MOD: update readme

README.md

@@ -49,25 +49,23 @@ export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 pipeline:
 # 1 Fully Fine‑tune image model with ZeRO
 accelerate launch --deepspeed_config_file=ds_config_zero3.json train_lora.py
-python train_lora.py
+fully_fine_tine_stablediffusion
 
-# 2 SFT 120B OSS 语言模型 with QLoRA
-python sft_train.py
+# 2 SFT 120B OSS 语言模型 with QLoRA 
+lauguage_model_fine_tuning
 
-# 3 distill 120B OSS模型给20B OSS模型
-用 Teacher 生成 Response，student模型用LoRA fine tuning
-
-# 4 Build RAG index
-python build_embeddings.py
+# 3 RLHF PPO 120B OSS 语言模型 with QLoRA : 训练 reward model 
+lauguage_model_fine_tuning
 
-# 5 收集偏好 → 训练 reward model 
-python reward_model.py
+# 4 distill 120B OSS模型给20B OSS模型
+lauguage_model_fine_tuning
+用 Teacher 生成 Response，student模型用LoRA fine tuning
 
-# 6 PPO RLHF 微调20B OSS model
-python ppo_tune.py
+# 5 Build RAG index embedding table
+retrieval_augmented_generation
 
-# 7 Inference with RAG
-python rag_infer.py
+# 6 Inference with RAG
+inference.py
 
 
 system flow: