-
Notifications
You must be signed in to change notification settings - Fork 27
/
README.Rmd
141 lines (101 loc) · 8.72 KB
/
README.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
---
output: github_document
editor_options:
markdown:
wrap: 72
---
<!-- README.md is generated from README.Rmd. Please edit that file -->
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-",
out.width = "100%"
)
```
# gpttools <a href="https://jameshwade.github.io/gpttools/"><img src="man/figures/logo.png" align="right" height="125"/></a>
<!-- badges: start -->
[](https://www.repostatus.org/#wip)
[](https://CRAN.R-project.org/package=gpttools)
[](https://jameshwade.r-universe.dev)
[](https://app.codecov.io/gh/JamesHWade/gpttools?branch=main)
[](https://lifecycle.r-lib.org/articles/stages.html#experimental)
[](https://github.com/JamesHWade/gpttools/actions/workflows/R-CMD-check.yaml)
[](https://github.com/pre-commit/pre-commit)
<!-- badges: end -->
The goal of gpttools is to extend gptstudio for R package developers to
more easily incorporate use of large language models (LLMs) into their
project workflows. These models appear to be a step change in our use of
text for knowledge work, but you should carefully consider ethical
implications of using these models. Ethics of LLMs (also called
[Foundation Models](https://arxiv.org/abs/2108.07258)) is an area of
very active discussion.
## Installation
### Install from GitHub with `{pak}`
``` r
# install.packages("pak")
pak::pak("JamesHWade/gpttools")
```
### Install from [R-Universe](https://r-universe.dev/)
``` r
# Enable repository from jameshwade
options(repos = c(
jameshwade = "https://jameshwade.r-universe.dev",
CRAN = "https://cloud.r-project.org"
))
# Download and install gpttools in R
install.packages("gpttools")
# Browse the gpttools manual pages
help(package = "gpttools")
```
### Available AI Services and Models
| AI Service | Models | Documentation | Setup |
| ---- | ---- | ---- | ---- |
| [OpenAI](https://platform.openai.com) | gpt-4-turbo, gpt-4, gpt-3.5-turbo ([latest models](https://platform.openai.com/docs/models)) | [OpenAI API Docs](https://platform.openai.com/docs/api-reference) | [OpenAI Setup](articles/openai.md) |
| [HuggingFace](https://huggingface.co/) | various | [HF Inference API Docs](https://huggingface.co/docs/hub/models-inference) | [HF Setup](articles/huggingface.md) |
| [Anthropic](https://docs.anthropic.com/claude/docs/guide-to-anthropics-prompt-engineering-resources) | claude-2.1, claude-instant-1.2 | [Anthropic API Docs](https://docs.anthropic.com/claude/reference/getting-started-with-the-api) | [Anthropic Setup](articles/anthropic.md) |
| [Ollama](https://ollama.ai/) | mistral, llama2, mixtral, phi ([latest models](https://ollama.ai/library)) | [Ollama API Docs](https://github.com/jmorganca/ollama/blob/main/docs/api.md) | [Ollama Setup](articles/ollama.md) |
| [Perplexity](https://www.perplexity.ai) | pplx-7b-chat, pplx-70b-chat, pplx-7b-online, pplx-70b-online, llama-2-70b-chat, codellama-34b-instruct, mistral-7b-instruct, and mixtral-8x7b-instruct | [Perplexity API Docs](https://docs.perplexity.ai/reference/post_chat_completions) | [Perplexity Setup](articles/perplexity.md) |
| [Google AI Studio](https://ai.google.dev/tutorials/ai-studio_quickstart) | Gemini and Palm (legacy) | [Google AI Studio Docs](https://ai.google.dev/docs) | [Google AI Studio Setup](articles/googleai.md) |
| [Azure OpenAI](https://learn.microsoft.com/en-us/azure/ai-services/openai/overview) | gpt-4, gpt-3.5-turbo ([latest models](https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models#gpt-4-and-gpt-4-turbo-preview)) | [Azure OpenAI API Docs](https://learn.microsoft.com/en-us/azure/ai-services/openai/quickstart?tabs=command-line,python&pivots=rest-api) | [Azure OpenAI Setup](articles/azure.md) |
### Default AI Service: OpenAI
To get started, you must first set up an API service. The package is configured to work with several AI service providers, allowing for flexibility and choice based on your specific needs. The default configuration is set to use OpenAI's services. To use it you need:
1. Make an OpenAI account. [Sign up here](https://platform.openai.com/).
2. [Create an OpenAI API key](https://platform.openai.com/account/api-keys) to use with the package.
3. Set the API key up in Rstudio. See the section below on configuring the API key.
#### Configuring OpenAI API Key
To interact with the OpenAI API, it's required to have a valid `OPENAI_API_KEY` environment variable. Here are the steps to configure it.
You can establish this environment variable globally by including it in your project's .Renviron file. This approach ensures that the environment variable persists across all sessions as the Shiny app runs in the background.
Here is a set of commands to open the .Renviron file for modification:
```{r}
#| eval: false
require(usethis)
edit_r_environ()
```
For a persistent setting that loads every time you launch this project, add the following line to .Renviron, replacing `"<APIKEY>"` with your actual API key:
``` bash
OPENAI_API_KEY="<APIKEY>"
```
**Caution:** If you're using version control systems like GitHub or GitLab, remember to include .Renviron in your .gitignore file to prevent exposing your API key!
**Important Note:** OpenAI API will not function without valid payment details entered into your OpenAI account. This is a restriction imposed by OpenAI and is unrelated to this package.
### Alternative AI Service Providers
While OpenAI is the default and currently considered one of the most robust options, `gpttools` is also compatible with other AI service providers. These include [Anthropic](articles/anthropic.md), [HuggingFace](articles/huggingface.md), [Google AI Studio](articles/googleai.md), [Azure OpenAI](articles/azure.md), and [Perplexity](articles/perplexity.md). You can select any of these providers based on your preference or specific requirements. You can also run local models with [Ollama](articles/ollama.md). This requires more setup but at the benefit of not sharing your data with any third party.
To use an alternative provider, you will need to obtain the relevant API key or access credentials from the chosen provider and configure them similarly.
## Privacy Notice for gpttools
This privacy notice is applicable to the R package that uses popular language models like gpt-4 turbo and claude-2.1. By using this package, you agree to adhere to the privacy terms and conditions set by the API service.
### Data Sharing with AI Services
When using this R package, any text or code you highlight/select with your cursor, or the prompt you enter within the built-in applications, will be sent to the selected AI service provider (e.g., OpenAI, Anthropic, HuggingFace, Google AI Studio, Azure OpenAI) as part of an API request. This data sharing is governed by the privacy notice, rules, and exceptions that you agreed to with the respective service provider when creating an account.
### Security and Data Usage by AI Service Providers
We cannot guarantee the security of the data you send via the API to any AI service provider, nor can we provide details on how each service processes or uses your data. However, these providers often state that they use prompts and results to enhance their AI models, as outlined in their terms of use. Be sure to review the terms of use of the respective AI service provider directly.
### Limiting Data Sharing
The R package is designed to share only the text or code that you specifically highlight/select or include in a prompt through our built-in applications. No other elements of your R environment will be shared unless you turn those features on. It is your responsibility to ensure that you do not accidentally share sensitive data with any AI service provider.
**IMPORTANT: To maintain the privacy of your data, do not highlight, include in a prompt, or otherwise upload any sensitive data, code, or text that should remain confidential.**
## Code of Conduct
Please note that the gpttools project is released with a [Contributor
Code of
Conduct](https://jameshwade.github.io/gpttools/CODE_OF_CONDUCT.html). By
contributing to this project, you agree to abide by its terms.