Tensorflow-estimator

Breaking changes

Both for Estimator and for main TensorFlow, `tf.contrib` has been deprecated, and functionality has been either migrated to the core TensorFlow API, to an ecosystem project such as https://www.github.com/tensorflow/addons or https://www.github.com/tensorflow/io, or removed entirely.

Thanks to our Contributors

This release contains contributions from many people at Google.

Changes

* Use tf.compat.v1.estimator.inputs instead of tf.estimator.inputs
* Replace contrib references with tf.estimator.experimental.* for apis in early_stopping.py

Thanks to our Contributors

This release contains contributions from many people at Google

This release is similar to 1.15.1 with the only addition of making `early_stopping` hook work again after TensorFlow released security patch 1.15.2

Thanks to our Contributors

This release contains contributions from many people at Google.

This release is the same as 1.15.0 but we needed a new release to update the version number inside `setup.py`

Changes

* `tf.keras.estimator.model_to_estimator` now supports exporting to tf.train.Checkpoint format, which allows the saved checkpoints to be compatible with `model.load_weights`.
* "Fix tests in canned estimators."
* Expose Head as public API.
* Fixes critical bugs that help with DenseFeatures usability in TF2

Thanks to our Contributors

This release contains contributions from many people at Google.

Changes

* `tf.keras.estimator.model_to_estimator` now supports exporting to tf.train.Checkpoint format, which allows the saved checkpoints to be compatible with `model.load_weights`.
* "Fix tests in canned estimators."
* Expose Head as public API.
* Fixes critical bugs that help with DenseFeatures usability in TF2

Thanks to our Contributors

This release contains contributions from many people at Google.

Changes

* Use `tf.compat.v1.estimator.inputs` instead of `tf.estimator.inputs`
* Replace `contrib` references with `tf.estimator.experimental.*` for APIs in `early_stopping.py`
* Determining the “correct” value of the `--iterations_per_loop` for TPUEstimator or DistributionStrategy continues to be a challenge for our users. We propose dynamically tuning the `--iterations_per_loop` variable, specifically for using TPUEstimator in training mode, based on a user target TPU execution time. Users might specify a value such as: `--iterations_per_loop=300s`, which will result in roughly 300 seconds being spent on the TPU between host side operations.

Thanks to our Contributors

This release contains contributions from many people at Google.