refine docs

README.md

@@ -51,7 +51,7 @@ python ./custom_op_test.py
 
 除少量 `deprecated` 相关的警告外，如果输出全部都是 True，则说明自定义算子安装成功并且运行正常。
 
-### 2.1 训练
+### 2.2 训练
 
 > [!NOTE]
 > 暂时只支持 water_se_e2_a 案例的训练
@@ -63,7 +63,7 @@ cd examples/water/se_e2_a
 dp train ./input.json
 ```
 
-### 2.2 评估
+### 2.3 评估
 
 ``` sh
 # 进入案例目录
@@ -74,7 +74,7 @@ WEIGHT_PATH="path/to/your_model.pdparams"
 dp test -m ${WEIGHT_PATH} -s ../data/data_3/ -n 30
 ```
 
-### 2.3 导出静态图模型
+### 2.4 导出静态图模型
 
 ``` sh
 # 进入案例目录
@@ -87,7 +87,7 @@ DUMP_PATH="path/to/your_dump"
 dp freeze -i ${WEIGHT_PATH} -o ${DUMP_PATH}
 ```
 
-### 2.4 在 LAMMPS(GPU) 中推理
+### 2.5 在 LAMMPS(GPU) 中推理
 
 1. 修改 `examples/water/lmp/in.lammps` 文件，将 `pair_style deepmd` 后面的路径改为 **2.3 导出静态图模型** 这一章节内设置好的 DUMP_PATH 的值
 
@@ -108,7 +108,7 @@ dp freeze -i ${WEIGHT_PATH} -o ${DUMP_PATH}
     pip install python/dist/paddlepaddle_gpu-0.0.0-cp39-cp39-linux_x86_64.whl
     ```
 
-3. 安装 LAMMPS 并运行推理
+3. Paddle 推理库和 LAMMPS 联合编译安装，并运行推理
 
     ``` sh
     # 下载并解压 lammps 源码
@@ -166,7 +166,9 @@ dp freeze -i ${WEIGHT_PATH} -o ${DUMP_PATH}
     lmp_serial -in in.lammps
     ```
 
-4. 直接运行推理
+4. [可选]直接运行推理
+
+    若已完成 **3. Paddle 推理库和 LAMMPS 联合编译安装，并运行推理**，且没有对 C++ 代码进行修改，则无需重新联合编译 Paddle 推理库和 LAMMPS，直接运行以下命令即可开始推理。
 
     ``` sh
     # 设置推理时的 GPU 卡号
@@ -182,7 +184,7 @@ dp freeze -i ${WEIGHT_PATH} -o ${DUMP_PATH}
     lmp_serial -in in.lammps
     ```
 
----
+--------------------------------------------------------------------------------
 
 <span style="font-size:larger;">DeePMD-kit Manual</span>
 ========
@@ -216,7 +218,8 @@ For more information, check the [documentation](https://deepmd.readthedocs.io/).
 
 # Highlights in DeePMD-kit v2.0
 
-* [Model compression](doc/freeze/compress.md). Accelerate the efficiency of model inference 4-15 times.
+- [Model compression](doc/freeze/compress.md). Accelerate the efficiency of model inference 4-15 times.
+
 - [New descriptors](doc/model/overall.md). Including [`se_e2_r`](doc/model/train-se-e2-r.md) and [`se_e3`](doc/model/train-se-e3.md).
 - [Hybridization of descriptors](doc/model/train-hybrid.md). Hybrid descriptor constructed from the concatenation of several descriptors.
 - [Atom type embedding](doc/model/train-se-e2-a-tebd.md). Enable atom-type embedding to decline training complexity and refine performance.
@@ -226,7 +229,8 @@ For more information, check the [documentation](https://deepmd.readthedocs.io/).
 
 ## Highlighted features
 
-* **interfaced with TensorFlow**, one of the most popular deep learning frameworks, making the training process highly automatic and efficient, in addition, Tensorboard can be used to visualize training procedures.
+- **interfaced with TensorFlow**, one of the most popular deep learning frameworks, making the training process highly automatic and efficient, in addition, Tensorboard can be used to visualize training procedures.
+
 - **interfaced with high-performance classical MD and quantum (path-integral) MD packages**, i.e., LAMMPS and i-PI, respectively.
 - **implements the Deep Potential series models**, which have been successfully applied to finite and extended systems including organic molecules, metals, semiconductors, insulators, etc.
 - **implements MPI and GPU supports**, making it highly efficient for high-performance parallel and distributed computing.