperOrigin-RevId: 304422256
-
- 12 3月, 2020 1 次提交
-
-
由 Jakub Konecny 创作于
PiperOrigin-RevId: 300347745
-
- 20 2月, 2020 3 次提交
- 14 2月, 2020 1 次提交
-
-
由 Zachary Garrett 创作于
Migrates `cmp` arguments in attr.s() to `eq` and `order` arguments: 1. If `cmp=False`, replacing it with `eq=False`. 2. If `cmp=True`, replacing it with `eq=True, order=True`. PiperOrigin-RevId: 294921678
-
- 18 1月, 2020 2 次提交
- 31 12月, 2019 1 次提交
-
-
由 Michael Reneer 创作于
PiperOrigin-RevId: 287599572
-
- 21 11月, 2019 1 次提交
-
-
由 Zachary Charles 创作于
PiperOrigin-RevId: 281621220
-
- 16 11月, 2019 2 次提交
-
-
由 A. Unique TensorFlower 创作于
PiperOrigin-RevId: 280784907
-
由 Michael Reneer 创作于
PiperOrigin-RevId: 280767713
-
- 07 11月, 2019 1 次提交
-
-
由 Zachary Garrett 创作于
PiperOrigin-RevId: 278962151
-
- 04 10月, 2019 1 次提交
-
-
由 Zachary Garrett 创作于
PiperOrigin-RevId: 272689129
-
- 13 9月, 2019 1 次提交
-
-
由 Michael Reneer 创作于
PiperOrigin-RevId: 268790969
-
- 12 9月, 2019 1 次提交
-
-
由 Michael Reneer 创作于
* Updated the `research` package imports. * Increased the BUILD visibility strictness. PiperOrigin-RevId: 268692104
-
- 08 7月, 2019 1 次提交
-
-
由 Jakub Konecny 创作于
PiperOrigin-RevId: 256903837
-
- 28 6月, 2019 1 次提交
-
-
由 Zhouyuan Huo 创作于
To modify the code, we can modify how client computes in "client_update", and modify how server computes in "server_update". PiperOrigin-RevId: 255438558
-