GitLab

PiperOrigin-RevId: 186541910

PiperOrigin-RevId: 186265033

PiperOrigin-RevId: 185783053

Replace tf.contrib.framework.get_or_create_global_step() with tf.train.get_or_create_global_step() in third_party/tensorflow_models

See https://www.tensorflow.org/api_docs/python/tf/contrib/framework/get_or_create_global_step and #2795

PiperOrigin-RevId: 176534973

PiperOrigin-RevId: 173414999

Add option for when there is only one label for classification, for speed. (The case for the standard datasets).

PiperOrigin-RevId: 172774825

PiperOrigin-RevId: 172665358

memory and fixes the OOM errors.

PiperOrigin-RevId: 172374935

Change logit and activation shapes to follow standard convention with batch size in the first dimension.

PiperOrigin-RevId: 172172495

PiperOrigin-RevId: 171498799

To simplify syncing this c ode back into Google.

Clamp the requested frequency range to [0, Nyquist].

PiperOrigin-RevId: 185469605

185215255  by Zhichao Lu:

    Stop populating image/object/class/text field when generating COCO tf record.

--
185213306  by Zhichao Lu:

    Use the params batch size and not the one from train_config in input_fn

--
185209081  by Zhichao Lu:

    Handle the case when there are no ground-truth masks for an image.

--
185195531  by Zhichao Lu:

    Remove unstack and stack operations on features from third_party/object_detection/model.py.

--
185195017  by Zhichao Lu:

    Matrix multiplication based gather op implementation.

--
185187744  by Zhichao Lu:

    Fix eval_util minor issue.

--
185098733  by Zhichao Lu:

    Internal change

185076656  by Zhichao Lu:

    Increment the amount of boxes for coco17.

--
185074199  by Zhichao Lu:

    Add config for SSD Resnet50 v1 with FPN.

--
185060199  by Zhichao Lu:

    Fix a bug in clear_detections.
    This method set detection_keys to an empty dictionary instead of an empty set. I've refactored so that this ...

TF Saver now requires specifying the checkpoint version even when restoring to avoid errors when we specify
a path to a checkpoint file instead of a directory.

For stitched models, the readin matrices and bias vectors are initialized to the
"alignment" matrix and bias specified in each dataset's .h5 file. If do_train_readin is
True, these will be trainable, and if not, they will be fixed.

`max_number_of_evaluation` -> `max_number_of_evaluations`.

Add python-tk in dependency list

184048729  by Zhichao Lu:

    Modify target_assigner so that it creates regression targets taking keypoints into account.

--
184027183  by Zhichao Lu:

    Resnet V1 FPN based feature extractors for SSD meta architecture in Object Detection V2 API.

--
184004730  by Zhichao Lu:

    Expose a lever to override the configured mask_type.

--
183933113  by Zhichao Lu:

    Weight shared convolutional box predictor as described in https://arxiv.org/abs/1708.02002

--
183929669  by Zhichao Lu:

    Expanding box list operations for future data augmentations.

--
183916792  by Zhichao Lu:

    Fix unrecognized assertion function in tests.

--
183906851  by Zhichao Lu:

    - Change ssd meta architecture to use regression weights to compute loss normalizer.

--
183871003  by Zhichao Lu:

    Fix config_util_test wrong dependency.

--
183782120  by Zhichao Lu:

    Add __init__ file to third_party directories.

--
183779109  by Zhichao Lu:

    Setup regular version s...

__six__ is used on lines 174 and 175 but is never imported or defined.

In Python 2 the string and the r-string behave the same:
$ __python2__
```
>>> '$\Uparrow$ '
'$\\Uparrow$ '
>>> r'$\Uparrow$ '
'$\\Uparrow$ '
>>> '$\Uparrow$ ' == r'$\Uparrow$ '
True
```

In Python 3 the string raises a Syntax Error while the r-string works just like Python 2.
```
>>> '$\Uparrow$ '
  File "<stdin>", line 1
SyntaxError: (unicode error) 'unicodeescape' codec can't decode bytes in position 1-2: truncated \UXXXXXXXX escape
>>> r'$\Uparrow$ '
'$\\Uparrow$ '
```