update code

Author: HydrogenSulfate

deploy/python_serving/readme.md

@@ -64,10 +64,7 @@ python3.7 -m pip install paddle-serving-server-gpu==0.7.0.post112  # GPU with CU
   unzip ppTSM.zip
   popd
   ```
-
-- 我们提供了转换好的[PP-TSM推理模型](https://videotag.bj.bcebos.com/PaddleVideo-release2.3/ppTSM.zip)
-
-- 用 paddle_serving_client 把下载的推理模型转换成易于 Server 部署的模型格式：
+- 用 paddle_serving_client 把转换好的推理模型再转换成易于 Server 部署的模型格式：
   ```bash
   python3.7 -m paddle_serving_client.convert \
   --dirname inference/ppTSM \

deploy/python_serving/readme_en.md

@@ -65,8 +65,7 @@ When using PaddleServing for service deployment, you need to convert the saved i
   popd
   ```
 
-- We provide the converted [PP-TSM inference model](https://videotag.bj.bcebos.com/PaddleVideo-release2.3/ppTSM.zip)
-- Use paddle_serving_client to convert the downloaded inference model into a model format that is easy for server deployment:
+- Use paddle_serving_client to convert the converted inference model into a model format that is easy for server deployment:
   ```bash
   python3.7 -m paddle_serving_client.convert \
   --dirname inference/ppTSM \