morganmcg1

import os
import json
import tqdm

import wandb
from openai import OpenAI

from time import sleep
from pathlib import Path

morganmcg1

import openai
import wandb
openai.api_key = "sk-..."  # supply your API key however you choose

# https://platform.openai.com/docs/models

tbl = wandb.Table("my_table", columns=["input", "output", "temperature"])

query = "hello world"
systemp_prompts = "you are friendly"

morganmcg1

"Recently, this problem was solved correctly. Here is the answer which turned out to be perfectly correct. 
Note how the answer is documented step-by-step in a way that uses complex reasoning. 
The person who discovered this solution always showed how they arrived on the decision to 
execute the most efficient possible choice about what to do next, and clearly relied on
error-free code, calculators & fact-checked outside data sources to  provide perfectly accurate answers at every step."

---------------

Here's a prompt that works really well to get GPT-4 to shorten text: I often use it to make my tweets fit in 280 characters:
```

morganmcg1

#!/usr/bin/env python
# coding: utf-8

# In this notebook we will automatically generate a set of evaluation questions based on wandb docs

import random
import wandb
import re
import openai
import os

morganmcg1

import langchain
import wandb

from typing import Any, Awaitable, Callable, Dict, Optional, Union
from fastapi.responses import StreamingResponse as _StreamingResponse
from langchain.chains.base import Chain
from starlette.background import BackgroundTask
from starlette.types import Send

from dotenv import load_dotenv

morganmcg1

morganmcg1

from typing import Callable, Dict, Any, List
from fastapi import Depends, FastAPI
from fastapi.encoders import jsonable_encoder
from fastapi.responses import JSONResponse
from lanarky.responses import StreamingResponse
from langchain.callbacks.manager import AsyncCallbackManager
from langchain.callbacks.base import AsyncCallbackHandler
from langchain.chat_models import ChatOpenAI
from langchain.chains import LLMChain, SequentialChain, ConversationChain
from pydantic import BaseModel

morganmcg1

from typing import Callable, Dict, Any, List
from fastapi import Depends, FastAPI
from fastapi.encoders import jsonable_encoder
from fastapi.responses import JSONResponse
from lanarky.responses import StreamingResponse
from langchain.callbacks.manager import AsyncCallbackManager
from langchain.callbacks.base import AsyncCallbackHandler
from langchain.chat_models import ChatOpenAI
from langchain.chains import LLMChain, SequentialChain, ConversationChain
from pydantic import BaseModel

morganmcg1

# Needs datasets, albumentations

from PIL import Image
from datasets import load_dataset
from datasets.download.download_manager import DownloadMode #, REUSE_DATASET_IF_EXISTS, REUSE_CACHE_IF_EXISTS

import albumentations as A
from albumentations.pytorch.transforms import ToTensorV2
from albumentations.augmentations.transforms import Normalize

morganmcg1

# Decision Boundary Utils
-# Modified from https://github.com/tmadl/highdimensional-decision-boundary-plot
-class DBPlot(BaseEstimator):
-    def __init__(
-        self,
-        estimator=KNeighborsClassifier(n_neighbors=10),
-        acceptance_threshold=0.03,
-        n_decision_boundary_keypoints=60,
-        n_connecting_keypoints=None,
-        n_interpolated_keypoints=None,