Skip to content

T
tensorflow

xxadev / tensorflow
与 Fork 源项目一致

从无法访问的项目Fork

通知 3
Star 1
Fork 0
T
tensorflow

前往新版Gitcode，体验更适合开发者的 AI 搜索 >>

切换分支/标签
  1. 26 3月, 2020 2 次提交
  3. 13 3月, 2020 1 次提交
  5. 06 3月, 2020 1 次提交
  7. 28 2月, 2020 1 次提交
    • R
      Remove passing experimental_run_tf_function in most tests. · b7014702
      由 Reed Wanderman-Milne 提交于 
      The experimental_run_tf_function parameter no longer has any effect.

I didn't remove the functionality from testing_util.py and keras_parameterized.py to run with experimental_run_tf_function being True and False. I will remove that functionality in a future change.

PiperOrigin-RevId: 297674422
Change-Id: I5b1e67f78b4c3b60242241fb4dc2018f0ace6013
      b7014702
  9. 25 2月, 2020 1 次提交
  11. 22 2月, 2020 2 次提交
    • A
      Makes Model compile run inside the captured distribution strategy, so that any... · ef5d540a
      由 A. Unique TensorFlower 提交于 
      Makes Model compile run inside the captured distribution strategy, so that any metrics/optimizers created by compile are created in the distribution strategy scope (e.g. when deserializing strings that are the metric names).

Also adds a correctness test that verifies the model successfully captures the distribution strategy.

It also raises an error if there are metrics that are created outside the scope and not in compile.

So e.g. this will help the following case because the optimizer/metrics get created by compile:
with strategy.scope():
  model = ...
model.compile(optimizer='sgd', metrics=['binary_accuracy'])

And, it will raise an error in the following case because the metrics are created in a different distribution strategy scope than the model:
with strategy.scope():
  model = ...
model.compile(optimizer=tf.keras.optimizers.Blah() metrics=[tf.keras.metrics.BinaryAccuracy()])

PiperOrigin-RevId: 296553610
Change-Id: I988a80c1863de732da45555c444fc5a237ecd425
      ef5d540a
    • H
      Limit strategy combinations in test · 23a0ca34
      由 Håkon Sandsmark 提交于 
      Some of the GPU strategies failed earlier with another error message.
      23a0ca34
  13. 21 2月, 2020 2 次提交
    • H
      optimizer_v2: Improve error when called in cross-replica context · 5e15d37d
      由 Håkon Sandsmark 提交于 
      When calling `Optimizer.apply_gradients()` in a cross-replica distribution
context (with a non-default distribution strategy),
`distribute_ctx.get_replica_context()` returns None, so it would fail with
the error

    [...]/optimizer_v2.py", line 448, in apply_gradients
        return distribute_ctx.get_replica_context().merge_call(
    AttributeError: 'NoneType' object has no attribute 'merge_call'

This commit changes the error to a `RuntimeError` with a more descriptive
error message (inspired by the error message in the v1 optimizer) guiding
the user how to fix the issue, by either calling the `_distributed_apply()`
function instead or by using `tf.distribute.Strategy.experimental_run_v2`.
      5e15d37d
    • R
      Supplement multi_worker_callback_tf2_test.py with tests that have the same... · 32ec0712
      由 Rick Chao 提交于 
      Supplement multi_worker_callback_tf2_test.py with tests that have the same file path for ModelCheckpoint and TensorBoard callbacks.

PiperOrigin-RevId: 296328002
Change-Id: I28bb1d4b60e1fb47c1570852fe82d71adc6ebffe
      32ec0712
  15. 20 2月, 2020 1 次提交
    • T
      Keras ideal fit and compile. · 10666c59
      由 Thomas O'Malley 提交于 
      Kept all new abstractions private for now. In a few weeks, if we're
comfortable that these abstractions are working and stable, we should expose
many of them publicly.

Capabilites added by this CL:

(1) Easy to create a custom training step via overriding Model._train_step
(2) Easy to create custom tf.function / DistStrat logic via overriding
Model._make_train_function
(3) Advanced users can override Model.compile and Model.fit
(4) Full support for dicts, nested structures, etc with Subclassed Models.
(5) "Power user" path (tf.data inputs) only modifies data in Model._train_step,
where this behavior is easy to override and disable. This applies even to
Keras's assumption that data is passed in (x, y, sample_weight) format.

Behavior changes:

(1) "loss" passed to Callbacks is now stateful (like all other metrics in
Callbacks). This greatly simplifies the training step logic and callback logic.
(2) ProgbarLogger always uses steps. If steps is not available, the
ProgbarLogger handles inferring the steps after the first epoch.
(3) validation_batch_size added in `fit`, rather than inferring from generator.
(4) Model.inputs, Model.outputs, Model.input_names, and Model.output_names are
no longer populated for subclassed Models. Instead, "pseudo" output names are
created for subclassed Models, which are only used for metrics names and
SavedModel's signature.
(5) Cast NumPy floats to backend.floatx(), otherwise leave
unchanged (this is likely not a change, we did something like this in our old
version but the logic was scattered in many places)

PiperOrigin-RevId: 296090972
Change-Id: Ia5ac833fd39085bddb016833bd338083d0dc5fc2
      10666c59
  17. 15 2月, 2020 1 次提交
  19. 14 2月, 2020 1 次提交
  21. 13 2月, 2020 1 次提交
  23. 12 2月, 2020 2 次提交
  25. 11 2月, 2020 1 次提交
  27. 08 2月, 2020 1 次提交
  29. 04 2月, 2020 2 次提交
  31. 02 2月, 2020 1 次提交
  33. 01 2月, 2020 1 次提交
  35. 25 1月, 2020 1 次提交
  37. 18 1月, 2020 1 次提交
  39. 16 1月, 2020 1 次提交
  41. 10 1月, 2020 2 次提交
  43. 21 12月, 2019 1 次提交
  45. 20 12月, 2019 1 次提交
  47. 19 12月, 2019 1 次提交
    • B
      Small adjustments on import spacing. · 66c48046
      由 Brian Atkinson 提交于 
      This is mostly the result of an internal cleanup and formatting pass.

PiperOrigin-RevId: 286318018
Change-Id: I8f9e2f7519070035da73f9f24d2fc90864abc51b
      66c48046
  49. 17 12月, 2019 1 次提交
  51. 13 12月, 2019 2 次提交
  53. 10 12月, 2019 1 次提交
  55. 06 12月, 2019 1 次提交
  57. 02 12月, 2019 2 次提交
  59. 23 11月, 2019 1 次提交
  61. 19 11月, 2019 1 次提交
    • R
      Remove usage of... · 2adcf837
      由 Rick Chao 提交于 
      Remove usage of multi_process_runner_util.try_run_and_except_connection_error() as setting FAIL_FAST=false fixes the connection error issue. Make grpc_fail_fast an arg for MultiProcessRunner's __init__().

PiperOrigin-RevId: 281193957
Change-Id: I1c99cb90a15fdb26892d0ad37c533c186297b09d
      2adcf837
  63. 12 11月, 2019 1 次提交