Skip to content

Commit

Permalink
feat: Word count for edge-ai Topic
Browse files Browse the repository at this point in the history
  • Loading branch information
jermnelson committed Sep 7, 2024
1 parent 4509580 commit 68adf3c
Show file tree
Hide file tree
Showing 3 changed files with 58 additions and 11 deletions.
2 changes: 1 addition & 1 deletion checklist.md
Original file line number Diff line number Diff line change
Expand Up @@ -122,7 +122,7 @@
- [ ] Technical Support for Orphaned Modules
- [ ] Next Steps for Adopting AI and ML in FOLIO
- [ ] edge-ai
- [ ] 250 words
- [x] 250 words
- [ ] LLMs copyedit
- [ ] References copied into resources
- [ ] Conclusion
Expand Down
32 changes: 26 additions & 6 deletions next-steps-for-adopting-ai-and-ml-in-folio/edge-ai.html
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>AI Workflows for FOLIO - Edge AI Module</title>
<title>AI Workflows for FOLIO - Edge AI Module and AI Workflows</title>
<link href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" rel="stylesheet" integrity="sha384-QWTKZyjpPEjISv5WaRU9OFeRpok6YctnYmDr5pNlyT2bRjXh0JMhjY6hW+ALEwIH" crossorigin="anonymous">
<style>

Expand All @@ -21,27 +21,47 @@ <h3>23 September 2024</h3>

<div class="row">
<article class="col-9">
<h1>Edge AI Module</h1>
<p>The FOLIO <a href="https://github.com/foliolabs/edge-ai">Edge AI Module</a> provides a set of
<h1>Edge AI Module and AI Workflows</h1>
<p>The FOLIO <a href="https://github.com/folio-labs/edge-ai">Edge AI Module</a> provides a set of
API endpoints for different FOLIO modules. This backend model supports the following
Generative AI services:</p>
<ul>
<li>OpenAI's ChatGPT (requires an API token)</li>
<li>Anthropic Claude (requires an API token)</li>
<li>Google Gemini (requires an API token)</li>
<li>Locally hosted Llama model (through a hosted <a href="">GPT4ALL</a> instance)</li>
<li>Locally hosted Llama model (through a hosted <a href="">GPT4ALL</a> or <a href="https://github.com/ggerganov/llama.cpp">LLaMA.cpp</a> instance)</li>
</ul>
<h2>Existing API endpoints</h2>
<h3>Inventory</h3>
<p>Most of the development so far on <a href="https://github.com/folio-labs/edge-ai">edge-ai</a> has been supporting the requirements from the
<a href="https://github.com/folio-labs/ai-workflows/wiki/Automated-Metadata-Generation-Enrichment">Automated Metadata Generation Enrichment</a>
use-case. </p>
<h3>Circulation</h3>
<h3>Finance</h3>
<h2>AI Workflows</h2>
<p>The <a href="https://github.com/folio-labs/ai-workflows">ai-workflows</a> is a proof-of-concept open-source repository that </p>
<h3>What about mod-workflow?</h3>
<p>There is nothing that would prevent <a href="https://github.com/folio-labs/edge-ai">edge-ai</a> to use the new [mod-workflows][MOD_WRKFLW]
module in FOLIO once it is available for use and release. The current Directed Acyclic Graphs (DAGs) in
<a href="https://github.com/folio-labs/ai-workflows">ai-workflows</a> are simple enough that could be ported to [mod-workflows][MOD_WRKFLW] directly.</p>
<p>Where the <a href="https://github.com/folio-labs/ai-workflows">ai-workflows</a> is positioned to support more complex DAGs and other features beyond
<a href="https://github.com/folio-labs/edge-ai">edge-ai</a> and the planned functionality of [mod-workflows][MOD_WRKFLW] and should be able to
compliment and support both. </p>
<h2>Connection to FOLIO AI Use-cases</h2>
<p>From the edge-ai's use-cases hosted in the repository's wiki, a set of requirements are
being formulated (thank-you for help!) and will help prioritize the development of edge-ai.</p>
being formulated (thank-you for help!) and will help prioritize the development of <a href="https://github.com/folio-labs/edge-ai">edge-ai</a>.</p>
<h2>TODO</h2>
<ul>
<li>Automate the deployment of the <a href="https://github.com/folio-labs/edge-ai">edge-ai</a> to a FOLIO environment API service,
either <a href="https://github.com/folio-org/okapi">Okapi</a> or <a href="https://konghq.com/">Kong</a></li>
<li>Increase Unit test coverage</li>
<li>Add Integration Tests</li>
</ul>
<p>[MOD_WRKFLW]: </p>
</article>
<div class="col-3">
<h4>Navigation</h4>
<ol><li><a href="/wolfcon-2024-ai-workshop/intro-pre-conference/index.html">Introduction to WOLFcon AI Pre-conference Workshop</a></li><li><a href="/wolfcon-2024-ai-workshop/types-of-ai-ml-relevant-to-libraries/index.html">General Overview of AI and Machine Learning for Libraries</a></li><li><a href="/wolfcon-2024-ai-workshop/ethical-considerations-ai-ml-for-libraries/index.html">Ethical Considerations on using AI and Machine Learning for Libraries</a></li><li><a href="/wolfcon-2024-ai-workshop/exploring-llms/index.html">Exploring Large Language Models</a></li><li><a href="/wolfcon-2024-ai-workshop/folio-ai-use-cases/index.html">FOLIO AI Use Cases</a></li><li><a href="/wolfcon-2024-ai-workshop/next-steps-for-adopting-ai-and-ml-in-folio/index.html">Next Steps for Adopting AI and Machine Learning in FOLIO</a><ul><li><strong>Edge AI Module</strong></li><li><a href="conclusion.html">Concluding Discussion</a></li></ul></li><li><a href="/wolfcon-2024-ai-workshop/recommended-resources-for-further-learning/index.html">Recommended Resources for Further Learning</a></li></ol>
<ol><li><a href="/wolfcon-2024-ai-workshop/intro-pre-conference/index.html">Introduction to WOLFcon AI Pre-conference Workshop</a></li><li><a href="/wolfcon-2024-ai-workshop/types-of-ai-ml-relevant-to-libraries/index.html">General Overview of AI and Machine Learning for Libraries</a></li><li><a href="/wolfcon-2024-ai-workshop/ethical-considerations-ai-ml-for-libraries/index.html">Ethical Considerations on using AI and Machine Learning for Libraries</a></li><li><a href="/wolfcon-2024-ai-workshop/exploring-llms/index.html">Exploring Large Language Models</a></li><li><a href="/wolfcon-2024-ai-workshop/folio-ai-use-cases/index.html">FOLIO AI Use Cases</a></li><li><a href="/wolfcon-2024-ai-workshop/next-steps-for-adopting-ai-and-ml-in-folio/index.html">Next Steps for Adopting AI and Machine Learning in FOLIO</a><ul><li><a href="edge-ai.html">Edge AI Module</a></li><li><a href="conclusion.html">Concluding Discussion</a></li></ul></li><li><a href="/wolfcon-2024-ai-workshop/recommended-resources-for-further-learning/index.html">Recommended Resources for Further Learning</a></li></ol>
</div>
</div>

Expand Down
35 changes: 31 additions & 4 deletions next-steps-for-adopting-ai-and-ml-in-folio/edge-ai.md
Original file line number Diff line number Diff line change
@@ -1,24 +1,51 @@
# Edge AI Module
# Edge AI Module and AI Workflows

The FOLIO [Edge AI Module](https://github.com/foliolabs/edge-ai) provides a set of
The FOLIO [Edge AI Module][EDGE_AI] provides a set of
API endpoints for different FOLIO modules. This backend model supports the following
Generative AI services:

- OpenAI's ChatGPT (requires an API token)
- Anthropic Claude (requires an API token)
- Google Gemini (requires an API token)
- Locally hosted Llama model (through a hosted [GPT4ALL]() instance)
- Locally hosted Llama model (through a hosted [GPT4ALL]() or [LLaMA.cpp](https://github.com/ggerganov/llama.cpp) instance)

## Existing API endpoints

### Inventory
Most of the development so far on [edge-ai][EDGE_AI] has been supporting the requirements from the
[Automated Metadata Generation Enrichment](https://github.com/folio-labs/ai-workflows/wiki/Automated-Metadata-Generation-Enrichment)
use-case.

### Circulation

### Finance

## AI Workflows
The [ai-workflows][AI_WRKFLW] is a proof-of-concept open-source repository that

### What about mod-workflow?
There is nothing that would prevent [edge-ai][EDGE_AI] to use the new [mod-workflows][MOD_WRKFLW]
module in FOLIO once it is available for use and release. The current Directed Acyclic Graphs (DAGs) in
[ai-workflows][AI_WRKFLW] are simple enough that could be ported to [mod-workflows][MOD_WRKFLW] directly.

Where the [ai-workflows][AI_WRKFLW] is positioned to support more complex DAGs and other features beyond
[edge-ai][EDGE_AI] and the planned functionality of [mod-workflows][MOD_WRKFLW] and should be able to
compliment and support both.

## Connection to FOLIO AI Use-cases
From the edge-ai's use-cases hosted in the repository's wiki, a set of requirements are
being formulated (thank-you for help!) and will help prioritize the development of edge-ai.
being formulated (thank-you for help!) and will help prioritize the development of [edge-ai][EDGE_AI].



## TODO
- Automate the deployment of the [edge-ai][EDGE_AI] to a FOLIO environment API service,
either [Okapi][OKAPI] or [Kong][KONG]
- Increase Unit test coverage
- Add Integration Tests

[AI_WRKFLW]: https://github.com/folio-labs/ai-workflows
[EDGE_AI]: https://github.com/folio-labs/edge-ai
[KONG]: https://konghq.com/
[MOD_WRKFLW]:
[OKAPI]: https://github.com/folio-org/okapi

0 comments on commit 68adf3c

Please sign in to comment.