Merge pull request #1377 from jaelgu/jael-rag

Update README

bootcamp/tutorials/quickstart/apps/rag_search_with_milvus/README.md

@@ -29,14 +29,14 @@ $ cd bootcamp/bootcamp/tutorials/quickstart/app/rag_search_with_milvus
 
   ```
   OPENAI_API_KEY=************
-  MILVUS_CONNECTION_URI=./milvus_demo.db
+  MILVUS_ENDPOINT=./milvus_demo.db
   # MILVUS_TOKEN=************
 
   COLLECTION_NAME=my_rag_collection
   ```
 
   - `OPENAI_API_KEY`: We will use OpenAI for LLM capabilities. You need to prepare the [OpenAI API Key](https://help.openai.com/en/articles/4936850-where-do-i-find-my-openai-api-key).
-  - `MILVUS_CONNECTION_URI`: The URI to connect Milvus or Zilliz Cloud service. By default, we use a local file "./milvus_demo.db" for convenience, as it automatically utilizes [Milvus Lite](https://milvus.io/docs/milvus_lite.md) to store all data at local.
+  - `MILVUS_ENDPOINT`: The URI to connect Milvus or Zilliz Cloud service. By default, we use a local file "./milvus_demo.db" for convenience, as it automatically utilizes [Milvus Lite](https://milvus.io/docs/milvus_lite.md) to store all data at local.
     > - If you have large scale of data, you can set up a more performant Milvus server on docker or kubernetes. In this setup, please use the server uri, e.g. http://localhost:19530, as your uri.
     > 
     > - If you want to use Zilliz Cloud, the fully managed cloud service for Milvus, adjust the uri and token, which correspond to the Public Endpoint and Api key in Zilliz Cloud.
@@ -90,26 +90,27 @@ Run the Streamlit application:
 ```bash
 $ streamlit run app.py
 ```
+
 ### Example Usage
 
+**Step 1:** Enter your question in the chat and click on 'submit' button.
 <div style="text-align: center;">
   <figure>
     <img src="./pics/step1.png" alt="Description of Image" width="700"/>
-    <figcaption>Step 1: Enter your question in the chat and click on 'submit' button.</figcaption>
   </figure>
 </div>
 
+**Step 2:** Response generated by LLM based on the prompt.
 <div style="text-align: center;">
   <figure>
     <img src="./pics/step2.png" alt="Description of Image" width="700"/>
-    <figcaption>Step 2: Response generated by LLM based on the prompt.</figcaption>
   </figure>
 </div>
 
+**Step 3:** Top 3 retrieved original quotes from text data are listed on the left along with their distances, indicating how related the quote and the prompt are.
 <div style="text-align: center;">
   <figure>
     <img src="./pics/step3.png" alt="Description of Image" width="700"/>
-    <figcaption>Step 3: Top 3 retrieved original quotes from text data are listed on the left along with their distances, indicating how related the quote and the prompt are. </figcaption>
   </figure>
 </div>
 
@@ -125,6 +126,6 @@ $ streamlit run app.py
 
 - **app.py:** The main file to run application by Streamlit, where the user interface is defined and the RAG chatbot is presented.
 - **encoder.py:** The text encoder is able to convert text inputs to text embeddings using a pretrained model.
-- **insert.py:** This script handles the retrieving text data required for the application and inserting text embeddings into data collection.
+- **insert.py:** This script creates collection and inserts document chunks with embeddings into the collection.
 - **milvus_utils.py:** Functions to interact with the Milvus database, such as creating collection and retrieving search results.