john-adeojo

query = "What's the best day to visit the Eiffel Tower for the best view of the city?"
query_1 = f"{query} Return an explanation alongside your response, and a description of the relevant weather conditions"

john-adeojo

from sqlalchemy import create_engine
import openai
import pandas as pd
from langchain import OpenAI, SQLDatabase, SQLDatabaseChain

# load data in memeory to sql lite database
def load_data(df):
  engine = create_engine('sqlite:///:memory:')

  # Write the data to the SQLite database

john-adeojo

<OpenAIObject chat.completion id=chatcmpl-7byJmyUNMNeYIIQhrBQ295u4JqM4C at 0x1f654544680> JSON: {
  "id": "chatcmpl-7byJmyUNMNeYIIQhrBQ295u4JqM4C",
  "object": "chat.completion",
  "created": 1689284354,
  "model": "gpt-4-0613",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",

john-adeojo

import openai
import pandas as pd
import json

def parse_location(query):
    
    # OpenAI function calling
    
    function_call = [
    {

john-adeojo

from geopy.geocoders import Nominatim
import pandas as pd
import requests
import uuid

# gets latitude and longitude
def get_lat_long(location):
    geolocator = Nominatim(user_agent="my_application_" + str(uuid.uuid4()))
    location = geolocator.geocode(location)
    return (location.latitude, location.longitude)

john-adeojo

import pandas as pd
import openai
from langchain import OpenAI, SQLDatabase, SQLDatabaseChain

# Please replace "C:/path/to/my_database.db" with the absolute path to your database
db = SQLDatabase.from_uri("sqlite:///C:/path/to/my_database.db")
openai_token = YOUR_API_KEY # Replace with your API secret key from openai

llm = OpenAI(temperature=0, verbose=True, openai_api_key=openai_token)
db_chain = SQLDatabaseChain.from_llm(llm, db, verbose=True, use_query_checker=True)

john-adeojo

conda env create -f environment.yml

john-adeojo

data_location: YOUR_DATA_STORE
model_yaml: "https://raw.githubusercontent.com/john-adeojo/Credit-Card-Fraud-Model-Registry/main/model%20yaml%20files/model_1a.yaml"
output_dir: "../data/06_models"
model_options:
  seed: 42
  test_size: 0.2

john-adeojo

"""
This is a boilerplate pipeline 'train_model'
generated using Kedro 0.18.10
"""
from kedro.pipeline import Pipeline, node, pipeline
from .nodes import read_data, split_data, run_experiment, run_predictions, model_training_diagnostics 

def create_pipeline(**kwargs) -> Pipeline:
    return pipeline(
        [

john-adeojo

import pandas as pd
from sklearn.model_selection import train_test_split
# declare the other imports

# Helper functions
def get_latest_experiment_dir(base_dir):
    # Logic to fetch latest directory

def delete_file():
    # Logic to delete files