Upload 4 files

README.md

@@ -1,14 +1,51 @@
----
-title: DialogData
-emoji: 🐨
-colorFrom: green
-colorTo: blue
-sdk: streamlit
-sdk_version: 1.42.0
-app_file: app.py
-pinned: false
-license: apache-2.0
-short_description: Analisis data menggunakan PandasAI
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Chat with your Dataset
+
+This project demonstrates a web-based application to query a dataset through natural language.
+
+![](./assets/image.png)
+
+For this purpose, it uses:
+
+- [Streamlit](https://streamlit.io/) to build a data science web app
+- [Pandasai](https://pandas-ai.com/) to generate Pandas code from a query through [OpenAI GPT-3.5](https://platform.openai.com/docs/api-reference)
+
+## Download dataset
+
+Download the [dataset](https://github.com/Fraud-Detection-Handbook/simulated-data-transformed) into the `data` folder at the root of the project.
+
+## Run the project
+
+If you don't have a Python environment available, you can use the [conda package manager](https://docs.conda.io/projects/conda/en/latest/index.html) which comes with the [Anaconda distribution](https://www.anaconda.com/download) to manage a clean Python environment.
+
+Create a new environment and activate it:
+
+```sh
+conda create -n streamlit-pandasai python=3.9
+conda activate streamlit-pandasai
+```
+
+Install Python dependencies in the activate Python environment:
+
+```sh
+pip install -r requirements.txt
+```
+
+Create a [new API key](https://platform.openai.com/account/api-keys) and set it to the `OPENAI_API_KEY` environment variable beforehand.
+
+On Windows:
+
+```bash
+set OPENAI_API_KEY="sk-..."
+```
+
+On Unix:
+
+```sh
+export OPENAI_API_KEY="sk-..."
+```
+
+Run the Streamlit project:
+
+```sh
+streamlit run streamlit_app.py
+```

data.py

@@ -0,0 +1,18 @@
+import pickle
+from pathlib import Path
+
+import pandas as pd
+import streamlit as st
+
+
+def load_file(path: str) -> pd.DataFrame:
+    with open(path, "rb") as f:
+        dataset = pickle.load(f)
+        return dataset
+
+
+@st.cache_data
+def load_data(folder: str) -> pd.DataFrame:
+    all_datasets = [load_file(file) for file in Path(folder).iterdir()]
+    df = pd.concat(all_datasets)
+    return df

requirements.txt

@@ -0,0 +1,3 @@
+openai
+pandasai==1.4.10
+streamlit

streamlit_app.py

@@ -0,0 +1,75 @@
+import os
+import streamlit as st
+from pandasai import SmartDataframe
+from pandasai.callbacks import BaseCallback
+from pandasai.llm import OpenAI
+from pandasai.responses.response_parser import ResponseParser
+from data import load_data
+
+# Fungsi untuk mendapatkan daftar folder dalam direktori
+def list_folders(directory):
+    return [name for name in os.listdir(directory) if os.path.isdir(os.path.join(directory, name))]
+
+class StreamlitCallback(BaseCallback):
+    def __init__(self, container) -> None:
+        """Initialize callback handler."""
+        self.container = container
+
+    def on_code(self, response: str):
+        self.container.code(response)
+
+class StreamlitResponse(ResponseParser):
+    def __init__(self, context) -> None:
+        super().__init__(context)
+
+    def format_dataframe(self, result):
+        st.dataframe(result["value"])
+        return
+
+    def format_plot(self, result):
+        st.image(result["value"])
+        return
+
+    def format_other(self, result):
+        st.write(result["value"])
+        return
+
+# Tampilkan judul aplikasi
+st.write("# Chat with Tabular Dataframe 🦙")
+
+# Menentukan direktori asal
+root_dir = "./data"
+
+# Mendapatkan daftar folder
+folders = list_folders(root_dir)
+
+# Menampilkan kotak pemilihan folder
+selected_folder = st.selectbox("Pilih folder data", folders)
+
+# Memuat data berdasarkan folder yang dipilih
+if selected_folder:
+    df = load_data(os.path.join(root_dir, selected_folder))
+
+    # Menampilkan preview data
+    with st.expander("🔎 Dataframe Preview"):
+        st.write(df.tail(3))
+
+    # Membuat input untuk query
+    query = st.text_area("🗣️ Chat with Dataframe")
+    container = st.container()
+
+    # Eksekusi query jika ada input
+    if query:
+        llm = OpenAI(api_token=os.environ["BIT_OPENAI_API_KEY"])
+        query_engine = SmartDataframe(
+            df,
+            config={
+                "llm": llm,
+                "response_parser": StreamlitResponse,
+                # "callback": StreamlitCallback(container),
+            },
+        )
+
+        answer = query_engine.chat(query)
+else:
+    st.write("Silakan pilih folder untuk memuat data.")