Commit Graph

48 Commits

Author SHA1 Message Date
extio-de
6e7db0c6d8 Reworked instruction generation tools; 2025-01-27 20:17:32 +01:00
extio-de
8be692c237 Fix Execution; Removed dep to commons-io; 2025-01-25 18:10:09 +01:00
extio-de
77637436f2 MergeJsonl: Separate outputs per filter 2025-01-25 17:36:05 +01:00
extio-de
36ba17ef23 GenConversation: Provide history in the prompt; 2025-01-12 11:36:53 +01:00
extio-de
538e92ca0a Randomize character names only once per conversation; 2025-01-12 10:57:37 +01:00
extio-de
dad9915870 Readme 2025-01-11 23:20:53 +01:00
extio-de
2628cce48e mergeJsonl.filter; 2025-01-11 22:00:04 +01:00
extio-de
282b5ffb7f New tool: Contextual conversation; 2025-01-11 21:59:52 +01:00
extio-de
c76b108c19 AbstractContextualPrompts minor refactoring; 2024-12-30 11:58:38 +01:00
extio-de
ad001dae84 Misc improvements; 2024-12-29 12:05:15 +01:00
extio-de
4d320fa141 Output to multiple streams; 2024-12-27 12:27:10 +01:00
extio-de
436e996e5d Rewrite: Use different prompts for first and next continuations; 2024-12-27 12:25:59 +01:00
extio-de
674c47dfe1 Rewrite: Add previous split as history for better context; 2024-12-27 11:17:12 +01:00
extio-de
685c740c5c New tool: Create; New tool: Contextual instructions; Rewrite: Configure cnt for both impr and enh; Task Execution: Reimplemented task queuing approach; Fixes; 2024-12-26 21:49:51 +01:00
extio-de
07e5a4779a Update readme 2024-12-20 01:21:32 +01:00
extio-de
dc5a4f3e25 Improve contextual prompt instruction; 2024-12-20 00:43:16 +01:00
extio-de
d44fd53323 Configurable model categories; 2024-12-20 00:14:02 +01:00
extio-de
74eaa5c72b Tool pipelines; 2024-12-19 19:20:45 +01:00
extio-de
bfb4e8be3b Text2Jsonl2HistoryInstructCompletion: Improved llm prompt; Utils: normalizeModelResponse(), streamOut(); 2024-12-19 14:04:01 +01:00
extio-de
7352f28b3c Readme updated; 2024-12-18 20:58:59 +01:00
extio-de
56b7dcb265 MergeJsonl tools accepts multiple directories; 2024-12-18 20:48:58 +01:00
extio-de
1c94b97b92 DatasetTool marker interface; 2024-12-18 20:42:12 +01:00
extio-de
54ebdb2e51 New tool: Text2Jsonl2HistoryInstructCompletion; 2024-12-18 20:07:07 +01:00
extio-de
75c2109d6d Switch dataset preparer to lmlib; 2024-12-18 20:05:03 +01:00
extio-de
fb34561bd5 Client: Collect and show more statistics; MergeJsonl: Shuffle merged json; Rewrite: Fix progress indicator; 2024-12-10 18:11:48 +01:00
extio-de
6648ce43ef Updated description; 2024-12-02 23:15:44 +01:00
extio-de
f0872fb7ee createCharacterNameMapping: Changed log level; 2024-12-02 11:36:03 +01:00
extio-de
57f7ec2e48 Text2Jsonl2WithContextualPrompts: Prompt improved; Name mapping requires at least 5 characters (otherwise we may replace random syllables); 2024-12-01 22:50:53 +01:00
extio-de
b027c2a892 Text2Jsonl2WithContextualPrompts: Rename persons; Prompt conversation contexts; 2024-12-01 18:32:36 +01:00
extio-de
dda9400cc1 Text2Jsonl2WithContextualPrompts: Overlap chunks; 2024-11-16 10:54:06 +01:00
extio-de
70822207dd splitParagraphs: Split by the closest occurence to chunk_norm instead of chunk_min; 2024-11-16 10:52:28 +01:00
extio-de
82f969d666 Fix paragraphToContextualPrompt response post-processing; 2024-11-03 22:53:35 +01:00
extio-de
cb8ac57580 Improved prompt for Text2Jsonl2WithContextualPrompts; 2024-11-03 20:37:22 +01:00
extio-de
509f36238d Fix MixtralPromptStrategy; Refactoring; 2024-11-02 12:24:28 +01:00
extio-de
1813ab8928 Client refactoring; MergeJsonl tool; 2024-11-01 19:41:03 +01:00
extio-de
94a20bd4ed Merge pull request #1 from stephan-birkl/patch-1
Update Client.java
2024-11-01 19:18:35 +01:00
stephan-birkl
3b6a0ee74a Update Client.java 2024-11-01 17:43:40 +01:00
extio-de
e40e648557 splitParagraphs 2024-11-01 15:48:28 +01:00
extio-de
14bd1c1915 splitParagraphs 2024-11-01 15:31:24 +01:00
extio-de
65e50e1dcc Sliding contexts for contextual prompts; 2024-11-01 15:08:26 +01:00
extio-de
7232387134 Switched remaining tools to streaming output;
Generate contextual prompt datasets as single files per input;
Added Alpaca prompt template;
Generate multiple contextual prompt lengths;
2024-11-01 13:43:59 +01:00
extio-de
2087405c64 Readme; 2024-10-30 00:22:54 +01:00
extio-de
dca4aa0a29 Refactoring; Readme; 2024-10-30 00:21:18 +01:00
extio-de
efb520fc07 New generator: Jsonl QA style contextual prompts;
JsonL: Add sliding windows;
Streamed jsonl output;
Memory optimization: Buffered, blocking task creation;
2024-10-29 23:51:41 +01:00
extio-de
4b93ed97be Fix splitParagraphs; Improved prompts; 2024-10-29 21:00:23 +01:00
extio-de
030bc5bee1 Improved rewriting; 2024-10-27 09:54:34 +01:00
extio-de
78a814a64e Readme 2024-10-20 22:01:19 +02:00
extio-de
59dc27a8ef initial commit 2024-10-20 21:04:32 +02:00