Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,114 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
task_categories:
|
4 |
+
- image-to-text
|
5 |
+
- text-to-image
|
6 |
+
language:
|
7 |
+
- en
|
8 |
+
pretty_name: simons ARC (abstraction & reasoning corpus) solve boundingbox version 13
|
9 |
+
size_categories:
|
10 |
+
- 1K<n<10K
|
11 |
+
configs:
|
12 |
+
- config_name: default
|
13 |
+
data_files:
|
14 |
+
- split: train
|
15 |
+
path: data.jsonl
|
16 |
+
---
|
17 |
+
|
18 |
+
# Version 1
|
19 |
+
|
20 |
+
ARC-AGI Tasks where the job is to identify the boundingbox.
|
21 |
+
|
22 |
+
example count: 2-4.
|
23 |
+
|
24 |
+
test count: 1-2.
|
25 |
+
|
26 |
+
image size: 3-8.
|
27 |
+
|
28 |
+
filled bounding box.
|
29 |
+
|
30 |
+
# Version 2
|
31 |
+
|
32 |
+
image size: 3-15.
|
33 |
+
|
34 |
+
filled bounding box.
|
35 |
+
|
36 |
+
# Version 3
|
37 |
+
|
38 |
+
image size: 3-15.
|
39 |
+
|
40 |
+
filled+hollow bounding box.
|
41 |
+
|
42 |
+
# Version 4
|
43 |
+
|
44 |
+
image size: 3-20.
|
45 |
+
|
46 |
+
filled+hollow bounding box.
|
47 |
+
|
48 |
+
# Version 5
|
49 |
+
|
50 |
+
image size: 3-30.
|
51 |
+
|
52 |
+
filled+hollow bounding box.
|
53 |
+
|
54 |
+
# Version 6
|
55 |
+
|
56 |
+
image size: 3-30.
|
57 |
+
|
58 |
+
Added more variants: filled_inner, filled_outer, hollow_inner, hollow_outer.
|
59 |
+
|
60 |
+
# Version 7
|
61 |
+
|
62 |
+
Earlier predictions added to some of the rows.
|
63 |
+
|
64 |
+
# Version 8
|
65 |
+
|
66 |
+
Added fields: `arc_task`, `test_index`, `earlier_output`.
|
67 |
+
|
68 |
+
# Version 9
|
69 |
+
|
70 |
+
Added `generate_task_inner_boundingbox` where it's about find the inner bounding box.
|
71 |
+
|
72 |
+
max image size = 12.
|
73 |
+
|
74 |
+
inner pattern size = 2-3.
|
75 |
+
|
76 |
+
# Version 10
|
77 |
+
|
78 |
+
Bigger images with `generate_task_inner_boundingbox`.
|
79 |
+
|
80 |
+
max image size = 14.
|
81 |
+
|
82 |
+
inner pattern size = 2-4.
|
83 |
+
|
84 |
+
# Version 11
|
85 |
+
|
86 |
+
Bigger images with `generate_task_inner_boundingbox`.
|
87 |
+
|
88 |
+
max image size = 16.
|
89 |
+
|
90 |
+
inner pattern size = 2-5.
|
91 |
+
|
92 |
+
# Version 12
|
93 |
+
|
94 |
+
Bigger images with `generate_task_inner_boundingbox`.
|
95 |
+
|
96 |
+
max image size = 18.
|
97 |
+
|
98 |
+
inner pattern size = 2-5.
|
99 |
+
|
100 |
+
# Version 13
|
101 |
+
|
102 |
+
Extended `generate_task_inner_boundingbox` with `swap_background_outer` transformation.
|
103 |
+
|
104 |
+
max image size = 10.
|
105 |
+
|
106 |
+
inner pattern size = 2-3.
|
107 |
+
|
108 |
+
# Version 14
|
109 |
+
|
110 |
+
Focus on `swap_background_outer` transformation.
|
111 |
+
|
112 |
+
max image size = 12.
|
113 |
+
|
114 |
+
inner pattern size = 2-4.
|