eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r370812398
##########
File path: python/mxnet/kvstore/kvstore.py
##########
@@ -104,8 +104,22 @@ def broadcast(self, key, value, out, priority=0):
[ 2. 2. 2.]]
"""
- self.init(key, value)
- self.pull(key, out=out, priority=priority)
+ cvkeys, cvals, use_str_keys = _ctype_key_value(key, value)
+ if out is not None:
+ cokeys, couts, _ = _ctype_key_value(key, out)
+ else:
+ cokeys = cvkeys
+ couts = cvals
+
+ if use_str_keys:
+ check_call(_LIB.MXKVStoreBroadcastEx(
+ self.handle, mx_uint(len(cvkeys)), cvkeys,
mx_uint(len(cokeys)), cokeys,
+ cvals, couts, ctypes.c_int(priority)))
+ else:
+ check_call(_LIB.MXKVStoreBroadcast(
+ self.handle, mx_uint(len(cvkeys)), cvkeys,
mx_uint(len(cokeys)), cokeys,
+ cvals, couts, ctypes.c_int(priority)))
+
Review comment:
Sounds good. We don't need to have separate class then. We are already
raising exceptions in the backend when setting optimizer for P3. What about
also changing `is_capable` in KVStore to return the value based on
`DMLC_PS_VAN_TYPE`?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
