Ii-robot!1957 fix import error and coc demo

3d5d5644创建于 2025年2月27日历史提交

文件	最后提交记录	最后更新时间
README.md	!760 【core_r0.6.0】ATB路径整改 Merge pull request !760 from MingzhenWang/atbpath_0.6.0	1 年前
ffn.md	!443 自定义算子FFN支持单算子模式 & 文档更新 Merge pull request !443 from chenjiazhong/ffn_cpp	1 年前
fusion_attention.md	!454 update doc Merge pull request !454 from 李晓君/master	1 年前
gmm.md	!1457 [1.0.0 060]readme修正；增加moe mcore拦截 Merge pull request !1457 from 赵一帆/zyfrc4060	1 年前
lcal_coc.md	!1957 fix import error and coc demo Merge pull request !1957 from wangyuansheng8/1.0.0_core_r0.6.0	1 年前
npu_all_to_all_all_gather_bmm.md	!1157 MOE迭代torch接口文档输出字段描述完善 Merge pull request !1157 from 陈建军/cherry-pick-1729932310	1 年前
npu_bmm_reduce_scatter_all_to_all.md	!1157 MOE迭代torch接口文档输出字段描述完善 Merge pull request !1157 from 陈建军/cherry-pick-1729932310	1 年前
npu_dropout_add_layer_norm.md	!371 MoE MLP not use mc2 linear and remove ops from init Merge pull request !371 from Jializheng/master	1 年前
npu_fused_moe_token_permute.md	!823 docs(docs/ops/npu_fused_moe_token_permute.md) Merge pull request !823 from 邓佳/core_r0.6.0_docs	1 年前
npu_fused_moe_token_unpermute.md	!695 perf: use-fused-moe-token-permute-and-unpermute Merge pull request !695 from 邓佳/dev_moe_core_r0.6.0	1 年前
npu_grouped_mat_mul_all_reduce.md	!359 change ascendspeed to mindspeed Merge pull request !359 from 邓佳/master	1 年前
npu_groupmatmul_add.md	!1340 GMM+ADD overlap适配 bugfix Merge pull request !1340 from zengshu/1.0.0_core_r0.6.0	1 年前
npu_matmul_add.md	!1307 bugfix_GmmAdd_MmAdd Merge pull request !1307 from zengshu/core_r0.6.0	1 年前
npu_mm_all_reduce_add_rms_norm.md	!1130 [BUGfix!&update!] 确定性计算开关BUG修复及readme更新 Merge pull request !1130 from 杨杰/core_r0.6.0	1 年前
npu_mm_all_reduce_add_rms_norm_.md	!1130 [BUGfix!&update!] 确定性计算开关BUG修复及readme更新 Merge pull request !1130 from 杨杰/core_r0.6.0	1 年前
npu_ring_attention_update.md	!678 feat: use-fused-ring-attention-update融合算子 Merge pull request !678 from 邓佳/dev_fused_core_r0.6.0	1 年前
npu_rotary_position_embedding.md	!812 修改资料：rotary_position_embedding支持确定性计算;RotaryHalf模式下增加约束 Merge pull request !812 from 赵臣臣/core_r0.6.0	1 年前
rms_norm.md	!359 change ascendspeed to mindspeed Merge pull request !359 from 邓佳/master	1 年前
swiglu.md	!359 change ascendspeed to mindspeed Merge pull request !359 from 邓佳/master	1 年前

How to run the ops?

previous installation

CANN
CANN-NNAL(Ascend-Transformer-Boost)
torch_npu

compile and install

1. set the environment variables

# Default path, change it if needed.
source /usr/local/Ascend/ascend-toolkit/set_env.sh

if use Ascend-Transformer-Boost

# Default path, change it if needed.
source /usr/local/Ascend/nnal/atb/set_env.sh

2. include head files

newest torch_npu
newest cann

3. install scripts

python3 setup.py build
python3 setup.py bdist_wheel
pip3 install dist/*.whl --force-reinstall