diff --git a/spawnerdockerfile/README.md b/spawnerdockerfile/README.md index d8576ff..635a383 100755 --- a/spawnerdockerfile/README.md +++ b/spawnerdockerfile/README.md @@ -1,5 +1,53 @@ # Base Jupyter Notebook Stack +## ds_report + +```shell +[2024-07-17 02:25:56,956] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) + [WARNING] async_io requires the dev libaio .so object and headers but these were not found. + [WARNING] async_io: please install the libaio-dev package with apt + [WARNING] If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found. + [WARNING] Please specify the CUTLASS repo directory as environment variable $CUTLASS_PATH + [WARNING] sparse_attn requires a torch version >= 1.5 and < 2.0 but detected 2.4 + [WARNING] using untested triton version (3.0.0), only 1.0.0 is known to be compatible + +(deepspeed) root@ubuntu-finetune:~/binbbt/train/pretrain# cat .deepspeed_env +CUDA_HOME=/usr/local/cuda/ +TORCH_USE_CUDA_DSA=1 +CUTLASS_PATH=/opt/cutlass +TORCH_CUDA_ARCH_LIST="80;89;90;90a" +LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/lib:/usr/local/mpi/lib:/usr/local/mpi/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64 +NCCL_DEBUG=WARN +NCCL_SOCKET_IFNAME=bond0 +NCCL_IB_HCA=mlx5_0:1,mlx5_2:1,mlx5_4:1,mlx5_6:1 +NCCL_IB_GID_INDEX=3 +NCCL_NET_GDR_LEVEL=2 +NCCL_P2P_DISABLE=0 +NCCL_IB_DISABLE=0 +``` + +## test command + +docker run -it --rm --network=host --privileged --ipc=host --ulimit memlock=-1 --gpus all hotwa/notebook:ngc +docker run --rm -it --gpus all --ipc=host --ulimit memlock=-1 --ulimit stack=67108864 hotwa/notebook:ngc /bin/bash + +```shell +nvidia-smi +nvcc -V +ninja --version +ds_report +python -c "import torch; print('torch:', torch.__version__, torch)" +python -c "import torch; print('CUDA available:', torch.cuda.is_available())" +python -c "import deepspeed; deepspeed.ops.op_builder.CPUAdamBuilder().load()" +python -c "from flash_attn import flash_attn_func, flash_attn_varlen_func" +python -c "import apex.amp; print('Apex is installed and the amp module is available.')" +python -c "from xformers import ops as xops" +ibstat +ofed_info -s # 如果输出显示了 OFED 版本号,则说明 OFED 驱动已安装。 +mst version +mpirun --version +``` + > **Images hosted on Docker Hub are no longer updated. Please, use [quay.io image](https://quay.io/repository/jupyter/base-notebook)** [![docker pulls](https://img.shields.io/docker/pulls/jupyter/base-notebook.svg)](https://hub.docker.com/r/jupyter/base-notebook/)