EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

port deepseekv2 and mtp to main branch (#429)

Browse Source 
### What this PR does / why we need it?
This PR ports all the deepseek graph mode code and mtp code from v0.7.3
to the main branch
---------

Signed-off-by: SidaoY <1024863041@qq.com>
Signed-off-by: linfeng-yuan <1102311262@qq.com>
Signed-off-by: Yizhou Liu <liuyizhou5@h-partners.com>
Signed-off-by: mengwei805 <mengwei25@huawei.com>
Signed-off-by: libaokui <libaokui@huawei.com>
Signed-off-by: q00832892 <qiaoyang19@huawei.com>
Signed-off-by: ganyi <pleaplusone.gy@gmail.com>
Co-authored-by: SidaoY <1024863041@qq.com>
Co-authored-by: linfeng-yuan <1102311262@qq.com>
Co-authored-by: Yizhou Liu <liuyizhou5@h-partners.com>
Co-authored-by: mengwei805 <mengwei25@huawei.com>
Co-authored-by: libaokui <libaokui@huawei.com>
This commit is contained in:
Pleaplusone
2025-04-19 17:38:18 +08:00
committed by GitHub
parent 086423dc35
commit 1a1f9a6d89
33 changed files with 3361 additions and 315 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
examples/dp_offline/run_dp.sh Normal file
View File

@@ -0,0 +1,21 @@
export HCCL_IF_IP=${local_ip}
export GLOO_SOCKET_IFNAME=${ifname}
export TP_SOCKET_IFNAME=${ifname}
export HCCL_SOCKET_IFNAME=${ifname}
# dp_size = node_size * dp_per_node
node_size=1
node_rank=0
dp_per_node=2
master_addr=127.0.0.1
master_port=12345
rm -rf ./.torchair_cache/
rm -rf ./dynamo_*
rm -rf /root/ascend/log/debug/plog/*
export VLLM_ENABLE_GRAPH_MODE=0
export VLLM_ENABLE_MC2=0
torchrun --nproc_per_node ${dp_per_node} --nnodes ${node_size} \
--node_rank ${node_rank} --master_addr ${master_addr} --master_port ${master_port} \
data_parallel.py