eric-haibin-lin commented on a change in pull request #14200: Bulked op
segments to allow Variable nodes
URL: https://github.com/apache/incubator-mxnet/pull/14200#discussion_r259555329
##########
File path: src/imperative/cached_op.h
##########
@@ -53,10 +53,12 @@ struct CachedOpConfig : public
dmlc::Parameter<CachedOpConfig> {
.set_default(2)
.describe("Maximum number of operators that can be inlined.");
DMLC_DECLARE_FIELD(forward_bulk_size)
- .set_default(dmlc::GetEnv("MXNET_EXEC_BULK_EXEC_MAX_NODE_TRAIN", 15))
Review comment:
I noticed that env_vars like
`MXNET_EXEC_BULK_EXEC_TRAIN`/`MXNET_EXEC_BULK_EXEC_INFER`=0 are not respected
by the cached_op. Would you have time to kindly fix it for cached op?
https://github.com/apache/incubator-mxnet/blob/54fd288c7a4bf59d37f793c26ef9a98ed40b0c40/src/imperative/cached_op.cc#L593-L596
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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
