GitLab

[mnist]: Updates

Update blog_custom_estimators.py

Added correct link to blogpost
Changed Sentosa to Setosa in a couple of places

https://github.com/tensorflow/models/pull/3024

Update README.md in adv_imagenet_models.

Add newly released adversarially trained (no ensemble) checkpoint for
Inception Resnet v2.

Add code for Custom Estimators blog post.

- Use the object-oriented tf.layers API instead of the functional one.
  The object-oriented API is particularly useful when using the model
  with eager execution.
- Update unittest to train, evaluate, and predict using the model.
- Add a micro-benchmark for measuring step-time.
  The parameters (batch_size, num_steps etc.) have NOT been tuned,
  the purpose with this code is mostly to illustrate how model
  benchmarks may be written.

These changes are made as a step towards consolidating model definitions
for different TensorFlow features (like eager execution and support for
TPUs in
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/eager/python/examples/mnist
and
https://github.com/tensorflow/tpu-demos/tree/master/cloud_tpu/models/mnist

With examples, and updates to the README

Fix the CODEOWNERS file syntax

Update data URL for CIFAR dataset (http->https).

Change docstring to refer to tf.estimator instead of tf.learn in official/wide_deep

Fix a bug in create_pet_tf_record.py related to mask creation.

Add a pre-trained ResNet-50 checkpoint

fix eval_training_data bug

Update tutorials/rnn/README.md with the correct Quick Draw link

[mnist]: Tweaks

Update exporting_models.md

Update path to tf_record script

* Split iris data

* move shared code to iris_data.py

* add minimal csv example

* remove unused pandas import

* Use sparse softmax loss to avoid warning

- Remove `convert_to_records.py` and instead create `tf.data.Dataset`
  objects directly from the numpy arrays.
- Format the Google Python Style (https://github.com/google/yapf/)

Adding TCN.

Add Quick, Draw! recurrent neural network-based classification model.