update docs

README.md

@@ -37,7 +37,7 @@ It's Yet Another TF wrapper, but different in:
 	+	Speed comes for free with tensorpack -- it uses TensorFlow in the __correct way__ with no extra overhead.
 	  On various CNNs, it runs 1.5~1.7x faster than the equivalent Keras code.
 
-	+ Data-parallel multi-GPU/distributed training is off-the-shelf to use. It is as fast as Google's [official benchmark](https://www.tensorflow.org/performance/benchmarks).
+	+ Data-parallel multi-GPU training is off-the-shelf to use. It is as fast as Google's [official benchmark](https://www.tensorflow.org/performance/benchmarks).
 		You cannot beat its speed unless you're a TensorFlow expert.
 
 	+ See [tensorpack/benchmarks](https://github.com/tensorpack/benchmarks) for some benchmark scripts.

tensorpack/graph_builder/distributed.py

@@ -24,7 +24,7 @@ class DistributedReplicatedBuilder(DataParallelBuilder):
     Then each worker copy the latest variables from PS back to local.
 
     It is an equivalent of `--variable_update=distributed_replicated` in
-    `tensorflow/benchmarks <https://github.com/tensorflow/benchmarks>`.
+    `tensorflow/benchmarks <https://github.com/tensorflow/benchmarks>`_.
 
     Note:
         Gradients are not averaged across workers, but applied to PS variables

tensorpack/graph_builder/training.py

@@ -107,7 +107,7 @@ class SyncMultiGPUParameterServerBuilder(DataParallelBuilder):
     from each tower, averages them and applies to the shared variables.
 
     It is an equivalent of `--variable_update=parameter_server` in
-    `tensorflow/benchmarks <https://github.com/tensorflow/benchmarks>`.
+    `tensorflow/benchmarks <https://github.com/tensorflow/benchmarks>`_.
     """
     def __init__(self, towers, ps_device=None):
         """
@@ -166,7 +166,7 @@ class SyncMultiGPUReplicatedBuilder(DataParallelBuilder):
     Each gradient update is averaged across or GPUs through NCCL.
 
     It is an equivalent of `--variable_update=replicated` in
-    `tensorflow/benchmarks <https://github.com/tensorflow/benchmarks>`.
+    `tensorflow/benchmarks <https://github.com/tensorflow/benchmarks>`_.
     """
 
     def build(self, get_grad_fn, get_opt_fn):

tensorpack/graph_builder/utils.py

@@ -47,7 +47,7 @@ class OverrideToLocalVariable(object):
         return getter(name, *args, **kwargs)
 
 
-# Copied from https://github.com/tensorflow/benchmarks/blob/master/scripts/tf_cnn_benchmarks/variable_mgr.py
+# https://github.com/tensorflow/benchmarks/blob/48cbef14a592e02a14beee8e9aef3ad22cadaed1/scripts/tf_cnn_benchmarks/variable_mgr_util.py#L192-L218
 class LeastLoadedDeviceSetter(object):
     """ Helper class to assign variables on the least loaded ps-device."""
     def __init__(self, worker_device, ps_devices):