mlc_chat convert_weight的时候报错/mlc-mini/build/libmlc_llm_module.dylib, 0x0006): Symbol not found: __ZN3tvm7runtime7NDArray10CreateViewENS0_10ShapeTupleE10DLDataType
zhb-code opened this issue · comments
是否已有关于该错误的issue或讨论? | Is there an existing issue / discussion for this?
- 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions
该问题是否在FAQ中有解答? | Is there an existing answer for this in FAQ?
- 我已经搜索过FAQ | I have searched FAQ
当前行为 | Current Behavior
我准备在Android客户端部署MiniCPM,现在遇到问题,希望官方大佬帮忙解决一下,感谢!
期望行为 | Expected Behavior
现在mlc_chat convert_weight时一直报错,期望可以执行通过;
复现方法 | Steps To Reproduce
1、配置好环境变量
2、按照官方文档执行
mkdir -p build && cd build
generate build configuration
python3 ../cmake/gen_cmake_config.py && cd ..
build mlc_chat_cli
cd build && cmake .. && cmake --build . --parallel $(nproc) && cd ..
install
cd python && pip install -e . && cd ..
到这一步都正常
3、model是从https://modelscope.cn/models/OpenBMB/miniCPM-bf16/files这里下载的
3、继续执行mlc_chat convert_weight --model-type ${MODEL_TYPE} ./dist/models/${MODEL_NAME}-hf/ --quantization $QUANTIZATION -o dist/$MODEL_NAME/ 这一步的时候就报错了
报错信息:
OSError: dlopen(/Users/zhb/Documents/StudioWorkSpace/mlc-mini/build/libmlc_llm_module.dylib, 0x0006): Symbol not found: __ZN3tvm7runtime7NDArray10CreateViewENS0_10ShapeTupleE10DLDataType
Referenced from: /Users/zhb/Documents/StudioWorkSpace/mlc-mini/build/libmlc_llm_module.dylib
Expected in: <533BA3B9-AEB8-3E4C-94D9-6E70457BEDF3> /opt/miniconda3/envs/mlc-mini/lib/python3.11/site-packages/tvm/libtvm.dylib
该如何解决?
运行环境 | Environment
- OS:macox m3pro
- Python:3.11
- Transformers:
- PyTorch:
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`):
备注 | Anything else?
No response