flux
flux copied to clipboard
[QUESTION]你好,flux/test/python/moe_ag_scatter/test_moe_ag.py这个需要在8gpu的时候设置tp size=2需要怎么改?
我看flux/test/python/moe_gather_rs/test_moe_gather_rs.py里面可以随意设置tp size?我理解tp size可以和gpu数量不相等的,请不吝赐教,感谢!
In test_moe_ag.py, the tp size you want is actually ffn_tp_size
所以TP_GROUP.size()指的是物理gpu的数量?
In the context of sequence parallelism, the input is partitioned into shards, with the number of shards being equal to TP_GROUP.size(), as is the situation in our example. In this scenario, the TP_GROUP.size() is equivalent to the number of devices.
Just change the code in launch.sh nproc_per_node=2