add readme

README.md

@@ -1,6 +1,6 @@
 # ExperienceMaker
 
-## 🌟 What is ExperienceMaker?
+# 🌟 What is ExperienceMaker?
 ExperienceMaker provides agents with robust capabilities for experience generation and reuse. 
 By summarizing agents' past trajectories into experiences, it enables these experiences to be applied to subsequent tasks. 
 Through the continuous accumulation of experience, agents can keep learning and progressively become more skilled in performing tasks.
@@ -14,29 +14,54 @@ Through the continuous accumulation of experience, agents can keep learning and
 - **Ease of Use**: An HTTP POST interface is provided, allowing one-click startup via the command line. Configuration can be quickly updated using configuration files and command-line arguments.
 - **Flexibility**: A rich library of operations is included. By composing atomic ops into pipelines, users can flexibly implement any summarization or retrieval task.
 - **Experience Store**: Ready-to-use out of the box — there's no need for you to manually summarize experiences. You can directly leverage existing, comprehensive experience datasets to greatly enhance your agent’s capabilities.
-
-
+- 
 ### Framework
-
-💾 VectorStore: MemoryScope is equipped with a vector database (default is *ElasticSearch*) to store all memory fragments recorded in the system.
-
-🔧 Worker Library: MemoryScope atomizes the capabilities of long-term memory into individual workers, including over 20 workers for tasks such as query information filtering, observation extraction, and insight updating.
-
-🛠️ Operation Library: Based on the worker pipeline, it constructs the operations for memory services, realizing key capabilities such as memory retrieval and memory consolidation.
-
-- Memory Retrieval: Upon arrival of a user query, this operation returns the semantically related memory pieces 
-and/or those from the corresponding time if the query involves reference to time.
-- Memory Consolidation: This operation takes in a batch of user queries and returns important user information
-extracted from the queries as consolidated *observations* to be stored in the memory database.
-- Reflection and Re-consolidation: At regular intervals, this operation performs reflection upon newly recorded *observations*
-to form and update *insights*. Then, memory re-consolidation is performed to ensure contradictions and repetitions
-among memory pieces are properly handled.
+- APIs:
+  - **Retriever API**: Interface for experience retrieval. The input can be a query or conversation messages, and the output includes a list of retrieved experiences and combined contextual content, aiming to facilitate experience reuse.
+  - **Summarizer API**: Interface for experience summarization. The input is a list of agent's historical trajectories, and the output is a list of summarized experiences that have been stored in the vector store.
+  - **Vector Store API**: Interface for experience management. The input consists of database operation actions, with optional dump/load paths for experience data import/export.
+- Pipeline & Operator: ExperienceMaker abstracts the capabilities of experience summarization and retrieval into atomic functions. By composing these atomic functions into pipelines or adding custom operators, users can easily build any experience summarization or retrieval pipeline.
+- Vector Store: ExperienceMaker is equipped with a vector database, with ElasticSearch as the default (due to its excellent performance and ease of use), though it also supports other vector databases.
+- LLM & Embedding Model: ExperienceMaker relies on large language models and embedding models to provide text generation and vectorization services. These are the core atomic capabilities of ExperienceMaker.
 
 # install
+## Installation
+```shell
+git clone https://github.com/modelscope/ExperienceMaker.git
 
+# Install the package
+pip install .
+```
+
+## Install From PyPi
+```shell
+pip install experiencemaker
+```
 
 ## quick start
 
+## vector store
+ExperienceMaker is equipped with a vector database. 
+If you just want to try out ExperienceMaker, you can use `backend=local_file` for testing purposes. Please note that this method may become time-consuming when dealing with large amounts of data. You can skip this step.
+If you are planning to deploy ExperienceMaker or expect a significant QPS (queries per second), we recommend using `backend=elasticsearch`.
+To set up [Elasticsearch](https://www.elastic.co/docs/solutions/search/run-elasticsearch-locally) and Kibana locally, run the start-local script in the command line:
+```shell
+curl -fsSL https://elastic.co/start-local | sh
+```
+- Elasticsearch [quick start](../vector_store/elasticsearch.md)
+- chroma
+
+
+## Environment Variables
+```shell
+export LLM_API_KEY="sk-xxx"
+export LLM_BASE_URL="xxx"
+export EMBEDDING_MODEL_API_KEY="sk-xxx"
+export EMBEDDING_MODEL_BASE_URL="xxx"
+```
+
+
+
 
 
 ## 💡 Contribute