2024:Fine-Tuning for AI Extract - Revision history

Dgreenwood at 19:47, 30 July 2025

2025-07-30T19:47:23Z

← Older revision		Revision as of 14:47, 30 July 2025
Line 61:		Line 61:
	=== Verify Prerequisites ===		=== Verify Prerequisites ===
	Before fine tuning you will need a few things in place.		Before fine tuning you will need a few things in place.
	* An established '''''[[~~LLM Connector (Repository Option)\|~~LLM Connector]]'''''.		* An established '''''[[LLM Connector]]'''''.
	* A fleshed out '''Data Model''' with all appropriate '''[[Data Element ~~(Concept)\|Data Elements~~]]'''.		* A fleshed out '''Data Model''' with all appropriate '''[[Data Element]]s'''.
	* A configured '''''Fill Method''''' on the '''Data Model'''.		* A configured '''''Fill Method''''' on the '''Data Model'''.
	* A '''[[Batch]]''' with at least ten (10) documents.		* A '''[[Batch]]''' with at least ten (10) documents.
Line 98:		Line 98:

	<big>Verify Batch with Classified, Extracted Documents</big>		<big>Verify Batch with Classified, Extracted Documents</big>
	# <li value=9>A '''Batch''' with ten documents is present. The documents are classified as a '''[[~~Content Type (Concept)\|~~Content Type]]''' using the '''Data Model''' configured with the appropriate '''''Fill Method'''''.</li>		# <li value=9>A '''Batch''' with ten documents is present. The documents are classified as a '''[[Content Type]]''' using the '''Data Model''' configured with the appropriate '''''Fill Method'''''.</li>
	# On the "Advanced" tab of one of the classified document, you can see there is a "DocumentData.json" file informing that extraction has been performed. This is true for each document.		# On the "Advanced" tab of one of the classified document, you can see there is a "DocumentData.json" file informing that extraction has been performed. This is true for each document.

Dgreenwood: Dgreenwood moved page 2024:Fine Tuning (LLM Construct) to 2024:Fine-Tuning for AI Extract

2025-07-25T14:14:59Z

Dgreenwood moved page 2024:Fine Tuning (LLM Construct) to 2024:Fine-Tuning for AI Extract

← Older revision	Revision as of 09:14, 25 July 2025
(No difference)

Dgreenwood at 15:40, 25 April 2025

2025-04-25T15:40:13Z

Dgreenwood: Dgreenwood moved page Fine Tuning (LLM Construct) to 2024:Fine Tuning (LLM Construct)

2025-03-04T21:04:29Z

Dgreenwood moved page Fine Tuning (LLM Construct) to 2024:Fine Tuning (LLM Construct)

← Older revision	Revision as of 16:04, 4 March 2025
(No difference)

Dgreenwood at 21:32, 28 February 2025

2025-02-28T21:32:35Z

← Older revision		Revision as of 16:32, 28 February 2025
Line 92:		Line 92:
	# <li value=6>On the '''Data Model'''...</li>		# <li value=6>On the '''Data Model'''...</li>
	# ...in the '''''Fill Methods''''' property...		# ...in the '''''Fill Methods''''' property...
	# ...an '''[[~~AI Extract (Fill Method)\|~~AI Extract]]''' '''''Fill Method''''' is established and configured.		# ...an '''[[AI Extract]]''' '''''Fill Method''''' is established and configured.

	[[image: 2024_Fine-Tuning_02_03.png]]		[[image: 2024_Fine-Tuning_02_03.png]]

Randallkinard: // via Wikitext Extension for VSCode

2024-10-01T17:21:10Z

// via Wikitext Extension for VSCode

← Older revision		Revision as of 12:21, 1 October 2024
Line 142:		Line 142:

	[[image: 2024_Fine-Tuning_02_08.png]]		[[image: 2024_Fine-Tuning_02_08.png]]


			# <li value=15>On the "Fine Tuning" section of your OpenAI platform...</li>
			# ...you will see a newly created Fine Tuned model.

			[[image: 2024_Fine-Tuning_02_09.png]]
	</div>		</div>

Randallkinard: corrected batch error // via Wikitext Extension for VSCode

2024-10-01T16:44:43Z

corrected batch error // via Wikitext Extension for VSCode

← Older revision		Revision as of 11:44, 1 October 2024
Line 97:		Line 97:


	<big>Verify Batch with Classified, Extracted ~~Document(s)~~</big>		<big>Verify Batch with Classified, Extracted Documents</big>
	# <li value=9>A '''Batch''' with ~~a document~~ is present. The ~~document is~~ classified as a '''[[Content Type (Concept)\|Content Type]]''' using the '''Data Model''' configured with the appropriate '''''Fill Method'''''.</li>		# <li value=9>A '''Batch''' with ten documents is present. The documents are classified as a '''[[Content Type (Concept)\|Content Type]]''' using the '''Data Model''' configured with the appropriate '''''Fill Method'''''.</li>
	# On the "Advanced" tab of the classified document, you can see there is a "DocumentData.json" file informing that extraction has been performed.		# On the "Advanced" tab of one of the classified document, you can see there is a "DocumentData.json" file informing that extraction has been performed. This is true for each document.

	[[image: 2024_Fine-Tuning_02_04.png]]		[[image: 2024_Fine-Tuning_02_04.png]]

Dgreenwood: /* Verify Prerequisites */

2024-09-30T21:29:43Z

Verify Prerequisites

← Older revision		Revision as of 16:29, 30 September 2024
Line 64:		Line 64:
	* A fleshed out '''Data Model''' with all appropriate '''[[Data Element (Concept)\|Data Elements]]'''.		* A fleshed out '''Data Model''' with all appropriate '''[[Data Element (Concept)\|Data Elements]]'''.
	* A configured '''''Fill Method''''' on the '''Data Model'''.		* A configured '''''Fill Method''''' on the '''Data Model'''.
	* A '''[[Batch (Object)\|Batch]]''' with at least ten documents.		* A '''[[Batch (Object)\|Batch]]''' with at least ten (10) documents.
	** The documents in the '''[[Batch (Object)\|Batch]]''' should be classified.		** The documents in the '''[[Batch (Object)\|Batch]]''' should be classified.
	** The documents in the '''Batch''' should have extracted data. Verify that the extracted data is correct.		** The documents in the '''Batch''' should have extracted data. Verify that the extracted data is correct.

Dgreenwood: /* Verify Prerequisites */

2024-09-30T21:29:31Z

Verify Prerequisites

← Older revision		Revision as of 16:29, 30 September 2024
Line 64:		Line 64:
	* A fleshed out '''Data Model''' with all appropriate '''[[Data Element (Concept)\|Data Elements]]'''.		* A fleshed out '''Data Model''' with all appropriate '''[[Data Element (Concept)\|Data Elements]]'''.
	* A configured '''''Fill Method''''' on the '''Data Model'''.		* A configured '''''Fill Method''''' on the '''Data Model'''.
	* A '''[[Batch (Object)\|Batch]]''' with at least ~~one document~~.		* A '''[[Batch (Object)\|Batch]]''' with at least ten documents.
	** The ~~document(s)~~ in the '''[[Batch (Object)\|Batch]]''' should be classified.		** The documents in the '''[[Batch (Object)\|Batch]]''' should be classified.
	** The ~~document(s)~~ in the '''Batch''' should have extracted data. Verify that the extracted data is correct.		** The documents in the '''Batch''' should have extracted data. Verify that the extracted data is correct.

	The '''Batch''' and '''[[Project (Object)\|Project]]''' associated with this article will suffice for these purposes. You will, however, need to supply your own API key for your '''''LLM Connector'''''.		The '''Batch''' and '''[[Project (Object)\|Project]]''' associated with this article will suffice for these purposes. You will, however, need to supply your own API key for your '''''LLM Connector'''''.

Randallkinard: added "Perform Fine Tuning" section // via Wikitext Extension for VSCode

2024-09-27T22:15:19Z

added "Perform Fine Tuning" section // via Wikitext Extension for VSCode

@@ Line 65: / Line 65: @@
 * A configured '''''Fill Method''''' on the '''Data Model'''.
 * A '''[[Batch (Object)|Batch]]''' with at least one document.
-** The document(s) in the '''Batch''' should be classified.
+** The document(s) in the '''[[Batch (Object)|Batch]]''' should be classified.
 ** The document(s) in the '''Batch''' should have extracted data. Verify that the extracted data is correct.
@@ Line 93: / Line 93: @@
 # ...in the '''''Fill Methods''''' property...
 # ...an '''[[AI Extract (Fill Method)|AI Extract]]''' '''''Fill Method''''' is established and configured.
 [[image: 2024_Fine-Tuning_02_03.png]]
@@ Line 117: / Line 107: @@
 [[image: 2024_Fine-Tuning_02_05.png]]
 === Perform Fine Tuning ===
 </div>

← Older revision		Revision as of 10:40, 25 April 2025
Line 27:		Line 27:

	=== Example Use Cases ===		=== Example Use Cases ===
	* Fine-tuning based on a '''[[~~Data Model (Object)\|~~Data Model]]''' in '''Grooper''' to facilitate better extraction.		* Fine-tuning based on a '''[[Data Model]]''' in '''Grooper''' to facilitate better extraction.
	* Training a model on legal documents to generate contracts or provide legal advice.		* Training a model on legal documents to generate contracts or provide legal advice.
	* Customizing a chatbot to respond in the tone and style preferred by a company's brand.		* Customizing a chatbot to respond in the tone and style preferred by a company's brand.
Line 64:		Line 64:
	* A fleshed out '''Data Model''' with all appropriate '''[[Data Element (Concept)\|Data Elements]]'''.		* A fleshed out '''Data Model''' with all appropriate '''[[Data Element (Concept)\|Data Elements]]'''.
	* A configured '''''Fill Method''''' on the '''Data Model'''.		* A configured '''''Fill Method''''' on the '''Data Model'''.
	* A '''[[~~Batch (Object)\|~~Batch]]''' with at least ten (10) documents.		* A '''[[Batch]]''' with at least ten (10) documents.
	** The documents in the '''[[~~Batch (Object)\|~~Batch]]''' should be classified.		** The documents in the '''[[Batch]]''' should be classified.
	** The documents in the '''Batch''' should have extracted data. Verify that the extracted data is correct.		** The documents in the '''Batch''' should have extracted data. Verify that the extracted data is correct.

	The '''Batch''' and '''[[~~Project (Object)\|~~Project]]''' associated with this article will suffice for these purposes. You will, however, need to supply your own API key for your '''''LLM Connector'''''.		The '''Batch''' and '''[[Project]]''' associated with this article will suffice for these purposes. You will, however, need to supply your own API key for your '''''LLM Connector'''''.


Line 104:		Line 104:


	# <li value=11>Using a '''[[~~Batch Process Step (Object)\|~~Batch Process Step]]''' with a '''[[~~Review (Activity)\|~~Review]]''' activity, established with a Data View, you can verify that the extracted data is accurately displaying what is on the document.</li>		# <li value=11>Using a '''[[Batch Process Step]]''' with a '''[[Review]]''' activity, established with a Data View, you can verify that the extracted data is accurately displaying what is on the document.</li>

	[[image: 2024_Fine-Tuning_02_05.png]]		[[image: 2024_Fine-Tuning_02_05.png]]