initial commit

README.md

@@ -10,4 +10,12 @@ pinned: false
 license: mit
 ---
 
+## Getting Started
+
+Please configure your OpenAI API key as an environment variable:
+```
+export OPENAI_API_KEY="yout-key-here"
+```
+And ensure you have your `credentials.json` in your working directory. More info [here](https://developers.google.com/sheets/api/quickstart/python)
+
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,4 @@
+from transformers.tools.base import launch_gradio_demo
+from sheet_extractor import GoogleSheetExtractorTool
+
+launch_gradio_demo(GoogleSheetExtractorTool)

requirements.txt

@@ -0,0 +1,9 @@
+openai
+ortools
+requests
+gspread
+huggingface_hub
+google-api-python-client
+google-auth-httplib2
+google-auth-oauthlib
+transformers

sheet_extractor.py

@@ -0,0 +1,92 @@
+import math
+import json
+import regex
+import inspect
+import gspread
+import guidance
+from ast import literal_eval
+from transformers import Tool
+from oauth2client.service_account import ServiceAccountCredentials
+from googleapiclient.discovery import build
+from google.oauth2.credentials import Credentials
+
+from google.auth.transport.requests import Request
+from google.oauth2.credentials import Credentials
+from google_auth_oauthlib.flow import InstalledAppFlow
+from googleapiclient.discovery import build
+from googleapiclient.errors import HttpError
+
+from ortools.linear_solver import pywraplp
+
+guidance.llm = guidance.llms.OpenAI("gpt-4")
+
+
+structure_program = guidance(
+'''
+{{#user~}}
+{{description}}
+Help me extract args from the data blob to apply the following algorithm:
+{{code}}
+
+----
+
+{{~#each examples}}
+Data Blob: {{this.input}}
+Result: {{this.output}}
+---
+{{~/each}}
+
+Please help me extract the input values from a given data blob into a JSON.
+Data Blob: {{data_blob}}
+Result: 
+{{~/user}}
+
+{{#assistant~}}
+{{gen 'output'}}
+{{~/assistant}}
+''')
+
+
+class DataExtractorTool(Tool):
+    def __init__(self):
+        super().__init__()
+
+    def fetch_data(self, source_link: str) -> str:
+        """
+        Abstract method to fetch raw data from the source.
+        """
+        raise NotImplementedError("Subclasses must implement this method.")
+
+    def __call__(self, source_link: str) -> dict:
+        raw_data = self.fetch_data(source_link)
+        return raw_data
+
+
+class GoogleSheetExtractorTool(DataExtractorTool):
+    name = "google_sheet_extractor_tool"
+    description = """
+    Tool to extract data from Google Sheets.
+    Input: source_link - str - a url or google sheets id
+    e.x. source_link='https://docs.google.com/spreadsheets/d/SHEETS_ID/'.
+    Output is a string.
+    """
+
+    def fetch_data(self, source_link: str) -> str:
+        # Set up the credentials
+        scope = ["https://www.googleapis.com/auth/spreadsheets.readonly"]
+        creds = Credentials.from_authorized_user_file('token.json', scope)
+        service = build('sheets', 'v4', credentials=creds)
+
+        # Open the spreadsheet and get all values
+        source_link = source_link.rstrip("/")
+        spreadsheet_id = source_link.split('/')[-1]
+        range_name = 'Sheet1!A1:Z1000'
+        sheet = service.spreadsheets()
+        result = sheet.values().get(spreadsheetId=spreadsheet_id,
+                                    range=range_name).execute()
+        data = result.get('values', [])
+        
+        # Convert the data to a string representation
+        raw_data = '\n'.join([','.join(row) for row in data])
+        print(raw_data)
+        return raw_data

tool_config.json

@@ -0,0 +1,6 @@
+{
+  "description": "Tool to extract data from Google Sheets. Input: source_link - str - a url or google sheets id e.x. source_link='https://docs.google.com/spreadsheets/d/SHEETS_ID/'. Output is a string",
+  "name": "google_sheet_extractor_tool",
+  "tool_class": "sheet_extractor.GoogleSheetExtractorTool"
+}
+