Update CHANGELOG.md

CHANGELOG.md

@@ -8,26 +8,32 @@
 
 ### Fixes and improvements
 
-## [0.3.0](https://github.com/OpenNMT/OpenNMT-py/tree/v0.3.0) (2018-09-27)
-Now requires Pytorch 0.4.1
+## [0.4.0](https://github.com/OpenNMT/OpenNMT-py/tree/v0.4.0) (2018-10-08)
+* Fixed Speech2Text training (thanks Yuntian)
+
+* Removed -max_shard_size, replaced by -shard_size = number of examples in a shard.
+  Default value = 1M which works fine in most Text dataset cases. (will avoid Ram OOM in most cases)
 
-Multi-node Multi-GPU with Torch Distributed
 
-New options are:
--master_ip: ip address of the master node
--master_port: port number of th emaster node
--world_size = total number of processes to be run (total GPUs accross all nodes)
--gpu_ranks = list of indices of processes accross all nodes
+## [0.3.0](https://github.com/OpenNMT/OpenNMT-py/tree/v0.3.0) (2018-09-27)
+* Now requires Pytorch 0.4.1
+
+* Multi-node Multi-GPU with Torch Distributed
 
--gpuid is deprecated
+  New options are:
+  -master_ip: ip address of the master node
+  -master_port: port number of th emaster node
+  -world_size = total number of processes to be run (total GPUs accross all nodes)
+  -gpu_ranks = list of indices of processes accross all nodes
 
+* gpuid is deprecated
 See examples in https://github.com/OpenNMT/OpenNMT-py/blob/master/docs/source/FAQ.md
 
-Fixes to img2text now working
+* Fixes to img2text now working
 
-New sharding based on number of examples
+* New sharding based on number of examples
 
-Fixes to avoid 0.4.1 deprecated functions.
+* Fixes to avoid 0.4.1 deprecated functions.
 
 
 ## [0.2.1](https://github.com/OpenNMT/OpenNMT-py/tree/v0.2.1) (2018-08-31)