feat: Update training to accomodate antiSMASH and BGCFlow schema

.gitignore

@@ -114,3 +114,4 @@ site/
 Chinook.sqlite
 chroma.sqlite3
 notebooks
+chatbgc_env

README.md

@@ -16,10 +16,46 @@
 
 </p>
 
-Ask questions about biosynthetic gene clusters in your genome dataset via LLMs using Retrieval-Augmented Generation.
+Ask questions about BGCs in your genome dataset generated by [`BGCFlow`](https://github.com/NBChub/bgcflow) using Large Language Models (LLMs). 
 
-* Free software: MIT
-* Documentation: <https://nbchub.github.io/chatBGC/>
+This python package utilizes vector-based Retrieval-Augmented Generation (RAG) to translate natural language (English) to SQL Queries and is trained to query information from `antiSMASH` and other genome mining tools included in the [`BGCFlow`](https://github.com/NBChub/bgcflow) pipelines.
+
+![RAG](chatbgc/assets/3_RAG.png)
+
+## Quickstart
+For a quick use of `chatBGC`, you will need an [OpenAI API Key](https://platform.openai.com/api-keys) and the DuckDB database of your `BGCFlow` run (generated using `bgcflow build database`). See [BGCFlow Wiki](https://github.com/NBChub/bgcflow/wiki/04-Building-and-Serving-OLAP-Database) for more details on creating the database.
+
+```bash
+# Setup API Key
+OPENAI_API_KEY="<change this to your API Key>"
+
+# Create new folder to set up duckdb and vector database
+mkdir chatbgc
+cd chatbgc
+
+# Copy the database build using BGCFlow to this directory
+BGCFLOW_DIR="<change this to your BGCFlow directory>"
+PROJECT_NAME="<change this to your BGCFlow project name>"
+ANTISMASH_VERSION="7.1.0" # change this to the correct antiSMASH version used in your BGCFlow run. Only supports version 7.1.0 or above
+cp $BGCFLOW_DIR/data/processed/$PROJECT_NAME/dbt/antiSMASH_7.1.0/dbt_bgcflow.duckdb dbt_bgcflow.duckdb -n
+
+# Create python environment and install ChatBGC
+python3 -m venv chatbgc_env
+source chatbgc_env/bin/activate
+python3 -m pip install --upgrade pip
+pip install git+https://github.com/NBChub/chatBGC.git
+
+# Setup variable environment / secrets
+touch .env
+echo "export OPENAI_API_KEY=$OPENAI_API_KEY" > .env
+source .env
+
+# Train ChatBGC (Do it once)
+chatbgc train --llm_type openai_chat --model gpt-4o dbt_bgcflow.duckdb
+
+# Run ChatBGC
+chatbgc run --llm_type openai_chat --model gpt-4o dbt_bgcflow.duckdb
+```
 
 ## Configuration
 
@@ -106,8 +142,9 @@ chatbgc run <path_to_duckdb>
 chatbgc run --llm_type openai_chat <path_to_duckdb>
 ```
 
-## Development guide (TO DO)
-- Create duckdb schema
+## Notes
+* Free software: MIT
+* Documentation: <https://nbchub.github.io/chatBGC/>
 
 ## Credits
 

chatbgc/__init__.py

@@ -2,4 +2,4 @@
 
 __author__ = """Matin Nuhamunada"""
 __email__ = "[email protected]"
-__version__ = "0.1.3"
+__version__ = "0.2.0"

chatbgc/cli.py

@@ -58,7 +58,7 @@ def train(
         self,
         duckdb_path,
         model="llama3",
-        training_folder=str((Path(__file__).parent / "data").resolve()),
+        training_folder=str((Path(__file__).parent / "data/").resolve()),
         llm_type="ollama",
     ):
         """