Hyperreal

Timeline
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

50 most recent check-ins

2026-05-18
06:41
Prepare a better demo of the transcript explorer tool with a simplified workflow Leaf check-in: 486b51bbae user: sam_hames tags: trunk
06:40
Expand instructions Leaf check-in: 5c3190b30d user: sam_hames tags: transcripts_demo
06:21
Try, try, again to get the jupyter config right check-in: ae3514cd9a user: sam_hames tags: transcripts_demo
05:59
Set kernel and tweak the jupytext configuration check-in: 635cc853b2 user: sam_hames tags: transcripts_demo
05:32
Actually add the helper module for transcript processing check-in: 1b6091b630 user: sam_hames tags: transcripts_demo
05:19
Tidy up the transcripts notebook to have a simpler interface check-in: f840ae6247 user: sam_hames tags: transcripts_demo
02:42
Move back to py:percent for notebook source, as we don't need the lab interface anymore check-in: f5d8874f42 user: sam_hames tags: transcripts_demo
02:37
Add jupyter and binder config for larger file uploads through the ipywidgets.FileUpload check-in: 882b6021e9 user: sam_hames tags: transcripts_demo
2026-05-12
11:05
Remove github workflows check-in: 26aabd8797 user: sam_hames tags: trunk
2026-05-10
23:21
Rework the transcripts notebook to work from a zip file check-in: 511e677d94 user: sam_hames tags: transcripts_demo
22:32
Rearrange to prepare for using jupyter notebook interface for upload of a zip file check-in: acffd12386 user: sam_hames tags: transcripts_demo
22:29
Move transcripts to their own self contained folder check-in: 6b53a41031 user: sam_hames tags: transcripts_demo
2026-04-29
23:54
Add kernel spec to the pypercent twenty_newsgroups notebook check-in: dda38f6c5c user: sam_hames tags: trunk, release, 0.7.2
23:50
Change license to AGPL check-in: 5fa33c30ec user: sam_hames tags: trunk
23:43
Prepare for release check-in: e3b368e27c user: sam_hames tags: trunk
23:36
Put in a placeholder and working notes for making the newsgroups notebook more useful as an instructional guide check-in: 6915e14b88 user: sam_hames tags: trunk
06:24
Convert notebook to complete pypercent format and fix up link display check-in: 59da00a2ce user: sam_hames tags: trunk
2026-04-21
03:55
Don't display doc totals in the visualisation view, just the matching hits and similarity check-in: cd748f6354 user: sam_hames tags: trunk
03:54
Cleanup the server launcher in the twenty newsgroups notebook to display the right URL check-in: 88926b2c83 user: sam_hames tags: trunk
2026-03-09
01:25
Use OR, AND as labels in web UI, instead of conjunction/disjunction check-in: 50cfd91138 user: sam_hames tags: trunk
2026-03-06
02:49
Expand search bar to allow creating a query piece by piece/by extending it one element at a time check-in: ca315690d7 user: sam_hames tags: trunk
2026-03-05
04:06
Remove dead code in webserver check-in: e91724edc7 user: sam_hames tags: trunk
2026-03-02
04:56
Collapse the edit form into a dropdown and a single button check-in: c79991929b user: sam_hames tags: trunk
01:22
Keep track of location and value count to be able to distinguish available information on a field (eg. to know if phrase search is meaningful) check-in: b897fa7f09 user: sam_hames tags: trunk
01:21
Binder: bake twenty newsgroups data into built image check-in: 92be22d134 user: sam_hames tags: trunk
2026-02-11
00:02
Bugfix: handle zero entries for web display in format_si_magnitude helper check-in: a1187d1801 user: sam_hames tags: trunk
2026-02-04
01:38
Add test file for corpus interface! check-in: 8dc78cfb78 user: sam_hames tags: trunk
2026-01-21
23:56
Back to next development release check-in: 2d39cbe18a user: sam_hames tags: trunk
23:48
Bump version for a bugfix binder related release check-in: 94c4994342 user: sam_hames tags: trunk, release, 0.7.1
23:41
Move the binder specific configuration to the handier binder/ config folder check-in: 8b258407ef user: sam_hames tags: trunk
05:33
Update indexing and display logic for conversations notebooks check-in: 16c3cacb2b user: sam_hames tags: trunk
05:32
Update dependencies to the correct versions for the jupyter notebooks check-in: 7b9a5ecf79 user: sam_hames tags: trunk
05:32
Update defaults for HTML document rendering - don't include the doc_key, as it will only be useful in some cases - better to leave this to customisation on the corpus object check-in: acde84ff2d user: sam_hames tags: trunk
05:31
Bugfix: not using the correct logic to mark fields as range encoded based on indexed parameters check-in: 292c251100 user: sam_hames tags: trunk
03:24
Cleanup and start on the next version check-in: 72078f640b user: sam_hames tags: trunk
03:11
New release at last incorporating the substantial reworking of the interface and workflow check-in: aee8e0b75b user: sam_hames tags: trunk, release, 0.7.0
02:55
Draft notebook for conversation aware indexing of transcripts in word docs check-in: 67bd67da0f user: sam_hames tags: trunk
02:54
Initial experimental setup for binderhub environment check-in: 562b229735 user: sam_hames tags: trunk
02:53
Bugfix - note generating URLs correctly for main nav in jupyter notebook context check-in: 714befd1c6 user: sam_hames tags: trunk
01:44
Tweak docstring for refine_clustering check-in: e9a53d065e user: sam_hames tags: trunk
2026-01-20
05:55
Merge extensive refactoring of the data model and other changes so we're back on trunk based development check-in: 5ddcd4c673 user: sam_hames tags: trunk
04:42
Add missing file :/ Closed-Leaf check-in: c9f2d5fb0e user: sam_hames tags: refactor-all
04:41
Add additional information around construction of indexable docs to more finely encode what is and isn't range encoded check-in: 0984c7ea5a user: sam_hames tags: refactor-all
01:21
Construct a term-frequency ordered index as a template for future impact scores This is tentative and will likely change in the future. check-in: 3647143ac2 user: sam_hames tags: refactor-all
2025-12-09
00:00
Add utility method for refining multiple clusterings at once check-in: 1fb88e7ba3 user: sam_hames tags: refactor-all
2025-11-30
23:31
Refactor clustering algorithm function enable efficient computation of multiple cluster refinements at the same time - this supports things like subdividing a clustering and generate multiple clusterings from different randomised starting points easily. check-in: 7ec95a80ea user: sam_hames tags: refactor-all
09:21
Plot cross posting groups first rather than last check-in: 5490579d3c user: sam_hames tags: refactor-all
2025-11-27
22:10
Also plot top cross posted groups in the main figure check-in: e3b52bfed4 user: sam_hames tags: refactor-all
02:44
Track logarithm base 2 term frequency for positional feature occurences check-in: 2f9443860a user: sam_hames tags: refactor-all
2025-11-24
22:31
Make a wide rather than long results table for twenty newsgroups check-in: f5bff605b3 user: sam_hames tags: refactor-all