updated langchain csv and sqlite / sql workflow

3_db.py

@@ -1,15 +1,18 @@
 import os 
 from dotenv import load_dotenv
-from langchain import OpenAI, SQLDatabase
-from langchain import SQLDatabaseChain
+from langchain import OpenAI, SQLDatabase, SQLDatabaseChain
 
 load_dotenv()
 
-dburi = os.getenv("DATABASE_URL")
+# dburi = os.getenv("DATABASE_URL")
+dburi = "sqlite:///academy/academy.db"
 db = SQLDatabase.from_uri(dburi)
 llm = OpenAI(temperature=0)
 
 db_chain = SQLDatabaseChain(llm=llm, database=db, verbose=True)
 
-db_chain.run("How many responses do we have this year?")
-db_chain.run("What is the average ratings for Samuel Chan in the responses table?")
+db_chain.run("How many rows is in the responses table of this db?")
+db_chain.run("Describe the responses table")
+db_chain.run("What are the top 3 countries where these responses are from?")
+db_chain.run("Give me a summary of how these customers come to hear about us. \
+    What is the most common way they hear about us?")

4_csv.py

@@ -0,0 +1,18 @@
+from dotenv import load_dotenv
+from langchain import OpenAI 
+from langchain.document_loaders.csv_loader import CSVLoader
+
+load_dotenv()
+
+filepath = "academy/academy.csv"
+loader = CSVLoader(filepath)
+data = loader.load()
+print(data)
+
+llm = OpenAI(temperature=0)
+
+from langchain.agents import create_csv_agent
+agent = create_csv_agent(llm, filepath, verbose=True)
+agent.run("What percentage of the respondents are students versus professionals?")
+agent.run("List the top 3 devices that the respondents use to submit their responses")
+agent.run("Consider iOS and Android as mobile devices. What is the percentage of respondents that discovered us through social media submitting this from a mobile device?")