scripts/run_ba_role_sft.sh

#!/usr/bin/env bash
set -euo pipefail

script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
exp_root="$(cd "${script_dir}/.." && pwd)"

role_name="${ROLE_NAME:-data_insight}"
base_model_path="${BASE_MODEL_PATH:-meta-llama/Llama-3.1-8B}"
deepspeed_config_name="${DEEPSPEED_CONFIG_NAME:-${exp_root}/config/ds.json}"
dataset_sub_name="${DATASET_SUB_NAME:-}"
dataset_split="${DATASET_SPLIT:-train}"
system_prompt="${SYSTEM_PROMPT:-}"

output_root="${OUTPUT_ROOT:-${exp_root}/output/${role_name}}"
sft_dataset_name="${SFT_DATASET_NAME:-${exp_root}/data/${role_name}_sft.jsonl}"
model_sft_lora_path="${output_root}/sft_lora"
model_sft_full_path="${output_root}/sft_full"

mkdir -p "${output_root}"

if [ ! -f "${sft_dataset_name}" ] && [ ! -d "${sft_dataset_name}" ]; then
  echo "Missing SFT dataset: ${sft_dataset_name}"
  echo "See ${exp_root}/README.md and the sample JSONL files."
  exit 1
fi

deepspeed "${exp_root}/sft.py" \
  --dataset_name="${sft_dataset_name}" \
  --dataset_sub_name="${dataset_sub_name}" \
  --dataset_split="${dataset_split}" \
  --system_prompt="${system_prompt}" \
  --model_name="${base_model_path}" \
  --seq_length="${SEQ_LENGTH:-1024}" \
  --output_name="${model_sft_lora_path}" \
  --use_QLora="${USE_QLORA:-True}" \
  --batch_size="${BATCH_SIZE:-8}" \
  --use_flash_attention_2="${USE_FLASH_ATTENTION_2:-True}" \
  --deepspeed_config_name="${deepspeed_config_name}" \
  --num_train_epochs="${NUM_TRAIN_EPOCHS:-2}" \
  --gradient_accumulation_steps="${GRADIENT_ACCUMULATION_STEPS:-8}" \
  --learning_rate="${LEARNING_RATE:-1e-5}"

python "${exp_root}/merge_adapter.py" \
  --base_model_name="${base_model_path}" \
  --model_name="${model_sft_lora_path}" \
  --merged_model_name="${model_sft_full_path}"