osmarks
c277b49dc1
fix resumption, oops
2024-11-07 20:43:26 +00:00
osmarks
b9bb629e6f
performance improvements
2024-11-07 16:52:58 +00:00
osmarks
7fa14d45ae
improve observability and fix up Reddit dump for full-scale run
2024-11-02 19:38:05 +00:00
1d0ff95955
Sparse autoencoder testing
2024-10-05 17:22:44 +01:00
fc6d0c9409
Fix crawler (rate limit changes)
2024-07-15 18:48:55 +01:00
43ff9215fb
File metadata storage.
...
The backend now knows how big images are, so the frontend can size images correctly before they're loaded.
This should significantly improve the UI on slow connections.
Also fix bug where videos weren't erased from the index properly.
2024-06-26 20:02:12 +01:00
1ab254ff1d
Adjust index storage for memory efficiency and fix SQLite interface type confusion
2024-06-25 08:23:30 +01:00
e7adf738f6
Fix typo, in the sense of application-killing bug.
2024-05-31 00:35:16 +01:00
747058e254
misc fixes
...
- thumbnails/OCR off was broken
- problematic video files caused segfaults (I blame ffmpeg for this)
2024-05-30 19:05:54 +01:00
3257521068
Video search
2024-05-30 15:58:31 +01:00
74d91d52e5
probably I should do better testing
2024-05-29 21:25:50 +01:00
5eae8674ce
video parsing basics
2024-05-28 22:28:41 +01:00
129b769a56
hackily patch horrifyingly nondeterministic-but-fast image encoder in
2024-05-27 20:21:44 +01:00
d4e136b6a7
AITemplate builds of the image encoder work, at great personal cost
2024-05-27 19:05:25 +01:00
a8329e43fc
more progress on Reddit
2024-05-27 15:22:28 +01:00
f8d68d9d54
WIP Reddit dump loader
2024-05-24 17:47:18 +01:00
978aadda6a
Improved UI for sliders
2024-05-22 20:26:23 +01:00
d8c147df52
Predefined embedding modes in search
2024-05-22 20:17:13 +01:00
14387a61a3
refactor configuration
2024-05-22 19:02:34 +01:00
ffc3d648a6
basic monitoring implementation
2024-05-22 18:49:32 +01:00
ce590298a7
concurrent index queries and fix database typo yet again
2024-05-22 18:25:50 +01:00
349fe802f7
meme interpretability
2024-05-22 16:18:45 +01:00
bd426a30ba
Port meme acquisition pipeline to new API, database
...
Also fix a really stupid oversight in crawling code.
2024-05-22 15:43:56 +01:00
30b1b72712
I really should test database queries better
2024-05-22 14:35:29 +01:00
9455438bab
frontend fixes
2024-05-21 20:17:23 +01:00
24fbc0dd1f
apparently, that quality value is too low
2024-05-21 20:09:28 +01:00
63a9b3d9a6
the consequences of my own actions
2024-05-21 12:39:04 +01:00
e705a9db21
I hate precedence
2024-05-21 12:33:32 +01:00
b7010b41dd
oops
2024-05-21 01:58:50 +01:00
7cb42e028f
Rewrite entire application (well, backend) in Rust and also Go
...
I decided I wanted to integrate the experimental OCR thing better, so I rewrote in Go and also integrated the thumbnailer.
However, Go is a bad langauge and I only used it out of spite.
It turned out to have a very hard-to-fix memory leak due to some unclear interaction between libvips and both sets of bindings I tried, so I had Claude-3 transpile it to Rust then spent a while fixing the several mistakes it made and making tweaks.
The new Rust version works, although I need to actually do something with the OCR data and make the index queryable concurrently.
2024-05-21 00:09:04 +01:00
fa863c2075
"release" unfinished scripts and miscellaneous JSON files
2024-05-18 14:34:30 +01:00
caa8306ff7
oops
2024-05-18 13:21:13 +01:00
6491e02e88
preliminary work on OCR
2024-05-18 00:39:05 +01:00
a3574674d0
"documentation"
2024-04-27 17:33:24 +01:00
2447e134ef
There were more memes than anticipated.
2024-04-27 17:24:01 +01:00
80db16d02a
full pipeline
2024-04-22 18:44:29 +01:00
7bae095384
accidentally lost some manual labels, oops, etc
2024-04-22 13:54:07 +01:00
cebb4f9d00
better evals
2024-04-22 13:43:06 +01:00
58ce70bb5e
meme rater model code (documentation "later")
2024-04-21 23:50:48 +01:00
0b0261f625
preliminary meme rater work
2024-04-20 16:55:11 +01:00
e9a7493343
stop requiring internal aiosqlite patch
2024-01-25 00:01:02 +00:00
e3ffc426b7
Actually delete missing files
2024-01-02 14:12:26 +00:00
4626f53bcb
Return to OpenCLIP
2023-11-13 17:31:43 +00:00
74bb1bc343
thumbnailer system
2023-10-27 15:50:21 +01:00
5b5ef271aa
""documentation""
2023-10-09 12:35:26 +01:00
68a14d7da9
unignore device
2023-10-08 22:54:06 +01:00
20fcc9317f
forgot the README
2023-10-08 22:52:57 +01:00
46fca3eb7f
faster indexing, SigLIP models
2023-10-08 22:52:17 +01:00
2c9ce67ab2
I really should do better testing
2023-09-30 21:10:27 +01:00
c235217751
read from query strings
2023-09-30 15:56:04 +01:00
