#!/bin/bash
export HSA_FORCE_FINE_GRAIN_PCIE=1
export MIOPEN_FIND_MODE=3
export MIOPEN_COMPILE_PARALLEL_LEVEL=1
export NCCL_PLUGIN_P2P=ucx
export RCCL_NCHANNELS=2
export NCCL_SOCKET_IFNAME=ib0
export NCCL_P2P_LEVEL=5
export NCCL_IB_HCA=mlx5_0
export NCCL_DEBUG=INFO
export NCCL_NET_GDR_LEVEL=SYS
export NCCL_NET_PLUGIN=none
unset RCCL_NCHANNELS
unset NCCL_NET_GDR_LEVEL
lrank=$OMPI_COMM_WORLD_LOCAL_RANK
echo "LRANK===============================$lrank"
RANK=$OMPI_COMM_WORLD_RANK
WORLD_SIZE=$OMPI_COMM_WORLD_SIZE
echo "WORLD_SIZE*************$WORLD_SIZE"

export HIP_VISIBLE_DEVICES=0,1,2,3


APP="python3 ../fine-tune.py \
    --deepspeed ../ds_config.json \
    --report_to "none" \
    --data_path "../data/belle_chat_ramdon_10k.json" \
    --model_name_or_path "../../baichuan2-7b-base" \
    --output_dir "output" \
    --model_max_length 64 \
    --num_train_epochs 4 \
    --per_device_train_batch_size 1 \
    --gradient_accumulation_steps 1 \
    --save_strategy epoch \
    --learning_rate 2e-5 \
    --lr_scheduler_type constant \
    --adam_beta1 0.9 \
    --adam_beta2 0.98 \
    --adam_epsilon 1e-8 \
    --max_grad_norm 1.0 \
    --weight_decay 1e-4 \
    --warmup_ratio 0.0 \
    --logging_steps 1 \
    --gradient_checkpointing False \
    --fp16 \
    --local_rank $lrank "

case ${lrank} in
[0])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_0:1
  export UCX_IB_PCI_BW=mlx5_0:50Gbs
  numactl --cpunodebind=0 --membind=0 ${APP}
  ;;
[1])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_1:1
  export UCX_IB_PCI_BW=mlx5_1:50Gbs
  numactl --cpunodebind=1 --membind=1 ${APP}
  ;;
[2])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_2:1
  export UCX_IB_PCI_BW=mlx5_2:50Gbs
  numactl --cpunodebind=2 --membind=2 ${APP}
  ;;
[3])
  export HIP_VISIBLE_DEVICES=0,1,2,3
  export UCX_NET_DEVICES=mlx5_3:1
  export UCX_IB_PCI_BW=mlx5_3:50Gbs
  numactl --cpunodebind=3 --membind=3 ${APP}
  ;;
esac