experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Document explicitly iterations_per_loop is not respected in TPUEstimator evaluation.

Browse Source 
PiperOrigin-RevId: 239215958
This commit is contained in:
Ruoxin Sang 2019-03-19 10:14:57 -07:00 committed by TensorFlower Gardener
parent a988f5e6fa
commit 5ae4902198
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tensorflow/python/tpu/tpu_config.py
View File

@ -62,6 +62,8 @@ class TPUConfig(
system before returning to CPU host for each `Session.run`. This means system before returning to CPU host for each `Session.run`. This means
global step is increased `iterations_per_loop` times in one `Session.run`. global step is increased `iterations_per_loop` times in one `Session.run`.
It is recommended to be set as number of global steps for next checkpoint. It is recommended to be set as number of global steps for next checkpoint.
Note that in evaluation don't use this value, instead we run total eval
`steps` on TPU for a single `Session.run`.
num_shards: (Deprecated, ignored by TPUEstimator). num_shards: (Deprecated, ignored by TPUEstimator).
The number of model replicas in the system. For non-model-parallelism The number of model replicas in the system. For non-model-parallelism
case, this number equals the total number of TPU cores. For case, this number equals the total number of TPU cores. For