sky.yaml

# Example: a distributed DeepSpeed job (DeepSpeed-Chat) on 2 VMs.
#
# This takes care constructing a "hostfile" to pass to DeepSpeed.
#
# Usage:
#
#   $ sky launch sky.yaml -r --down -c ds
#
#   # Optional: After the job starts running, you can log into the two nodes and
#   # check gpustat:
#   $ ssh ds
#   $ gpustat -i
#   $ ssh ds-worker1
#   $ gpustat -i

resources:
  accelerators: A100:1  # GCP, Lambda
  # accelerators: A100-80GB:1  # Azure, GCP, SCP
  # accelerators: A10G:1  # AWS. Will OOM for (1) single_node/run_1.3b_lora.sh (2) multi_node/run_66b.sh.
  # accelerators: T4:1  # AWS, Azure, GCP. Will OOM for (1) single_node/run_1.3b_lora.sh (2) multi_node/run_66b.sh.
num_nodes: 2

setup: |
  git clone https://github.com/microsoft/DeepSpeedExamples.git || true
  cd DeepSpeedExamples
  git checkout d7c42b4f34df91035e7ed3e0c51500bb53d0bc71

  conda activate deepspeed
  if [ $? -eq 0 ]; then
    echo 'conda env exists'
  else
    conda create -n deepspeed python=3.8 -y
    conda activate deepspeed

    pip install deepspeed

    cd applications/DeepSpeed-Chat
    pip install -r requirements.txt

    # Required by DeepSpeed in multi-node settings.
    #
    # NOTE(skypilot): DeepSpeed uses `pdsh` to log into each node and calls
    # `ninja --version`; so it has to be installed system-wide rather than in
    # the above 'deepspeed' conda env.
    sudo apt-get -y install pdsh ninja-build
  fi

file_mounts:
  # Required for DeepSpeed's passwordless SSH (run commands on nodes).
  ~/.ssh/id_rsa: ~/.ssh/sky-key

run: |
  cd DeepSpeedExamples
  conda activate deepspeed

  # Launch on the first node only
  if [ "${SKYPILOT_NODE_RANK}" == "0" ]; then

    # Prepare a hostfile.
    HOSTFILE_PATH=/tmp/hostfile.${SKYPILOT_TASK_ID}
    python -c "import os;n_gpus=os.environ['SKYPILOT_NUM_GPUS_PER_NODE'];print('\n'.join([f'{ip} slots={n_gpus}' for ip in os.environ['SKYPILOT_NODE_IPS'].splitlines()]))" > ${HOSTFILE_PATH}

    echo "*******************************************"
    echo "Hostfile: ${HOSTFILE_PATH}"
    cat ${HOSTFILE_PATH}
    echo "*******************************************"

    ################ Your launch command goes here ################

    cd applications/DeepSpeed-Chat/training/step1_supervised_finetuning/

    # Adapted from: training_scripts/single_node/run_1.3b_lora.sh
    # Note the additional argument: --hostfile $HOSTFILE_PATH
    # Alternatively, you can move HOSTFILE_PATH to /job/hostfile:
    #   sudo mkdir -p /job; sudo chmod 777 /job; mv ${HOSTFILE_PATH} /job/hostfile

    OUTPUT_PATH=./output
    mkdir -p $OUTPUT_PATH
    deepspeed \
      --hostfile $HOSTFILE_PATH \
      main.py \
      --data_path Dahoas/rm-static Dahoas/full-hh-rlhf Dahoas/synthetic-instruct-gptj-pairwise yitingxie/rlhf-reward-datasets \
      --data_split 2,4,4 \
      --model_name_or_path facebook/opt-1.3b \
      --per_device_train_batch_size 8 \
      --per_device_eval_batch_size 8 \
      --max_seq_len 512 \
      --learning_rate 1e-3 \
      --weight_decay 0.1 \
      --num_train_epochs 16 \
      --gradient_accumulation_steps 1 \
      --lr_scheduler_type cosine \
      --num_warmup_steps 0 \
      --seed 1234 \
      --zero_stage 0 \
      --lora_dim 128 \
      --lora_module_name decoder.layers. \
      --only_optimize_lora \
      --deepspeed \
      --output_dir $OUTPUT_PATH \
      | tee $OUTPUT_PATH/training.log

  fi