summary refs log tree commit diff
path: root/tasks/johannesm.gmi
blob: 840bd3ec24201a5393d02a599120eeab42414fa9 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
# Tasks for Johannes

## Tags

* kanban: johannesm
* assigned: johannesm
* status: in progress

## Tasks

### Ongoing

* [] Get system into use in GN

* [] Draft outline paper

* [] Talk with Hao about publication on agent system and gain over GNQA

* [] Read literature for paper



### Later

* [~] Catch up on SPARQL

* [~] Catch up on LMDB


### Past

* [X] Make RAG script available for reuse and clean it with Bonface

* [X] Discuss with Bonface on how to get metadata for RAG

* [X] Pickle RAG -> not successful :)

* [X] Try out SPARQLWrapper

* [X] Optimize RAG

* [X] Integrate with RAG

* [X] Get actual metadata with SPARQL endpoint

* [X] Make RAG agentic -> AI system

* [X] Visit precompute issue

* [X] Find more affordable options for GNQA

* [X] Replace GNQA backend with Bonz

* [X] Test GPU

* [X] Test new AI system

* [X] Use GPU to make naturalization faster

* [X] Fix bugs and optimize AI system

* [X] Test and validate AI system working

* [X] Read up on performance evaluation for AI systems

* [X] Refetch all data from SPARQL

* [X] Preproces RDF for improved naturalization

* [X] Make asynchronous requests to server for naturalization

* [X] Document work on issue tracker

* [X] Package code AI system

* [X] Draft API endpoint

* [X] Test package and share with Bonz

* [X] Test AI system with descriptions, qtl and real biology questions

* [X] Compare performance of Claude and open model on GN data before finetuning

* [X] Look into system finetuning

* [X] Test performance gain

* [X] Think about how to show responses of agent system and GNQA in one UI

* [X] Get API working