OpenGrok
Home
Sort by relevance
Sort by last modified time
Full Search
Definition
Symbol
File Path
History
|
|
Help
Searched
refs:TPUEstimator
(Results
1 - 4
of
4
) sorted by null
/external/tensorflow/tensorflow/contrib/gan/python/estimator/python/
tpu_gan_estimator_impl.py
38
class TPUGANEstimator(tpu_estimator.
TPUEstimator
):
96
#
TPUEstimator
options.
151
model_dir: Same as `
TPUEstimator
`: Directory to save model parameters,
157
config: Same as `
TPUEstimator
`: An `tpu_config.RunConfig` configuration
159
params: Same as `
TPUEstimator
`: An optional `dict` of hyper parameters
162
`
TPUEstimator
`, including 'batch_size'.
163
use_tpu: Same as `
TPUEstimator
`: A bool indicating whether TPU support is
167
train_batch_size: Same as `
TPUEstimator
`: An int representing the global
168
training batch size.
TPUEstimator
transforms this global batch size to a
172
eval_batch_size: Same as `
TPUEstimator
`: An int representing evaluatio
[
all
...]
/external/tensorflow/tensorflow/python/tpu/
tpu_estimator.py
15
"""
TPUEstimator
class."""
161
In
TPUEstimator
, the user provided computation, the model_fn, is wrapped
167
TPUEstimator
adapt the TPU training iterations according to the final steps
169
in TPUConfig and steps as 10 in
TPUEstimator
.train(), the iterations_per_loop
267
"""Ops and objects returned from a `model_fn` and passed to `
TPUEstimator
`.
286
update_op)` tuple. See `
TPUEstimator
` for MNIST example how to specify the
720
# batch 100: images, labels, stop = 1 (
TPUEstimator
appended)
722
# where the final batch (id = 100) is appended by
TPUEstimator
, so we
[
all
...]
/external/tensorflow/tensorflow/contrib/eager/python/examples/revnet/
main_estimator_tpu.py
102
"""Model function required by the `tf.contrib.tpu.
TPUEstimator
` API.
219
revnet_classifier = tf.contrib.tpu.
TPUEstimator
(
/external/tensorflow/tensorflow/contrib/learn/python/learn/
experiment.py
248
if isinstance(estimator, tpu_estimator.
TPUEstimator
):
250
"`Experiment` class cannot work with `tf.contrib.tpu.
TPUEstimator
`. "
251
"Please call `
TPUEstimator
` train/evaluate directly. \n"
253
"distributed training, while `
TPUEstimator
` is working in in-graph "
[
all
...]
Completed in 546 milliseconds